paperID
stringlengths 36
36
| pwc_id
stringlengths 8
47
| arxiv_id
stringlengths 6
16
⌀ | nips_id
float64 | url_abs
stringlengths 18
329
| url_pdf
stringlengths 18
742
| title
stringlengths 8
325
| abstract
stringlengths 1
7.27k
⌀ | authors
stringlengths 2
7.06k
| published
stringlengths 10
10
⌀ | conference
stringlengths 12
47
⌀ | conference_url_abs
stringlengths 16
198
⌀ | conference_url_pdf
stringlengths 27
199
⌀ | proceeding
stringlengths 6
47
⌀ | taskID
stringlengths 7
1.44k
| areaID
stringclasses 688
values | embedding
stringlengths 9.26k
12.5k
| umap_embedding
stringlengths 29
44
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
40a9d33b-4ecd-40a8-90dc-43c0cac780b5 | cylks-unsupervised-cycle-lucas-kanade-network | 1811.11325 | null | https://arxiv.org/abs/1811.11325v4 | https://arxiv.org/pdf/1811.11325v4.pdf | CyLKs: Unsupervised Cycle Lucas-Kanade Network for Landmark Tracking | Across a majority of modern learning-based tracking systems, expensive annotations are needed to achieve state-of-the-art performance. In contrast, the Lucas-Kanade (LK) algorithm works well without any annotation. However, LK has a strong assumption of photometric (brightness) consistency on image intensity and is easy to drift because of large motion, occlusion, and aperture problem. To relax the assumption and alleviate the drift problem, we propose CyLKs, a data-driven way of training Lucas-Kanade in an unsupervised manner. CyLKs learns a feature transformation through CNNs, transforming the input images to a feature space which is especially favorable to LK tracking. During training, we perform differentiable Lucas-Kanade forward and backward on the convolutional feature maps, and then minimize the re-projection error. During testing, we perform the LK tracking on the learned features. We apply our model to the task of landmark tracking and perform experiments on datasets of THUMOS and 300VW. | ['Wentao Han', 'Xinshuo Weng'] | 2018-11-28 | null | null | null | null | ['landmark-tracking'] | ['computer-vision'] | [-2.96636999e-01 -2.18071327e-01 -3.96919966e-01 -4.88987267e-01
-6.78021252e-01 -5.74497461e-01 5.86357951e-01 -3.67147356e-01
-6.82991803e-01 4.84231710e-01 -1.51887476e-01 -2.04299659e-01
1.79396167e-01 -5.31977773e-01 -9.44038987e-01 -8.48803639e-01
1.57725647e-01 4.08891231e-01 6.45007908e-01 1.27157480e-01
2.00395584e-02 6.65964365e-01 -1.17548740e+00 -2.14766040e-01
6.06685579e-01 8.81728292e-01 -1.31524391e-02 7.02778876e-01
-1.25268891e-01 7.82606363e-01 -2.69058317e-01 -1.60925597e-01
5.19021511e-01 -2.97493488e-01 -7.45507181e-01 -3.81896943e-02
9.31950808e-01 -3.42741609e-01 -4.04196501e-01 1.06752169e+00
2.81422675e-01 8.38664696e-02 4.97277915e-01 -1.46362448e+00
-6.86185300e-01 3.56807448e-02 -6.59455717e-01 1.47050709e-01
-1.19904317e-01 2.48716965e-01 6.16750479e-01 -1.03631222e+00
6.82278574e-01 9.90104675e-01 1.05689502e+00 7.29324281e-01
-1.22543442e+00 -7.73508132e-01 1.88972697e-01 -2.67677963e-01
-1.50227857e+00 -3.82220298e-01 5.91819584e-01 -6.28355563e-01
5.22747099e-01 -5.49767986e-02 8.00572157e-01 6.18185461e-01
2.18243107e-01 9.12493110e-01 7.87999392e-01 -4.03229296e-01
-2.83851363e-02 2.91845705e-02 -2.62459397e-01 1.14626276e+00
7.69591108e-02 1.62183106e-01 -5.24303317e-01 -4.16705161e-02
1.02217484e+00 6.91274703e-02 -1.96852431e-01 -1.04442310e+00
-1.42406344e+00 6.86897457e-01 9.37070012e-01 3.27478424e-02
-5.61389467e-03 8.15170288e-01 1.63095176e-01 7.32996464e-02
3.12198937e-01 1.99691311e-01 -5.66022098e-01 1.32655315e-02
-1.06064367e+00 1.77357331e-01 3.52112651e-01 9.41977084e-01
1.09276855e+00 6.86300769e-02 -7.56592825e-02 3.43271375e-01
7.27862418e-01 7.58360922e-01 4.45474982e-01 -1.03509092e+00
1.24230407e-01 5.47340572e-01 2.79841840e-01 -7.98084080e-01
-6.11449242e-01 -4.77115750e-01 -5.94511807e-01 5.85372627e-01
7.51293182e-01 -9.97341052e-02 -1.16026950e+00 1.69974577e+00
7.29693592e-01 4.82129812e-01 -7.22816586e-02 1.03282261e+00
7.02723861e-01 5.36567509e-01 6.42683879e-02 1.43682674e-01
9.55052495e-01 -1.25879145e+00 -5.47415137e-01 -1.29692912e-01
1.16764212e+00 -6.10007286e-01 1.12371171e+00 4.42971326e-02
-8.22296321e-01 -5.98417819e-01 -9.03593123e-01 -2.66349107e-01
-2.73109943e-01 4.11988109e-01 6.04366779e-01 6.63919508e-01
-1.26667571e+00 3.72921437e-01 -1.21478713e+00 -4.30510759e-01
4.44528759e-01 4.66854930e-01 -4.46514308e-01 1.95366055e-01
-7.22111046e-01 9.00788784e-01 2.83105582e-01 3.58379215e-01
-9.91608977e-01 -8.57993484e-01 -9.29259717e-01 -6.57419711e-02
3.16301376e-01 -6.08679414e-01 1.24963641e+00 -9.84458327e-01
-1.62271357e+00 9.90585506e-01 -1.72804341e-01 -4.00972962e-01
7.78184056e-01 -5.91330707e-01 2.72461846e-02 -3.28650057e-01
5.11641316e-02 1.11467922e+00 8.67296278e-01 -1.10705495e+00
-7.27346420e-01 -1.69833731e-02 -9.10889357e-02 4.82636653e-02
-2.65287429e-01 -1.07165091e-01 -6.95369899e-01 -5.06680131e-01
1.92629963e-01 -1.22898066e+00 -1.96401820e-01 7.44664431e-01
-1.57530695e-01 -2.11104780e-01 1.06978285e+00 -2.62484759e-01
9.79888201e-01 -2.38481331e+00 -4.86147776e-02 2.38065962e-02
1.86945513e-01 2.82223850e-01 -6.04933836e-02 -5.81451692e-02
2.69837737e-01 -2.96288669e-01 -2.18291000e-01 -4.81531143e-01
-2.78170966e-02 1.86928391e-01 -2.30855450e-01 9.80854332e-01
6.16531484e-02 1.14471388e+00 -1.02864134e+00 -6.27872229e-01
3.59982222e-01 4.16122943e-01 -5.92803180e-01 7.44553804e-02
-3.23858410e-01 6.82041943e-01 -3.61318827e-01 6.13080144e-01
6.88116431e-01 -3.27165067e-01 -2.41845086e-01 -6.39356226e-02
-4.06329125e-01 1.64309278e-01 -1.00799000e+00 1.90898502e+00
-3.28618169e-01 1.02550447e+00 -3.21347058e-01 -7.14761257e-01
8.89737606e-01 -6.76779449e-02 6.28364265e-01 -4.16542351e-01
1.89638257e-01 1.78407595e-01 -1.34566635e-01 -8.16362351e-02
4.18390155e-01 1.40116841e-01 1.50104314e-01 2.98374712e-01
2.36144122e-02 1.44606102e-02 -2.38556936e-01 1.16229080e-01
8.27571571e-01 6.62333071e-01 1.37571618e-01 -5.21157146e-01
4.66839522e-01 2.59281665e-01 6.65871441e-01 6.37231469e-01
-2.96237737e-01 5.57123125e-01 4.61210124e-02 -9.49514329e-01
-8.50629032e-01 -1.07349718e+00 -1.02446117e-02 1.02579701e+00
2.76990622e-01 -2.59205967e-01 -7.55067348e-01 -8.95719767e-01
1.31255597e-01 1.39445961e-01 -8.24095547e-01 -1.48440376e-01
-6.83950782e-01 -4.10404652e-01 7.82591760e-01 7.63636470e-01
6.65455341e-01 -9.76997852e-01 -8.45924199e-01 1.23572975e-01
3.88789177e-02 -1.09270954e+00 -7.79628932e-01 1.48324490e-01
-7.22240448e-01 -9.57043469e-01 -6.49774730e-01 -8.55628908e-01
9.87484574e-01 4.59721386e-01 8.09724629e-01 3.35375547e-01
-2.80693144e-01 3.61289203e-01 -9.94110703e-02 -3.98561090e-01
8.14101566e-03 1.51027381e-01 1.67720318e-01 -1.55158397e-02
3.70419204e-01 -2.76367143e-02 -5.35342038e-01 5.13092279e-01
-5.76282978e-01 2.97739096e-02 3.89188379e-01 8.07500899e-01
7.24724293e-01 -1.70067191e-01 4.38566208e-02 -6.00929320e-01
-2.13828892e-01 1.02252491e-01 -1.32160687e+00 1.83082089e-01
-4.85737354e-01 1.30781503e-02 3.46955121e-01 -6.50460482e-01
-9.17514741e-01 6.98917568e-01 1.38251796e-01 -6.82011545e-01
9.08346474e-02 6.84932247e-02 6.38247579e-02 -6.49288774e-01
6.62038505e-01 2.53390193e-01 -2.64952511e-01 -2.66601205e-01
5.47524154e-01 2.73661107e-01 8.14648569e-01 -4.13883597e-01
1.34782898e+00 9.30866480e-01 -1.11901909e-01 -6.95534587e-01
-1.10813951e+00 -5.89255214e-01 -9.47334111e-01 -3.30823183e-01
1.11335146e+00 -9.87282276e-01 -6.94911718e-01 5.09528697e-01
-1.01877177e+00 -8.97329390e-01 -3.42723399e-01 6.47370279e-01
-6.34169936e-01 1.75122380e-01 -3.65803987e-01 -5.88917911e-01
-1.78956300e-01 -1.09741795e+00 1.12258697e+00 4.29186463e-01
-2.40968298e-02 -1.03138411e+00 3.11285526e-01 -2.03589484e-01
3.31997693e-01 7.81766623e-02 3.66780579e-01 -1.27701268e-01
-8.93094003e-01 -1.54596642e-01 -4.04401243e-01 -1.81755759e-02
1.69252858e-01 9.82703418e-02 -9.90274727e-01 -5.25472283e-01
-4.00655776e-01 -2.33000323e-01 6.55957401e-01 6.37309670e-01
1.07213330e+00 8.71362463e-02 -5.47854543e-01 1.18980265e+00
1.30651975e+00 -3.80137265e-02 4.39684063e-01 7.64656305e-01
9.44306552e-01 2.19512388e-01 8.02573681e-01 -6.28959667e-03
5.39721251e-01 8.03068221e-01 4.03784752e-01 -4.52571571e-01
-2.27151781e-01 -4.91868675e-01 4.15616542e-01 5.00148833e-01
1.35796219e-01 5.43440096e-02 -9.51012135e-01 5.98350823e-01
-1.93968213e+00 -5.75409949e-01 -2.61665851e-01 2.15790081e+00
7.00998604e-01 2.33048826e-01 1.38741910e-01 -4.26313579e-01
4.18263704e-01 1.06112356e-03 -5.18910348e-01 3.70871164e-02
1.06416471e-01 -2.23044470e-01 1.04758894e+00 4.96521652e-01
-1.40813708e+00 1.59250462e+00 6.66546869e+00 4.40114796e-01
-1.49460781e+00 9.34981704e-02 2.28723928e-01 1.09398440e-02
2.88328510e-02 1.77203089e-01 -1.02330828e+00 2.78038889e-01
4.47310060e-01 1.93689674e-01 2.86373734e-01 1.11441648e+00
-8.91446844e-02 -3.35984863e-02 -1.05765760e+00 1.00203097e+00
-5.57982251e-02 -1.37861812e+00 -3.99478346e-01 -3.09184082e-02
9.27329421e-01 4.64029312e-01 1.29279092e-01 2.81884581e-01
7.87858129e-01 -8.04729760e-01 8.49481344e-01 3.42751324e-01
7.56604612e-01 -2.87228554e-01 5.72425604e-01 4.12481397e-01
-1.50969660e+00 1.86582908e-01 -6.16575539e-01 2.76029229e-01
8.19748342e-02 2.66901970e-01 -7.30741203e-01 1.00343876e-01
1.07294226e+00 8.50998521e-01 -6.01841033e-01 1.30118656e+00
-3.98733288e-01 4.26984608e-01 -5.04677653e-01 1.99303418e-01
4.73743349e-01 -8.38647485e-02 1.89750969e-01 1.23144269e+00
2.31938407e-01 -1.60160869e-01 3.64863932e-01 6.36923432e-01
-2.17309043e-01 5.78896375e-04 -6.13390625e-01 3.17090839e-01
4.73514259e-01 1.29352665e+00 -9.60848868e-01 -3.62697154e-01
-4.48067635e-01 9.93604124e-01 4.63000327e-01 4.00109828e-01
-9.61985826e-01 -9.07337591e-02 5.21855414e-01 8.31993669e-02
5.01427174e-01 -4.13752437e-01 -1.41787887e-01 -1.15034580e+00
-3.03657595e-02 -3.09586436e-01 2.83677638e-01 -7.21122205e-01
-8.22437882e-01 3.34257036e-01 -1.95950434e-01 -1.55423903e+00
6.82685822e-02 -5.84652841e-01 -5.92724681e-01 5.98439157e-01
-1.83104563e+00 -1.32346141e+00 -6.17622316e-01 5.11617362e-01
5.14091671e-01 1.74883276e-01 4.45496380e-01 4.43570793e-01
-2.71358043e-01 6.72392726e-01 1.48589760e-01 4.80946630e-01
1.02874506e+00 -1.26212156e+00 6.66642129e-01 7.59665728e-01
2.44197324e-01 5.52609026e-01 4.32157099e-01 -5.76254904e-01
-1.37745488e+00 -1.33041286e+00 6.49945080e-01 -7.95737624e-01
7.27502048e-01 -4.40679252e-01 -8.70254874e-01 1.19929886e+00
-9.77301970e-02 7.04185128e-01 3.48196715e-01 -1.17756426e-01
-4.41929549e-01 -8.78509209e-02 -8.62662494e-01 5.34980297e-01
1.14117813e+00 -4.29611593e-01 -2.81197876e-01 2.06737518e-01
6.33847713e-01 -8.34736049e-01 -7.11721182e-01 1.51771888e-01
7.47897387e-01 -5.98699868e-01 1.03546596e+00 -5.24829984e-01
-2.02952161e-01 -9.58051741e-01 5.40198237e-02 -1.27436197e+00
-4.19462204e-01 -6.29064083e-01 7.29231611e-02 1.04406667e+00
1.85217395e-01 -5.27250350e-01 1.06679785e+00 3.26087832e-01
-1.83778759e-02 -4.42345113e-01 -9.81342912e-01 -9.60044563e-01
6.61790594e-02 -1.43912509e-01 4.55303788e-01 1.03842711e+00
-3.86750787e-01 -8.35133567e-02 -6.41372859e-01 3.69342804e-01
7.40722358e-01 -1.92694291e-02 1.50161290e+00 -1.25690913e+00
5.96315861e-02 -3.66465390e-01 -5.23551404e-01 -1.43394923e+00
1.40767455e-01 -7.02710390e-01 5.47664404e-01 -1.24250698e+00
-3.95818278e-02 -8.08389306e-01 -1.65898070e-01 6.19152904e-01
-1.34363070e-01 5.23901284e-01 2.16601968e-01 4.86168981e-01
-9.91863191e-01 5.48242211e-01 1.12011921e+00 -5.95262349e-02
-2.52954274e-01 -1.83726922e-01 -1.51971862e-01 9.49699581e-01
6.56636953e-01 -7.77172387e-01 -9.45245624e-02 -7.87247300e-01
1.19295023e-01 -3.29899907e-01 4.19216692e-01 -9.63854849e-01
4.89564627e-01 -1.84850082e-01 5.64524412e-01 -7.48698652e-01
2.35325381e-01 -9.74213719e-01 9.59951431e-02 6.87507033e-01
-1.51296780e-01 1.35411590e-01 1.26990974e-01 4.83362734e-01
-2.89485808e-02 1.88783389e-02 9.80173945e-01 1.66072160e-01
-8.10918629e-01 6.82445526e-01 -1.11280262e-01 -7.74477571e-02
1.09571242e+00 -2.39848554e-01 -2.95901716e-01 -1.01881020e-01
-4.30941612e-01 4.58246469e-01 8.96421134e-01 4.38916475e-01
4.04811412e-01 -1.66775131e+00 -3.86240453e-01 3.17314327e-01
2.45733663e-01 4.17879373e-01 -9.18561369e-02 1.16281164e+00
-9.15869772e-01 3.16105932e-01 -6.44949228e-02 -1.06139421e+00
-1.09736407e+00 5.53543806e-01 5.78980565e-01 -7.37504810e-02
-1.07876027e+00 8.63006353e-01 5.83101213e-01 -6.40875399e-01
3.83399099e-01 -4.14272308e-01 2.42804294e-03 -4.53062743e-01
5.33466399e-01 2.41042152e-02 -1.69793129e-01 -7.35589087e-01
-5.59661269e-01 1.00726712e+00 -7.08813220e-02 1.07113354e-01
9.99895334e-01 -1.62881091e-01 1.43928915e-01 1.87059522e-01
1.14410436e+00 -1.53005207e-02 -1.74454701e+00 -4.77361977e-01
2.71000087e-01 -6.99454486e-01 3.82633656e-02 -2.45243192e-01
-1.15487719e+00 9.73733604e-01 8.29714715e-01 -1.64734840e-01
7.87596583e-01 6.96861371e-03 7.79686511e-01 4.70780015e-01
2.48430714e-01 -9.23077941e-01 7.73479640e-02 5.06786227e-01
3.62131208e-01 -1.35467744e+00 -9.06194653e-03 -1.88886419e-01
-4.00886267e-01 1.00799763e+00 9.44896519e-01 -3.00535083e-01
6.70341015e-01 3.78278196e-01 6.26013875e-01 -2.87047535e-01
-3.54873836e-01 -2.10426554e-01 3.44278991e-01 6.47009194e-01
2.12405816e-01 -1.83909610e-01 1.59509048e-01 -3.09294611e-01
-2.50505824e-02 1.22362688e-01 2.04357088e-01 1.02280319e+00
-5.97718894e-01 -8.14126134e-01 -5.01676679e-01 -3.42161185e-03
-2.02389807e-01 1.44906238e-01 -1.74214303e-01 1.01197970e+00
3.90662253e-02 2.47588694e-01 1.15686223e-01 -1.93865597e-01
2.28246808e-01 -1.55850559e-01 4.39545125e-01 -3.99929821e-01
-3.33118081e-01 2.01314971e-01 -5.03843725e-01 -7.21473098e-01
-6.80313110e-01 -6.95751667e-01 -1.52583706e+00 -1.35547593e-01
-5.53527594e-01 9.54893306e-02 7.62859046e-01 1.02110815e+00
2.74822232e-03 5.09365797e-01 4.11245257e-01 -9.66709852e-01
-3.05639416e-01 -6.76053286e-01 -2.38014981e-01 2.79131770e-01
6.86587691e-01 -1.01380026e+00 -1.39311612e-01 1.93622887e-01] | [6.5254998207092285, -2.060889959335327] |
6cf50415-3996-42e6-85d7-1a618cabdd6e | semantic-dense-reconstruction-with-consistent | 2109.14821 | null | https://arxiv.org/abs/2109.14821v1 | https://arxiv.org/pdf/2109.14821v1.pdf | Semantic Dense Reconstruction with Consistent Scene Segments | In this paper, a method for dense semantic 3D scene reconstruction from an RGB-D sequence is proposed to solve high-level scene understanding tasks. First, each RGB-D pair is consistently segmented into 2D semantic maps based on a camera tracking backbone that propagates objects' labels with high probabilities from full scans to corresponding ones of partial views. Then a dense 3D mesh model of an unknown environment is incrementally generated from the input RGB-D sequence. Benefiting from 2D consistent semantic segments and the 3D model, a novel semantic projection block (SP-Block) is proposed to extract deep feature volumes from 2D segments of different views. Moreover, the semantic volumes are fused into deep volumes from a point cloud encoder to make the final semantic segmentation. Extensive experimental evaluations on public datasets show that our system achieves accurate 3D dense reconstruction and state-of-the-art semantic prediction performances simultaneously. | ['Federico Tombari', 'Lijin Fang', 'Cheng Guo', 'Yingxuan You', 'Yanyan Li', 'Yingcai Wan'] | 2021-09-30 | null | null | null | null | ['3d-scene-reconstruction'] | ['computer-vision'] | [ 3.54050100e-01 4.10326630e-01 -1.79393008e-01 -8.24950814e-01
-6.49719656e-01 -3.60776335e-01 2.58071184e-01 -2.28710085e-01
1.34299174e-01 1.44033954e-01 -1.65249363e-01 3.54152694e-02
2.20740214e-01 -1.17794919e+00 -1.16844285e+00 -2.42183492e-01
4.25905794e-01 1.10068429e+00 8.47686589e-01 2.42722139e-01
3.07591558e-01 7.07647741e-01 -1.55284846e+00 2.03314781e-01
6.61218584e-01 1.43678892e+00 8.85789275e-01 3.14367622e-01
-8.56620014e-01 5.34070730e-01 2.44453594e-01 -1.03185281e-01
5.20846844e-01 -8.42632949e-02 -8.26909006e-01 6.98327661e-01
2.83764452e-01 -6.10648096e-01 -2.76435375e-01 1.04790056e+00
-1.30793661e-01 -5.27428649e-02 3.95754993e-01 -1.13398182e+00
-1.65118739e-01 1.40608028e-02 -7.13625550e-01 -5.25447369e-01
5.77831268e-01 -6.34787828e-02 5.82406878e-01 -1.04454255e+00
8.98946285e-01 1.42452013e+00 6.09493315e-01 3.27830285e-01
-9.32580352e-01 -5.83201766e-01 4.26600844e-01 -5.12315966e-02
-1.22157395e+00 1.10810466e-01 1.22810388e+00 -3.78744304e-01
9.43558872e-01 -1.02014886e-02 1.21639824e+00 7.37705171e-01
-2.23346651e-01 8.88415158e-01 9.84086394e-01 1.79671030e-02
4.00491267e-01 3.54086957e-03 -1.55500785e-01 8.85113776e-01
-1.32298142e-01 -1.00309685e-01 -6.46943867e-01 1.48729742e-01
1.15496516e+00 4.31306332e-01 -1.01415284e-01 -1.09961033e+00
-1.20037234e+00 4.51778412e-01 6.91826761e-01 -1.53531596e-01
-5.91396928e-01 2.71962732e-01 -8.54902342e-02 -3.54480177e-01
7.18569398e-01 -2.88953274e-01 -6.28190517e-01 1.80872589e-01
-8.35204840e-01 1.58277839e-01 5.32781243e-01 1.40165639e+00
1.28058636e+00 -2.83114642e-01 4.62581754e-01 4.63797599e-01
7.03394949e-01 9.84196723e-01 -2.60445625e-01 -1.43040633e+00
5.36526144e-01 1.05467391e+00 1.56023726e-01 -8.87852192e-01
-3.69139493e-01 -1.72695816e-01 -5.47640979e-01 -3.48074883e-02
-7.12339208e-02 7.92985141e-01 -1.18212032e+00 1.03416288e+00
9.50154662e-01 4.87648427e-01 -2.26739675e-01 1.07764745e+00
6.77347720e-01 7.54219055e-01 -1.19119242e-01 1.37169749e-01
1.09658206e+00 -8.22435617e-01 -1.04441956e-01 -4.76427644e-01
8.39951113e-02 -3.71105373e-01 6.92681432e-01 2.68080205e-01
-1.06120002e+00 -4.84889418e-01 -6.60731494e-01 -5.08276045e-01
-6.56320676e-02 -4.04568970e-01 5.01048446e-01 2.13905916e-01
-8.83579850e-01 2.44304717e-01 -1.03153837e+00 -2.26190612e-01
8.15664649e-01 2.41508819e-02 -3.82016629e-01 -5.86846948e-01
-5.43088853e-01 6.15923405e-01 5.81586659e-01 -1.82217345e-01
-1.22622526e+00 -9.72151041e-01 -1.05985034e+00 -2.81413317e-01
4.03797150e-01 -9.91476476e-01 1.12019336e+00 -5.41870594e-01
-1.19661713e+00 1.41403663e+00 -4.34013605e-01 -1.09448493e-01
5.07494152e-01 -2.64034897e-01 1.98160589e-01 5.28175890e-01
3.52030516e-01 9.05629337e-01 6.28688574e-01 -1.98877025e+00
-8.89143348e-01 -8.98471653e-01 -2.94840038e-01 4.44526166e-01
3.57032120e-01 -5.34782708e-01 -1.00120234e+00 -3.15809958e-02
1.12114465e+00 -7.43608892e-01 -4.98575330e-01 3.68802160e-01
-6.56401455e-01 1.87267631e-01 9.24252570e-01 -8.49500597e-01
3.91371936e-01 -1.83870268e+00 5.77818811e-01 4.59823221e-01
2.12106496e-01 -3.41022879e-01 1.29140124e-01 -2.41621390e-01
3.74275386e-01 -2.57103562e-01 -7.64237702e-01 -7.53591359e-01
-9.32492223e-03 4.94281679e-01 -3.65232378e-01 5.17599165e-01
-1.77135587e-01 9.91201580e-01 -1.00937247e+00 -6.61073029e-01
8.62421811e-01 5.67074120e-01 -5.20276129e-01 4.78954136e-01
-7.58244395e-01 7.47412801e-01 -9.15430486e-01 1.00645268e+00
1.11322641e+00 -5.71062505e-01 -1.53333709e-01 -3.33261371e-01
-1.65788591e-01 1.30958065e-01 -8.54811609e-01 2.68160200e+00
-3.23597074e-01 3.40423137e-02 3.32893757e-03 -9.84230697e-01
1.02673125e+00 -1.75103977e-01 8.44056487e-01 -8.20139349e-01
1.05549552e-01 1.90117106e-01 -1.12670338e+00 -2.18800172e-01
5.06340623e-01 -1.65545270e-01 -2.08428115e-01 2.47776821e-01
5.65463817e-03 -1.19677472e+00 -6.31840646e-01 2.78529555e-01
7.52884567e-01 6.17875278e-01 -9.39438790e-02 2.05487162e-01
4.53534633e-01 5.00760794e-01 6.31569386e-01 2.17590705e-01
1.56671211e-01 9.03184474e-01 6.64261505e-02 -5.59576392e-01
-1.51978970e+00 -1.63162923e+00 -1.08212955e-01 1.07155532e-01
9.41747904e-01 -5.51788099e-02 -7.35795319e-01 -7.71417022e-01
2.47834742e-01 8.41011882e-01 -4.17345464e-01 4.80636209e-02
-4.25328225e-01 -1.49891973e-01 -3.09049040e-01 4.40638423e-01
6.78472757e-01 -7.87416399e-01 -8.05067003e-01 1.42064422e-01
-4.09874529e-01 -1.42891526e+00 -8.96016806e-02 6.24658391e-02
-1.26990342e+00 -1.21348226e+00 -4.78957683e-01 -6.98644042e-01
8.74847889e-01 7.53901184e-01 1.12243593e+00 -2.42002279e-01
-4.78148423e-02 7.01872349e-01 -3.72171581e-01 -7.60940015e-02
-4.44148630e-01 -2.70922571e-01 -3.23749334e-01 -7.11096600e-02
3.51017088e-01 -6.68340683e-01 -7.12370813e-01 4.11627263e-01
-6.86441481e-01 7.72389889e-01 4.45921361e-01 1.82957023e-01
1.48871922e+00 -6.76794574e-02 -1.88116238e-01 -7.02820718e-01
-6.24451637e-01 -5.70474803e-01 -9.08013880e-01 8.72456804e-02
-3.96192551e-01 -1.51134923e-01 5.58004268e-02 1.52406707e-01
-1.26272714e+00 6.50744736e-01 -1.74857020e-01 -1.12733722e+00
-4.46095496e-01 -1.26762331e-01 -4.00919467e-01 1.42461330e-01
-9.25329700e-02 6.10998988e-01 2.09452156e-02 -6.78386331e-01
6.80718362e-01 4.57155406e-01 5.45904756e-01 -3.69682640e-01
8.85821819e-01 9.91485417e-01 3.29912677e-02 -5.74951589e-01
-1.25622106e+00 -5.82136333e-01 -1.04480052e+00 -5.07636487e-01
1.32099915e+00 -1.28273547e+00 -3.72146606e-01 5.94439268e-01
-1.35000682e+00 -4.89834994e-01 -4.09516364e-01 4.25969988e-01
-9.49159145e-01 2.19063476e-01 -4.00954962e-01 -5.78686833e-01
-6.29398599e-02 -1.18600047e+00 1.88635218e+00 1.69402152e-01
1.63749218e-01 -6.56918883e-01 -2.66947169e-02 7.81501949e-01
-2.18394071e-01 3.99481952e-01 7.56774247e-01 1.81487873e-01
-1.49878788e+00 1.15695432e-01 -4.79015172e-01 4.17250842e-01
-1.01914063e-01 -4.48792666e-01 -9.96223092e-01 4.61976439e-01
2.95200318e-01 -1.21851943e-01 5.43064594e-01 5.32855213e-01
1.43213117e+00 2.44812593e-01 -7.10829258e-01 1.07393229e+00
1.64124346e+00 6.71586916e-02 5.63014328e-01 2.15243623e-01
1.32443321e+00 6.18405581e-01 7.49478161e-01 4.30758506e-01
1.02146733e+00 4.60863441e-01 1.07841074e+00 3.73460613e-02
-1.93819612e-01 -8.61951768e-01 1.73788462e-02 8.82222235e-01
2.49749094e-01 4.49481681e-02 -9.72448170e-01 3.90652686e-01
-1.54732811e+00 -5.91727197e-01 -4.95738000e-01 2.04280162e+00
4.64137942e-01 6.93581477e-02 -2.00041145e-01 -8.71678814e-02
6.46473527e-01 2.52722025e-01 -1.01626503e+00 4.05086607e-01
-1.37132555e-01 -6.66883448e-03 5.76281071e-01 5.70562840e-01
-7.36562669e-01 1.16496873e+00 4.95357466e+00 6.15263343e-01
-6.62576914e-01 4.14499193e-01 6.50864959e-01 -1.02856047e-01
-7.78045893e-01 1.60116240e-01 -7.59081960e-01 5.50390542e-01
3.03955466e-01 2.44089335e-01 1.88405573e-01 1.06266165e+00
3.14899459e-02 -5.15025735e-01 -9.77434576e-01 1.36611736e+00
8.66624042e-02 -1.45450330e+00 1.44150525e-01 2.12911397e-01
9.74651694e-01 5.13108134e-01 -3.47967267e-01 -2.44593516e-01
4.67504174e-01 -5.47036946e-01 1.33259833e+00 8.56802881e-01
8.65228713e-01 -6.84844911e-01 1.90924555e-01 5.95368505e-01
-1.38762665e+00 6.91206753e-02 -4.52916861e-01 3.24873000e-01
8.06572437e-01 9.77088273e-01 -6.03041291e-01 7.35548615e-01
8.98657918e-01 1.16632104e+00 -9.90009755e-02 7.82059193e-01
-2.86252201e-01 6.94695264e-02 -5.16463459e-01 4.83049214e-01
1.50958374e-01 -4.51112181e-01 4.33194697e-01 5.03653884e-01
6.79098666e-01 5.40773451e-01 1.71773076e-01 1.24544275e+00
-9.72278789e-03 -2.97751218e-01 -7.08266854e-01 5.30652881e-01
6.56233728e-01 1.05255508e+00 -1.09596884e+00 -4.36067969e-01
-4.35952216e-01 1.32908726e+00 2.35729173e-01 1.39655575e-01
-7.61886418e-01 3.26877594e-01 5.50794065e-01 3.31190646e-01
3.97690326e-01 -3.58964950e-01 -6.38738513e-01 -1.14951479e+00
7.79928416e-02 8.41775909e-02 -1.41667221e-02 -1.41608524e+00
-1.06264901e+00 2.90128440e-01 -2.25217808e-02 -1.39702809e+00
2.29296565e-01 -2.69876063e-01 -5.54057322e-02 7.97132015e-01
-1.60128558e+00 -1.22783315e+00 -7.78824329e-01 6.52076781e-01
6.29468322e-01 4.50320959e-01 4.26818818e-01 -4.11492307e-03
-5.09452559e-02 -4.18510109e-01 -2.20984891e-01 -1.87178046e-01
-1.44300774e-01 -9.69041049e-01 5.40733159e-01 6.77136362e-01
6.85024485e-02 -3.08607757e-01 1.40170187e-01 -1.05737293e+00
-1.53724241e+00 -1.46434236e+00 5.38750708e-01 -7.90023446e-01
8.57711136e-02 -5.68368495e-01 -8.48667502e-01 6.63058639e-01
-5.59798300e-01 1.60436854e-01 8.58811736e-02 -6.28904700e-01
-2.80670792e-01 3.15498076e-02 -1.32700634e+00 -4.07563709e-03
1.73409343e+00 -6.71329737e-01 -6.18895888e-01 3.39644849e-01
1.25784993e+00 -9.88539815e-01 -9.26450372e-01 4.79740053e-01
3.09096426e-01 -1.18433487e+00 1.49864042e+00 -6.97135925e-04
7.07579553e-01 -4.39825922e-01 -7.06820548e-01 -1.06832242e+00
1.35171130e-01 1.18129022e-01 -3.34941864e-01 7.81892776e-01
-2.09920630e-01 -1.56872123e-01 1.23496795e+00 6.63400590e-01
-6.14458740e-01 -4.83785808e-01 -8.91507089e-01 -4.94447708e-01
-3.60312760e-01 -1.03258598e+00 8.83059621e-01 8.11264694e-01
-7.35206127e-01 5.32879531e-02 3.62569317e-02 5.54775238e-01
1.31968129e+00 6.85237110e-01 9.95112062e-01 -1.40413463e+00
2.06029803e-01 -1.11914508e-01 -5.12785077e-01 -1.73832703e+00
3.11429739e-01 -1.16885912e+00 2.06006676e-01 -1.92757845e+00
2.56957918e-01 -8.57373834e-01 9.57198739e-02 1.34127989e-01
1.87653214e-01 2.33988807e-01 9.06370133e-02 1.90370142e-01
-7.95201480e-01 9.31792319e-01 1.42318058e+00 -4.09263931e-02
-4.12563570e-02 -2.12378293e-01 -2.09588885e-01 8.98165226e-01
3.72420341e-01 -6.05767727e-01 -5.07384360e-01 -7.57703722e-01
9.89509374e-02 4.96739894e-01 6.91411495e-01 -9.16398764e-01
2.66210921e-02 -2.67546445e-01 6.02311909e-01 -1.46457112e+00
7.39449859e-01 -1.26200140e+00 5.05021989e-01 2.43415549e-01
2.17153966e-01 -4.18222517e-01 -1.45981297e-01 9.32137668e-01
3.15771252e-02 2.49332592e-01 6.89495683e-01 -5.56558013e-01
-1.10462201e+00 9.87439036e-01 4.43623096e-01 -1.14119776e-01
1.24357426e+00 -6.19450212e-01 3.38861495e-01 9.07133073e-02
-6.84430897e-01 3.83092970e-01 1.17918169e+00 5.04329681e-01
1.21566045e+00 -1.29601896e+00 -3.38264465e-01 4.83899206e-01
2.27517992e-01 1.29032588e+00 8.13717902e-01 4.70091522e-01
-7.72165537e-01 4.42315072e-01 -8.21046233e-02 -1.35621083e+00
-8.48470926e-01 4.20098275e-01 3.48255128e-01 2.49014795e-01
-1.24402404e+00 1.02917814e+00 6.22313023e-01 -8.62819016e-01
3.89005952e-02 -5.16897798e-01 5.03863990e-01 -3.50670367e-01
2.64566243e-01 2.10127294e-01 3.49778086e-02 -9.98687327e-01
-4.13164675e-01 1.09634423e+00 4.43447202e-01 -1.68281570e-01
1.64119804e+00 -6.09961987e-01 -1.38544321e-01 5.24029970e-01
1.34631467e+00 -5.05919278e-01 -1.83319020e+00 -4.72671181e-01
-3.45074356e-01 -1.01214206e+00 2.61957884e-01 -4.66739655e-01
-1.39564419e+00 9.35200393e-01 4.17335063e-01 -3.12769055e-01
1.02333140e+00 7.43460357e-01 1.07724273e+00 -9.92627963e-02
1.31686020e+00 -7.59841919e-01 5.35022244e-02 3.50556374e-01
6.74854517e-01 -1.22007060e+00 8.81788041e-03 -7.35729694e-01
-5.43128490e-01 8.24465811e-01 6.25233829e-01 -8.26648548e-02
7.13361681e-01 9.57608595e-03 -3.88055712e-01 -6.59112334e-01
-2.45785996e-01 6.65434543e-03 2.55612992e-02 8.55428934e-01
-5.64908326e-01 1.74496189e-01 6.37409151e-01 3.12661171e-01
-1.95530891e-01 3.55084427e-02 2.07386866e-01 7.46872425e-01
-6.61277294e-01 -7.52604425e-01 -4.86310214e-01 4.21317339e-01
4.66414243e-01 3.52302134e-01 -1.13643453e-01 5.73244274e-01
2.46483788e-01 4.85866785e-01 5.56495249e-01 -4.71850604e-01
3.77521485e-01 -1.87312122e-02 8.06629002e-01 -8.28494370e-01
2.21068799e-01 -4.46947590e-02 -3.79967898e-01 -1.29955471e+00
-5.75700343e-01 -7.60501206e-01 -1.82897019e+00 -1.26470074e-01
1.56432196e-01 -2.74167329e-01 1.08507991e+00 8.51269662e-01
3.69245559e-01 2.90487409e-01 7.30253458e-01 -1.33884251e+00
1.58880323e-01 -4.40115720e-01 -7.73897111e-01 3.49419981e-01
5.54248057e-02 -8.81605327e-01 -2.52882212e-01 2.80066073e-01] | [8.505988121032715, -2.928380012512207] |
ed14bcbb-3458-4c73-a0f8-b556a5361a9a | spectral-unmixing-of-hyperspectral-images | 2204.04638 | null | https://arxiv.org/abs/2204.04638v2 | https://arxiv.org/pdf/2204.04638v2.pdf | Spectral Unmixing of Hyperspectral Images Based on Block Sparse Structure | Spectral unmixing (SU) of hyperspectral images (HSIs) is one of the important areas in remote sensing (RS) that needs to be carefully addressed in different RS applications. Despite the high spectral resolution of the hyperspectral data, the relatively low spatial resolution of the sensors may lead to mixture of different pure materials within the image pixels. In this case, the spectrum of a given pixel recorded by the sensor can be a combination of multiple spectra each belonging to a unique material in that pixel. Spectral unmixing is then used as a technique to extract the spectral characteristics of the different materials within the mixed pixels and to recover the spectrum of each pure spectral signature, called endmember. Block-sparsity exists in hyperspectral images as a result of spectral similarity between neighboring pixels. In block-sparse signals, the nonzero samples occur in clusters and the pattern of the clusters is often supposed to be unavailable as prior information. This paper presents an innovative spectral unmixing approach for HSIs based on block-sparse structure. Hyperspectral unmixing problem is solved using pattern coupled sparse Bayesian learning strategy (PCSBL). To evaluate the performance of the proposed SU algorithm, it is tested on both synthetic and real hyperspectral data and the quantitative results are compared to those of other state-of-the-art methods in terms of abundance angle distance and mean squared error. The achieved results show the superiority of the proposed algorithm over the other competing methods by a significant margin. | ['Amin Zehtabian', 'Hadi Zayyani', 'Roozbeh Rajabi', 'Seyed Hossein Mosavi Azarang'] | 2022-04-10 | null | null | null | null | ['hyperspectral-unmixing'] | ['computer-vision'] | [ 9.48190808e-01 -6.72912180e-01 6.50981292e-02 1.59364194e-01
-5.32748103e-01 -5.16069353e-01 3.44562620e-01 -1.39733762e-01
9.08015370e-02 8.99961352e-01 -2.93762004e-03 1.30584452e-03
-5.77206075e-01 -7.43543148e-01 -4.00266975e-01 -1.46885061e+00
2.13241100e-01 2.56469786e-01 -1.43378034e-01 -7.10865930e-02
5.53954206e-02 5.62539399e-01 -1.85946226e+00 2.00410128e-01
1.13678741e+00 9.89358783e-01 6.45638525e-01 2.51149386e-01
-9.00226235e-02 3.07504296e-01 -3.20882201e-01 5.69735289e-01
5.81862628e-01 -5.59137583e-01 -1.59443825e-01 6.19658887e-01
5.16359031e-01 6.99253827e-02 -2.38429513e-02 1.69934630e+00
1.86399236e-01 2.61278689e-01 7.75798440e-01 -8.37012708e-01
-1.53488070e-01 4.03593779e-01 -1.12815094e+00 -3.20859924e-02
-1.73748150e-01 -2.70343810e-01 7.26954699e-01 -7.84267008e-01
1.76055089e-01 8.74134421e-01 3.44434619e-01 -2.16837630e-01
-1.30932641e+00 -6.57611370e-01 -1.41832307e-01 2.61902243e-01
-1.73614335e+00 -3.95665050e-01 9.20407891e-01 -5.88551819e-01
2.33114004e-01 5.92799127e-01 6.10186219e-01 2.37794101e-01
-1.40133336e-01 2.65295446e-01 1.64978802e+00 -4.24271643e-01
1.73918083e-01 7.87852407e-02 2.96496540e-01 2.88738370e-01
6.51228428e-01 2.38076970e-01 -1.98636279e-01 -1.63838416e-01
3.79109591e-01 3.60056728e-01 -7.62968659e-01 -4.76763844e-01
-1.03640389e+00 5.94914556e-01 5.50813079e-01 4.80247498e-01
-7.00786650e-01 -6.07569456e-01 -1.66869119e-01 -7.26135373e-02
2.33889714e-01 -1.77607447e-01 -7.20099406e-03 6.49768710e-01
-1.31522012e+00 -4.28501442e-02 3.47001612e-01 4.63802785e-01
1.11825335e+00 3.42233896e-01 1.91907957e-01 9.64383125e-01
6.50710762e-01 1.12883747e+00 4.65576172e-01 -5.73734999e-01
2.43121877e-01 5.36257744e-01 3.95381063e-01 -1.20637298e+00
-7.84502998e-02 -6.22901380e-01 -1.20437407e+00 3.00954103e-01
1.46882236e-01 7.82688335e-02 -8.92062783e-01 1.28432119e+00
3.18341494e-01 4.46126044e-01 3.39161932e-01 1.04108226e+00
6.59901679e-01 1.33729625e+00 -5.81554621e-02 -8.23551238e-01
9.91933703e-01 -6.82365179e-01 -7.33810186e-01 -4.24662918e-01
-1.25857726e-01 -9.55212057e-01 3.42583895e-01 4.91522878e-01
-4.86007780e-01 -5.15104890e-01 -1.24681222e+00 8.53249371e-01
-3.81594151e-02 3.51132959e-01 3.56128871e-01 5.87005794e-01
-4.31856543e-01 3.40797871e-01 -6.22980297e-01 -2.88958371e-01
1.23155273e-01 7.36422837e-02 -4.12284762e-01 -5.16307116e-01
-7.41364360e-01 5.17008007e-01 7.04000115e-01 4.57805455e-01
-6.06958210e-01 -4.54829544e-01 -5.57411551e-01 2.57625822e-02
2.97843754e-01 -1.38387993e-01 4.18396801e-01 -1.45849550e+00
-1.03132939e+00 4.55569774e-01 -3.70662302e-01 5.08434512e-02
-1.11643951e-02 2.28720546e-01 -8.47395957e-01 2.58235186e-01
1.12204596e-01 -6.27183318e-02 9.24032092e-01 -1.58364236e+00
-6.32135510e-01 -6.03833377e-01 -7.05839038e-01 4.59752649e-01
-1.51255026e-01 -2.45932892e-01 2.18929619e-01 -4.97436434e-01
8.11893523e-01 -9.70158696e-01 -5.08834645e-02 -4.24042135e-01
-2.74898171e-01 3.98605645e-01 1.07452643e+00 -7.92476475e-01
1.00240409e+00 -2.53825808e+00 2.22144619e-01 5.90099871e-01
-2.37176120e-01 4.14575011e-01 5.95428748e-03 4.54059213e-01
-5.23544371e-01 -3.77526671e-01 -8.03517878e-01 4.05075848e-01
-4.11121130e-01 7.70134181e-02 -8.72542262e-02 9.80737031e-01
-3.15138012e-01 1.99626144e-02 -8.14695954e-01 -2.39393607e-01
4.04389232e-01 3.90856117e-01 3.51314843e-01 1.91501886e-01
-6.86706677e-02 5.57103217e-01 -2.15948939e-01 8.24334681e-01
1.22899759e+00 -1.32780552e-01 3.10956955e-01 -6.48646533e-01
-4.32438791e-01 -4.02198166e-01 -1.92630970e+00 1.33638346e+00
8.28680471e-02 1.67104036e-01 6.26358151e-01 -1.26847851e+00
8.59906316e-01 4.85017300e-01 6.34214759e-01 -1.38952926e-01
-1.26662701e-01 5.51332772e-01 1.96536139e-01 -4.57955599e-01
3.96263421e-01 -4.90941256e-01 6.43583417e-01 2.25370631e-01
-2.86684752e-01 -1.16596811e-01 2.65063047e-01 -2.93753833e-01
2.98042566e-01 -1.00510970e-01 5.49144030e-01 -5.35304070e-01
8.45254660e-01 1.87926635e-01 5.00196338e-01 3.37715775e-01
6.56111315e-02 2.04106316e-01 -4.31543529e-01 4.12720665e-02
-1.01266074e+00 -9.34040844e-01 -5.20905793e-01 4.51848090e-01
3.27828079e-01 3.74483466e-01 -2.18929037e-01 8.38817209e-02
-1.52568758e-01 4.73777413e-01 -7.78575912e-02 2.52710581e-01
1.06298169e-02 -1.35106552e+00 -7.21684620e-02 -1.82422504e-01
6.59453034e-01 -6.89050376e-01 -2.17961922e-01 1.86947376e-01
-2.98127711e-01 -8.02069962e-01 1.18280835e-01 8.87341127e-02
-9.97174978e-01 -1.15637386e+00 -6.13291025e-01 -6.80404603e-01
7.58898973e-01 1.06544399e+00 4.17928874e-01 -3.39801311e-01
-1.10290870e-01 4.54139290e-03 -5.21071553e-01 -2.74282664e-01
-5.17436624e-01 -3.62660348e-01 -7.23484680e-02 7.93703020e-01
1.83213025e-01 -6.45844340e-01 -4.89179015e-01 3.27964276e-01
-1.31931853e+00 -4.29005036e-03 8.18570912e-01 8.18719268e-01
7.49133945e-01 9.99140382e-01 1.52651817e-01 -6.19511425e-01
5.15478961e-02 -7.67057717e-01 -9.21151996e-01 3.60415310e-01
-4.10034180e-01 -3.59669775e-01 4.81878966e-01 -2.18408719e-01
-1.22921097e+00 2.85526484e-01 5.12151062e-01 -2.72965848e-01
-4.57335413e-01 1.05560422e+00 -4.54773813e-01 -2.11289227e-01
6.33641899e-01 7.06310272e-01 1.36035517e-01 -5.93397439e-01
-3.39355133e-02 9.74418819e-01 5.25067925e-01 -3.34551364e-01
9.71699476e-01 6.70249760e-01 3.89504403e-01 -1.52526307e+00
-6.88633263e-01 -9.04562950e-01 -3.32852423e-01 -1.95916623e-01
6.65858388e-01 -1.11644530e+00 -2.66296901e-02 6.44635439e-01
-5.78532100e-01 1.66241005e-01 2.22952008e-01 7.36061633e-01
2.35037412e-02 8.24858725e-01 -1.25739262e-01 -1.07290447e+00
-2.24650651e-01 -1.18963647e+00 4.83880997e-01 2.54991829e-01
3.60478014e-01 -6.76416099e-01 -4.73909751e-02 5.49864590e-01
1.95904344e-01 3.45767289e-01 9.61822212e-01 -3.95299345e-01
-5.21852970e-01 -2.46358737e-01 -3.47450793e-01 5.99708796e-01
5.95111609e-01 1.21079475e-01 -8.45564187e-01 -4.87220556e-01
2.74946660e-01 6.74152598e-02 7.98646033e-01 7.13038206e-01
6.19787097e-01 -2.85565674e-01 -3.22289884e-01 4.61806327e-01
2.05821037e+00 4.32517648e-01 5.66191196e-01 3.03150445e-01
8.09022784e-01 7.72089303e-01 6.77515507e-01 4.67823297e-01
-4.44037259e-01 2.44265765e-01 6.14283979e-01 -1.57907888e-01
2.66353160e-01 1.93218991e-01 2.73915708e-01 7.35373735e-01
-1.07542306e-01 -1.93153843e-01 -7.02570558e-01 4.90748465e-01
-1.78250837e+00 -1.40437233e+00 -7.26834893e-01 2.44617581e+00
4.73996460e-01 -5.93199074e-01 -1.91622183e-01 6.92376971e-01
1.25992227e+00 4.54343557e-01 -4.04414535e-01 3.75000089e-01
-5.90803802e-01 1.31800920e-01 7.92590678e-01 5.32794714e-01
-1.15268433e+00 4.95438129e-01 5.02632570e+00 9.23004031e-01
-1.26592863e+00 -5.46363853e-02 2.94362456e-01 2.78304964e-01
-1.79446995e-01 1.81177273e-01 -3.70780289e-01 4.74862725e-01
6.00086510e-01 8.47937912e-02 7.98443913e-01 3.10656667e-01
4.34450179e-01 -5.75088501e-01 -4.51221555e-01 1.15153635e+00
1.09956667e-01 -9.08881247e-01 7.31586218e-02 7.05974400e-02
1.24006093e+00 6.14449866e-02 1.03379143e-02 -5.42661548e-01
-7.52930641e-02 -9.29063320e-01 5.99467099e-01 6.02054119e-01
6.75364017e-01 -6.58551097e-01 7.48382449e-01 5.86842418e-01
-1.17747962e+00 -2.49012023e-01 -5.66024244e-01 4.15554307e-02
1.24677625e-02 1.05037034e+00 -4.83471185e-01 9.58642542e-01
4.89927471e-01 8.97405744e-01 -8.35471153e-02 1.24679387e+00
-1.16199106e-01 7.53997266e-01 -5.03286600e-01 4.00412500e-01
2.96743572e-01 -1.11045611e+00 8.52844238e-01 6.81679368e-01
9.17964160e-01 4.20151025e-01 2.19950303e-01 9.00979280e-01
4.23203111e-01 4.10386354e-01 -5.53599954e-01 -4.37952548e-01
5.82041383e-01 1.33593929e+00 -5.62552631e-01 -3.44846934e-01
-4.25174743e-01 5.69794953e-01 -5.75722456e-01 5.03192246e-01
-3.17562133e-01 1.75765097e-01 4.31130528e-01 1.70817569e-01
4.38531309e-01 -1.83695689e-01 -7.20474869e-02 -1.09467316e+00
-9.29570198e-02 -1.10653234e+00 3.79434079e-01 -9.23987269e-01
-1.28418458e+00 3.93245041e-01 3.93900461e-02 -1.59124017e+00
2.91800171e-01 -5.24246275e-01 -3.00320297e-01 1.33450854e+00
-1.67325592e+00 -1.13095272e+00 -7.23123491e-01 7.27806866e-01
1.45794019e-01 -3.32058966e-01 8.51797402e-01 1.60983175e-01
-5.71764112e-01 -4.42954510e-01 8.89794469e-01 -2.75248319e-01
4.91487145e-01 -8.02152693e-01 -9.09759104e-01 1.26869893e+00
5.15376925e-02 3.70906919e-01 8.64044011e-01 -7.68046439e-01
-1.18987167e+00 -1.01785100e+00 3.02611232e-01 5.71503401e-01
6.68196738e-01 4.81568187e-01 -8.54751766e-01 3.30960363e-01
2.15360522e-01 -4.87039350e-02 1.13027549e+00 -4.15604383e-01
-1.23207219e-01 -4.65439558e-01 -1.07257473e+00 1.73082292e-01
2.04859152e-01 -1.77999794e-01 -4.15750027e-01 3.94086659e-01
-7.09379166e-02 1.27299353e-01 -8.17422390e-01 6.00237608e-01
4.18646991e-01 -1.11847377e+00 8.78652573e-01 4.13188301e-02
1.59925118e-01 -1.05189502e+00 -6.56971872e-01 -1.39297748e+00
-5.47053874e-01 -5.01694456e-02 3.79806012e-01 1.10008860e+00
2.04161555e-01 -5.69192648e-01 5.06051719e-01 3.95400785e-02
-8.20119008e-02 1.22973233e-01 -7.06103027e-01 -6.52736425e-01
-4.29543912e-01 4.16529411e-03 5.70943236e-01 1.21231222e+00
-2.07606539e-01 2.39588127e-01 -4.61042255e-01 8.61389995e-01
1.09897614e+00 6.05710864e-01 3.08672607e-01 -1.62007034e+00
-3.39641958e-01 -3.84239405e-01 -1.35468170e-01 -5.00459194e-01
1.55907601e-01 -8.66229594e-01 8.55824426e-02 -1.56506097e+00
3.30782741e-01 -5.32582939e-01 -3.82204801e-01 2.22037420e-01
-2.33116657e-01 2.66832530e-01 5.43667143e-03 7.50302792e-01
2.70078778e-01 3.82397234e-01 9.97210741e-01 -5.52858889e-01
-2.92790562e-01 1.42363548e-01 -5.09408295e-01 4.52031076e-01
6.69646680e-01 -3.22451234e-01 -3.74265581e-01 -1.84164315e-01
1.11477554e-01 2.71806538e-01 2.61968464e-01 -1.23971021e+00
7.85259753e-02 -4.68194425e-01 2.29927287e-01 -7.96295166e-01
2.77314812e-01 -1.46057963e+00 9.23441947e-01 3.34213644e-01
2.62718618e-01 -5.79701781e-01 5.29601462e-02 5.92686176e-01
-4.60075527e-01 -5.60586989e-01 1.08611298e+00 -2.95091808e-01
-8.16628158e-01 1.80838853e-01 -4.49501693e-01 -7.33861208e-01
9.85273957e-01 -5.26759326e-01 -1.96876913e-01 -1.12970695e-01
-5.48062444e-01 -3.06558073e-01 5.02237320e-01 -2.61298627e-01
4.30796683e-01 -1.16076028e+00 -9.01180208e-01 2.90680557e-01
3.30466211e-01 -2.74705496e-02 7.20595777e-01 9.57696497e-01
-7.13232696e-01 3.37395072e-01 -3.91662300e-01 -6.44250333e-01
-1.46914709e+00 4.91444528e-01 4.29524660e-01 1.09814972e-01
-1.64162740e-02 4.07020301e-01 8.18528533e-02 -1.20738417e-01
-1.91436112e-01 1.40542030e-01 -5.48910677e-01 1.72852904e-01
4.81464565e-01 6.53178871e-01 7.16038868e-02 -1.18727612e+00
-1.04005612e-01 6.66139543e-01 5.85083425e-01 -5.93264773e-02
1.42567146e+00 -5.46417758e-02 -9.72965121e-01 5.73883533e-01
9.45273578e-01 2.59849638e-01 -9.92507815e-01 -3.45215917e-01
-3.75067413e-01 -6.05498433e-01 5.06535172e-01 -8.29148889e-01
-9.36064839e-01 6.79604471e-01 8.94648194e-01 2.04062387e-01
1.28941524e+00 -4.90091115e-01 2.33012795e-01 1.49165064e-01
1.98220089e-01 -9.23336148e-01 -5.96180737e-01 -7.15478463e-03
5.63370287e-01 -1.26757956e+00 3.28206360e-01 -6.54164970e-01
-3.36196989e-01 1.09732699e+00 1.37881264e-01 1.34432956e-03
7.23971605e-01 -2.13607445e-01 1.02985285e-01 -1.01346582e-01
1.47601590e-01 -2.78315574e-01 4.05688494e-01 4.48562890e-01
3.41082484e-01 3.63795698e-01 -4.30528075e-01 1.15290219e-02
2.21526891e-01 -2.80237347e-01 3.87145907e-01 7.26763904e-01
-7.87488461e-01 -9.64187086e-01 -1.28116846e+00 4.49128896e-01
-1.67027205e-01 -8.27745497e-02 1.75349005e-02 2.89872319e-01
2.95860350e-01 1.25415671e+00 -2.45680690e-01 -9.24765393e-02
-8.24429393e-02 1.55257225e-01 3.45835239e-01 -5.34627855e-01
1.77291766e-01 7.43457854e-01 -2.58886397e-01 4.05491106e-02
-1.17682827e+00 -9.07730639e-01 -8.99885952e-01 -5.80738820e-02
-4.57740277e-01 4.49759454e-01 7.24764228e-01 8.41815472e-01
-2.04546928e-01 1.86194777e-01 8.35329294e-01 -8.60088348e-01
-5.35404205e-01 -1.16529942e+00 -1.54068708e+00 2.68289775e-01
3.36839497e-01 -6.08902454e-01 -6.45860791e-01 1.98665291e-01] | [10.075997352600098, -2.062713384628296] |
d8c59517-297d-4efe-89f0-6f5247a72b37 | learning-and-exploiting-multiple-subgoals-for | 1905.05180 | null | https://arxiv.org/abs/1905.05180v1 | https://arxiv.org/pdf/1905.05180v1.pdf | Learning and Exploiting Multiple Subgoals for Fast Exploration in Hierarchical Reinforcement Learning | Hierarchical Reinforcement Learning (HRL) exploits temporally extended actions, or options, to make decisions from a higher-dimensional perspective to alleviate the sparse reward problem, one of the most challenging problems in reinforcement learning. The majority of existing HRL algorithms require either significant manual design with respect to the specific environment or enormous exploration to automatically learn options from data. To achieve fast exploration without using manual design, we devise a multi-goal HRL algorithm, consisting of a high-level policy Manager and a low-level policy Worker. The Manager provides the Worker multiple subgoals at each time step. Each subgoal corresponds to an option to control the environment. Although the agent may show some confusion at the beginning of training since it is guided by three diverse subgoals, the agent's behavior policy will quickly learn how to respond to multiple subgoals from the high-level controller on different occasions. By exploiting multiple subgoals, the exploration efficiency is significantly improved. We conduct experiments in Atari's Montezuma's Revenge environment, a well-known sparse reward environment, and in doing so achieve the same performance as state-of-the-art HRL methods with substantially reduced training time cost. | ['Libo Xing'] | 2019-05-13 | null | null | null | null | ['montezumas-revenge'] | ['playing-games'] | [-8.55481774e-02 1.09496005e-01 -4.33463275e-01 6.80109635e-02
-8.09529245e-01 -5.17896235e-01 3.54102343e-01 1.53656527e-01
-6.63926005e-01 1.17778385e+00 2.80586362e-01 -3.15093011e-01
-1.91741884e-01 -6.31223142e-01 -4.93897647e-01 -7.76834905e-01
-5.21049798e-01 6.79187775e-01 2.04543054e-01 -4.35475707e-01
3.28754336e-01 3.51448178e-01 -1.41433811e+00 -1.33561373e-01
6.21031106e-01 8.37750137e-01 4.56815541e-01 5.66695750e-01
2.69907713e-01 1.07961464e+00 -5.79620361e-01 7.03615189e-01
5.08002818e-01 -4.95272547e-01 -8.10499907e-01 1.64105311e-01
-2.04833627e-01 -5.88147938e-01 -1.99206278e-01 9.54411447e-01
4.40297335e-01 6.84469223e-01 1.33231357e-01 -1.10958290e+00
-2.60159448e-02 7.43123293e-01 -4.81217474e-01 -3.89138609e-02
3.58221769e-01 7.55405784e-01 8.88015151e-01 -2.66777188e-01
5.62047005e-01 1.48811555e+00 -3.46061513e-02 7.12055862e-01
-1.53104854e+00 -5.62656581e-01 7.08127677e-01 7.36300647e-02
-8.11033666e-01 -1.61102355e-01 5.94903529e-01 -1.76060528e-01
8.29314053e-01 -2.58381903e-01 7.94165134e-01 9.76318300e-01
2.21537217e-01 7.85376012e-01 1.52297950e+00 -1.09541491e-01
1.00158703e+00 -2.19829261e-01 -2.92670101e-01 7.26783574e-01
-1.08798482e-01 8.08023453e-01 -5.05306840e-01 -2.85432637e-01
1.00072956e+00 1.63511019e-02 -8.41627941e-02 -7.41441369e-01
-1.13752198e+00 9.31304157e-01 5.19353449e-01 7.83989057e-02
-8.49771261e-01 3.65055323e-01 3.42596769e-01 6.69158340e-01
-3.91532183e-01 9.72671568e-01 -4.34599578e-01 -5.07501423e-01
-3.88224125e-01 6.85417831e-01 6.73627138e-01 6.74100757e-01
8.64253044e-01 5.10599673e-01 -3.97704124e-01 5.59633493e-01
1.16711959e-01 2.36815825e-01 5.75424671e-01 -1.48341882e+00
6.01946294e-01 5.53032577e-01 5.93201816e-01 -4.07136858e-01
-5.19113600e-01 -4.79221642e-01 -3.20921987e-01 1.12805569e+00
2.17109442e-01 -5.69941998e-01 -9.86963928e-01 2.00481653e+00
5.73786318e-01 -5.60019463e-02 4.07048613e-01 1.05248034e+00
-4.34134677e-02 5.55959821e-01 1.11158393e-01 -3.44156116e-01
1.03702819e+00 -9.66796517e-01 -4.65620905e-01 -4.94362265e-01
5.81958592e-01 -1.43616617e-01 1.14788210e+00 4.38520104e-01
-1.07470334e+00 -4.87453103e-01 -9.85174835e-01 5.74849844e-01
6.83116466e-02 -1.01442233e-01 6.25808120e-01 -6.41989857e-02
-8.02375853e-01 8.51028144e-01 -9.62625504e-01 4.28125821e-02
1.79305568e-01 4.43677187e-01 -1.25309244e-01 3.68369967e-02
-1.13228834e+00 9.43152487e-01 5.74503183e-01 -1.59156799e-01
-1.76151133e+00 -1.18872322e-01 -8.21745694e-01 1.50579885e-01
1.23304009e+00 -2.72672743e-01 1.84292758e+00 -6.36264682e-01
-1.88324678e+00 -3.63499448e-02 2.37809122e-01 -5.49748898e-01
3.39744538e-01 -2.61400700e-01 1.36413909e-02 8.87356047e-03
3.01725835e-01 7.63854265e-01 9.15956795e-01 -1.19443679e+00
-1.03554928e+00 -3.44910771e-01 3.81536275e-01 7.20128298e-01
1.53510958e-01 -3.80768925e-01 -1.14218220e-01 -4.55516368e-01
-1.56936184e-01 -1.11376154e+00 -9.85746205e-01 -4.66714174e-01
-1.06583387e-01 -3.72240007e-01 6.38108730e-01 -1.22992814e-01
1.03637922e+00 -1.95183432e+00 4.56267953e-01 1.55788392e-01
6.47436827e-02 1.96255893e-01 -3.53115648e-01 5.91581523e-01
1.30414024e-01 -3.28955203e-01 -1.12022109e-01 -1.15898721e-01
-4.20652708e-04 4.34287786e-01 -3.64825189e-01 1.42440945e-01
-9.50587019e-02 6.72870815e-01 -1.26445866e+00 -1.15089789e-01
1.71675190e-01 -2.96786278e-01 -6.15275085e-01 6.68569565e-01
-8.13680649e-01 7.55688488e-01 -8.21226418e-01 4.87279892e-01
-5.30631207e-02 -1.14136703e-01 4.17630076e-01 5.59626102e-01
-3.34128410e-01 5.13322413e-01 -1.42362547e+00 1.67381346e+00
-4.16163683e-01 4.04355749e-02 2.86217600e-01 -9.01388705e-01
8.09775949e-01 3.84850979e-01 5.56561649e-01 -9.90479827e-01
-2.40570512e-02 7.42410198e-02 1.73150420e-01 -3.71589839e-01
4.89713073e-01 -7.06299171e-02 -3.90395999e-01 6.57094836e-01
-2.83193439e-01 -1.55346051e-01 3.89121503e-01 8.35207924e-02
1.37858689e+00 5.28609037e-01 6.07922196e-01 4.77981567e-02
1.19249858e-01 2.73508221e-01 1.01448643e+00 1.19133294e+00
-3.43589336e-01 -2.11434111e-01 7.01539695e-01 -4.49977577e-01
-7.49992847e-01 -8.34922850e-01 5.35569131e-01 1.31882167e+00
1.50552645e-01 -3.22948903e-01 -3.40077817e-01 -8.21357846e-01
1.75537124e-01 8.21048260e-01 -6.72100246e-01 -1.20695285e-01
-8.10220420e-01 -2.75871992e-01 -1.01312347e-01 5.88331103e-01
5.50207019e-01 -1.60892665e+00 -1.56767452e+00 6.75627410e-01
2.83718258e-01 -7.51143157e-01 -5.64945281e-01 7.55273998e-01
-9.05861676e-01 -1.03814769e+00 -3.62284571e-01 -5.19475102e-01
5.36022365e-01 3.26214135e-02 1.01553512e+00 -1.89709544e-01
-2.68597692e-01 3.56582880e-01 -2.19486177e-01 -4.24012132e-02
-2.82500237e-01 -1.96570411e-01 3.11609238e-01 -5.60466707e-01
-3.22480500e-02 -5.45756161e-01 -4.83882070e-01 2.87266463e-01
-5.95563352e-01 6.23858608e-02 7.08456397e-01 1.09574306e+00
6.20925725e-01 3.80750299e-01 6.81774616e-01 -6.24296069e-01
9.08447444e-01 -4.09310043e-01 -1.17913938e+00 -1.24051519e-01
-6.71675265e-01 5.96575618e-01 8.84516299e-01 -7.17798173e-01
-9.58644688e-01 3.31447691e-01 2.82142311e-01 -4.75119233e-01
-2.66252160e-01 5.51210761e-01 5.55632897e-02 2.77872741e-01
6.55898988e-01 2.47975409e-01 1.55221656e-01 -4.08874512e-01
3.35402846e-01 2.35676527e-01 4.34324831e-01 -9.93216157e-01
6.43786311e-01 -6.80856407e-02 1.15563180e-02 -1.83482692e-01
-6.34752154e-01 -1.42781451e-01 -5.00954948e-02 -1.94956854e-01
6.03756666e-01 -9.25776184e-01 -1.16143847e+00 3.93980667e-02
-5.03330886e-01 -1.04214489e+00 -6.52716398e-01 4.69407618e-01
-9.68154490e-01 -2.10450932e-01 -4.47726220e-01 -9.32931244e-01
4.77134921e-02 -1.45869136e+00 6.74727678e-01 4.64327455e-01
-1.47517756e-01 -5.79890966e-01 3.00550669e-01 -3.94981429e-02
4.18225378e-01 1.80317685e-01 8.85289609e-01 -4.22427148e-01
-6.96687222e-01 2.91932821e-01 3.57693851e-01 -1.79029740e-02
1.90862820e-01 -7.83084095e-01 -2.95099735e-01 -7.60958254e-01
1.05469368e-01 -9.84824598e-01 5.27991474e-01 1.08531356e-01
1.04510224e+00 -6.17053866e-01 -2.42564648e-01 7.62662739e-02
1.28348625e+00 7.06588030e-01 1.47956342e-01 5.57773829e-01
2.19249591e-01 4.47398663e-01 1.24159884e+00 8.66490483e-01
2.88680613e-01 7.94726133e-01 9.64109361e-01 1.27088666e-01
2.56107628e-01 -4.36212510e-01 6.68795168e-01 -4.88270223e-02
8.98620263e-02 2.74734855e-01 -5.25661290e-01 3.64459306e-01
-2.36084628e+00 -1.00535250e+00 6.94564104e-01 2.39226055e+00
1.05031478e+00 2.58391887e-01 4.52781767e-01 -2.47600719e-01
2.61783123e-01 2.07445592e-01 -1.31423807e+00 -3.15941125e-01
5.07746220e-01 -1.78668089e-02 3.94601375e-01 5.90137661e-01
-8.58957350e-01 1.23834503e+00 6.13529110e+00 5.79537988e-01
-1.05423963e+00 -3.78643274e-01 3.94740641e-01 -4.66868013e-01
-9.03876685e-03 2.26753235e-01 -9.19070721e-01 2.89739579e-01
6.60140991e-01 -1.54087752e-01 1.12097907e+00 1.15582073e+00
5.30980885e-01 -5.75673819e-01 -1.26413345e+00 5.99873364e-01
-4.40571994e-01 -1.08674073e+00 -5.12754500e-01 2.23779023e-01
7.08570123e-01 3.98824317e-03 -7.71876145e-03 8.58464479e-01
1.21210253e+00 -1.03776407e+00 5.32850564e-01 1.80391908e-01
5.08231759e-01 -8.92539740e-01 1.42074659e-01 7.79497266e-01
-1.15228546e+00 -7.49773145e-01 -1.21968850e-01 -4.19409603e-01
5.14096878e-02 -8.64158422e-02 -9.97612774e-01 1.78354800e-01
4.36844558e-01 3.93558353e-01 -9.37053561e-02 7.20050216e-01
-6.14382207e-01 3.46579283e-01 -2.45529160e-01 -2.78753936e-01
8.20458353e-01 -2.20128953e-01 6.37434244e-01 4.12238002e-01
1.22499391e-01 2.76328713e-01 1.09493983e+00 7.34806001e-01
2.33047485e-01 -3.13705087e-01 -7.08508492e-01 -1.87996611e-01
5.88950753e-01 1.07246792e+00 -3.18470329e-01 -3.44956517e-01
-5.29488623e-02 5.96971869e-01 7.59104192e-01 4.98292923e-01
-5.65821409e-01 -1.02953888e-01 8.86659980e-01 -2.28232935e-01
2.97855854e-01 -4.38315481e-01 1.11248910e-01 -8.78919601e-01
-3.26428831e-01 -1.37279546e+00 4.63233054e-01 -5.79290032e-01
-7.49477863e-01 4.34712350e-01 2.28544492e-02 -1.18991673e+00
-9.46793616e-01 -2.03534797e-01 -4.32162404e-01 7.93259442e-01
-1.44450796e+00 -2.95231313e-01 9.52471048e-02 7.37529576e-01
8.38518679e-01 -4.30377603e-01 9.79945123e-01 -4.00703520e-01
-5.62983453e-01 4.00817357e-02 6.42109811e-02 -2.90035844e-01
4.92935300e-01 -1.53695524e+00 -1.67702548e-02 5.38033545e-01
-2.81314373e-01 3.67143869e-01 7.98684180e-01 -7.95706570e-01
-1.47538245e+00 -8.18648279e-01 5.08029275e-02 1.35613486e-01
5.80611169e-01 -1.93249229e-02 -6.96221769e-01 6.16912127e-01
1.11650147e-01 -2.64273107e-01 2.19450161e-01 1.90995768e-01
1.77828036e-02 -5.85911646e-02 -9.67054427e-01 1.06134772e+00
6.95229888e-01 -9.76044834e-02 -7.61758804e-01 1.83672130e-01
6.43870354e-01 -6.77223027e-01 -5.81262290e-01 2.81177759e-01
2.16148347e-01 -9.15045559e-01 6.73152924e-01 -9.51556385e-01
3.22200596e-01 -4.79559630e-01 5.87155223e-02 -1.78135824e+00
-4.91091371e-01 -8.82901132e-01 -3.06121349e-01 5.07494152e-01
7.30980411e-02 -5.99248707e-01 7.66109467e-01 6.31606817e-01
-4.39259186e-02 -1.03307676e+00 -9.76253211e-01 -8.96529496e-01
-1.56859070e-01 -1.60250768e-01 6.24223709e-01 5.15734076e-01
3.39836031e-01 4.69125509e-01 -7.15744853e-01 1.07401580e-01
6.00928843e-01 6.44221902e-01 9.43768144e-01 -8.08761299e-01
-8.18797529e-01 -3.58771801e-01 3.87480199e-01 -1.25322437e+00
7.21856654e-02 -4.33278471e-01 6.00759625e-01 -1.38577747e+00
-6.60402477e-02 -7.12126255e-01 -4.08623576e-01 1.03575051e+00
-1.28492579e-01 -6.24435663e-01 4.41708773e-01 1.44973859e-01
-9.59073722e-01 8.26787472e-01 1.59373188e+00 8.65613595e-02
-8.23477864e-01 1.51849583e-01 -7.36209214e-01 6.47374213e-01
9.85785127e-01 -5.26349247e-01 -7.05056608e-01 -1.66073725e-01
-2.06302162e-02 9.82435942e-01 9.51610655e-02 -9.53463733e-01
1.84214115e-01 -8.41376007e-01 2.29013011e-01 -2.86479592e-01
4.91149127e-01 -8.56685460e-01 -4.34288718e-02 9.97737169e-01
-8.37274730e-01 3.44104677e-01 1.12418063e-01 6.91634595e-01
3.51828821e-02 -1.88791379e-01 8.50403965e-01 -5.68939805e-01
-8.40001464e-01 4.36432302e-01 -8.12745750e-01 -2.87447684e-03
1.23717237e+00 1.88218817e-01 -2.14023337e-01 -4.98805821e-01
-8.39039624e-01 1.05102277e+00 4.01335180e-01 4.11201417e-01
6.11965001e-01 -1.21388519e+00 -2.71097243e-01 4.13152464e-02
-1.29652604e-01 1.54953584e-01 1.39850136e-02 3.99766177e-01
3.25899720e-01 2.17046693e-01 -5.25066674e-01 -2.48904064e-01
-1.06161070e+00 6.62786543e-01 3.68992418e-01 -8.80068660e-01
-8.58693302e-01 3.69010657e-01 5.40478975e-02 -2.63146818e-01
5.37224710e-01 -1.21268556e-01 -4.20178145e-01 -1.69595033e-01
4.82289135e-01 3.44398409e-01 -4.62975025e-01 1.11751020e-01
-8.40930641e-02 2.11892784e-01 -2.68875122e-01 -4.86804873e-01
1.31919801e+00 3.69335338e-02 2.81157196e-01 4.02681172e-01
4.52052563e-01 -3.93165976e-01 -2.00793409e+00 -3.81579876e-01
2.49881446e-02 -5.75945973e-01 5.89972399e-02 -1.10261512e+00
-6.43740058e-01 4.95819300e-01 4.82963592e-01 9.53633189e-02
1.05424082e+00 -2.94830739e-01 5.04933655e-01 8.34328234e-01
8.36483300e-01 -1.55834258e+00 6.52617157e-01 8.02742958e-01
9.80443001e-01 -1.28297961e+00 -1.15218088e-01 3.94921243e-01
-1.13625503e+00 9.06669378e-01 1.12233222e+00 -1.77290365e-01
-1.84057597e-02 1.85431540e-01 -5.03907502e-02 -1.00198790e-01
-1.29426956e+00 -3.73380214e-01 -3.87274623e-01 5.36228180e-01
-3.29759419e-01 1.56156093e-01 -3.39553803e-02 8.80168080e-02
1.57313421e-02 3.98244597e-02 4.93196547e-01 1.32168055e+00
-9.77140605e-01 -1.26819193e+00 -3.61899436e-01 4.09835964e-01
1.51918055e-02 3.28927755e-01 7.74514228e-02 6.47920787e-01
-2.31762350e-01 8.07586789e-01 -1.08980320e-01 -1.31843999e-01
2.52989054e-01 -1.32842436e-01 3.57791960e-01 -9.78742540e-01
-6.30741596e-01 3.22417587e-01 1.22040100e-01 -1.18784070e+00
9.04803500e-02 -7.10734129e-01 -1.79103374e+00 1.19490951e-01
2.67004639e-01 3.65287602e-01 4.37197655e-01 9.57176268e-01
3.84564519e-01 7.83610642e-01 9.76483285e-01 -8.89557123e-01
-1.28132546e+00 -6.23665452e-01 -5.77994704e-01 2.20768347e-01
6.88063025e-01 -9.47888255e-01 -1.23979123e-02 -5.89400232e-01] | [4.100903034210205, 1.703028917312622] |
9665cf4d-e35f-42a7-a52e-aeb767a2ba19 | adaptive-face-recognition-using-adversarial | 2305.13605 | null | https://arxiv.org/abs/2305.13605v1 | https://arxiv.org/pdf/2305.13605v1.pdf | Adaptive Face Recognition Using Adversarial Information Network | In many real-world applications, face recognition models often degenerate when training data (referred to as source domain) are different from testing data (referred to as target domain). To alleviate this mismatch caused by some factors like pose and skin tone, the utilization of pseudo-labels generated by clustering algorithms is an effective way in unsupervised domain adaptation. However, they always miss some hard positive samples. Supervision on pseudo-labeled samples attracts them towards their prototypes and would cause an intra-domain gap between pseudo-labeled samples and the remaining unlabeled samples within target domain, which results in the lack of discrimination in face recognition. In this paper, considering the particularity of face recognition, we propose a novel adversarial information network (AIN) to address it. First, a novel adversarial mutual information (MI) loss is proposed to alternately minimize MI with respect to the target classifier and maximize MI with respect to the feature extractor. By this min-max manner, the positions of target prototypes are adaptively modified which makes unlabeled images clustered more easily such that intra-domain gap can be mitigated. Second, to assist adversarial MI loss, we utilize a graph convolution network to predict linkage likelihoods between target data and generate pseudo-labels. It leverages valuable information in the context of nodes and can achieve more reliable results. The proposed method is evaluated under two scenarios, i.e., domain adaptation across poses and image conditions, and domain adaptation across faces with different skin tones. Extensive experiments show that AIN successfully improves cross-domain generalization and offers a new state-of-the-art on RFW dataset. | ['Weihong Deng', 'Mei Wang'] | 2023-05-23 | null | null | null | null | ['face-recognition', 'unsupervised-domain-adaptation'] | ['computer-vision', 'methodology'] | [ 4.85631734e-01 1.11714229e-01 -2.77584642e-01 -6.72716558e-01
-3.36805373e-01 -6.00108862e-01 3.67218167e-01 -4.19928610e-01
-8.46584514e-02 7.31800854e-01 -2.23331332e-01 3.39266397e-02
-1.86544418e-01 -8.31980526e-01 -5.61980009e-01 -9.56805170e-01
1.02453902e-01 3.50872755e-01 -4.40839976e-02 2.85338126e-02
-3.61997373e-02 6.91727459e-01 -1.33272958e+00 1.92000881e-01
1.10337591e+00 1.10240567e+00 3.20727355e-03 -9.30098817e-02
-3.11911911e-01 3.43357950e-01 -8.28953087e-01 -5.65744221e-01
5.26084244e-01 -5.40731966e-01 -4.33311135e-01 2.38714159e-01
3.94441277e-01 -1.60611182e-01 -4.24766213e-01 1.39971161e+00
6.32556915e-01 7.22711235e-02 9.43799436e-01 -1.55805004e+00
-8.75460863e-01 4.32232589e-01 -9.21921968e-01 -1.09500326e-01
8.05603266e-02 1.62333220e-01 4.65824336e-01 -1.00360727e+00
5.59226751e-01 1.36591792e+00 5.17060757e-01 9.24083769e-01
-1.31030333e+00 -1.27324581e+00 2.06324473e-01 2.08760828e-01
-1.65996552e+00 -4.87217396e-01 1.22314477e+00 -2.31415421e-01
1.21661007e-01 6.80923313e-02 5.95441759e-02 1.33645952e+00
-1.25570476e-01 5.95922947e-01 1.13207126e+00 -2.88085282e-01
2.19518065e-01 4.04195219e-01 -2.44857505e-01 4.86936122e-01
9.54154804e-02 2.47949094e-01 -4.93862510e-01 -1.17399156e-01
6.38847232e-01 2.51674000e-02 -3.76536191e-01 -5.99208176e-01
-6.94470406e-01 6.78337514e-01 5.49804151e-01 5.62813953e-02
-1.57371923e-01 -5.49620271e-01 1.35470927e-01 2.61671603e-01
2.67132998e-01 6.19545169e-02 -3.20158988e-01 4.76268083e-01
-7.26398528e-01 -1.87689617e-01 6.53743088e-01 9.88038957e-01
8.79226565e-01 2.23406717e-01 -4.02236246e-02 1.27300251e+00
4.83269244e-01 6.17114782e-01 6.49249792e-01 -4.75558519e-01
6.02137923e-01 7.66232848e-01 -2.21737429e-01 -1.31878710e+00
-2.02792183e-01 -5.50663888e-01 -1.07945013e+00 3.00724149e-01
4.32566345e-01 -2.41790488e-01 -1.11322522e+00 2.04443669e+00
5.66676438e-01 5.38556993e-01 1.67567030e-01 1.02161193e+00
7.64018536e-01 4.42741901e-01 1.97842836e-01 -3.70983750e-01
1.05819225e+00 -5.59760511e-01 -5.45749664e-01 -5.00812352e-01
2.19848886e-01 -6.95216477e-01 8.29434276e-01 1.54524982e-01
-5.98178506e-01 -8.02574694e-01 -1.28012609e+00 5.25956392e-01
-3.23654741e-01 8.80671740e-02 2.94106662e-01 8.09632838e-01
-6.43535376e-01 4.53986406e-01 -5.43043435e-01 -1.69686928e-01
7.17540145e-01 5.56431711e-01 -5.58788896e-01 -3.17147523e-01
-1.17401421e+00 4.71859694e-01 4.01151121e-01 1.43851876e-01
-7.83736169e-01 -6.53778076e-01 -6.76038265e-01 -2.26132661e-01
4.81622517e-01 -2.51612425e-01 5.96854925e-01 -1.52920461e+00
-1.45511913e+00 7.95661688e-01 7.96795860e-02 1.05008334e-02
5.22155225e-01 2.16291234e-01 -9.11905944e-01 2.35648248e-02
5.25522307e-02 7.78966308e-01 1.21063995e+00 -1.42964041e+00
-3.60667050e-01 -7.32045293e-01 -3.21819395e-01 2.76001036e-01
-7.71601081e-01 -3.80703926e-01 -3.66203159e-01 -8.90153110e-01
1.53835297e-01 -8.61591935e-01 2.01619059e-01 1.77795947e-01
-4.80836660e-01 -1.61725268e-01 1.23454046e+00 -4.86582577e-01
9.14133251e-01 -2.37792659e+00 5.88788465e-02 5.01871467e-01
-2.97685023e-02 5.94582558e-01 -4.20139283e-01 -1.83379631e-02
-3.57521027e-01 1.12500034e-01 -4.91060704e-01 4.42051813e-02
-2.41080344e-01 3.18562746e-01 -2.67469734e-01 3.89643908e-01
5.80875993e-01 5.06018519e-01 -7.33157694e-01 -5.45716941e-01
5.92198744e-02 4.82406884e-01 -3.33701104e-01 3.18571717e-01
-3.48499790e-02 6.76346362e-01 -5.88636518e-01 7.60035038e-01
1.33109009e+00 8.75567179e-03 3.66942674e-01 -3.87593657e-01
5.31061769e-01 -3.46709549e-01 -1.36403990e+00 1.41139996e+00
-2.95333266e-01 1.94666028e-01 1.29242942e-01 -1.21317399e+00
1.31007087e+00 1.34475276e-01 4.02166158e-01 -6.16735041e-01
2.96228174e-02 4.88202050e-02 1.27522826e-01 -2.25602806e-01
-2.53296822e-01 -1.26096457e-01 2.50016749e-01 2.74391353e-01
2.13326111e-01 2.57051915e-01 -2.67574906e-01 -1.63873866e-01
7.47434735e-01 1.82087570e-01 8.52685645e-02 -1.54401973e-01
7.40061879e-01 -3.20209801e-01 9.72664118e-01 3.14463437e-01
-3.97279888e-01 6.26110375e-01 3.45182508e-01 -1.97726101e-01
-6.65969431e-01 -1.24173224e+00 -2.89334476e-01 7.33098924e-01
4.21603113e-01 2.52519161e-01 -7.49086142e-01 -1.21136367e+00
2.48262972e-01 5.34102857e-01 -5.96387684e-01 -6.65384829e-01
-3.81906658e-01 -7.27412820e-01 7.09431291e-01 4.39274609e-01
8.56268942e-01 -1.03491890e+00 1.83989272e-01 -2.63064336e-02
1.06601082e-01 -9.25202429e-01 -5.09222329e-01 4.39463519e-02
-7.38300622e-01 -1.17355585e+00 -6.58216655e-01 -8.72350335e-01
1.10140657e+00 1.83456108e-01 7.37716734e-01 -1.74229681e-01
-3.07524741e-01 9.18907821e-02 -3.49234909e-01 -1.95931092e-01
-3.27452123e-01 -1.92230806e-01 2.62155920e-01 5.57345748e-01
6.59969985e-01 -5.93584120e-01 -6.73235476e-01 7.99945116e-01
-1.04350698e+00 -3.83215100e-01 6.73874199e-01 1.14339137e+00
5.37182271e-01 2.81123370e-01 9.72477436e-01 -9.46115315e-01
4.11421001e-01 -6.87976003e-01 -4.83936161e-01 4.64751154e-01
-6.17006063e-01 -6.17596433e-02 1.00811219e+00 -9.44573343e-01
-1.29480004e+00 2.09666789e-01 1.83173016e-01 -6.46567762e-01
-2.05381945e-01 1.79895118e-01 -8.68484199e-01 -3.77350241e-01
7.63371825e-01 1.85829595e-01 2.43192643e-01 -3.18766952e-01
1.86139822e-01 7.95759320e-01 4.34001148e-01 -5.18166542e-01
1.18985128e+00 2.73051739e-01 9.30441990e-02 -7.01662898e-01
-4.73050743e-01 -7.59138465e-02 -5.97220421e-01 -1.58117801e-01
4.95714992e-01 -8.14043581e-01 -4.73899990e-01 5.95976830e-01
-7.07628787e-01 -6.67527542e-02 8.24885257e-03 3.64888161e-01
-7.30331019e-02 4.19013172e-01 -1.67756110e-01 -7.02027023e-01
-1.21885501e-01 -1.05767977e+00 6.94323957e-01 6.35353684e-01
1.43507063e-01 -8.97189260e-01 -4.98212695e-01 2.43754283e-01
1.68110207e-01 2.03660131e-01 8.76443148e-01 -9.17128026e-01
-2.83497870e-01 -1.97685644e-01 -3.46755087e-01 7.46002316e-01
5.69215655e-01 -1.37653202e-01 -1.03554988e+00 -4.13358033e-01
2.19919588e-02 -3.78865212e-01 5.07850647e-01 3.66804795e-03
1.26833475e+00 -4.13879633e-01 -5.52791119e-01 7.45190501e-01
1.23304772e+00 4.42973703e-01 5.42118728e-01 -1.02609716e-01
7.54712462e-01 8.11233699e-01 7.82227695e-01 3.21051806e-01
-3.47095355e-02 6.74175143e-01 3.25778782e-01 -6.52256757e-02
-2.43497461e-01 -4.32239562e-01 2.76984870e-01 2.65068084e-01
4.02273178e-01 -2.66975075e-01 -7.20238924e-01 2.24176049e-01
-1.42285275e+00 -6.62975073e-01 4.76173550e-01 2.35369349e+00
9.45673823e-01 -5.41413799e-02 -3.67779583e-02 1.08163767e-02
1.15801215e+00 -5.52091841e-03 -1.02506018e+00 6.39015734e-02
-1.78205341e-01 2.92776972e-01 4.40083981e-01 1.18875891e-01
-1.06198752e+00 8.47116470e-01 5.01515341e+00 1.14403915e+00
-1.41378725e+00 -8.83725956e-02 9.20069218e-01 1.74338445e-01
-1.12788804e-01 -3.32913846e-01 -7.58260250e-01 7.58122146e-01
4.26764548e-01 6.53549507e-02 5.51496446e-01 9.88538325e-01
-1.80365160e-01 2.84776866e-01 -1.01300049e+00 9.66511011e-01
1.47608966e-01 -6.65311635e-01 1.60223082e-01 -1.28002400e-02
7.58828998e-01 -5.05019605e-01 4.26271498e-01 2.24935055e-01
2.01004058e-01 -1.01851594e+00 2.56421834e-01 2.09775865e-01
1.06193042e+00 -9.95179355e-01 6.13693893e-01 2.49130204e-01
-1.01370656e+00 -1.43842727e-01 -5.85799932e-01 5.18189132e-01
-3.77639830e-01 4.95470762e-01 -1.01217103e+00 6.45100296e-01
4.87422794e-01 5.07457078e-01 -5.42300642e-01 7.17703223e-01
-2.82898366e-01 4.77027178e-01 -3.04996789e-01 2.47961178e-01
-2.32639119e-01 -2.88257331e-01 5.18333137e-01 7.92779922e-01
3.23731631e-01 8.07242692e-02 2.12646395e-01 8.93888116e-01
-3.21254730e-01 1.09888077e-01 -7.45305419e-01 4.42808904e-02
9.10013378e-01 1.25355029e+00 -5.44356465e-01 -3.89570408e-02
-3.11128289e-01 1.13684404e+00 2.79587448e-01 6.54394925e-01
-8.10898125e-01 -4.29356992e-01 7.34502196e-01 -5.49706183e-02
9.53279957e-02 1.34571776e-01 -2.19653055e-01 -9.68480766e-01
1.51907071e-01 -9.72446442e-01 4.74627346e-01 -2.85002232e-01
-1.72702444e+00 6.01316094e-01 -1.59828112e-01 -1.42698038e+00
-1.43846214e-01 -6.24615788e-01 -5.72604060e-01 9.77194309e-01
-1.38848722e+00 -1.17630899e+00 -4.35404480e-01 8.18502307e-01
2.29559869e-01 -5.31566203e-01 6.79046929e-01 5.31902194e-01
-7.47870445e-01 1.36464047e+00 1.49264336e-01 4.36347783e-01
1.08234394e+00 -8.31001341e-01 -5.18410988e-02 8.26240838e-01
6.35981262e-02 6.52882934e-01 2.31896475e-01 -7.31055021e-01
-1.27208233e+00 -1.40914297e+00 2.20346689e-01 3.34508047e-02
2.25665361e-01 -2.96117634e-01 -1.06544006e+00 4.41958189e-01
-3.62428665e-01 2.85846591e-01 8.45212758e-01 -1.56313255e-01
-6.20091558e-01 -6.75801575e-01 -1.66311347e+00 6.21860027e-01
1.11162233e+00 -4.49660867e-01 -2.93058485e-01 2.39141539e-01
4.00117368e-01 -2.24593297e-01 -8.36557329e-01 7.95823574e-01
3.84497017e-01 -7.94846058e-01 1.05996358e+00 -5.24879694e-01
1.25714809e-01 -4.59583819e-01 -9.80169605e-03 -1.36025262e+00
-2.27938011e-01 -3.31286550e-01 6.70238510e-02 1.70795298e+00
3.16001564e-01 -8.60214412e-01 9.73614752e-01 6.39548659e-01
5.61197475e-02 -5.17236888e-01 -1.04192913e+00 -8.76492739e-01
-6.30016699e-02 7.47665316e-02 6.69610679e-01 1.17311776e+00
-3.20813924e-01 1.42712384e-01 -2.30792418e-01 3.71012926e-01
8.06164443e-01 -5.90049066e-02 6.18236065e-01 -1.21276307e+00
-6.64013028e-02 -2.89310485e-01 -5.04067659e-01 -8.32717061e-01
4.73287523e-01 -1.00598562e+00 -4.52223010e-02 -7.58467436e-01
-4.25677374e-02 -8.03227544e-01 -4.67256933e-01 5.84652603e-01
-2.45687082e-01 4.37919110e-01 -5.68374321e-02 2.00887576e-01
-4.01020944e-02 5.64207673e-01 1.31563652e+00 -3.44337732e-01
-2.13661268e-01 1.00525789e-01 -7.65999615e-01 4.91756797e-01
7.90953815e-01 -3.79174829e-01 -6.09387577e-01 -2.18863785e-01
-5.42886555e-01 -2.35041194e-02 2.91501671e-01 -1.06773138e+00
1.82476282e-01 -2.41253480e-01 9.54718590e-01 -1.55548185e-01
2.67153025e-01 -1.06536305e+00 1.97733074e-01 2.28448972e-01
-1.05232731e-01 -4.80610520e-01 1.75109491e-01 7.06345618e-01
-2.37874478e-01 -1.85387522e-01 1.14514554e+00 1.76013336e-01
-7.20927238e-01 5.60865700e-01 2.42470816e-01 -1.92773119e-02
1.31658518e+00 -4.53281432e-01 -2.90384263e-01 -1.86141178e-01
-5.54865301e-01 2.89660275e-01 4.26891595e-01 5.92334330e-01
6.72027111e-01 -1.60819042e+00 -6.30962133e-01 6.21189654e-01
2.43764311e-01 8.56916246e-04 4.04984921e-01 4.01721925e-01
-2.41319276e-02 3.82104106e-02 -3.94076496e-01 -6.03920758e-01
-1.18823731e+00 7.00056314e-01 2.66574144e-01 8.35271254e-02
-9.59864408e-02 9.75335777e-01 4.96626258e-01 -5.84333420e-01
2.28757337e-01 4.47802693e-01 -2.32230902e-01 8.10072273e-02
4.83529627e-01 1.54490083e-01 -8.82915705e-02 -6.54656827e-01
-4.67449158e-01 6.21952534e-01 -1.76404223e-01 2.33160615e-01
8.87924314e-01 -9.09343287e-02 -9.12194327e-03 -6.53574094e-02
1.35082376e+00 -2.55731679e-02 -1.47950375e+00 -3.86109859e-01
-1.38530105e-01 -6.86038613e-01 -3.48561138e-01 -9.60741282e-01
-1.42115700e+00 8.30415130e-01 9.59211290e-01 -1.33842021e-01
1.44525206e+00 -2.74990290e-01 5.44337034e-01 6.81163445e-02
2.55839139e-01 -1.15435982e+00 3.90351146e-01 3.23736668e-02
6.61013186e-01 -1.40593243e+00 -2.65691847e-01 -7.39740014e-01
-7.10427403e-01 1.03543055e+00 1.13409877e+00 6.53461143e-02
5.75133502e-01 6.41547665e-02 1.99888304e-01 2.12640822e-01
-1.42213345e-01 1.54929385e-01 3.77896875e-01 1.12262523e+00
1.40763462e-01 6.68814555e-02 -1.22031458e-02 7.21152008e-01
1.17998146e-01 -1.93093851e-01 -7.14008361e-02 6.47398889e-01
-1.79009810e-01 -1.38876152e+00 -5.30297637e-01 3.95280153e-01
-2.93542773e-01 1.36904791e-01 -5.57564437e-01 7.73215652e-01
3.00095260e-01 1.12919688e+00 1.10605331e-02 -6.94557548e-01
2.22798988e-01 2.15399995e-01 4.63558257e-01 -4.71714467e-01
-1.55448616e-01 -1.75667316e-01 -3.94730091e-01 -2.47441202e-01
-1.80491969e-01 -2.87654847e-01 -1.17243671e+00 -1.20502003e-01
-4.15373176e-01 1.09245978e-01 4.74172443e-01 7.15104282e-01
4.78950649e-01 4.17964131e-01 1.21520066e+00 -4.89154130e-01
-8.42843175e-01 -9.15135860e-01 -6.27841294e-01 6.53360844e-01
1.48123652e-02 -9.05185521e-01 -3.66903514e-01 2.05681962e-03] | [13.092738151550293, 1.0093599557876587] |
328cdc21-fd60-4f41-8fe8-8408090a9b88 | generating-diverse-and-consistent-qa-pairs | 2005.13837 | null | https://arxiv.org/abs/2005.13837v5 | https://arxiv.org/pdf/2005.13837v5.pdf | Generating Diverse and Consistent QA pairs from Contexts with Information-Maximizing Hierarchical Conditional VAEs | One of the most crucial challenges in question answering (QA) is the scarcity of labeled data, since it is costly to obtain question-answer (QA) pairs for a target text domain with human annotation. An alternative approach to tackle the problem is to use automatically generated QA pairs from either the problem context or from large amount of unstructured texts (e.g. Wikipedia). In this work, we propose a hierarchical conditional variational autoencoder (HCVAE) for generating QA pairs given unstructured texts as contexts, while maximizing the mutual information between generated QA pairs to ensure their consistency. We validate our Information Maximizing Hierarchical Conditional Variational AutoEncoder (Info-HCVAE) on several benchmark datasets by evaluating the performance of the QA model (BERT-base) using only the generated QA pairs (QA-based evaluation) or by using both the generated and human-labeled pairs (semi-supervised learning) for training, against state-of-the-art baseline models. The results show that our model obtains impressive performance gains over all baselines on both tasks, using only a fraction of data for training. | ['Seanie Lee', 'Donghwan Kim', 'Sung Ju Hwang', 'Dong Bok Lee', 'Woo Tae Jeong'] | 2020-05-28 | generating-diverse-and-consistent-qa-pairs-1 | https://aclanthology.org/2020.acl-main.20 | https://aclanthology.org/2020.acl-main.20.pdf | acl-2020-6 | ['question-answer-generation'] | ['natural-language-processing'] | [ 3.50636393e-02 4.93627578e-01 5.20923257e-01 -4.03177559e-01
-1.73612905e+00 -7.65755057e-01 6.75266981e-01 1.84929520e-02
-4.72478509e-01 9.54445302e-01 2.91848391e-01 -2.70422459e-01
2.73455471e-01 -8.08196008e-01 -7.29931712e-01 -4.02173281e-01
7.51734734e-01 1.34027600e+00 2.97514528e-01 -5.00428557e-01
-1.27519593e-01 -6.09769464e-01 -1.55968463e+00 5.94303370e-01
1.51149881e+00 9.40670371e-01 4.90133852e-01 6.53371155e-01
-3.92340928e-01 1.12847376e+00 -7.87104130e-01 -9.91576195e-01
-6.18810579e-02 -7.82672226e-01 -1.38395798e+00 -2.68352211e-01
4.93121386e-01 -3.66465837e-01 -1.28776701e-02 8.92041683e-01
4.23442870e-01 2.99285471e-01 1.06877387e+00 -1.17194426e+00
-8.04104269e-01 5.38343787e-01 -3.72815914e-02 -1.54258385e-01
4.55617130e-01 2.36158744e-02 1.47726071e+00 -9.28436816e-01
8.57713282e-01 1.25932479e+00 2.59503603e-01 7.85885453e-01
-1.18818104e+00 -1.43825904e-01 -4.60213184e-01 5.32397211e-01
-1.11832142e+00 -2.66672283e-01 5.78528464e-01 -4.73792762e-01
9.34490979e-01 7.81381205e-02 -6.80995882e-02 1.30322480e+00
-6.56071026e-03 1.00152969e+00 9.50882256e-01 -7.01894522e-01
3.83458942e-01 2.22771376e-01 4.66590613e-01 5.05970955e-01
-3.31528336e-01 -2.80332536e-01 -2.09088624e-01 -5.31620562e-01
2.18719408e-01 -6.91549480e-01 -4.84745890e-01 -3.00028473e-01
-1.03755343e+00 1.29286337e+00 2.45767221e-01 3.23241293e-01
-4.45620418e-01 -1.43181220e-01 2.16046199e-01 4.14124578e-01
2.69506484e-01 6.46768808e-01 -6.79767549e-01 -1.58281028e-01
-6.92753553e-01 5.96077263e-01 9.79250550e-01 7.61977196e-01
9.58651423e-01 -2.85041600e-01 -8.16759408e-01 1.02810657e+00
4.23917621e-01 5.35522223e-01 6.11612141e-01 -1.20497012e+00
9.00018394e-01 7.02696681e-01 4.19694930e-01 -5.67306697e-01
6.01173677e-02 -1.68606564e-02 -6.78561747e-01 -3.68145168e-01
7.54404664e-01 -5.27393699e-01 -8.25756848e-01 1.88091886e+00
5.81177115e-01 -2.61119634e-01 5.69702864e-01 7.21878827e-01
1.20932221e+00 1.17156148e+00 1.99863806e-01 -1.11802176e-01
1.47851038e+00 -1.29075658e+00 -1.00351965e+00 -1.89993575e-01
8.00543249e-01 -5.24104059e-01 1.30849802e+00 8.95692706e-02
-9.50348198e-01 -5.81141293e-01 -6.66698813e-01 -5.85856795e-01
-2.11805657e-01 1.63960323e-01 -1.64847061e-01 3.95223200e-01
-9.34842229e-01 3.40041041e-01 -4.84285265e-01 5.21274246e-02
4.27917857e-03 5.60496934e-03 -2.41036385e-01 -2.51961708e-01
-1.79265916e+00 9.63780880e-01 3.50129336e-01 -1.74298078e-01
-1.19502985e+00 -6.36681199e-01 -1.21129000e+00 2.42221251e-01
3.56584370e-01 -9.10986662e-01 1.73072112e+00 -9.16180611e-01
-1.40120900e+00 5.90131104e-01 -3.97849143e-01 -5.29081702e-01
5.78226805e-01 -2.30921641e-01 2.53651440e-02 2.21693411e-01
3.41646284e-01 8.60762239e-01 8.61813664e-01 -1.43276501e+00
-3.87068748e-01 -2.54879385e-01 2.47092232e-01 2.74977714e-01
-1.42647430e-01 -3.23208541e-01 -2.32326999e-01 -3.36052291e-02
-3.67545605e-01 -7.53327370e-01 -1.70200527e-01 -6.74415410e-01
-3.13651294e-01 -1.07376909e+00 7.44527340e-01 -1.35884845e+00
9.65908825e-01 -1.55088782e+00 4.16047335e-01 -2.66921014e-01
5.68897389e-02 3.17583799e-01 -2.66430527e-01 5.31027079e-01
3.08542669e-01 -1.17226802e-01 -7.70196021e-01 -5.79898298e-01
1.42058164e-01 3.26908380e-01 -4.77130890e-01 -1.40932947e-01
4.11379844e-01 1.04293251e+00 -1.16552293e+00 -8.98387074e-01
-2.26629968e-03 3.40131789e-01 -5.93301177e-01 7.96753228e-01
-9.54753280e-01 6.52610421e-01 -3.94726306e-01 3.11873138e-01
4.75241780e-01 -5.98229349e-01 4.75233823e-01 -2.16069579e-01
6.23150647e-01 4.40075219e-01 -7.91032553e-01 1.62470114e+00
-6.95079625e-01 6.38091981e-01 -8.66245553e-02 -8.42874050e-01
7.00653791e-01 8.20410669e-01 4.99316454e-02 -8.63271892e-01
2.24132568e-01 1.80295840e-01 -2.68708259e-01 -7.30323434e-01
6.52378619e-01 -5.77129722e-02 -1.89844251e-01 6.40347719e-01
6.42308414e-01 -3.60138804e-01 5.18575072e-01 5.54798305e-01
9.14467335e-01 2.40541443e-01 1.40102029e-01 1.71105620e-02
7.39468277e-01 4.01659161e-01 4.56278414e-01 6.80697083e-01
-3.95260155e-01 7.65616298e-01 5.58139384e-01 1.13543784e-02
-1.02788854e+00 -1.05439627e+00 -3.42869461e-02 1.00348353e+00
-1.43089548e-01 -4.33264524e-01 -1.25753999e+00 -1.13271618e+00
-2.76577979e-01 1.19728732e+00 -8.11105430e-01 2.39725754e-01
-5.67304730e-01 -4.17608172e-01 5.15457928e-01 3.27319235e-01
5.97479999e-01 -1.19320738e+00 -4.66785043e-01 2.73339033e-01
-1.14678764e+00 -1.13100946e+00 -1.68971494e-01 -2.97758281e-01
-5.89010775e-01 -1.16891015e+00 -8.45346689e-01 -8.74133945e-01
3.77338648e-01 -1.83052376e-01 1.88633990e+00 -8.70253444e-02
2.65842885e-01 5.10814786e-01 -5.72621405e-01 -2.42045864e-01
-8.05848658e-01 4.20122221e-02 -2.65302658e-01 5.60014695e-02
4.25142497e-01 -1.97757691e-01 -3.78731728e-01 1.28860727e-01
-1.04115617e+00 -1.04406133e-01 3.02356482e-01 1.33381557e+00
6.08049154e-01 -3.43090653e-01 7.73940325e-01 -9.69792724e-01
9.75794435e-01 -7.34859347e-01 -4.99190241e-01 6.98861122e-01
-3.01384479e-01 3.33455503e-01 6.16358221e-01 -8.77396092e-02
-1.55493593e+00 -2.15686500e-01 -4.76936996e-01 -2.97726303e-01
-1.40220344e-01 5.87102830e-01 -2.28688106e-01 6.74067378e-01
9.84159410e-01 1.61894277e-01 -2.01199889e-01 -4.39783454e-01
7.84610569e-01 1.08279562e+00 4.83040005e-01 -7.18763471e-01
4.92242694e-01 1.31156757e-01 -5.68003535e-01 -6.74087226e-01
-1.24262178e+00 -3.99441004e-01 -6.12071157e-01 -6.50519505e-02
1.18099535e+00 -9.10047174e-01 -3.89433652e-01 1.82916149e-02
-1.58327425e+00 -3.75362039e-01 -4.99515980e-02 6.53509051e-02
-4.83022332e-01 4.97355103e-01 -6.40735865e-01 -9.10243928e-01
-7.31039941e-01 -1.16958046e+00 1.17832577e+00 2.68937081e-01
-3.26302111e-01 -1.14153028e+00 6.33326411e-01 1.25365841e+00
2.46112972e-01 1.34011060e-01 1.05989492e+00 -9.17035460e-01
-3.14425766e-01 3.79689112e-02 -4.98615094e-02 5.60117722e-01
-9.17068496e-02 -2.44180188e-01 -1.15312243e+00 -5.47520667e-02
1.63909152e-01 -1.17339158e+00 7.34067619e-01 9.39061120e-02
8.02250326e-01 -2.60541707e-01 -6.56624213e-02 -3.10944498e-01
1.21245801e+00 -1.43132538e-01 5.52017272e-01 -2.34543439e-02
5.95192015e-01 1.01295054e+00 6.82372630e-01 -1.42231630e-02
8.52337241e-01 4.78474528e-01 3.35308790e-01 4.76928711e-01
1.05534993e-01 -3.85384768e-01 2.79893577e-01 1.14046955e+00
3.22003067e-01 -5.28707087e-01 -1.18824494e+00 1.06628978e+00
-1.87982976e+00 -7.19279945e-01 -2.67284691e-01 1.94578969e+00
1.22315466e+00 -4.80554760e-01 -1.76001370e-01 -2.35538453e-01
6.34096980e-01 -3.26553956e-02 -2.92119265e-01 -1.24330387e-01
8.78486484e-02 3.65346134e-01 -4.22506630e-01 7.39094615e-01
-9.39862132e-01 8.22326362e-01 5.76045275e+00 8.18741798e-01
-2.92110175e-01 5.69340169e-01 7.11026847e-01 3.58415961e-01
-5.52863657e-01 -1.75574478e-02 -5.09559095e-01 4.18082982e-01
1.49451065e+00 1.30504593e-01 7.06189871e-02 9.44237649e-01
-2.77195871e-01 -1.53410897e-01 -1.12665391e+00 7.57383406e-01
2.53611296e-01 -1.26102507e+00 3.13325822e-01 -3.70543480e-01
9.77596343e-01 4.50090989e-02 -1.82307750e-01 9.17048752e-01
8.31474602e-01 -9.56693947e-01 4.68595147e-01 3.10071528e-01
4.34519678e-01 -4.50943351e-01 1.16611242e+00 7.53504276e-01
-7.47222781e-01 8.44308659e-02 -4.34926629e-01 7.16409758e-02
3.96053672e-01 4.13594395e-01 -9.79453444e-01 8.46505523e-01
8.37509692e-01 5.91346510e-02 -6.10143781e-01 4.39799815e-01
-5.10327280e-01 9.17890549e-01 -1.37463287e-01 -2.34473124e-01
3.46159756e-01 -3.06617945e-01 2.17085421e-01 8.48777294e-01
9.08544008e-03 7.88152516e-02 -1.91811845e-01 1.24219644e+00
-4.74888295e-01 2.98666984e-01 -3.10891539e-01 -1.30079865e-01
5.06151915e-01 1.19951546e+00 1.88570127e-01 -6.30178392e-01
-4.22142535e-01 1.00226068e+00 8.15287292e-01 2.90350586e-01
-5.16438961e-01 -3.13037276e-01 1.71602424e-02 -3.90229017e-01
1.66513786e-01 1.47528248e-02 6.32392522e-03 -1.37041533e+00
1.12423107e-01 -1.21055245e+00 8.31994474e-01 -1.04804611e+00
-1.60349929e+00 9.02215719e-01 1.12201516e-02 -9.63222742e-01
-8.86968136e-01 -4.67403233e-01 -4.53045607e-01 1.11986017e+00
-1.43459201e+00 -1.01097810e+00 -3.62482429e-01 7.33634830e-01
7.91866660e-01 -7.44623616e-02 8.82390320e-01 1.98116630e-01
-2.36426398e-01 5.41820824e-01 2.46082410e-01 4.14101303e-01
7.53329158e-01 -1.62285411e+00 4.47897494e-01 6.14302993e-01
5.56252360e-01 4.19606119e-01 7.79254735e-01 -5.58117688e-01
-1.07428420e+00 -8.42076480e-01 1.33037233e+00 -1.25627208e+00
3.80027592e-01 -3.56118143e-01 -1.24581754e+00 7.94390082e-01
8.88030469e-01 -4.53455478e-01 8.64603221e-01 -7.39964247e-02
-2.97236472e-01 4.59938407e-01 -1.16308415e+00 5.09781659e-01
2.62357354e-01 -6.13563359e-01 -1.31448817e+00 6.08338356e-01
1.07880354e+00 -5.41432738e-01 -9.40716147e-01 3.55317742e-01
4.69906740e-02 -8.37153554e-01 5.99235356e-01 -6.92006767e-01
8.40251327e-01 -2.77962148e-01 -2.66534954e-01 -1.47769797e+00
1.90660775e-01 -2.65661508e-01 -3.44421387e-01 1.27215409e+00
7.57838190e-01 -2.55860001e-01 6.14032090e-01 7.78616130e-01
6.34049550e-02 -5.96364737e-01 -1.05124366e+00 -3.96915734e-01
6.01646781e-01 -1.41203269e-01 5.14057100e-01 9.29634929e-01
-1.46737814e-01 9.99691784e-01 -2.65466183e-01 1.69910155e-02
4.84734833e-01 8.16044882e-02 9.09870207e-01 -1.11365676e+00
-4.53681797e-01 2.69520521e-01 2.93750376e-01 -1.08179891e+00
3.40236217e-01 -6.84299648e-01 3.30907613e-01 -1.94297171e+00
9.16487202e-02 -7.80029148e-02 1.52137756e-01 2.41364107e-01
-7.25034237e-01 -7.09338486e-02 -1.45529369e-02 1.67159930e-01
-8.17838252e-01 1.21316540e+00 1.14989758e+00 -3.31217855e-01
-1.23634681e-01 -3.68229002e-01 -4.09253091e-01 3.17793965e-01
5.70487559e-01 -4.98474747e-01 -5.42592645e-01 -8.23914945e-01
2.88551360e-01 3.72950166e-01 1.64950877e-01 -7.51425266e-01
1.02168910e-01 9.88556221e-02 7.36283213e-02 -6.79012120e-01
4.53187823e-01 -4.88886952e-01 -4.52706933e-01 8.53420608e-03
-3.59003693e-01 6.11967593e-02 -7.20084086e-02 6.79866493e-01
-6.02512538e-01 -7.35638440e-01 4.91692126e-01 -3.11937362e-01
-5.17242610e-01 -7.63350129e-02 -2.25383550e-01 8.08728456e-01
6.54784203e-01 5.31214178e-01 -5.08173168e-01 -8.62809241e-01
-5.48346579e-01 7.20755756e-01 1.01422213e-01 3.74487877e-01
4.05712038e-01 -1.18354094e+00 -1.09289956e+00 -2.94375569e-01
3.22428167e-01 3.10341388e-01 5.63965857e-01 3.55784863e-01
-6.19470298e-01 7.49783874e-01 1.39472365e-01 -7.02667117e-01
-9.22758639e-01 3.21230888e-01 4.82968569e-01 -8.58357191e-01
-9.41080973e-02 9.09923017e-01 9.30071026e-02 -1.19466782e+00
-2.88352277e-02 1.16248399e-01 -5.18328607e-01 1.44126564e-01
5.04761636e-01 1.99517012e-01 1.77645400e-01 -6.66379750e-01
5.43420911e-02 2.88511962e-02 -1.23646282e-01 -5.90845048e-01
9.85371351e-01 -1.19758956e-01 -1.08116500e-01 3.28082204e-01
1.05944502e+00 -2.53339708e-01 -1.07139802e+00 -5.23836970e-01
3.85635719e-02 -7.94665888e-02 -1.01052307e-01 -7.59132385e-01
-5.92493534e-01 1.23936546e+00 3.95766377e-01 2.38721892e-01
5.29565275e-01 1.93169773e-01 1.04000771e+00 8.03731143e-01
3.07812005e-01 -1.15977597e+00 3.43994230e-01 7.66227722e-01
9.77541566e-01 -1.59686780e+00 -5.82727373e-01 -1.55554518e-01
-1.15932345e+00 7.10662246e-01 8.98026347e-01 5.12321070e-02
4.17755723e-01 -6.44565225e-01 5.30176640e-01 -1.81164846e-01
-9.45699453e-01 -1.67617053e-01 4.13391501e-01 6.79729819e-01
5.55923164e-01 -5.55773973e-02 -2.33214229e-01 8.92148733e-01
-3.42892587e-01 -4.06999439e-01 4.66105819e-01 8.53771865e-01
-2.81228960e-01 -1.25137353e+00 -2.57080913e-01 5.72344601e-01
-4.11708176e-01 -2.25450188e-01 -5.54194152e-01 6.00033581e-01
-2.57028818e-01 1.52001607e+00 5.52955270e-02 -1.48106039e-01
7.08246008e-02 5.86073518e-01 2.88683653e-01 -6.16100490e-01
-6.04133546e-01 -4.97693241e-01 3.66507560e-01 -2.46833086e-01
-4.23760206e-01 -6.04530752e-01 -9.26172018e-01 2.47356728e-01
-4.57302392e-01 7.95704961e-01 3.71139258e-01 1.35962510e+00
4.46744651e-01 3.80252659e-01 2.30940878e-01 -3.06449085e-01
-9.58263040e-01 -1.43029368e+00 -1.24865860e-01 7.99221337e-01
1.89267024e-01 -6.42809331e-01 -3.82261187e-01 1.81776196e-01] | [11.370841026306152, 8.206477165222168] |
cfe9ce4d-f6c3-4320-936a-af6dfff1b84b | learning-srgb-to-raw-rgb-de-rendering-with-1 | 2206.01813 | null | https://arxiv.org/abs/2206.01813v1 | https://arxiv.org/pdf/2206.01813v1.pdf | Learning sRGB-to-Raw-RGB De-rendering with Content-Aware Metadata | Most camera images are rendered and saved in the standard RGB (sRGB) format by the camera's hardware. Due to the in-camera photo-finishing routines, nonlinear sRGB images are undesirable for computer vision tasks that assume a direct relationship between pixel values and scene radiance. For such applications, linear raw-RGB sensor images are preferred. Saving images in their raw-RGB format is still uncommon due to the large storage requirement and lack of support by many imaging applications. Several "raw reconstruction" methods have been proposed that utilize specialized metadata sampled from the raw-RGB image at capture time and embedded in the sRGB image. This metadata is used to parameterize a mapping function to de-render the sRGB image back to its original raw-RGB format when needed. Existing raw reconstruction methods rely on simple sampling strategies and global mapping to perform the de-rendering. This paper shows how to improve the de-rendering results by jointly learning sampling and reconstruction. Our experiments show that our learned sampling can adapt to the image content to produce better raw reconstructions than existing methods. We also describe an online fine-tuning strategy for the reconstruction network to improve results further. | ['Michael S. Brown', 'Marcus A. Brubaker', 'Abhijith Punnappurath', 'Seonghyeon Nam'] | 2022-06-03 | learning-srgb-to-raw-rgb-de-rendering-with | http://openaccess.thecvf.com//content/CVPR2022/html/Nam_Learning_sRGB-to-Raw-RGB_De-Rendering_With_Content-Aware_Metadata_CVPR_2022_paper.html | http://openaccess.thecvf.com//content/CVPR2022/papers/Nam_Learning_sRGB-to-Raw-RGB_De-Rendering_With_Content-Aware_Metadata_CVPR_2022_paper.pdf | cvpr-2022-1 | ['raw-reconstruction'] | ['computer-vision'] | [ 6.72561824e-01 -3.00460815e-01 -2.33667549e-02 -7.14603126e-01
-6.22578502e-01 -1.54436857e-01 1.95373476e-01 -6.50568306e-01
-5.24427593e-01 5.17962933e-01 -9.17655975e-02 -2.81733185e-01
2.59119153e-01 -9.61650729e-01 -9.15131032e-01 -8.58619809e-01
4.71340507e-01 8.17924645e-03 3.44176531e-01 4.60890755e-02
4.89677899e-02 6.93747163e-01 -1.50450075e+00 2.11499661e-01
5.60557067e-01 1.27001321e+00 8.43971848e-01 8.81418765e-01
-4.53321695e-01 7.97466576e-01 -4.33397144e-01 1.12157077e-01
7.63699174e-01 -5.95399380e-01 -3.84515464e-01 2.58367628e-01
5.19776225e-01 -8.43215346e-01 -5.19544303e-01 1.19474459e+00
3.74430776e-01 2.27541670e-01 -4.68527749e-02 -9.95673478e-01
-5.24287105e-01 8.64200294e-02 -5.79006255e-01 -2.56405294e-01
3.23140472e-01 1.85013264e-01 3.87267292e-01 -6.67690754e-01
3.90806496e-01 1.14859462e+00 8.07161391e-01 5.54557681e-01
-1.19967961e+00 -6.82726562e-01 -1.37216181e-01 2.31609493e-01
-1.38223517e+00 -5.00186384e-01 9.86784697e-01 7.79637769e-02
8.79051030e-01 5.20123065e-01 1.03607047e+00 7.76056707e-01
-9.31894556e-02 1.79307133e-01 1.37300217e+00 -5.65650403e-01
2.53154457e-01 2.21188515e-02 -3.94757867e-01 6.33326411e-01
1.96902845e-02 5.59576005e-02 -7.72212565e-01 2.42006689e-01
1.50419939e+00 2.34381914e-01 -5.84778011e-01 -3.05972219e-01
-1.19907010e+00 5.16754389e-01 8.08131635e-01 -1.50183126e-01
-2.91237026e-01 6.62303448e-01 -4.99010086e-02 3.01861852e-01
3.07757705e-01 2.38164186e-01 -4.02694702e-01 -1.74242899e-01
-1.00289345e+00 -3.24781895e-01 4.86553460e-01 1.01033998e+00
1.19671595e+00 7.53569975e-02 4.12640333e-01 8.96106064e-01
3.97285074e-01 8.56249273e-01 5.42519271e-01 -1.42550755e+00
5.12918949e-01 3.63774419e-01 1.11020450e-02 -9.82771099e-01
-2.25551397e-01 -3.83183546e-02 -1.04083204e+00 5.39536715e-01
1.81065187e-01 3.45308989e-01 -9.78511751e-01 1.23118532e+00
3.86138141e-01 2.68470764e-01 -8.29620138e-02 1.14136314e+00
8.27966332e-01 1.11428082e+00 -4.77106631e-01 -1.92410320e-01
1.04979265e+00 -8.38849962e-01 -6.77319646e-01 -3.45759898e-01
1.38388038e-01 -7.11696684e-01 1.82368672e+00 6.23017013e-01
-8.60177457e-01 -6.14215553e-01 -1.10316825e+00 -5.79025805e-01
-9.80203822e-02 2.87114441e-01 6.42826915e-01 7.26642013e-01
-1.17612708e+00 4.47767317e-01 -1.04899323e+00 -2.91837752e-01
2.99700439e-01 2.57722884e-01 -2.72123665e-01 -2.84105778e-01
-7.82000661e-01 6.77413285e-01 3.05237234e-01 3.42548877e-01
-5.72791278e-01 -5.58742940e-01 -4.94365543e-01 -1.92838773e-01
3.09113562e-01 -6.72943652e-01 1.00207031e+00 -1.27237916e+00
-2.16873002e+00 6.70418978e-01 7.96407610e-02 -1.13035522e-01
3.34213316e-01 9.51500461e-02 -6.48178682e-02 3.15612525e-01
-2.95159668e-01 7.02087820e-01 1.08583403e+00 -1.29231226e+00
-5.69640577e-01 -2.23912820e-01 1.12326197e-01 3.73377770e-01
-4.17814791e-01 -3.92441988e-01 -9.25377011e-01 -3.47411841e-01
5.99841058e-01 -6.80169523e-01 -2.55954444e-01 5.74622571e-01
-2.61895090e-01 4.67077404e-01 9.32483137e-01 -4.90388960e-01
8.03594410e-01 -2.07362318e+00 -1.17152922e-01 1.02783844e-01
-7.67289475e-02 2.54498702e-02 -1.06438525e-01 1.53529905e-02
5.76387607e-02 -3.13805401e-01 -3.59045602e-02 -6.10792756e-01
-4.73911226e-01 4.75340039e-01 -4.35716212e-01 6.21320128e-01
-3.24835688e-01 4.29496616e-01 -7.37947881e-01 -4.36614275e-01
7.24070787e-01 8.78156960e-01 -4.37534869e-01 5.98748803e-01
-1.12765148e-01 5.92311919e-01 -1.73254445e-01 5.96416175e-01
8.77746880e-01 -2.98049301e-01 1.29207790e-01 -8.49130452e-01
-2.13116467e-01 3.75465125e-01 -1.26156473e+00 1.93156719e+00
-8.85620773e-01 8.16445827e-01 3.33966911e-01 -7.10373700e-01
9.25260246e-01 -2.49807537e-01 4.88125116e-01 -8.60231638e-01
-2.12708190e-02 1.63323611e-01 -5.72503567e-01 -3.40567321e-01
7.27141917e-01 -1.09612919e-01 3.60393554e-01 2.49595821e-01
-4.13421363e-01 -7.26060331e-01 -3.13720256e-01 -9.13209245e-02
7.17723548e-01 5.69392085e-01 2.07348555e-01 6.77495897e-02
2.90421903e-01 1.45592868e-01 4.94976968e-01 6.95790946e-01
2.74724573e-01 8.94098341e-01 1.53761739e-02 -5.10048211e-01
-1.24568307e+00 -1.07026088e+00 -1.10222429e-01 8.05529952e-01
4.73423272e-01 -2.41017401e-01 -6.13616467e-01 3.34567614e-02
-3.96414876e-01 5.46211362e-01 -3.33455890e-01 1.89863462e-02
-6.42336547e-01 -5.16758680e-01 1.75898522e-01 3.54723334e-01
1.07642806e+00 -8.02291453e-01 -1.16174555e+00 1.39334043e-02
-4.25789766e-02 -1.11164188e+00 -3.30800503e-01 3.04559857e-01
-1.07406318e+00 -9.09831166e-01 -3.49880368e-01 -3.94701242e-01
8.41957867e-01 8.00750136e-01 7.42679119e-01 1.79874808e-01
-2.85300016e-01 4.67414588e-01 -3.15825939e-01 -8.38776603e-02
-3.17339212e-01 -1.05014361e-01 -2.40967155e-01 -4.25454713e-02
7.34409643e-03 -4.96477664e-01 -7.84160674e-01 1.99237406e-01
-1.29290104e+00 9.04236495e-01 4.35658723e-01 6.79994643e-01
1.12225258e+00 1.94540992e-01 -2.94657260e-01 -6.22079670e-01
6.96862787e-02 8.77859965e-02 -1.21738279e+00 1.85946390e-01
-6.38914466e-01 -5.34042343e-02 6.78621829e-01 -3.57192695e-01
-1.14415216e+00 4.30685878e-01 -1.33840203e-01 -7.45752275e-01
1.23642437e-01 1.19978160e-01 -3.49463165e-01 -3.21572393e-01
3.89863044e-01 4.16226238e-01 3.25685084e-01 -4.35673267e-01
3.05339217e-01 7.38377929e-01 8.34030926e-01 -2.95194387e-01
7.97094464e-01 8.42013597e-01 8.46397355e-02 -9.25522804e-01
-7.98406661e-01 -3.46200407e-01 -3.50786895e-01 -4.15801197e-01
8.67483079e-01 -9.53491092e-01 -5.65810502e-01 5.98407507e-01
-9.63566244e-01 -9.49044287e-01 -4.05772865e-01 4.49012399e-01
-6.66533351e-01 3.29888999e-01 -5.18467426e-01 -4.70682770e-01
-2.46918917e-01 -1.29643881e+00 1.21556950e+00 3.33736151e-01
5.90274870e-01 -7.64553845e-01 -8.80308896e-02 6.24542683e-02
7.58835375e-01 -1.14670642e-01 4.25264090e-01 8.25657606e-01
-1.07092547e+00 9.17400271e-02 -5.92796385e-01 4.95499641e-01
4.82885897e-01 3.58705036e-02 -1.01755309e+00 -1.70190707e-01
2.52388120e-01 -1.25335559e-01 6.25101328e-01 4.08680290e-01
1.76575375e+00 -1.14192851e-01 -1.63825024e-02 1.56447554e+00
2.02311707e+00 8.82352516e-03 9.18464124e-01 7.66071379e-01
1.04490006e+00 1.73240468e-01 6.02635801e-01 4.11405712e-01
4.84603614e-01 8.65302205e-01 8.08363497e-01 -1.86745510e-01
-3.85955095e-01 -3.48515928e-01 5.00462890e-01 8.34105074e-01
-2.12102439e-02 -1.70844849e-02 -5.91211617e-01 -1.29020780e-01
-1.51398635e+00 -7.57944524e-01 -2.33262822e-01 2.44328833e+00
1.16974902e+00 -2.94451565e-01 -3.19465369e-01 -2.56738991e-01
4.36235875e-01 2.89179325e-01 -6.96633935e-01 -1.40154049e-01
-2.43175343e-01 8.08795244e-02 1.22657108e+00 6.60325348e-01
-7.05609143e-01 9.09524798e-01 6.40225267e+00 6.72906280e-01
-1.63459337e+00 1.92426875e-01 6.39811158e-01 -2.86307395e-01
-4.11678433e-01 1.18012518e-01 -7.19155669e-01 5.33347189e-01
7.60452092e-01 3.66635263e-01 1.11138117e+00 8.20921540e-01
3.16611648e-01 -4.57005590e-01 -8.07084322e-01 1.70770311e+00
1.79110523e-02 -1.31125355e+00 -5.43735325e-02 5.14991619e-02
6.50653362e-01 1.19919084e-01 -6.42586276e-02 -2.71039784e-01
1.34603083e-01 -7.72975326e-01 8.22293282e-01 7.71560252e-01
1.37199819e+00 -3.74286652e-01 3.38798016e-01 2.42209271e-01
-1.14957523e+00 8.81518573e-02 -1.05691612e+00 5.40628172e-02
3.32654804e-01 8.31595302e-01 -7.43813097e-01 1.79788813e-01
1.04425991e+00 7.25391150e-01 -5.86053193e-01 6.68843865e-01
-2.68791914e-01 3.21998596e-01 -6.44338548e-01 2.10541323e-01
-9.39755291e-02 -7.21317887e-01 1.24396840e-02 7.31192172e-01
7.62345135e-01 1.83413401e-01 -1.51916007e-02 8.12228680e-01
-2.06944812e-02 -1.80185452e-01 -5.01683652e-01 2.23205775e-01
4.52674955e-01 1.26703882e+00 -6.25598192e-01 -4.48213965e-01
-5.42080581e-01 1.27639651e+00 1.32876914e-02 3.86810064e-01
-8.55092704e-01 -3.06266785e-01 4.87350345e-01 2.68013030e-01
2.49933749e-01 -4.48406100e-01 -3.39796215e-01 -1.22908795e+00
-9.57569852e-02 -6.62787199e-01 -1.40530735e-01 -1.46752989e+00
-9.48437035e-01 7.27001667e-01 3.98740619e-02 -1.40546381e+00
-6.38859719e-02 -7.89331973e-01 -1.32937640e-01 7.99867451e-01
-1.72124135e+00 -1.08183455e+00 -1.05575860e+00 9.99084592e-01
4.27553266e-01 1.57232776e-01 6.15692914e-01 3.67183447e-01
-3.88654798e-01 2.27461934e-01 2.97896504e-01 -2.32053220e-01
6.44008815e-01 -1.01314020e+00 6.01754487e-02 8.91452849e-01
-1.07840367e-01 4.32047725e-01 5.92625916e-01 -4.26101983e-01
-1.97213554e+00 -1.10402584e+00 9.40122351e-04 5.97989745e-02
2.35056996e-01 -3.30056518e-01 -7.49718130e-01 6.44878983e-01
9.79486331e-02 4.52154458e-01 2.21778065e-01 -5.75228751e-01
-2.30177417e-01 -9.21134949e-01 -1.16456246e+00 5.27563691e-01
9.94732499e-01 -6.78614974e-01 3.16652842e-02 2.73113847e-01
7.65824556e-01 -7.93391466e-01 -6.46666944e-01 -1.10574886e-01
6.30617797e-01 -1.28679574e+00 1.15649152e+00 3.30967069e-01
2.78874546e-01 -7.18724191e-01 -4.86263931e-01 -1.30988014e+00
4.08407561e-02 -4.46230233e-01 8.08075070e-02 7.68810451e-01
-1.47003233e-01 -7.93832839e-01 7.90435135e-01 6.54322207e-01
-7.12939426e-02 -3.52132767e-01 -1.04438972e+00 -4.79037017e-01
-6.44863009e-01 -6.34474695e-01 9.15567458e-01 6.42697811e-01
-7.54719615e-01 -6.43141493e-02 -5.13685763e-01 2.58668631e-01
8.57186317e-01 2.93721378e-01 1.12283981e+00 -7.52751470e-01
-5.23139715e-01 -1.89993829e-01 -2.82875597e-01 -1.54578078e+00
-2.87282109e-01 -5.19920886e-01 3.58165741e-01 -1.49286771e+00
-3.08001470e-02 -8.68782938e-01 7.41928518e-02 2.38682419e-01
1.46713287e-01 6.37846351e-01 2.60323226e-01 4.45344776e-01
-3.54519188e-01 3.22570831e-01 1.27616894e+00 8.13924000e-02
-2.81759262e-01 -2.99935192e-01 -2.93786705e-01 5.23387730e-01
7.87969351e-01 -3.13680619e-01 -6.24318779e-01 -9.17677879e-01
6.07352912e-01 6.86435997e-02 4.47108716e-01 -1.10839784e+00
2.36137316e-01 -4.35153961e-01 4.97968316e-01 -6.77174091e-01
7.47969508e-01 -1.20225990e+00 7.75101900e-01 1.39595866e-01
-1.20901637e-01 2.78939418e-02 -2.14764059e-01 4.45596546e-01
-2.93219555e-02 -2.63710141e-01 9.66481984e-01 -3.32077146e-01
-7.53797174e-01 1.80525318e-01 6.75050095e-02 -6.13808632e-01
7.90148139e-01 -6.67149961e-01 -1.24325782e-01 -5.80719233e-01
-2.21738890e-01 -3.65752131e-01 1.02620268e+00 -1.40440091e-01
9.28599834e-01 -1.43710196e+00 -7.85531327e-02 6.19656444e-01
-1.54211178e-01 5.80759346e-01 1.74079672e-01 7.76101470e-01
-1.33997774e+00 1.67028531e-02 -3.23515058e-01 -9.01464164e-01
-1.13716209e+00 3.39698166e-01 4.66077507e-01 1.25708461e-01
-1.00931728e+00 6.92605734e-01 -6.40288219e-02 -3.93339276e-01
2.25470692e-01 -7.27607727e-01 4.86520350e-01 -4.87015098e-01
6.42814755e-01 3.39892685e-01 8.08771923e-02 -4.55124170e-01
-3.65041867e-02 7.19404101e-01 5.16261697e-01 -1.56228885e-01
1.51689208e+00 -6.03133500e-01 -3.62789482e-01 4.57157642e-01
1.43881357e+00 -2.28331611e-02 -1.70898080e+00 -2.32319400e-01
-6.68171525e-01 -1.12646282e+00 6.70926750e-01 -3.57455224e-01
-1.64402688e+00 9.04090822e-01 8.87015522e-01 3.41574848e-02
1.64852202e+00 -1.84415102e-01 7.21344292e-01 5.76404870e-01
7.56924212e-01 -1.19265103e+00 -6.12291433e-02 8.11191946e-02
7.57688403e-01 -1.14363468e+00 3.28721642e-01 -3.10939252e-01
-2.74644524e-01 1.57175815e+00 5.21949291e-01 -7.18828142e-02
5.07643878e-01 3.93918961e-01 2.84739524e-01 9.60877016e-02
-3.51392090e-01 1.71095520e-01 -1.55590385e-01 6.08944178e-01
9.83197987e-02 1.64308697e-02 2.73353785e-01 -1.94762886e-01
-1.92943394e-01 1.38506368e-01 7.32244432e-01 6.93524003e-01
-4.53369617e-01 -1.12989724e+00 -8.38941336e-01 4.70768273e-01
2.27631908e-02 -5.39055765e-02 1.23384833e-01 6.27475262e-01
-5.55419438e-02 5.48928499e-01 3.32320273e-01 -1.88325390e-01
1.38979837e-01 -5.11794329e-01 7.36273825e-01 -4.21477258e-01
-1.04779772e-01 2.60563225e-01 -3.39211941e-01 -1.19539368e+00
-7.75079310e-01 -3.70554268e-01 -1.17736828e+00 -3.50950480e-01
-3.34622979e-01 -4.08885360e-01 1.28105962e+00 4.18425232e-01
1.68030560e-01 4.13639516e-01 7.68162489e-01 -1.36658561e+00
-4.06370848e-01 -7.09677935e-01 -6.63652480e-01 2.18327567e-01
4.55307722e-01 -3.79713714e-01 -4.16086197e-01 3.08233351e-01] | [10.24644947052002, -2.531107187271118] |
b46fed99-ed98-4b6a-8f44-ea6008f95def | leveraging-structured-biological-knowledge | 2101.05136 | null | https://arxiv.org/abs/2101.05136v1 | https://arxiv.org/pdf/2101.05136v1.pdf | Leveraging Structured Biological Knowledge for Counterfactual Inference: a Case Study of Viral Pathogenesis | Counterfactual inference is a useful tool for comparing outcomes of interventions on complex systems. It requires us to represent the system in form of a structural causal model, complete with a causal diagram, probabilistic assumptions on exogenous variables, and functional assignments. Specifying such models can be extremely difficult in practice. The process requires substantial domain expertise, and does not scale easily to large systems, multiple systems, or novel system modifications. At the same time, many application domains, such as molecular biology, are rich in structured causal knowledge that is qualitative in nature. This manuscript proposes a general approach for querying a causal biological knowledge graph, and converting the qualitative result into a quantitative structural causal model that can learn from data to answer the question. We demonstrate the feasibility, accuracy and versatility of this approach using two case studies in systems biology. The first demonstrates the appropriateness of the underlying assumptions and the accuracy of the results. The second demonstrates the versatility of the approach by querying a knowledge base for the molecular determinants of a severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)-induced cytokine storm, and performing counterfactual inference to estimate the causal effect of medical countermeasures for severely ill patients. | ['Olga Vitek', 'Robert Ness', 'Kristie Oxford', 'Charles Tapley Hoyt', 'Jeremy Teuton', 'Craig Bakker', 'Pallavi Kolambkar', 'Somya Bhargava', 'Sara Mohammad-Taheri', 'Kaushal Paneri', 'Jeremy Zucker'] | 2021-01-13 | null | null | null | null | ['counterfactual-inference'] | ['miscellaneous'] | [ 5.48224151e-01 1.82782754e-01 -4.14695770e-01 -4.06164825e-02
-3.16799492e-01 -7.83568323e-01 6.16403937e-01 5.14089048e-01
8.50958154e-02 1.29226649e+00 1.89428404e-01 -1.05399156e+00
-6.35054588e-01 -8.54589522e-01 -1.06379771e+00 -5.57190597e-01
-4.68935102e-01 5.59278071e-01 -7.31780976e-02 6.88872486e-02
2.45766923e-01 7.11237609e-01 -1.30944860e+00 3.63779038e-01
7.51468539e-01 4.54295427e-01 -1.76966935e-03 7.73189902e-01
-9.71092563e-03 6.91375852e-01 -4.95100617e-01 -1.44830689e-01
-1.77083850e-01 -5.38513422e-01 -6.16952837e-01 -5.45340538e-01
-1.15434960e-01 -4.47368473e-02 6.56095147e-02 4.82364476e-01
3.89192909e-01 -3.91202360e-01 8.16579998e-01 -1.61666024e+00
-2.29933590e-01 4.23048019e-01 -3.14388096e-01 1.69876769e-01
5.06367207e-01 3.28366220e-01 8.21994483e-01 -2.66469389e-01
7.39574790e-01 1.60504127e+00 7.45746434e-01 3.36168855e-01
-1.79797578e+00 -6.05956435e-01 1.20491117e-01 -1.92742318e-01
-1.07381797e+00 -2.26978809e-01 2.42344469e-01 -9.36474383e-01
8.51961851e-01 5.04488766e-01 8.61410499e-01 8.15735042e-01
6.94939911e-01 6.03942052e-02 1.31342208e+00 -2.18606278e-01
5.79702437e-01 -1.50909692e-01 -2.32971892e-01 6.28628373e-01
7.42071152e-01 6.48473561e-01 -4.50395375e-01 -9.59219992e-01
7.23469853e-01 3.28715920e-01 -2.64673471e-01 -5.07011533e-01
-1.14555705e+00 9.38959837e-01 2.66651902e-02 5.77420890e-02
-4.60183203e-01 6.19525552e-01 2.85582900e-01 8.43887255e-02
1.77881464e-01 6.20799422e-01 -6.68539107e-01 2.97997534e-01
-7.08618283e-01 4.56830323e-01 1.03394985e+00 5.99722385e-01
3.34861785e-01 -2.32466355e-01 -9.27267820e-02 -1.27282692e-02
4.45754945e-01 6.32052243e-01 -1.59806222e-01 -9.39556897e-01
-1.84679523e-01 5.24023235e-01 5.29623687e-01 -8.06149483e-01
-4.18878555e-01 -4.37273504e-03 -6.25025034e-01 2.46804520e-01
4.92477238e-01 -3.78833234e-01 -7.21797347e-01 1.91631985e+00
5.82301497e-01 2.49927700e-01 -6.09887429e-02 4.98205394e-01
3.89619410e-01 5.66299200e-01 5.39389849e-01 -8.19733143e-01
1.36020446e+00 1.70906544e-01 -7.35689342e-01 2.85682738e-01
5.58189571e-01 -5.49200356e-01 7.02464998e-01 1.70712546e-01
-8.75076950e-01 1.54703125e-01 -7.87723005e-01 4.20601189e-01
-4.08498943e-01 -4.73405629e-01 7.70514250e-01 7.30262458e-01
-9.41281199e-01 5.66596508e-01 -7.44164824e-01 -3.97845268e-01
2.96389729e-01 4.20324057e-01 -1.30714476e-01 -1.72217116e-02
-1.47561860e+00 9.40837562e-01 3.17151904e-01 -2.66583741e-01
-9.97850478e-01 -1.43605459e+00 -5.71680903e-01 1.76951349e-01
3.05580348e-01 -1.36838412e+00 9.78600621e-01 -3.89733046e-01
-1.08780575e+00 2.92808622e-01 -2.40932569e-01 -3.25296968e-01
3.21315348e-01 4.75025147e-01 -1.91844419e-01 -6.57994822e-02
9.98996869e-02 2.20540091e-01 2.39988402e-01 -1.29459429e+00
-4.77036655e-01 -5.40334105e-01 1.56981468e-01 -1.74761161e-01
4.49347794e-01 1.18945092e-01 3.13094199e-01 -5.26893437e-01
-3.36569428e-01 -8.15355480e-01 -5.43597043e-01 -8.01375434e-02
-3.62231046e-01 -4.67746854e-02 5.89924812e-01 -3.80658537e-01
1.24895549e+00 -1.80387914e+00 -2.35613212e-01 3.51583213e-01
1.24639101e-01 -2.43751273e-01 7.20167533e-02 9.73969579e-01
-3.82503867e-01 7.39741266e-01 -4.26430911e-01 7.40946770e-01
-1.65005818e-01 2.25082003e-02 -5.20662129e-01 3.87932926e-01
3.64184380e-01 1.00675857e+00 -1.05475068e+00 -4.74303901e-01
1.47514641e-01 3.52149427e-01 -5.84856749e-01 2.49204546e-01
-6.10696018e-01 4.15011495e-01 -5.78918874e-01 1.98721260e-01
3.53823960e-01 -4.67953950e-01 9.74388897e-01 -1.09062076e-01
-2.55681008e-01 2.47184828e-01 -1.12567592e+00 9.65678751e-01
-3.11321080e-01 6.31253645e-02 -5.72692901e-02 -8.55573773e-01
3.91993910e-01 6.42213881e-01 7.52818525e-01 -1.79190621e-01
-1.00731831e-02 -1.00390306e-02 1.93192530e-02 -6.31015658e-01
-4.28454041e-01 -8.07030320e-01 -4.56204116e-02 7.56279349e-01
-2.60675460e-01 -3.42512429e-01 1.70936763e-01 2.06213325e-01
1.30101180e+00 -1.20823517e-01 7.95646608e-01 -5.03429770e-01
1.03096198e-02 3.92994046e-01 5.69056988e-01 6.72600448e-01
1.72467142e-01 7.07012564e-02 1.13441193e+00 -4.46784914e-01
-9.92061734e-01 -1.14424324e+00 -1.28003970e-01 4.57591206e-01
-1.84380874e-01 -7.36185238e-02 -5.38484573e-01 -4.00731832e-01
4.98185098e-01 8.93905520e-01 -9.68026876e-01 -2.08606869e-01
-2.90870965e-01 -9.95066285e-01 5.14790237e-01 2.92285800e-01
-2.75174916e-01 -5.69829762e-01 -7.81224608e-01 2.31779709e-01
-1.25544325e-01 -5.58280170e-01 -1.72723845e-01 1.15113622e-02
-9.63247418e-01 -1.74481297e+00 -1.52735040e-01 -2.05028981e-01
5.79846501e-01 -1.11743342e-02 1.03905046e+00 7.07891863e-03
-5.36925375e-01 2.77095169e-01 3.39654386e-01 -7.99473345e-01
-6.91856921e-01 -8.63511026e-01 1.60804003e-01 -4.17169392e-01
-7.04915747e-02 -4.90338147e-01 -6.82775199e-01 1.75058559e-01
-1.02039564e+00 7.16591254e-04 4.57949907e-01 1.01460242e+00
6.02078378e-01 1.97315872e-01 9.58107352e-01 -1.23431194e+00
8.34859550e-01 -7.80833125e-01 -8.56988072e-01 4.85482752e-01
-8.50001156e-01 5.10963611e-02 5.93541920e-01 -2.93148071e-01
-1.13906515e+00 1.89368278e-01 3.60310733e-01 -5.46959080e-02
-9.84994099e-02 9.82831240e-01 -2.23254040e-01 4.85235065e-01
7.90300190e-01 -3.02716851e-01 3.37617218e-01 -1.55323833e-01
5.65144897e-01 3.01520586e-01 6.76989630e-02 -6.33789062e-01
2.96920270e-01 7.32408941e-01 5.20502687e-01 -3.75458658e-01
-3.26152623e-01 -6.08776920e-02 -1.96864992e-01 3.86661477e-02
7.28180170e-01 -6.51671708e-01 -1.37965763e+00 -1.16591223e-01
-1.16343260e+00 -4.10718918e-01 -2.85776734e-01 5.61806500e-01
-6.85141385e-01 -1.42355204e-01 -2.70854592e-01 -9.57649171e-01
9.50006172e-02 -9.31025445e-01 8.15469265e-01 -3.22873712e-01
-4.88703430e-01 -1.17913055e+00 4.93190616e-01 4.52301465e-02
2.45194227e-01 7.88673699e-01 1.54779887e+00 -4.31577116e-01
-5.80041885e-01 -1.72101751e-01 -2.30934381e-01 -4.50017482e-01
3.64709258e-01 4.18314219e-01 -6.13120079e-01 -7.00338706e-02
-1.43230245e-01 -6.28922284e-02 4.17262733e-01 8.16940606e-01
9.05504525e-01 -7.71030366e-01 -8.01460505e-01 -1.62692577e-01
1.38740695e+00 4.59858567e-01 4.23640460e-01 -3.67427647e-01
2.60548890e-01 9.51461971e-01 5.82696378e-01 5.00462770e-01
2.47561827e-01 5.33504605e-01 3.55758250e-01 -1.91817507e-01
2.93173432e-01 -3.51529151e-01 2.25172974e-02 -6.33997023e-02
1.60900488e-01 -2.75979459e-01 -1.02269173e+00 5.31059861e-01
-1.80926859e+00 -1.10038888e+00 -3.45526487e-01 2.30742884e+00
1.16826534e+00 -2.78371722e-01 2.10241079e-01 -2.37394437e-01
7.50783861e-01 -4.88687128e-01 -6.84703767e-01 -5.80627918e-01
1.08942591e-01 1.13377094e-01 5.31273425e-01 5.52935183e-01
-5.48422635e-01 2.93261170e-01 7.78539133e+00 4.87841725e-01
-8.92361581e-01 -1.06936902e-01 7.88769126e-01 -8.12377483e-02
-8.09720874e-01 5.45337975e-01 -2.80977428e-01 3.46450567e-01
1.37127423e+00 -6.67616785e-01 3.58038813e-01 1.32358834e-01
9.95663881e-01 -2.93766648e-01 -1.33291197e+00 2.16169551e-01
-5.37303627e-01 -1.88271832e+00 2.35973462e-01 6.63363636e-02
6.51999950e-01 -3.61857325e-01 -2.31567696e-01 -3.05881023e-01
9.48124588e-01 -1.24107945e+00 3.12661260e-01 7.92747080e-01
9.74078536e-01 -6.08045161e-01 5.34731925e-01 2.63545185e-01
-6.19232893e-01 -1.04636019e-02 1.22016832e-01 -2.12998733e-01
2.60300070e-01 9.16903079e-01 -1.21824193e+00 4.13584590e-01
5.54977834e-01 1.33170158e-01 2.31149763e-01 8.99963260e-01
-9.93572921e-02 7.85900712e-01 -3.25423121e-01 -1.40114620e-01
-3.81351322e-01 1.05383217e-01 3.62966269e-01 1.09104800e+00
1.40901178e-01 4.63883191e-01 -2.30146050e-01 1.01900220e+00
2.41007179e-01 -5.56143150e-02 -1.06217122e+00 -4.81790960e-01
5.56537271e-01 6.13541543e-01 -5.20656586e-01 -4.45134640e-01
-9.62876156e-02 5.11721820e-02 -1.28992245e-01 5.25664806e-01
-7.04412818e-01 3.36484648e-02 6.05267167e-01 3.76282305e-01
2.99618039e-02 1.86328918e-01 -3.77959609e-01 -6.50690377e-01
-4.67862397e-01 -1.09827828e+00 7.91626155e-01 -7.96682000e-01
-1.19515610e+00 -3.19984078e-01 4.80576396e-01 -7.11163521e-01
-3.42337638e-01 -4.03279305e-01 -3.69190931e-01 1.12743211e+00
-1.02123857e+00 -8.80558908e-01 2.64337897e-01 2.82472104e-01
-1.76800832e-01 4.46721375e-01 9.14139926e-01 -1.19227543e-01
-4.61530179e-01 -4.74690050e-02 1.57775804e-01 -6.15231454e-01
5.43817163e-01 -1.17895412e+00 1.48814797e-01 4.39633757e-01
-6.33956313e-01 1.15428376e+00 1.17098916e+00 -1.17240536e+00
-1.51336122e+00 -1.13077402e+00 9.88045096e-01 -6.02486372e-01
1.00351846e+00 -2.31047675e-01 -6.18884742e-01 5.47005832e-01
8.02024752e-02 -3.13243687e-01 9.62490141e-01 1.56917468e-01
-3.49826604e-01 1.29190758e-01 -1.40860415e+00 8.80141854e-01
6.44958913e-01 -1.76320970e-01 -5.14659882e-01 4.89485860e-01
9.56411123e-01 3.05773243e-02 -1.13560009e+00 5.08139133e-01
7.84402072e-01 -6.14019454e-01 9.03412759e-01 -1.25549328e+00
3.53047550e-01 -6.99515700e-01 -1.52382404e-02 -1.17005336e+00
-2.49152124e-01 -7.72016346e-01 1.00526221e-01 7.53839731e-01
6.97177410e-01 -8.30777228e-01 3.81950021e-01 8.21859717e-01
4.24714446e-01 -7.77974367e-01 -8.56434226e-01 -6.22820318e-01
7.84742907e-02 -1.56364664e-01 7.92340100e-01 1.13278687e+00
2.67735690e-01 3.51862043e-01 -5.05242720e-02 3.71036381e-01
5.44401169e-01 3.89999330e-01 6.23809934e-01 -1.27511239e+00
-3.75899434e-01 -2.37449110e-01 -2.24280298e-01 -4.12916578e-02
-1.61787197e-01 -5.13251841e-01 -2.69129544e-01 -1.69682372e+00
5.65325022e-01 -5.65543294e-01 -1.64202616e-01 4.40722495e-01
-2.51637578e-01 -3.29681247e-01 -2.81107247e-01 2.66451146e-02
-1.04541704e-02 1.78390920e-01 9.59402502e-01 -7.50445127e-02
-1.69322789e-01 -1.48786977e-01 -8.15478981e-01 5.41977227e-01
7.63069510e-01 -7.41533875e-01 -7.30499029e-01 3.56648684e-01
6.04298651e-01 7.19286978e-01 8.61794114e-01 1.53486291e-02
3.94891538e-02 -9.41108823e-01 1.66554257e-01 -3.46846372e-01
-1.69614986e-01 -8.29779863e-01 1.12888098e+00 1.08860099e+00
-3.84453624e-01 8.72124732e-02 5.01851678e-01 9.49024379e-01
7.35136569e-02 1.20461188e-01 5.31867266e-01 -1.16235882e-01
-7.67198801e-02 5.50902784e-02 -5.67110181e-01 5.74913248e-02
1.11311913e+00 -7.13511482e-02 -5.59355915e-01 -2.90912867e-01
-7.19170988e-01 2.82112926e-01 5.07147372e-01 8.55589435e-02
4.22114164e-01 -8.96633267e-01 -7.30686188e-01 -2.15154216e-01
4.60310467e-02 -4.71000075e-01 2.65363157e-01 1.01502514e+00
-3.60592932e-01 7.06755221e-01 -6.19889330e-03 -4.43687916e-01
-1.06575370e+00 1.09255707e+00 3.69290054e-01 -2.31328905e-01
5.98494411e-02 1.11302719e-01 7.25346446e-01 -3.17079842e-01
-5.17575264e-01 -1.17180459e-01 5.83206043e-02 -5.03578112e-02
3.23311657e-01 4.08332795e-01 -1.93506211e-01 8.03713128e-02
-5.64632833e-01 -1.00627821e-03 3.63149226e-01 -1.33760244e-01
1.26089895e+00 -9.80382040e-02 -5.28671443e-01 5.60609877e-01
7.24754393e-01 7.18644932e-02 -1.05766535e+00 3.27934682e-01
-2.98199039e-02 -1.87874049e-01 -2.95986414e-01 -1.19956577e+00
-4.72117901e-01 5.98840058e-01 4.65725631e-01 3.71322513e-01
9.09895718e-01 9.19296443e-02 -3.70141082e-02 7.69905522e-02
2.98898995e-01 -5.19560516e-01 -4.08522427e-01 -2.23417804e-01
1.01644051e+00 -7.48850107e-01 2.44840086e-01 -6.56974971e-01
-2.72733241e-01 7.51161695e-01 8.55645537e-02 2.53773600e-01
7.78680325e-01 5.31014144e-01 -1.49638638e-01 -5.33842802e-01
-1.39369309e+00 2.00181544e-01 -3.93845141e-03 5.50987422e-01
6.53804898e-01 4.15181488e-01 -7.21591413e-01 2.95917422e-01
1.49691537e-01 3.84143680e-01 6.08718157e-01 9.35988367e-01
-1.33520678e-01 -1.02605391e+00 -4.31794554e-01 7.52499938e-01
-5.01424432e-01 -2.59024501e-01 -6.27959609e-01 1.04710960e+00
-3.11151557e-02 1.12861216e+00 -2.17236921e-01 6.93212971e-02
4.63910490e-01 3.30147557e-02 4.14170116e-01 -5.13328254e-01
-3.13641429e-01 1.11121181e-02 2.70655990e-01 -5.80191612e-01
-4.35401320e-01 -9.27994132e-01 -1.34096515e+00 -5.22702217e-01
-3.54440421e-01 3.16560864e-01 5.52061975e-01 7.80098796e-01
8.10342014e-01 5.59376478e-01 5.00565469e-01 -7.08299875e-02
-2.96688288e-01 -4.93394494e-01 -3.48389745e-01 1.88072830e-01
3.50588202e-01 -8.15752447e-01 -3.51211429e-01 3.51166308e-01] | [7.938242435455322, 5.393410682678223] |
e3ad2fd3-c942-49e1-ad9d-3879c5226c8b | a-compositional-feature-embedding-and | 2109.12380 | null | https://arxiv.org/abs/2109.12380v3 | https://arxiv.org/pdf/2109.12380v3.pdf | A Compositional Feature Embedding and Similarity Metric for Ultra-Fine-Grained Visual Categorization | Fine-grained visual categorization (FGVC), which aims at classifying objects with small inter-class variances, has been significantly advanced in recent years. However, ultra-fine-grained visual categorization (ultra-FGVC), which targets at identifying subclasses with extremely similar patterns, has not received much attention. In ultra-FGVC datasets, the samples per category are always scarce as the granularity moves down, which will lead to overfitting problems. Moreover, the difference among different categories is too subtle to distinguish even for professional experts. Motivated by these issues, this paper proposes a novel compositional feature embedding and similarity metric (CECS). Specifically, in the compositional feature embedding module, we randomly select patches in the original input image, and these patches are then replaced by patches from the images of different categories or masked out. Then the replaced and masked images are used to augment the original input images, which can provide more diverse samples and thus largely alleviate overfitting problem resulted from limited training samples. Besides, learning with diverse samples forces the model to learn not only the most discriminative features but also other informative features in remaining regions, enhancing the generalization and robustness of the model. In the compositional similarity metric module, a new similarity metric is developed to improve the classification performance by narrowing the intra-category distance and enlarging the inter-category distance. Experimental results on two ultra-FGVC datasets and one FGVC dataset with recent benchmark methods consistently demonstrate that the proposed CECS method achieves the state of-the-art performance. | ['Yongsheng Gao', 'Yi Liao', 'Xiaohan Yu', 'Miaohua Zhang', 'Yajie Sun'] | 2021-09-25 | null | null | null | null | ['fine-grained-visual-categorization'] | ['computer-vision'] | [ 1.22049026e-01 -4.39560741e-01 -9.99166965e-02 -4.27839369e-01
-3.16630483e-01 -4.10183817e-01 5.22463977e-01 1.71959028e-01
-2.40330920e-01 4.65656757e-01 8.65052342e-02 2.09677145e-01
-2.20526829e-01 -8.21334243e-01 -1.88557699e-01 -9.43753719e-01
3.35880518e-01 -1.64041027e-01 5.68619668e-01 4.39327061e-02
3.02751809e-01 2.93390632e-01 -1.85140681e+00 3.39801341e-01
1.00642991e+00 1.45316184e+00 2.81361252e-01 -5.24454042e-02
-4.09548849e-01 4.89157557e-01 -5.14838278e-01 -1.05918080e-01
1.70784190e-01 -3.53563488e-01 -4.63689923e-01 3.30432534e-01
4.08947796e-01 -8.61752182e-02 -2.63110816e-01 1.16018963e+00
3.01561028e-01 1.30878344e-01 7.32660174e-01 -1.27899802e+00
-1.11047602e+00 2.50039965e-01 -7.46562600e-01 1.63513556e-01
-5.71226962e-02 1.21841304e-01 9.01868701e-01 -1.08762813e+00
3.26764703e-01 1.37068427e+00 5.56264699e-01 4.45061952e-01
-1.18258238e+00 -7.85151482e-01 5.91541648e-01 4.98741865e-01
-1.81938374e+00 -1.17970658e-02 8.93784225e-01 -5.59715509e-01
3.11447948e-01 4.31389481e-01 4.62069422e-01 7.74925351e-01
-7.43789151e-02 5.07980227e-01 1.19792449e+00 -2.62210488e-01
1.73922077e-01 4.19593006e-01 2.89194345e-01 4.27687168e-01
3.51773471e-01 -3.59467305e-02 5.09049278e-03 9.26993638e-02
5.01961291e-01 4.70938712e-01 -4.97753590e-01 -5.48028469e-01
-1.15226519e+00 8.85718465e-01 8.69365633e-01 5.85409999e-01
-1.86788887e-01 -3.89004499e-01 5.04638791e-01 1.66482851e-01
2.84471273e-01 1.06416695e-01 -2.46143237e-01 2.47583553e-01
-6.63246989e-01 -7.05290958e-02 1.35571510e-01 8.02645385e-01
1.00355971e+00 -5.16534373e-02 -4.39609259e-01 1.21077847e+00
1.73351020e-01 2.73010641e-01 8.93033445e-01 -4.43442523e-01
4.02421743e-01 1.09642637e+00 -1.77626595e-01 -1.55763853e+00
-1.78212270e-01 -7.37649858e-01 -1.21337914e+00 3.68544348e-02
2.06320658e-01 4.11527544e-01 -8.36421311e-01 1.55539024e+00
4.32350665e-01 4.56923284e-02 -2.44454980e-01 1.15673280e+00
9.20316160e-01 7.21896887e-01 1.26955658e-01 -2.12796092e-01
1.32346082e+00 -9.60902929e-01 -4.14125919e-01 -2.96143413e-01
2.95465559e-01 -6.78527534e-01 1.47449887e+00 9.57167149e-02
-3.56974781e-01 -1.12280488e+00 -1.19831157e+00 1.17411219e-01
-5.38672507e-01 2.14125872e-01 4.62989092e-01 4.59837794e-01
-5.15436530e-01 4.21955347e-01 -3.71475011e-01 -8.53079408e-02
6.20224655e-01 3.88519689e-02 -4.66830641e-01 -4.11486089e-01
-1.04702425e+00 3.94203722e-01 5.93298733e-01 -2.77844444e-03
-4.69363302e-01 -6.26533747e-01 -7.11046159e-01 2.19708666e-01
3.13701332e-01 -1.78251028e-01 5.83977759e-01 -1.07204425e+00
-1.02878845e+00 6.74469233e-01 1.10951819e-01 6.11153953e-02
4.30939406e-01 4.04539734e-01 -6.69967532e-01 8.86205435e-02
3.03226799e-01 5.39770246e-01 1.03576112e+00 -1.19940209e+00
-8.64601195e-01 -5.32936275e-01 -1.31269738e-01 2.39852313e-02
-6.64576828e-01 -3.62896502e-01 -4.41306204e-01 -9.83308971e-01
2.91302741e-01 -6.34642541e-01 7.15955719e-02 2.13988125e-01
-2.18986437e-01 -4.79706317e-01 1.10002303e+00 -3.72498214e-01
1.32017589e+00 -2.53893852e+00 9.62808505e-02 2.65141964e-01
2.35294133e-01 2.97129750e-01 -2.28943691e-01 1.04724228e-01
-4.31355909e-02 7.77331144e-02 -2.06306428e-01 1.26151949e-01
-5.25516458e-02 6.95160776e-02 -1.42890900e-01 3.94246906e-01
3.16939920e-01 6.48080468e-01 -7.18824387e-01 -5.89346111e-01
3.27127844e-01 2.71380633e-01 -3.57034385e-01 1.27623662e-01
1.53034300e-01 2.23386556e-01 -6.27808034e-01 7.12909400e-01
1.01348960e+00 -3.36468279e-01 -8.20996389e-02 -4.76490498e-01
-7.21189603e-02 -4.52910095e-01 -1.40678906e+00 1.23539460e+00
-3.96983802e-01 2.55658776e-01 -2.73631871e-01 -1.22288918e+00
1.14776492e+00 -1.71327353e-01 2.01541245e-01 -8.24341476e-01
1.82111278e-01 1.58995837e-01 6.12410009e-02 -3.35731149e-01
1.70043230e-01 -1.52950868e-01 -3.90891768e-02 1.39006779e-01
-7.40312561e-02 2.03932241e-01 1.46568686e-01 -1.04133964e-01
5.34406066e-01 -2.64412940e-01 4.30332422e-01 -3.53887528e-01
9.84643221e-01 -1.35651842e-01 8.02890837e-01 3.46258640e-01
-4.27328557e-01 5.60393810e-01 1.44521501e-02 -3.90609950e-01
-7.76340723e-01 -1.15342224e+00 -4.80568111e-01 9.68674719e-01
8.02287400e-01 -3.22330415e-01 -5.50742805e-01 -9.42714036e-01
2.87786931e-01 2.62901336e-01 -8.87259603e-01 -5.11056662e-01
-2.36260176e-01 -6.80805504e-01 1.76387310e-01 6.46475315e-01
7.60520995e-01 -8.96991432e-01 -2.14057460e-01 1.52922168e-01
-1.98783159e-01 -7.41677642e-01 -6.95928574e-01 -1.69844180e-01
-6.82055593e-01 -1.13971758e+00 -7.83234596e-01 -1.07242191e+00
8.02967250e-01 7.70431101e-01 6.16316319e-01 1.94108188e-01
-4.74574178e-01 -1.34910092e-01 -6.34447455e-01 -1.38614867e-02
-6.25958964e-02 -2.30074003e-01 3.45099270e-02 6.06732607e-01
5.64391017e-01 -3.36677730e-01 -7.79430330e-01 6.74961627e-01
-8.32311809e-01 -1.06434099e-01 6.97934687e-01 1.22691989e+00
8.73644173e-01 3.59160393e-01 7.96069205e-01 -5.39964795e-01
3.38266075e-01 -3.98067385e-01 -3.87199908e-01 2.05935389e-01
-6.87745094e-01 -1.88100129e-01 1.27251840e+00 -7.76065290e-01
-9.10573542e-01 -3.95472467e-01 1.51513055e-01 -5.75657248e-01
-2.03070760e-01 2.09484726e-01 -4.30735558e-01 -2.96395868e-01
5.32157302e-01 4.83607113e-01 -6.85375035e-02 -6.20421290e-01
2.75235623e-01 1.03613460e+00 4.11128730e-01 -2.19036207e-01
9.62123036e-01 3.77004445e-01 -2.38492101e-01 -7.29429305e-01
-6.79746389e-01 -5.25372028e-01 -5.32828093e-01 1.02316402e-02
6.29393816e-01 -8.44538987e-01 -4.88044083e-01 4.08710063e-01
-6.18258059e-01 2.78162181e-01 -2.75348365e-01 4.64524537e-01
-1.01819009e-01 5.57554603e-01 -3.42684656e-01 -4.53316957e-01
-1.95986181e-01 -1.11299384e+00 9.53676581e-01 5.98200858e-01
5.50718457e-02 -6.44320667e-01 -4.08632636e-01 1.85687259e-01
4.23526913e-01 8.96933228e-02 1.20146024e+00 -5.31149566e-01
-4.77251768e-01 -3.34124207e-01 -6.65123284e-01 7.34357119e-01
6.27611220e-01 -1.18574373e-01 -8.75676930e-01 -4.89816606e-01
-8.43588933e-02 -3.22360992e-01 8.63756716e-01 6.75712153e-02
1.68495667e+00 -2.86385715e-01 -4.50321525e-01 6.32226467e-01
1.54117835e+00 3.34804237e-01 3.41942430e-01 4.53877181e-01
8.27308834e-01 6.14363909e-01 9.43698943e-01 3.27086717e-01
2.34045461e-01 6.33367598e-01 2.52990514e-01 7.52903596e-02
-3.37556273e-01 -2.71728247e-01 -6.20165840e-02 8.31266820e-01
6.33802786e-02 1.74810916e-01 -4.37156707e-01 5.07709563e-01
-1.54874182e+00 -9.25885439e-01 6.74683601e-02 2.10193229e+00
7.69005001e-01 1.31945005e-02 1.64289342e-03 4.76287454e-01
1.04802454e+00 2.83896178e-01 -6.64367557e-01 -7.07505718e-02
-1.91365704e-01 1.20357769e-02 7.83303902e-02 -6.59672916e-02
-1.26583815e+00 6.65731668e-01 4.44619131e+00 1.48893535e+00
-1.37787902e+00 7.09567666e-02 6.67287767e-01 2.25484118e-01
-1.33129835e-01 -2.02313915e-01 -7.10933208e-01 9.47481930e-01
2.34161973e-01 -1.84356883e-01 3.62233818e-01 1.06539035e+00
-3.30279283e-02 2.15486377e-01 -8.44264746e-01 1.26172829e+00
1.10957704e-01 -1.12487423e+00 2.96801805e-01 -1.09627634e-01
7.70523965e-01 -5.03545225e-01 1.17080443e-01 4.98304695e-01
-2.69542783e-01 -8.92793059e-01 7.30387986e-01 2.59119302e-01
1.04765463e+00 -8.88692796e-01 8.67712736e-01 3.21668595e-01
-1.60868645e+00 -4.61208105e-01 -9.73963201e-01 1.62313148e-01
-3.80481690e-01 6.15261316e-01 -2.28224352e-01 7.48322248e-01
8.28352273e-01 8.72513950e-01 -9.88488436e-01 1.06093776e+00
1.53167352e-01 3.14554214e-01 3.24430726e-02 -1.52085096e-01
3.10193509e-01 -3.00165385e-01 1.82804912e-01 9.28323090e-01
3.53608161e-01 2.59602256e-03 4.41393107e-01 7.27067351e-01
-1.30209057e-02 4.53834742e-01 -3.68173778e-01 1.48391396e-01
6.10093117e-01 1.35211504e+00 -6.52460039e-01 -3.37679535e-01
-5.52384853e-01 9.79735315e-01 2.87237167e-01 2.44046256e-01
-7.42635071e-01 -8.21214795e-01 5.83201885e-01 -2.92629153e-02
5.08659661e-01 3.23656887e-01 -1.99779406e-01 -1.24153602e+00
2.24545866e-01 -1.00579417e+00 6.11462653e-01 -3.83893222e-01
-1.61048234e+00 6.82218552e-01 -3.11287045e-01 -1.72581363e+00
1.95350617e-01 -4.80790228e-01 -5.13985455e-01 9.08421397e-01
-1.33749712e+00 -1.19917333e+00 -8.34365308e-01 6.76325440e-01
5.85608840e-01 -2.15176806e-01 6.35664165e-01 4.39993054e-01
-6.75317228e-01 9.36402738e-01 4.40176964e-01 1.24367282e-01
6.37261152e-01 -9.32968974e-01 -1.88386127e-01 7.85405815e-01
-1.68737143e-01 6.61721647e-01 2.23764271e-01 -3.70431840e-01
-1.03098547e+00 -1.58367324e+00 3.66875321e-01 4.14103764e-04
4.89845335e-01 -3.25629234e-01 -1.25271821e+00 2.17373326e-01
-3.70589912e-01 5.00238597e-01 6.15013123e-01 -2.53553629e-01
-6.94676936e-01 -6.53891921e-01 -1.24375212e+00 5.16380727e-01
1.02263868e+00 -5.42554140e-01 -6.78208411e-01 9.19704288e-02
6.95965707e-01 1.50547355e-01 -7.91974187e-01 5.31216443e-01
5.73480785e-01 -9.55781639e-01 9.59295690e-01 -3.73765439e-01
2.50938565e-01 -7.36703277e-01 -4.16870326e-01 -1.43336618e+00
-8.27703714e-01 2.40885660e-01 2.30397046e-01 1.51330471e+00
-1.99528709e-01 -8.18536639e-01 5.12721837e-01 1.04646394e-02
-1.91749528e-01 -8.88734221e-01 -8.59005928e-01 -1.02739429e+00
6.38141483e-02 -3.43546346e-02 8.78909409e-01 8.83177102e-01
-2.75052845e-01 1.33543313e-01 -1.05196968e-01 7.56906494e-02
7.19677210e-01 7.48170733e-01 5.05050600e-01 -1.40648890e+00
-1.17072307e-01 -5.62955797e-01 -7.87110031e-01 -7.51967072e-01
-5.51454350e-02 -1.00157535e+00 6.40253574e-02 -1.28942156e+00
3.62402588e-01 -7.00749695e-01 -5.05904734e-01 3.06054235e-01
-4.82547194e-01 6.49530828e-01 2.55772650e-01 3.83324593e-01
-4.07533377e-01 9.22624886e-01 1.42350376e+00 -4.88547027e-01
3.10878009e-02 -9.89178568e-02 -9.97486591e-01 6.64795816e-01
5.74613333e-01 -1.32138774e-01 -4.97318596e-01 1.62595119e-02
-6.25412583e-01 -4.79946196e-01 3.80466044e-01 -1.13940203e+00
-5.84766082e-03 -2.51458883e-01 8.74089122e-01 -5.34177840e-01
1.31021813e-01 -8.56563330e-01 6.28206357e-02 5.04485786e-01
-9.81569886e-02 -3.65656137e-01 1.12854607e-01 7.55882144e-01
-5.07550597e-01 -1.17938094e-01 1.18858671e+00 1.21115834e-01
-1.07240415e+00 5.37809968e-01 8.68827030e-02 1.43721208e-01
1.29289067e+00 -5.86226225e-01 -4.92227584e-01 2.56371111e-01
-4.14046854e-01 1.40479922e-01 5.39767623e-01 5.86674571e-01
6.91967487e-01 -1.77414620e+00 -4.98413175e-01 4.95377928e-01
7.10857987e-01 -8.93028975e-02 7.25203276e-01 6.35248423e-01
-1.44144133e-01 4.06594038e-01 -3.99354309e-01 -6.51563644e-01
-1.21132958e+00 1.00813830e+00 1.27050981e-01 4.17690352e-02
-5.27396977e-01 7.24971831e-01 7.26161838e-01 -2.98827499e-01
9.05830711e-02 -2.25095630e-01 -5.21644771e-01 2.63014466e-01
8.38561237e-01 3.42606276e-01 2.38159634e-02 -8.49431038e-01
-5.65470040e-01 1.05961180e+00 -2.60980994e-01 6.03461206e-01
1.02875733e+00 -3.66312206e-01 -1.14488583e-02 3.42317641e-01
1.58514309e+00 3.18736024e-02 -1.18247712e+00 -5.05682051e-01
-2.68927604e-01 -9.31791067e-01 -1.01485653e-02 -6.51260853e-01
-1.16466570e+00 1.05579841e+00 9.73266900e-01 3.25504303e-01
1.32461953e+00 -4.22886536e-02 6.42596662e-01 1.29672792e-02
4.65377152e-01 -9.73389745e-01 1.95623398e-01 1.55255035e-01
8.52698863e-01 -1.27079487e+00 -1.60039544e-01 -4.95025545e-01
-5.76571345e-01 1.07442772e+00 8.63117576e-01 -1.41620547e-01
6.47152007e-01 -3.36839020e-01 -2.80965660e-02 1.23526968e-01
-2.72988468e-01 -1.74813822e-01 4.99697983e-01 7.36913025e-01
1.50463786e-02 1.92499086e-01 -4.11587715e-01 1.00561845e+00
6.48151413e-02 -1.61813304e-01 1.20000511e-01 5.39595723e-01
-6.50186062e-01 -8.54328632e-01 -4.34506029e-01 7.86433339e-01
-1.64552629e-01 9.33944806e-02 -2.07020417e-01 7.36070395e-01
5.95025659e-01 9.30698812e-01 1.48050845e-01 -6.69863999e-01
4.14430201e-01 -2.68873096e-01 2.10612744e-01 -5.44519424e-01
-1.97569907e-01 -6.16308339e-02 -3.71669978e-01 -3.87963265e-01
-1.53046876e-01 -3.97292167e-01 -1.15206432e+00 -2.32903853e-01
-3.57978463e-01 2.72411406e-01 1.01399064e-01 7.19308197e-01
3.11601490e-01 5.49403846e-01 1.20464432e+00 -6.96859479e-01
-7.77198851e-01 -9.99801517e-01 -8.12304735e-01 6.00646257e-01
2.23570883e-01 -1.04790485e+00 -5.84995925e-01 -2.17474833e-01] | [9.715106964111328, 2.045559883117676] |
210305bc-f537-49d1-839a-f1961091b46e | multi-contrast-computed-tomography-atlas-of | 2306.01853 | null | https://arxiv.org/abs/2306.01853v1 | https://arxiv.org/pdf/2306.01853v1.pdf | Multi-Contrast Computed Tomography Atlas of Healthy Pancreas | With the substantial diversity in population demographics, such as differences in age and body composition, the volumetric morphology of pancreas varies greatly, resulting in distinctive variations in shape and appearance. Such variations increase the difficulty at generalizing population-wide pancreas features. A volumetric spatial reference is needed to adapt the morphological variability for organ-specific analysis. Here, we proposed a high-resolution computed tomography (CT) atlas framework specifically optimized for the pancreas organ across multi-contrast CT. We introduce a deep learning-based pre-processing technique to extract the abdominal region of interests (ROIs) and leverage a hierarchical registration pipeline to align the pancreas anatomy across populations. Briefly, DEEDs affine and non-rigid registration are performed to transfer patient abdominal volumes to a fixed high-resolution atlas template. To generate and evaluate the pancreas atlas template, multi-contrast modality CT scans of 443 subjects (without reported history of pancreatic disease, age: 15-50 years old) are processed. Comparing with different registration state-of-the-art tools, the combination of DEEDs affine and non-rigid registration achieves the best performance for the pancreas label transfer across all contrast phases. We further perform external evaluation with another research cohort of 100 de-identified portal venous scans with 13 organs labeled, having the best label transfer performance of 0.504 Dice score in unsupervised setting. The qualitative representation (e.g., average mapping) of each phase creates a clear boundary of pancreas and its distinctive contrast appearance. The deformation surface renderings across scales (e.g., small to large volume) further illustrate the generalizability of the proposed atlas template. | ['Bennett A. Landman', 'Yuankai Huo', 'Jeffrey M. Spraggins', 'Shunxing Bao', 'Qi Yang', 'Xin Yu', 'Yucheng Tang', 'Ho Hin Lee', 'Yinchi Zhou'] | 2023-06-02 | null | null | null | null | ['computed-tomography-ct', 'anatomy'] | ['methodology', 'miscellaneous'] | [-1.21403530e-01 1.19812243e-01 -1.69137537e-01 -5.17519534e-01
-9.15358961e-01 -9.06353772e-01 4.89543796e-01 4.96660531e-01
-2.00101197e-01 1.99409887e-01 5.98378062e-01 4.29058485e-02
-1.48043320e-01 -6.12000525e-01 -5.89271069e-01 -8.86118472e-01
-5.18152416e-01 7.05914974e-01 6.37874752e-02 2.80135542e-01
-1.78473935e-01 5.52164435e-01 -5.76744258e-01 -2.87166797e-04
9.66042817e-01 7.63558745e-01 2.78941691e-02 3.51414144e-01
2.13409644e-02 -6.94153458e-02 -1.28328810e-02 -4.09803241e-01
6.31651402e-01 -5.59246421e-01 -4.43366855e-01 2.38443948e-02
7.28729248e-01 -2.89640039e-01 -3.51415217e-01 9.52729762e-01
5.36027789e-01 -1.63890585e-01 8.66453886e-01 -8.14484119e-01
-6.44928813e-01 7.59792209e-01 -5.00914156e-01 2.81762302e-01
1.11721657e-01 4.27160889e-01 3.85022700e-01 -6.16602182e-01
7.50028908e-01 6.42128944e-01 1.04722250e+00 2.84862608e-01
-1.34790027e+00 -7.21527636e-01 -8.61356929e-02 -5.94955862e-01
-1.25411570e+00 1.34849370e-01 6.01501822e-01 -7.13687479e-01
3.37739944e-01 2.11549670e-01 1.16924286e+00 8.58343124e-01
6.77498698e-01 1.34123443e-02 1.49269378e+00 -6.98438752e-03
-7.28000179e-02 -2.82815576e-01 -3.60659733e-02 7.93358207e-01
6.20471001e-01 1.02593005e-01 -3.32978927e-02 -3.35094839e-01
1.19206274e+00 2.30548173e-01 -2.94475645e-01 -9.25414681e-01
-1.76274419e+00 7.05714941e-01 6.47559404e-01 2.14242175e-01
-7.23672450e-01 -2.03725621e-01 4.71774697e-01 -7.20724836e-02
3.48518752e-02 2.79672265e-01 -3.08274388e-01 2.87652791e-01
-8.75367820e-01 -1.52102038e-01 7.11232066e-01 9.21746492e-01
1.24125339e-01 4.62381802e-02 -2.27324128e-01 6.51708126e-01
7.12190032e-01 5.07042468e-01 8.04164529e-01 -7.96381652e-01
1.49051726e-01 6.93516731e-01 -4.65807617e-01 -5.73770404e-01
-8.77014518e-01 -5.62966168e-01 -1.13607013e+00 8.52639899e-02
7.59279609e-01 -1.00238085e-01 -1.18262470e+00 1.68670416e+00
6.04583442e-01 -7.09613636e-02 -1.32372111e-01 1.26117218e+00
1.05539787e+00 -3.88394743e-02 7.41985261e-01 -1.42943561e-01
1.90470099e+00 -5.89964509e-01 -2.48555735e-01 2.43539929e-01
3.18368345e-01 -6.37833774e-01 8.35865915e-01 -6.73370212e-02
-1.16043723e+00 1.08069465e-01 -8.68070900e-01 3.48824352e-01
-6.60861209e-02 -1.01769961e-01 6.07709348e-01 7.28337765e-01
-6.78259075e-01 5.10979652e-01 -1.26146615e+00 -7.36542523e-01
5.23110271e-01 4.47874248e-01 -6.33394897e-01 1.32800683e-01
-6.84481561e-01 1.06005669e+00 2.98997104e-01 -1.14131905e-01
-8.84290576e-01 -1.40161431e+00 -8.72374833e-01 -1.60342529e-01
-2.40908355e-01 -1.16063702e+00 7.52620816e-01 -8.33075941e-01
-1.62531245e+00 1.22977233e+00 4.63863045e-01 -2.53466159e-01
8.86173844e-01 3.00001740e-01 -4.74784076e-01 3.73383582e-01
-1.96109265e-02 4.30839479e-01 5.85628808e-01 -1.30046511e+00
-7.50131160e-02 -5.79935431e-01 -6.80666745e-01 3.48046809e-01
2.87438452e-01 3.30265433e-01 -2.30620891e-01 -9.91316974e-01
6.89806283e-01 -1.22337532e+00 -4.00845528e-01 3.94449502e-01
-6.60130531e-02 5.11727214e-01 2.74166018e-01 -1.02943838e+00
7.08757222e-01 -1.95232928e+00 -5.14754094e-02 5.19728661e-01
4.51424599e-01 -3.46337318e-01 1.48754671e-01 -1.72099277e-01
-3.22036147e-01 1.29549623e-01 -6.25039458e-01 1.04122594e-01
2.63247006e-02 1.58168331e-01 3.30404669e-01 1.14231122e+00
-2.20348999e-01 1.16856980e+00 -9.51834559e-01 -8.56935620e-01
3.12976032e-01 5.69371104e-01 -5.71334124e-01 2.68796593e-01
4.21633661e-01 1.00214529e+00 -3.59024554e-01 8.37416708e-01
7.80131698e-01 -3.01111072e-01 4.50589150e-01 -5.81436098e-01
-2.39675775e-01 -1.08794011e-02 -8.17275286e-01 2.03594518e+00
-1.58028916e-01 -3.73428166e-02 4.43795025e-01 -4.35028076e-01
8.28827262e-01 4.25346404e-01 1.17855000e+00 -6.44389212e-01
2.36712351e-01 3.32750171e-01 4.37037021e-01 -1.98033229e-01
-1.04131788e-01 -5.26034951e-01 -1.00382036e-02 2.63492465e-01
2.70604044e-01 -1.94075927e-01 -8.61555114e-02 -2.58415103e-01
7.19811976e-01 2.12668508e-01 3.71736139e-01 -8.16543281e-01
3.04028034e-01 1.29184052e-01 5.84082663e-01 3.91908318e-01
-5.08818090e-01 1.09503353e+00 1.68075964e-01 -5.78560293e-01
-1.16527486e+00 -1.71334457e+00 -7.96254933e-01 6.41380966e-01
1.35983288e-01 -5.16888546e-03 -5.39049447e-01 -7.78790712e-01
2.74293929e-01 1.55192316e-01 -5.93224466e-01 1.09570496e-01
-9.65779424e-01 -1.25021446e+00 7.04705477e-01 4.13550705e-01
2.84710288e-01 -7.21978784e-01 -6.75102949e-01 2.26327479e-01
2.07396839e-02 -8.62460792e-01 -6.85603797e-01 2.15748519e-01
-1.29907119e+00 -1.08800387e+00 -1.22780001e+00 -8.96783948e-01
9.29607451e-01 -4.45945710e-01 1.41787457e+00 -2.96742141e-01
-3.14782470e-01 6.08693421e-01 -4.40035239e-02 -1.25999987e-01
-6.51008368e-01 3.56842205e-02 4.79139201e-02 -3.49934667e-01
-1.42376751e-01 -6.79611742e-01 -1.18757653e+00 4.14050281e-01
-7.26927400e-01 -9.19027850e-02 8.46589267e-01 5.13982713e-01
9.63569641e-01 -7.42366672e-01 1.53589204e-01 -5.51938891e-01
3.54363710e-01 -7.59918869e-01 -5.75070977e-01 3.27508479e-01
-7.03040004e-01 -1.80377200e-01 3.58867645e-01 -6.84789181e-01
-8.11451972e-01 9.26185921e-02 2.30384707e-01 -2.25115761e-01
-2.91322768e-01 2.72503823e-01 2.34067470e-01 -4.44795430e-01
4.44057971e-01 1.88945696e-01 3.75942558e-01 -3.08455110e-01
1.84260726e-01 -5.88364564e-02 6.16034448e-01 -7.27909565e-01
7.26666868e-01 3.81398082e-01 3.19398195e-01 -2.20842674e-01
2.73038056e-02 -1.10191405e-01 -1.14324915e+00 -1.49822086e-01
1.15520370e+00 -7.90723681e-01 -6.90859914e-01 3.03667963e-01
-4.77191806e-01 -4.07699436e-01 -4.48225588e-01 1.10665524e+00
-4.25262451e-01 4.65384752e-01 -7.81336308e-01 1.50688827e-01
-8.98828149e-01 -1.39819443e+00 7.53104746e-01 2.10063174e-01
-3.41092318e-01 -1.23253763e+00 4.40388858e-01 -1.51891068e-01
7.99460709e-01 8.80265534e-01 9.89400387e-01 -7.41556227e-01
-3.50082785e-01 1.14613980e-01 -2.78695256e-01 -9.07588080e-02
2.08259985e-01 -1.95304796e-01 -3.83021355e-01 -2.51081437e-01
-1.03389055e-01 2.60079175e-01 2.77695030e-01 9.09843981e-01
8.27100396e-01 -2.60390509e-02 -1.49490073e-01 1.25217843e+00
1.53150141e+00 8.46859142e-02 2.52929598e-01 3.56361330e-01
8.18675637e-01 2.69420266e-01 4.03109156e-02 4.70513403e-01
6.05563402e-01 5.27005911e-01 4.30974126e-01 -5.22616446e-01
-3.27237278e-01 -1.17999531e-01 1.52188763e-01 1.01871824e+00
-3.39151949e-01 5.52443862e-01 -1.15786278e+00 4.25967723e-01
-9.94808257e-01 -4.74300712e-01 -3.13711315e-01 2.40199995e+00
8.09375286e-01 -4.73966837e-01 2.91793972e-01 -9.86825526e-01
7.45340168e-01 -4.75683920e-02 -5.31327665e-01 -1.02469265e-01
-5.79565689e-02 2.04009026e-01 9.57644045e-01 2.53726363e-01
-1.21185613e+00 2.69128710e-01 6.84191418e+00 -1.17073774e-01
-1.43679595e+00 3.01239938e-01 3.86759132e-01 -2.42614816e-03
-2.68428922e-01 -1.13872318e-02 -1.77901909e-01 4.72225547e-01
9.11628723e-01 -3.00707519e-01 3.47937256e-01 4.96037096e-01
-9.38794836e-02 1.28197461e-01 -1.15667510e+00 7.97587633e-01
5.83381876e-02 -9.40741956e-01 -1.94347706e-02 2.05078244e-01
6.09891951e-01 4.36252892e-01 1.03418835e-01 2.06713319e-01
3.60330194e-01 -1.09821105e+00 5.61208367e-01 5.30626237e-01
1.17670560e+00 -1.88514471e-01 5.11159062e-01 -3.31030339e-01
-1.41324019e+00 3.29324961e-01 6.60121143e-02 6.81677997e-01
2.99011767e-01 2.19622225e-01 -8.70805740e-01 7.02939570e-01
6.46063566e-01 3.19770932e-01 -4.99220222e-01 1.27440381e+00
1.88876837e-01 3.93194884e-01 -6.18983924e-01 6.66210771e-01
-5.42516857e-02 -8.03122044e-01 6.25629365e-01 1.50403976e+00
6.41365051e-01 2.66262054e-01 2.65003860e-01 1.09828627e+00
-3.04240137e-02 5.22935450e-01 -2.48244733e-01 3.87162924e-01
7.81157985e-02 1.71722543e+00 -1.01167464e+00 -2.84687191e-01
-6.84029877e-01 5.65079808e-01 -1.68795928e-01 1.59261242e-01
-1.02221608e+00 3.90230298e-01 4.02217239e-01 3.31824660e-01
-2.42810342e-02 -2.05138296e-01 -4.14903641e-01 -1.31950378e+00
-1.19855314e-01 -8.00883532e-01 8.42841566e-01 -3.23531449e-01
-1.63371146e+00 5.05628943e-01 1.98684812e-01 -1.41155076e+00
5.60035221e-02 -3.12101454e-01 -6.53022707e-01 1.11034930e+00
-1.38694382e+00 -1.42611110e+00 -8.65789115e-01 5.12539208e-01
-2.41445545e-02 1.49890244e-01 1.03723681e+00 4.10136968e-01
-2.30563819e-01 6.03948712e-01 -8.31810981e-02 3.34265828e-01
1.03503311e+00 -1.34428799e+00 -1.50436267e-01 4.14420873e-01
-5.07338107e-01 8.51856291e-01 2.42276117e-01 -8.61351371e-01
-1.68532181e+00 -1.23004305e+00 8.90374482e-02 -5.03756106e-01
6.43707573e-01 -1.88257676e-02 -8.63578141e-01 9.59948242e-01
3.22142661e-01 7.01008439e-01 9.52291429e-01 -2.96142966e-01
-2.27519795e-01 4.65773493e-02 -1.69902408e+00 5.05681753e-01
8.20913136e-01 8.36315528e-02 -7.98916876e-01 1.50746936e-02
1.81765199e-01 -1.00131547e+00 -2.07650423e+00 5.70574284e-01
1.04789734e+00 -6.26118481e-01 1.21199977e+00 -2.93187827e-01
1.60386607e-01 -1.76924467e-01 5.87251745e-02 -1.17431962e+00
-5.56593955e-01 -4.80041593e-01 2.06685498e-01 1.06106889e+00
1.54752865e-01 -9.52399492e-01 5.64989924e-01 1.03463733e+00
-4.66865689e-01 -5.45594215e-01 -9.60867107e-01 -4.27279323e-01
6.89949155e-01 2.88308978e-01 5.14895916e-01 1.34711480e+00
-3.67259271e-02 -4.52297658e-01 4.67674792e-01 3.23026150e-01
9.88036931e-01 2.36079335e-01 4.66966808e-01 -1.40196574e+00
2.81463005e-02 -6.97314024e-01 -2.71556705e-01 -3.97718787e-01
-9.43922698e-02 -1.57109237e+00 -3.76172185e-01 -1.40290272e+00
7.09058940e-01 -8.01781118e-01 -5.77521205e-01 2.54388779e-01
-8.25657407e-05 1.78913563e-01 2.53223211e-01 4.27395135e-01
-1.63251042e-01 2.14640632e-01 1.73886085e+00 1.04535371e-01
-2.58756012e-01 -2.89929181e-01 -4.40907329e-01 7.14702249e-01
7.15111613e-01 -5.81231117e-01 1.51186556e-01 -7.62490928e-02
-5.78113258e-01 2.32914209e-01 5.29429793e-01 -8.05176914e-01
4.24829051e-02 3.81665602e-02 9.00828123e-01 -2.95193791e-01
-1.96092561e-01 -1.14815438e+00 7.31677771e-01 7.87438095e-01
-6.66072816e-02 4.77589607e-01 2.55877465e-01 3.64046660e-03
6.17001727e-02 5.57172857e-02 1.12674594e+00 -4.63774621e-01
-2.99317297e-02 7.05806136e-01 -2.10732698e-01 2.82621622e-01
9.77651238e-01 -2.21400380e-01 -7.93757066e-02 1.04313992e-01
-9.31302011e-01 5.99248242e-03 8.36704433e-01 5.30753955e-02
3.22845191e-01 -1.40283608e+00 -9.63354349e-01 2.67556995e-01
4.43238132e-02 9.82601643e-02 3.41233224e-01 1.73023391e+00
-8.80391240e-01 1.18727461e-01 -6.28361046e-01 -1.00231934e+00
-1.05296659e+00 3.43177855e-01 8.20527136e-01 -4.93295699e-01
-1.22056890e+00 2.15961933e-01 4.75768477e-01 -7.26940215e-01
-2.13079840e-01 -8.69078934e-01 2.28303149e-02 -1.08250901e-01
-4.88781594e-02 1.48673683e-01 1.51605187e-02 -9.75623071e-01
-4.70838159e-01 1.10791421e+00 2.07833946e-01 1.75087914e-01
1.37662160e+00 -1.48316532e-01 -1.05324097e-01 -1.54733863e-02
1.03074348e+00 2.16854170e-01 -1.27940857e+00 -7.70654157e-02
-2.59256184e-01 -3.03323597e-01 -1.48898968e-02 -8.89920592e-01
-1.35479796e+00 2.86863714e-01 1.19791579e+00 -1.44112900e-01
9.21758711e-01 5.47445789e-02 5.46343803e-01 -5.91542482e-01
3.65674138e-01 -3.08213145e-01 -5.14815211e-01 2.64412194e-01
8.99226129e-01 -1.31321669e+00 2.31601581e-01 -3.89773756e-01
-7.38246083e-01 1.13600802e+00 4.24651951e-01 -3.43460321e-01
5.54410994e-01 5.95742583e-01 3.96278530e-01 -3.80263686e-01
7.13690296e-02 3.56888086e-01 6.27381325e-01 6.34864628e-01
7.55517364e-01 3.09692889e-01 -4.50257152e-01 6.38826072e-01
-2.93200761e-01 -3.00010562e-01 1.75674483e-02 7.62964904e-01
1.25426054e-01 -8.74953508e-01 -5.71895659e-01 4.25020725e-01
-7.45238245e-01 4.14420664e-02 1.26751244e-01 1.13771749e+00
-9.80050191e-02 3.27819437e-02 1.49802431e-01 4.48066920e-01
4.64553237e-01 2.12562695e-01 7.93586493e-01 -2.89002657e-01
-1.24820280e+00 2.10898951e-01 -3.11543971e-01 -5.41148901e-01
-3.28140259e-01 -9.88382995e-01 -1.64189947e+00 -5.84018491e-02
1.75415680e-01 -3.95147562e-01 7.86049962e-01 4.18312848e-01
1.75060451e-01 5.59787571e-01 2.81518072e-01 -9.27372932e-01
-4.43932682e-01 -8.75517190e-01 -5.23004353e-01 9.47783887e-01
7.54876435e-02 -6.52231932e-01 -2.64338672e-01 1.80671543e-01] | [14.480670928955078, -2.668203115463257] |
e8f744ea-0ad3-40d2-aaae-a97a167b37f8 | deep-mr-fingerprinting-with-total-variation | 1902.10205 | null | http://arxiv.org/abs/1902.10205v1 | http://arxiv.org/pdf/1902.10205v1.pdf | Deep MR Fingerprinting with total-variation and low-rank subspace priors | Deep learning (DL) has recently emerged to address the heavy storage and
computation requirements of the baseline dictionary-matching (DM) for Magnetic
Resonance Fingerprinting (MRF) reconstruction. Fed with non-iterated
back-projected images, the network is unable to fully resolve
spatially-correlated corruptions caused from the undersampling artefacts. We
propose an accelerated iterative reconstruction to minimize these artefacts
before feeding into the network. This is done through a convex regularization
that jointly promotes spatio-temporal regularities of the MRF time-series.
Except for training, the rest of the parameter estimation pipeline is
dictionary-free. We validate the proposed approach on synthetic and in-vivo
datasets. | ['Pedro A. Gómez', 'Marion I. Menzel', 'Carolin M. Pirkl', 'Guido Buonincontri', 'Mohammad Golbabaee'] | 2019-02-26 | null | null | null | null | ['magnetic-resonance-fingerprinting'] | ['medical'] | [ 5.06667733e-01 -2.12181564e-02 1.79170251e-01 -3.93035620e-01
-7.50379086e-01 -1.54431537e-01 3.05870622e-01 8.25498477e-02
-6.47278547e-01 6.26487792e-01 2.84201503e-01 -2.14922979e-01
-2.73903579e-01 -4.25255775e-01 -7.82398164e-01 -7.27190733e-01
-3.72054905e-01 2.66338378e-01 -8.67297407e-03 2.20863596e-01
7.37940669e-02 7.24967360e-01 -1.08847928e+00 3.83967340e-01
5.61478019e-01 8.55874956e-01 6.60005271e-01 3.89819533e-01
3.45208049e-01 8.27491343e-01 -1.44098476e-01 1.46543384e-01
3.82794321e-01 -2.69249916e-01 -6.47361577e-01 -8.59627351e-02
5.47245145e-01 -7.90580869e-01 -6.59942448e-01 1.15992880e+00
9.93879437e-01 2.46432528e-01 2.07855195e-01 -5.78017592e-01
-1.68930858e-01 3.44963640e-01 -4.73046422e-01 7.04586029e-01
1.87509395e-02 -8.08796436e-02 3.05784047e-01 -1.32057083e+00
1.04158306e+00 7.08051622e-01 9.24980819e-01 4.24753755e-01
-1.46874976e+00 -3.31727147e-01 -4.32736605e-01 9.05653462e-02
-1.23267400e+00 -7.04749167e-01 8.13289940e-01 -4.52042431e-01
8.17233622e-01 1.10496439e-01 5.57947516e-01 1.09647083e+00
4.38029170e-01 4.66545284e-01 1.33564520e+00 -3.62375557e-01
2.09550321e-01 -3.22129697e-01 7.51383007e-02 4.81865197e-01
7.95928389e-02 5.39120853e-01 -5.76982796e-01 -4.48822141e-01
1.15517664e+00 4.78896312e-02 -4.30946916e-01 -3.98027539e-01
-1.60443771e+00 6.27156675e-01 1.94482878e-01 3.84236097e-01
-7.60065019e-01 1.41442820e-01 5.43539703e-01 1.59891650e-01
4.36425060e-01 3.54030132e-01 -2.00526029e-01 8.09430405e-02
-1.40813482e+00 1.35007173e-01 2.94768989e-01 5.00075698e-01
4.09579128e-01 3.10235828e-01 -2.09035888e-01 9.20469165e-01
2.12601155e-01 3.69945318e-01 6.69221580e-01 -1.10487747e+00
3.94151844e-02 -1.21074058e-01 -1.22895345e-01 -1.37612915e+00
-7.37385511e-01 -8.06115150e-01 -1.22692204e+00 1.42024374e-02
3.96475941e-01 7.39773214e-02 -8.80270660e-01 1.63265359e+00
5.66917181e-01 6.79512382e-01 -5.48718035e-01 1.43634915e+00
8.01896334e-01 2.86303073e-01 -1.69280931e-01 -4.14435387e-01
1.17494464e+00 -5.39215624e-01 -9.68796134e-01 -1.46933347e-02
4.98251915e-01 -7.13496923e-01 7.32228518e-01 3.60528558e-01
-1.13996494e+00 -5.32425702e-01 -1.15213561e+00 -1.35660693e-01
2.64103144e-01 6.72065988e-02 4.46201444e-01 2.18874231e-01
-9.99459743e-01 9.86194432e-01 -1.28227222e+00 2.45988414e-01
4.77349967e-01 4.47255135e-01 -5.21275342e-01 -2.00556204e-01
-1.18781924e+00 9.78792906e-01 5.97473569e-02 3.89186144e-01
-1.16741872e+00 -1.09399092e+00 -6.34987950e-01 -4.05446321e-01
1.62885576e-01 -6.58878207e-01 7.08086729e-01 -5.82213998e-01
-1.46348822e+00 9.23310459e-01 -1.69698402e-01 -6.13290966e-01
6.09212041e-01 -1.80379510e-01 -4.37959522e-01 4.94638056e-01
1.48421660e-01 4.06369418e-01 1.08976579e+00 -9.96307909e-01
1.90944895e-01 -4.45844352e-01 -4.69530970e-01 -7.80999884e-02
1.15032025e-01 -7.69416243e-02 -2.17928901e-01 -1.17383492e+00
5.20606220e-01 -1.00847125e+00 -4.87373859e-01 2.19762579e-01
-1.93700790e-01 7.37693369e-01 5.06694317e-01 -1.16486561e+00
1.14173579e+00 -2.07445574e+00 -7.65969381e-02 4.00279611e-01
5.41457772e-01 2.65690178e-01 -2.20190763e-01 1.41994983e-01
-4.58149314e-01 -6.67659521e-01 -3.49044800e-01 -4.29240078e-01
-4.24720377e-01 2.19501898e-01 -3.00731927e-01 1.19809508e+00
-1.81163475e-01 7.76423037e-01 -9.47382390e-01 -3.19101959e-01
2.51791000e-01 6.92483425e-01 -7.35252142e-01 2.55621046e-01
2.04093680e-01 1.10876489e+00 -1.56330749e-01 3.86864156e-01
9.75614190e-01 -3.75110924e-01 5.20833969e-01 -7.14154303e-01
-7.37452209e-02 4.29649025e-01 -1.16317701e+00 2.39161301e+00
-3.40226740e-01 4.38840061e-01 4.25557137e-01 -1.50085354e+00
6.10239625e-01 3.40117723e-01 9.93046761e-01 -1.05101752e+00
1.32288471e-01 4.85489070e-01 -1.70710031e-02 -5.13565540e-01
2.12090492e-01 -2.35619307e-01 5.33401132e-01 5.68195462e-01
1.28308341e-01 4.82097477e-01 -1.14123508e-01 7.08769560e-02
1.16649389e+00 2.71945179e-01 3.88341839e-03 -7.86527693e-01
5.22051096e-01 -2.47386277e-01 6.59421086e-01 8.16663325e-01
-2.41875142e-01 9.18066382e-01 -5.45971356e-02 -7.68315971e-01
-1.35288024e+00 -1.00185895e+00 -6.74370408e-01 6.19816124e-01
3.15769166e-02 -1.47492841e-01 -5.67663491e-01 -3.14233691e-01
4.05074507e-02 1.96668133e-01 -6.19836509e-01 -1.37145057e-01
-1.12713218e+00 -8.38302493e-01 4.03422028e-01 1.79877415e-01
1.06407918e-01 -8.08452666e-01 -8.29985559e-01 7.40918279e-01
-5.23868918e-01 -1.31947100e+00 -4.71015424e-01 3.18558425e-01
-1.23576343e+00 -8.31408381e-01 -8.42638314e-01 -6.07883275e-01
7.07870126e-01 2.76867658e-01 1.01427734e+00 1.71620965e-01
-5.87593496e-01 5.92155010e-02 5.07260785e-02 2.31136113e-01
-3.51835072e-01 -3.54429245e-01 1.89276651e-01 1.68266997e-01
-7.36356974e-02 -9.80364084e-01 -9.75594878e-01 2.24600121e-01
-9.03457105e-01 1.62698418e-01 4.83078271e-01 1.35936332e+00
1.11576736e+00 -3.89715254e-01 4.63286161e-01 -7.89173424e-01
3.74519169e-01 -4.74548906e-01 -5.55353642e-01 -5.47177643e-02
-4.12389129e-01 1.10529453e-01 5.63061237e-01 -7.34610260e-01
-5.76250136e-01 8.17297474e-02 -1.32369548e-01 -7.49553561e-01
1.23237923e-01 6.05841398e-01 4.09833163e-01 -4.70739871e-01
5.97298026e-01 4.87938315e-01 3.37590635e-01 -6.48927927e-01
1.85154416e-02 1.82007208e-01 9.47080731e-01 -5.45717001e-01
3.82531017e-01 8.55928123e-01 2.39438713e-01 -9.23740625e-01
-6.33367062e-01 -3.20732236e-01 -6.55145705e-01 -3.44030917e-01
6.23630404e-01 -8.55266571e-01 -5.10967791e-01 4.28820580e-01
-1.19026756e+00 -2.17531219e-01 -3.11115295e-01 8.43213320e-01
-6.86770082e-01 7.37411201e-01 -7.58599639e-01 -3.05349588e-01
-5.96063316e-01 -1.33733761e+00 9.02426362e-01 -4.73297656e-01
-1.23006187e-01 -8.65320563e-01 4.66385454e-01 2.05495983e-01
7.33574748e-01 4.52037066e-01 6.68012917e-01 -3.39900464e-01
-3.88435125e-01 6.14797957e-02 -9.33986753e-02 3.51256311e-01
-1.78394347e-01 -7.61802495e-01 -8.89279127e-01 -5.62158287e-01
5.89871109e-01 -1.59660727e-01 6.24682665e-01 8.77917171e-01
1.36957550e+00 -1.87514156e-01 -1.09813564e-01 1.09893048e+00
1.34538972e+00 -1.65817425e-01 6.78875983e-01 3.18617195e-01
6.07311010e-01 4.18982148e-01 3.22398663e-01 5.75573206e-01
8.97127241e-02 9.10371721e-01 1.96209088e-01 -2.81674713e-01
-4.97998774e-01 -1.76921576e-01 8.43371302e-02 9.27309453e-01
7.18817636e-02 3.49660039e-01 -9.66872156e-01 5.27242005e-01
-1.64838815e+00 -9.22896326e-01 -1.18885309e-01 2.18734360e+00
9.43691671e-01 -2.07907245e-01 -1.32124305e-01 3.21329273e-02
6.01364672e-01 2.75317281e-01 -7.31129766e-01 1.41753376e-01
-1.71199203e-01 4.73442107e-01 5.69683790e-01 5.27468622e-01
-1.04686689e+00 3.16962451e-01 6.80547333e+00 7.22467124e-01
-1.42020452e+00 5.36097765e-01 5.40969968e-01 -1.76827908e-01
-3.29987228e-01 -1.77746326e-01 -1.56566605e-01 4.59332138e-01
8.93781185e-01 1.91150159e-01 7.72018969e-01 4.41768169e-01
4.50145185e-01 -8.52688998e-02 -9.32609677e-01 1.26992297e+00
-1.63985416e-01 -1.74562180e+00 -2.24969879e-01 1.01174831e-01
5.13671100e-01 4.20625806e-01 -8.11585337e-02 -2.65871942e-01
-1.84566051e-01 -8.99628520e-01 6.65922582e-01 6.95180118e-01
1.01931870e+00 -4.27841872e-01 4.59149778e-01 2.70388752e-01
-8.20966601e-01 1.62294433e-01 -4.54994470e-01 3.38155001e-01
3.81434858e-01 1.30117118e+00 -7.10197985e-01 4.50496286e-01
6.08080208e-01 7.20885336e-01 -7.24798143e-02 7.03039765e-01
2.42673188e-01 3.74741852e-01 -4.17366385e-01 8.18960071e-01
4.57032397e-02 -3.65341544e-01 7.54262865e-01 1.22451389e+00
1.62150323e-01 1.88975498e-01 5.40980361e-02 1.03841972e+00
2.33393431e-01 -2.48418733e-01 -4.52182561e-01 1.99370220e-01
2.79830188e-01 1.15624416e+00 -6.64717734e-01 -1.95233598e-01
-2.94774711e-01 8.69775712e-01 3.33757460e-01 3.07612896e-01
-6.75968885e-01 1.17505267e-01 4.19338495e-01 5.86576760e-01
2.62918532e-01 -4.14642334e-01 -2.61468589e-01 -1.31049252e+00
1.70831785e-01 -9.91810203e-01 1.76536486e-01 -5.67210019e-01
-1.21669519e+00 5.58673561e-01 -5.97702786e-02 -1.20673466e+00
-6.64726719e-02 -1.59131199e-01 -1.12155192e-01 8.92949641e-01
-1.60459661e+00 -6.92105353e-01 -8.74634236e-02 8.35066080e-01
1.71828102e-02 1.04280636e-01 8.16322207e-01 1.07040811e+00
-3.13064635e-01 5.26453614e-01 1.77883312e-01 -2.11381633e-02
6.02303743e-01 -8.58644605e-01 1.34639174e-01 1.01339769e+00
-1.79229036e-01 9.59213138e-01 7.29969859e-01 -7.06705868e-01
-1.53760970e+00 -8.92417729e-01 7.62961924e-01 4.15136181e-02
6.21038496e-01 -3.73514980e-01 -1.02998686e+00 4.16541100e-01
-1.91255167e-01 7.59567738e-01 5.85161269e-01 -5.21627367e-01
-1.93672955e-01 -9.51981470e-02 -1.29012370e+00 4.85091358e-02
9.37860191e-01 -8.28566015e-01 -4.45603520e-01 5.57002306e-01
3.05733621e-01 -7.66470313e-01 -1.14743078e+00 4.65169340e-01
7.25373685e-01 -9.43288445e-01 1.26400208e+00 -4.24399137e-01
3.53248030e-01 -3.06085259e-01 -1.37460783e-01 -9.78911221e-01
-4.15736169e-01 -7.96566010e-01 -3.40111732e-01 4.43096966e-01
-1.43357471e-01 -5.15752733e-01 6.91905737e-01 3.17349643e-01
-4.85316068e-01 -7.19706833e-01 -1.35995781e+00 -6.46353483e-01
-2.68003851e-01 -4.21970427e-01 2.16011584e-01 1.25214899e+00
-1.40736192e-01 -1.95015594e-01 -6.87155604e-01 2.83535928e-01
1.15744174e+00 1.16800465e-01 3.28515589e-01 -7.94771910e-01
-5.38067400e-01 4.70306315e-02 -1.95262715e-01 -1.22616351e+00
-1.01735108e-01 -1.13977861e+00 1.59200169e-02 -1.04719698e+00
1.28769323e-01 -6.16362989e-01 -4.55067337e-01 1.58550546e-01
9.57138017e-02 2.82768697e-01 2.46097073e-02 5.47600448e-01
-2.77650476e-01 3.92120421e-01 1.64592874e+00 1.61419407e-01
1.86035529e-01 -3.92810732e-01 -7.96273127e-02 4.45030481e-01
3.42033118e-01 -8.08811307e-01 -2.85950899e-01 -6.54352963e-01
1.39494613e-01 6.62315547e-01 6.07007921e-01 -9.88603175e-01
2.79782057e-01 3.13651115e-01 3.82472545e-01 -7.56684482e-01
1.59743324e-01 -1.05597198e+00 6.44847751e-01 7.08891928e-01
-4.84225690e-01 1.87804386e-01 1.33440405e-01 4.04684186e-01
-2.63596147e-01 3.59911993e-02 1.22275257e+00 -3.73305418e-02
-3.35020609e-02 5.50917685e-01 -3.29402059e-01 5.53233027e-02
4.68484938e-01 -1.86819881e-01 1.08522348e-01 -1.38526950e-02
-1.17301798e+00 -3.14200103e-01 2.08294347e-01 -1.16803804e-02
8.25772405e-01 -1.48319948e+00 -7.13621318e-01 5.77731073e-01
-3.96382540e-01 -2.90815793e-02 7.61592746e-01 1.48764443e+00
-6.71930015e-01 3.41371894e-01 -4.54854429e-01 -7.00298190e-01
-6.98665559e-01 5.98392248e-01 7.55834162e-01 -7.02422976e-01
-1.21694076e+00 6.00022376e-01 -4.67815734e-02 -5.85171819e-01
1.65141687e-01 -1.47559807e-01 7.42689939e-03 -2.09030181e-01
6.80567443e-01 4.67887372e-01 5.19702911e-01 -5.68590760e-01
-5.27614474e-01 3.22688282e-01 -4.14647758e-02 -1.56659007e-01
1.64789975e+00 -1.42287448e-01 -2.04833880e-01 4.63635400e-02
1.51988149e+00 -2.27147937e-01 -1.13904333e+00 -5.19647121e-01
-8.36941451e-02 -3.57065052e-01 7.53601551e-01 -6.36060178e-01
-1.38581812e+00 7.78870285e-01 9.90791619e-01 -3.57914329e-01
1.02398825e+00 -6.21323705e-01 1.34788191e+00 9.22838692e-03
5.61868310e-01 -9.41717923e-01 -2.17132762e-01 3.22103262e-01
9.61925685e-01 -9.12835598e-01 3.29584241e-01 -5.78270992e-04
-9.85243320e-02 1.23252654e+00 -9.03428867e-02 -4.27514523e-01
8.80358636e-01 5.85761845e-01 -3.85166034e-02 -4.87644464e-01
-3.40747476e-01 4.70157415e-01 2.56384313e-01 6.22698963e-01
2.72757590e-01 -2.33599499e-01 -4.93578076e-01 2.18441024e-01
2.15431318e-01 2.89573252e-01 3.39396000e-01 1.02839291e+00
1.38539458e-02 -8.93911898e-01 -4.21299696e-01 4.41581607e-01
-6.15584791e-01 -2.64361531e-01 4.09370512e-01 3.64721894e-01
-5.71354963e-02 4.26370353e-01 2.94636115e-02 -1.83409631e-01
2.27680072e-01 -4.34946954e-01 6.63916349e-01 -2.16958076e-01
-6.46751761e-01 3.99673164e-01 -8.70893970e-02 -1.20145786e+00
-6.19394124e-01 -7.58575201e-01 -1.23981535e+00 -2.44238287e-01
-2.28373602e-01 -1.99901178e-01 7.43632317e-01 6.59965277e-01
6.18366718e-01 4.72986788e-01 6.94691837e-01 -9.91479158e-01
-6.89293921e-01 -7.39517391e-01 -6.28452063e-01 6.53374016e-01
5.68356991e-01 -7.23331153e-01 -3.79757911e-01 -8.93475488e-02] | [13.428775787353516, -2.4409239292144775] |
01ef5639-5f79-481a-bde8-0da82516a4da | meta-learning-for-low-resource-neural-machine-1 | null | null | https://openreview.net/forum?id=S1g5ylbm1Q | https://openreview.net/pdf?id=S1g5ylbm1Q | Meta-Learning for Low-Resource Neural Machine Translation | In this paper, we propose to extend the recently introduced model-agnostic meta-learning algorithm (MAML, Finn et al., 2017) for low resource neural machine translation (NMT). We frame low-resource translation as a meta-learning problem, and we learn to adapt to low-resource languages based on multilingual high-resource language tasks. We use the universal lexical representation (Gu et al., 2018b) to overcome the input-output mismatch across different languages. We evaluate the proposed meta-learning strategy using eighteen European languages (Bg, Cs, Da, De, El, Es, Et, Fr, Hu, It, Lt, Nl, Pl, Pt, Sk, Sl, Sv and Ru) as source tasks and five diverse languages (Ro, Lv, Fi, Tr, and Ko) as target tasks. We show that the proposed approach significantly outperforms the multilingual, transfer learning based approach (Zoph et al., 2016) and enables us to train a competitive NMT system with only a fraction of training examples. For instance, the proposed approach can achieve as high as 22.04 BLEU on Romanian-English WMT’16 by seeing only 16,000 translated words (~600 parallel sentences). | ['Anonymous'] | 2018-05-23 | null | null | null | null | ['low-resource-neural-machine-translation'] | ['natural-language-processing'] | [ 3.08444537e-02 -1.91575050e-01 -5.11181474e-01 7.85730779e-02
-1.38857615e+00 -7.23096132e-01 1.11882770e+00 -9.69930068e-02
-8.75879705e-01 1.38875651e+00 -1.03367325e-02 -9.22835350e-01
2.60146320e-01 -4.43021387e-01 -9.78759289e-01 -1.49213418e-01
4.22622442e-01 7.76682258e-01 -2.32899606e-01 -5.89276731e-01
3.96245793e-02 1.70280248e-01 -8.95960510e-01 5.99971831e-01
1.15119660e+00 4.33082044e-01 5.51255226e-01 5.95456243e-01
-2.19621688e-01 5.65755546e-01 -4.78781223e-01 -7.28333175e-01
1.32096589e-01 -7.05493391e-01 -9.66377020e-01 -4.92056310e-01
3.97483408e-01 2.08708450e-01 -8.75006616e-02 9.17468548e-01
6.67079449e-01 -1.06022224e-01 7.69676626e-01 -6.55684590e-01
-9.23654377e-01 1.17668891e+00 -6.82912290e-01 2.20290303e-01
4.02361564e-02 3.06430757e-02 8.14471304e-01 -1.51910663e+00
1.00892198e+00 1.26038480e+00 5.32463849e-01 7.71654665e-01
-9.86614943e-01 -8.37251663e-01 -7.49672428e-02 2.80367672e-01
-1.40268517e+00 -6.57855034e-01 2.67748684e-01 -1.22638807e-01
1.49141610e+00 1.58349037e-01 4.67510074e-01 1.56701195e+00
5.23410320e-01 7.56323814e-01 1.59368563e+00 -1.00365067e+00
3.44640687e-02 3.59730840e-01 -3.99843812e-01 5.36187589e-01
8.31975713e-02 1.35317841e-03 -6.61796629e-01 -1.65328175e-01
3.94128561e-01 -4.71998066e-01 4.57959063e-02 1.56440064e-02
-1.69926202e+00 7.24225998e-01 6.30510151e-02 6.50909483e-01
-3.23134571e-01 -2.23001391e-01 5.20803154e-01 9.09030259e-01
8.88297379e-01 2.37776563e-01 -1.08529711e+00 -2.40579829e-01
-8.17170203e-01 -1.46776035e-01 6.78476155e-01 1.35734904e+00
7.77132869e-01 2.78473198e-01 -1.49954036e-01 1.18546772e+00
5.64966686e-02 8.32406402e-01 1.07639205e+00 -3.92615080e-01
1.03122211e+00 4.58720326e-02 3.24705499e-03 -1.77527830e-01
-3.25217664e-01 -6.13152444e-01 -1.06385469e+00 -5.13974786e-01
-4.18756157e-03 -3.54669273e-01 -8.51347804e-01 1.90116882e+00
1.81501973e-02 -6.23392724e-02 6.11717880e-01 3.99365813e-01
6.17317855e-01 8.95271480e-01 6.18267730e-02 -6.01310074e-01
1.13764954e+00 -1.58421373e+00 -5.72063208e-01 -4.82190669e-01
9.16644096e-01 -1.27719116e+00 1.20145535e+00 2.51835972e-01
-1.22583234e+00 -6.91797614e-01 -8.44825923e-01 2.88159009e-02
-5.40323913e-01 5.57667255e-01 3.31853658e-01 4.77167070e-01
-1.11374593e+00 4.58277911e-01 -6.17697775e-01 -8.57695937e-01
7.88791701e-02 1.95533723e-01 -3.34652275e-01 -1.52384982e-01
-1.54742479e+00 1.36014748e+00 8.63913476e-01 -6.92117065e-02
-8.46543849e-01 -4.01358396e-01 -5.31096160e-01 -5.30028522e-01
2.12314010e-01 -8.46470058e-01 1.19453526e+00 -1.30285168e+00
-2.04980636e+00 1.00480402e+00 -6.20109253e-02 -4.37352687e-01
5.57501972e-01 -2.38989711e-01 -7.62944996e-01 -2.58519024e-01
4.33524400e-02 6.20343626e-01 5.82232833e-01 -9.28529441e-01
-8.56040299e-01 -1.50973663e-01 -3.34832549e-01 2.99917072e-01
-3.21205318e-01 3.85342151e-01 -1.26199514e-01 -8.79925013e-01
-1.92619130e-01 -1.08036220e+00 -4.73369658e-02 -8.46944749e-01
-9.37393010e-02 -7.63103887e-02 2.56464005e-01 -1.09720743e+00
1.10911894e+00 -1.66169059e+00 5.91152132e-01 -1.92940488e-01
-3.24081779e-01 3.99919808e-01 -5.09609163e-01 8.64600778e-01
1.36103377e-01 2.12248102e-01 -3.37992549e-01 -3.52701277e-01
-1.21050313e-01 3.56112212e-01 -1.01652488e-01 2.82073557e-01
9.72337089e-03 1.30617881e+00 -9.42024112e-01 -3.03453445e-01
-4.10152450e-02 2.18235195e-01 -1.45318702e-01 1.25042046e-03
-2.17794821e-01 7.43394196e-01 4.82789986e-02 6.47592545e-01
4.05727297e-01 1.29430681e-01 4.24392462e-01 2.90147644e-02
-4.62028265e-01 4.56416458e-01 -5.41271329e-01 2.03398132e+00
-1.16302657e+00 5.99435747e-01 -1.76271543e-01 -6.78685784e-01
9.46229517e-01 6.10547543e-01 1.32493287e-01 -9.56361890e-01
3.32047977e-02 1.03799713e+00 1.50953099e-01 -2.65312552e-01
5.87745845e-01 -1.05090819e-01 -2.98882425e-01 7.20612764e-01
5.73338807e-01 1.20981455e-01 1.85743630e-01 -2.22915694e-01
5.98691046e-01 4.27696288e-01 5.95960140e-01 -3.65800530e-01
6.94895566e-01 3.71328853e-02 4.24997330e-01 5.97561419e-01
-4.67387512e-02 1.76644459e-01 -1.97543472e-01 -2.33310610e-01
-1.51491773e+00 -9.54317689e-01 -4.89044823e-02 1.40914989e+00
-4.59973991e-01 -4.21850264e-01 -8.35796237e-01 -5.61535418e-01
-4.28131610e-01 8.19634974e-01 -2.26453528e-01 -2.21525639e-01
-1.06352270e+00 -7.88967371e-01 7.68786609e-01 1.04514368e-01
5.85358500e-01 -1.31740153e+00 -6.40044361e-02 4.24164832e-01
-5.69026053e-01 -1.15160167e+00 -5.52241862e-01 1.26170352e-01
-9.56322551e-01 -3.22150618e-01 -1.03034234e+00 -9.32682037e-01
3.25789958e-01 -1.12042010e-01 1.23776388e+00 -4.72630709e-01
1.47148862e-01 -6.21727249e-03 -3.90071422e-01 -1.97721213e-01
-1.01944983e+00 7.45988727e-01 4.00047034e-01 -2.17711389e-01
2.28996307e-01 -3.43664169e-01 5.14988750e-02 1.65729046e-01
-6.72977507e-01 3.78590673e-01 1.10446846e+00 1.08212900e+00
7.13847220e-01 -6.74829781e-01 8.97314847e-01 -9.84022558e-01
6.05633616e-01 -6.32102668e-01 -3.75135899e-01 7.03796327e-01
-7.94589818e-01 5.59782423e-02 9.03362215e-01 -7.65098393e-01
-9.82075751e-01 -3.21814299e-01 -1.27578676e-01 -2.22804427e-01
7.42397606e-02 7.60371149e-01 -2.13593006e-01 -8.74659792e-02
6.49977505e-01 5.81552386e-01 -6.61498189e-01 -6.70688152e-01
6.88784361e-01 1.11128867e+00 3.91453087e-01 -8.53833973e-01
8.05344105e-01 -1.48358375e-01 -3.97253036e-01 -4.73884970e-01
-5.86674631e-01 4.06780653e-03 -1.09720647e+00 7.41819590e-02
4.82689142e-01 -1.20459294e+00 9.12752897e-02 3.75390112e-01
-1.33610845e+00 -6.58041298e-01 -7.50873238e-02 7.76004195e-01
-7.96995163e-01 6.36439845e-02 -9.99926448e-01 -4.70768362e-01
-8.77310455e-01 -9.97252345e-01 8.79386127e-01 -2.35519871e-01
-3.44073735e-02 -1.21599233e+00 2.43432119e-01 1.35799125e-01
7.78331459e-01 -7.10669458e-02 1.09054542e+00 -7.28366613e-01
-9.20727327e-02 3.57576072e-01 -8.40684697e-02 3.67752224e-01
2.76926249e-01 -2.62119025e-01 -5.91821134e-01 -7.58362114e-01
-1.66911542e-01 -5.53125322e-01 6.91531539e-01 -6.37822226e-02
4.47363317e-01 -5.71852565e-01 4.25412990e-02 6.77306533e-01
1.45575309e+00 6.56671003e-02 3.33885044e-01 6.29569054e-01
6.52600110e-01 3.00960511e-01 6.89699352e-01 -8.33921507e-02
4.39402550e-01 7.15126276e-01 -3.55690151e-01 -3.47712338e-02
-2.69449413e-01 -3.89062524e-01 9.88180816e-01 1.84684932e+00
-3.35084140e-01 -2.35925063e-01 -1.00776660e+00 4.18690205e-01
-1.83840966e+00 -5.84450960e-01 1.44413233e-01 2.43408442e+00
1.13125944e+00 3.57652567e-02 1.46026323e-02 -5.35846233e-01
6.75097346e-01 -1.70072585e-01 -2.56900162e-01 -9.35706496e-01
-5.63313305e-01 5.65749884e-01 7.47498214e-01 5.54935634e-01
-7.46748924e-01 1.72668123e+00 5.43790674e+00 1.17935538e+00
-1.25515652e+00 8.01573455e-01 5.25484443e-01 9.45453942e-02
-3.21423024e-01 -7.32531846e-02 -6.92264080e-01 3.33131492e-01
1.73769033e+00 -4.82124180e-01 8.19000483e-01 3.73958141e-01
-1.07737273e-01 4.83449310e-01 -9.80243206e-01 8.35819602e-01
3.36956412e-01 -1.09074235e+00 3.41744095e-01 -1.51739582e-01
1.10677814e+00 6.35852993e-01 1.91199332e-01 8.07344437e-01
3.32131088e-01 -9.30550218e-01 8.98396969e-01 2.62194067e-01
1.16686416e+00 -9.86812115e-01 6.12866759e-01 6.69893086e-01
-9.70151484e-01 1.94620475e-01 -7.68857956e-01 1.67954460e-01
-2.17022505e-02 1.48000434e-01 -7.89425194e-01 1.25207293e+00
4.64020073e-01 6.08539820e-01 -6.26972973e-01 2.79802084e-01
-2.95883626e-01 8.41395736e-01 5.01345051e-03 1.35741513e-02
3.74625504e-01 -2.56114811e-01 5.21714985e-01 1.45460081e+00
6.58908844e-01 -5.10748804e-01 2.24739939e-01 4.43816096e-01
-4.74284500e-01 8.37813139e-01 -5.82425475e-01 -5.29854484e-02
3.50817472e-01 1.19057846e+00 -4.31920201e-01 -4.99859959e-01
-4.30851132e-01 1.34288168e+00 7.64787674e-01 4.50863719e-01
-5.51143110e-01 -1.15008958e-01 2.00007394e-01 -4.86382782e-01
9.28772911e-02 -3.25049371e-01 5.43733884e-04 -1.46629286e+00
-4.61701378e-02 -1.29632020e+00 3.15632492e-01 -5.36404073e-01
-1.31313336e+00 9.89941418e-01 -7.83712491e-02 -1.39723933e+00
-6.66150928e-01 -5.62654257e-01 -8.47845748e-02 1.36651039e+00
-1.61860454e+00 -1.72416902e+00 4.77257997e-01 5.22807300e-01
9.14245129e-01 -8.06178510e-01 1.00947618e+00 5.68760753e-01
-5.30204713e-01 9.37343597e-01 6.53797328e-01 3.98593508e-02
1.10611081e+00 -9.36834276e-01 8.52382362e-01 6.51125908e-01
4.32342708e-01 6.41738772e-01 3.25215787e-01 -6.59318209e-01
-1.54009593e+00 -1.40618122e+00 1.45380282e+00 -5.10921836e-01
8.82082820e-01 -5.25288641e-01 -5.81783116e-01 9.31028724e-01
6.97200894e-01 -2.51573443e-01 5.50818324e-01 -4.91293706e-02
-4.48521793e-01 1.29982948e-01 -1.01568925e+00 8.86455536e-01
1.06235015e+00 -6.21658504e-01 -5.35591066e-01 5.54649472e-01
8.82310033e-01 -4.80397910e-01 -1.00796068e+00 4.20903295e-01
5.73230505e-01 -4.24195290e-01 7.75156200e-01 -1.02563798e+00
4.45987374e-01 1.60535097e-01 -3.28108728e-01 -1.85628402e+00
-3.10874999e-01 -6.47887111e-01 -2.10470445e-02 1.34932101e+00
9.29194987e-01 -8.29948008e-01 1.17437281e-02 -4.43405271e-01
-3.31175923e-01 -6.03994250e-01 -1.28582644e+00 -1.16175759e+00
9.67588902e-01 -1.29647136e-01 6.31101847e-01 1.24087083e+00
-6.44109398e-02 6.84773386e-01 -7.65016437e-01 -1.77420169e-01
4.34354007e-01 2.79517695e-02 5.85446656e-01 -9.38662410e-01
-5.53605795e-01 -5.58411121e-01 2.70112723e-01 -5.95993459e-01
5.64140141e-01 -1.55132258e+00 -3.16428363e-01 -1.37967980e+00
2.98306495e-01 -2.77719885e-01 -6.67785108e-01 4.28610593e-01
-1.50515243e-01 4.26757663e-01 3.11034679e-01 4.94347632e-01
-4.53265160e-01 3.59073907e-01 1.18722558e+00 -2.24356860e-01
3.00595295e-02 -7.42001906e-02 -3.03872526e-01 3.82704347e-01
9.84106541e-01 -4.84359205e-01 1.24738947e-01 -8.92092347e-01
3.13751340e-01 -6.46334961e-02 -2.94129431e-01 -5.70102870e-01
1.92709789e-01 -2.40067586e-01 2.99658477e-01 -2.91785330e-01
-4.99139987e-02 -5.93446255e-01 2.73531646e-01 7.26827085e-01
-4.42528814e-01 7.37001300e-01 4.20577228e-01 2.43608244e-02
-1.60175055e-01 -1.29391640e-01 7.33560443e-01 -3.34008813e-01
-4.41801250e-01 2.32325301e-01 -3.36022556e-01 8.35274458e-02
5.18738687e-01 3.42140675e-01 -4.20982957e-01 -8.01246762e-02
-3.49389225e-01 -2.44441889e-02 3.00212264e-01 7.52741635e-01
3.73636186e-01 -1.62014031e+00 -1.36628008e+00 5.11114821e-02
3.97272885e-01 -5.77670217e-01 6.71456084e-02 1.02322280e+00
-4.50648695e-01 6.29080117e-01 -4.42409843e-01 -2.82451183e-01
-7.35524833e-01 5.06154180e-01 2.36075029e-01 -6.76821232e-01
-3.04651648e-01 3.39283198e-01 -3.26639533e-01 -1.12421060e+00
-3.16430211e-01 6.37762994e-02 1.17371544e-01 -9.26838368e-02
1.28851876e-01 4.52653885e-01 3.63179058e-01 -9.52986836e-01
-1.34426951e-01 5.02675712e-01 -2.34339431e-01 -6.86258018e-01
1.07666612e+00 -3.15251052e-01 -2.97848731e-01 8.93787086e-01
1.21477425e+00 2.30379894e-01 -3.44257683e-01 -6.22009039e-01
3.53686899e-01 -4.27247360e-02 -3.82444620e-01 -1.12379694e+00
-6.22478366e-01 9.20210481e-01 6.52316391e-01 -5.12569308e-01
1.00747085e+00 2.83034835e-02 9.63447154e-01 4.85655785e-01
8.86637747e-01 -1.29935360e+00 -3.41458231e-01 9.78124201e-01
8.28789353e-01 -1.11352873e+00 -2.41646916e-01 2.59281188e-01
-5.16073525e-01 1.20655906e+00 3.88746798e-01 3.35279822e-01
2.61645354e-02 1.50501668e-01 2.87360609e-01 4.78247166e-01
-1.04061651e+00 -1.11552618e-01 3.56562316e-01 3.51951361e-01
7.28868067e-01 3.62248033e-01 -7.89689541e-01 3.13000172e-01
-3.92772287e-01 -2.36348793e-01 3.47923070e-01 7.40992367e-01
-3.08155626e-01 -1.58273947e+00 -2.52515048e-01 2.41272017e-01
-5.85364521e-01 -6.12629592e-01 -4.58301216e-01 9.84872758e-01
-6.94246888e-02 8.52764308e-01 -1.68584332e-01 -3.97329062e-01
1.73112378e-01 4.49659735e-01 6.92103148e-01 -6.66692257e-01
-9.61717427e-01 2.64086515e-01 1.52757779e-01 3.42344232e-02
-3.91833514e-01 -5.72523415e-01 -7.08615065e-01 -1.52240857e-01
-9.17749628e-02 2.47779369e-01 8.12793732e-01 1.08030570e+00
1.88014954e-01 3.14380497e-01 8.65619063e-01 -5.77254593e-01
-6.89212561e-01 -1.61551142e+00 -1.22017868e-01 7.19235763e-02
2.19897144e-02 -2.50012428e-01 -1.92612201e-01 -1.45865738e-01] | [11.609410285949707, 10.316668510437012] |
b2418421-8ad3-428c-a0c7-060a68d0584a | conformal-quantitative-predictive-monitoring | 2211.02375 | null | https://arxiv.org/abs/2211.02375v2 | https://arxiv.org/pdf/2211.02375v2.pdf | Conformal Quantitative Predictive Monitoring of STL Requirements for Stochastic Processes | We consider the problem of predictive monitoring (PM), i.e., predicting at runtime the satisfaction of a desired property from the current system's state. Due to its relevance for runtime safety assurance and online control, PM methods need to be efficient to enable timely interventions against predicted violations, while providing correctness guarantees. We introduce \textit{quantitative predictive monitoring (QPM)}, the first PM method to support stochastic processes and rich specifications given in Signal Temporal Logic (STL). Unlike most of the existing PM techniques that predict whether or not some property $\phi$ is satisfied, QPM provides a quantitative measure of satisfaction by predicting the quantitative (aka robust) STL semantics of $\phi$. QPM derives prediction intervals that are highly efficient to compute and with probabilistic guarantees, in that the intervals cover with arbitrary probability the STL robustness values relative to the stochastic evolution of the system. To do so, we take a machine-learning approach and leverage recent advances in conformal inference for quantile regression, thereby avoiding expensive Monte-Carlo simulations at runtime to estimate the intervals. We also show how our monitors can be combined in a compositional manner to handle composite formulas, without retraining the predictors nor sacrificing the guarantees. We demonstrate the effectiveness and scalability of QPM over a benchmark of four discrete-time stochastic processes with varying degrees of complexity. | ['Luca Bortolussi', 'Nicola Paoletti', 'Francesca Cairoli'] | 2022-11-04 | null | null | null | null | ['prediction-intervals'] | ['miscellaneous'] | [ 4.73941743e-01 2.79103428e-01 -3.58987957e-01 -1.80552989e-01
-1.22704017e+00 -6.20596528e-01 6.58349335e-01 4.81777966e-01
1.93237692e-01 6.95112109e-01 -2.77656019e-01 -7.89608300e-01
-6.17052734e-01 -8.68917465e-01 -8.82055342e-01 -4.72470790e-01
-7.02694356e-01 4.03584778e-01 5.17592430e-01 2.41144672e-01
1.35251373e-01 5.42134881e-01 -1.40252817e+00 2.93491662e-01
3.21043611e-01 1.48195148e+00 -7.05655038e-01 8.51861417e-01
5.84941328e-01 1.01333582e+00 -3.77619773e-01 -1.83218971e-01
9.13732946e-02 -1.75099671e-01 -3.66455913e-01 -2.19792798e-01
-1.56268314e-01 -2.53206730e-01 5.97633757e-02 1.01411915e+00
-8.89859647e-02 -2.06406981e-01 6.26864731e-01 -1.86883080e+00
3.33951712e-01 8.86175513e-01 -4.98774052e-01 -6.20372519e-02
4.62062538e-01 4.15927112e-01 1.04721129e+00 -2.15923563e-01
3.91815007e-02 1.25869286e+00 7.57013977e-01 3.02034080e-01
-1.85165501e+00 -5.36155641e-01 2.77684778e-01 -2.42596731e-01
-1.36717236e+00 -4.23623979e-01 4.27435189e-01 -6.24772906e-01
1.01409960e+00 4.45662349e-01 2.71498173e-01 9.00249302e-01
8.13607633e-01 4.73026991e-01 1.16171336e+00 -2.37087533e-01
7.44044662e-01 -1.37849793e-01 2.46750131e-01 4.17428911e-01
2.75839746e-01 7.36856937e-01 -4.98466194e-01 -9.01780128e-01
3.29352319e-01 2.96443217e-02 -2.70459056e-02 -2.71737725e-01
-1.08775091e+00 4.47442085e-01 -6.90570712e-01 -3.25281531e-01
-2.21236154e-01 8.62603426e-01 6.11433864e-01 4.73456323e-01
2.95954287e-01 1.81392625e-01 -8.70352030e-01 -4.28810090e-01
-8.35417271e-01 3.12163025e-01 1.15555990e+00 9.93248224e-01
3.55119407e-01 1.09103076e-01 -5.21372855e-01 -2.33628646e-01
4.74843055e-01 8.08169186e-01 -3.17523539e-01 -1.27757990e+00
2.79185712e-01 2.69924998e-01 6.45906389e-01 -5.61974466e-01
-1.23816714e-01 -2.05777779e-01 -4.68586177e-01 4.94985431e-01
3.17314506e-01 -1.84382796e-01 -4.60092753e-01 1.79021811e+00
1.37293518e-01 4.23859119e-01 -1.91728324e-01 1.39609575e-01
-7.80938506e-01 8.16818953e-01 -6.88291043e-02 -8.53448272e-01
1.00270474e+00 -1.16816312e-02 -4.66082692e-01 -2.60205884e-02
4.53911722e-01 -2.27205083e-01 8.57209802e-01 8.08520436e-01
-1.06605065e+00 6.29716367e-02 -1.14432693e+00 9.71629560e-01
3.54378998e-01 -5.45131505e-01 3.34856927e-01 8.57789159e-01
-8.49528193e-01 8.56101811e-01 -1.49522102e+00 4.98519510e-01
1.35159433e-01 4.24332887e-01 1.72663048e-01 4.79631007e-01
-9.08091187e-01 6.63976967e-01 7.79594183e-02 -6.77176416e-02
-1.46413183e+00 -9.78261769e-01 -7.15813041e-01 2.25151941e-01
7.05023706e-01 -2.09686443e-01 1.71060586e+00 -5.05722046e-01
-1.56348503e+00 8.22404698e-02 -1.23809427e-01 -8.72244239e-01
6.65230453e-01 -1.02549128e-01 -5.48662782e-01 6.79179132e-02
-1.96697906e-01 -2.61061043e-01 1.04745781e+00 -8.81823540e-01
-7.79244602e-01 -1.81324214e-01 6.31581172e-02 -7.28920400e-01
1.73670575e-01 1.50918663e-01 2.07464740e-01 -1.82650443e-02
-1.49041772e-01 -9.07752395e-01 -4.99452949e-01 -2.60357797e-01
-5.32883823e-01 -9.67358649e-02 4.56251770e-01 -3.73040378e-01
1.35028720e+00 -2.03255272e+00 -2.72525102e-01 8.42958570e-01
-1.49465427e-01 -2.54909456e-01 4.48390752e-01 7.78068244e-01
7.35632479e-02 6.43870607e-02 -4.91889417e-01 -2.39114210e-01
5.94619811e-01 1.24999024e-01 -9.30687308e-01 7.50525534e-01
3.91487658e-01 5.95921218e-01 -7.05472052e-01 -3.53231102e-01
2.40289897e-01 -1.08335115e-01 -3.95484149e-01 1.78304344e-01
-9.51979637e-01 3.66532266e-01 -3.58536661e-01 5.58724225e-01
1.61476716e-01 -1.37469381e-01 3.51005435e-01 3.21406603e-01
-5.79024628e-02 2.39781186e-01 -1.39808977e+00 7.36276627e-01
-6.33268714e-01 1.09573901e-01 2.70042755e-02 -6.78540766e-01
7.40815282e-01 4.29803580e-01 5.64253569e-01 -3.57549012e-01
1.64295033e-01 8.39321762e-02 -4.01198000e-01 6.84600277e-03
8.30608085e-02 -3.17695528e-01 -5.24940133e-01 8.32771242e-01
-5.90681851e-01 -2.86312521e-01 -1.30214468e-01 -1.30975544e-01
1.49640024e+00 2.76171297e-01 4.18890834e-01 -3.69141459e-01
4.60486978e-01 -5.83864190e-02 9.11224782e-01 8.24613690e-01
-1.03432678e-01 -2.06731066e-01 1.27968669e+00 -1.23522595e-01
-9.60741043e-01 -1.26615012e+00 9.70626697e-02 8.70754957e-01
-1.47298098e-01 -3.67031455e-01 -4.58454490e-01 -3.02708238e-01
2.19746366e-01 1.19576442e+00 -4.94277894e-01 -3.99112731e-01
-3.93258065e-01 -4.29288030e-01 5.80186307e-01 5.90889812e-01
-2.62779534e-01 -5.22398353e-01 -8.58543396e-01 4.05887574e-01
5.33741117e-01 -8.97254944e-01 -4.02850986e-01 3.46328080e-01
-7.46318042e-01 -1.08949482e+00 5.05642354e-01 2.71432698e-02
3.95553023e-01 -5.38319826e-01 1.01316679e+00 -2.83173025e-01
3.36862542e-02 6.69448733e-01 2.80997843e-01 -4.17073876e-01
-9.49793935e-01 -6.20373428e-01 1.58420652e-01 1.97074842e-02
-6.84833899e-02 -8.64120364e-01 -1.52539060e-01 3.64550829e-01
-8.46411049e-01 -2.66231716e-01 1.21636882e-01 6.69812500e-01
9.66238737e-01 4.92312163e-01 4.68726546e-01 -7.96132803e-01
6.53618455e-01 -3.39680552e-01 -1.55706418e+00 3.58743012e-01
-9.79778647e-01 2.50402242e-01 9.14749444e-01 -5.61961830e-01
-7.16048062e-01 6.50760233e-02 2.77351528e-01 -3.97647053e-01
6.55647367e-02 5.32768965e-01 -2.14534938e-01 4.45519686e-01
3.68255615e-01 1.10498078e-01 -2.22296193e-02 1.43115848e-01
1.72336176e-01 3.12951058e-01 5.80838084e-01 -1.27680695e+00
6.80581868e-01 5.22408068e-01 6.52710378e-01 -5.65337911e-02
-7.89054811e-01 -4.13880944e-02 -1.62595529e-02 -1.95924848e-01
1.22060150e-01 -5.66717267e-01 -1.80974495e+00 2.07023062e-02
-7.68753350e-01 -5.57512164e-01 -5.60983181e-01 2.01763079e-01
-1.14947438e+00 1.26885474e-01 -6.92820311e-01 -1.84698260e+00
-2.27408946e-01 -9.82023954e-01 1.19772589e+00 -4.59377170e-01
-5.14903367e-01 -7.15040505e-01 2.16761678e-01 -4.07212138e-01
1.97020218e-01 7.80279636e-01 1.06091261e+00 -7.02665567e-01
-6.18318617e-01 -6.63911402e-01 1.44257262e-01 3.50328863e-01
-1.86717629e-01 4.04045403e-01 -9.23865259e-01 -3.74478996e-01
2.83850610e-01 7.67290518e-02 2.32712448e-01 3.02896112e-01
1.17951858e+00 -8.24861825e-01 -2.78424531e-01 5.78459054e-02
1.44685245e+00 3.00765663e-01 3.83888990e-01 2.50871778e-02
-1.75079644e-01 3.71329606e-01 9.23275709e-01 1.01163948e+00
-3.33150961e-02 3.70878249e-01 6.06261492e-01 7.51502991e-01
7.15679467e-01 -4.61500973e-01 9.02666450e-01 1.74429610e-01
3.53570431e-01 2.86702633e-01 -1.01700521e+00 4.44328398e-01
-1.87592566e+00 -1.14596629e+00 2.79000737e-02 2.84529471e+00
1.15234101e+00 8.52012336e-01 2.31368214e-01 4.40177530e-01
5.15378892e-01 -2.78531343e-01 -7.05965281e-01 -7.37088680e-01
4.44151759e-01 3.90934795e-01 8.76815617e-01 7.41970718e-01
-8.31870496e-01 1.36609510e-01 6.38911057e+00 7.56326318e-01
-8.08469057e-01 -3.16424668e-02 7.29470372e-01 -2.11258624e-02
-4.03131485e-01 4.32864070e-01 -8.32065821e-01 4.97522503e-01
1.88227284e+00 -5.34973025e-01 4.38706487e-01 1.07820070e+00
5.81991673e-01 -1.17690243e-01 -1.61625004e+00 4.72103089e-01
-7.25589812e-01 -1.06586838e+00 -6.29835963e-01 6.27896935e-02
6.67810619e-01 -3.27002048e-01 7.06637949e-02 3.42586547e-01
7.60780215e-01 -1.06128764e+00 1.03584814e+00 8.46497715e-01
7.40665257e-01 -1.25944829e+00 5.34694970e-01 6.35897636e-01
-1.05438054e+00 -4.30807233e-01 2.26806194e-01 -2.09851354e-01
3.45793843e-01 1.05179358e+00 -8.41718674e-01 3.31758231e-01
3.67624909e-01 2.52967954e-01 -7.77090061e-03 6.25315964e-01
-3.72625947e-01 1.00118101e+00 -6.43035889e-01 -1.45059424e-02
-2.78449476e-01 6.77483454e-02 5.96546769e-01 1.00163007e+00
2.56374836e-01 -3.63577068e-01 2.83762306e-01 9.10991490e-01
5.30489147e-01 -6.09794557e-01 -3.91492367e-01 -9.62183177e-02
7.19942391e-01 7.57379174e-01 -3.76013666e-01 -3.09830248e-01
-1.91098258e-01 1.54308379e-01 -2.36103430e-01 2.35707119e-01
-1.22968459e+00 -6.87952116e-02 7.29390621e-01 1.11306667e-01
2.45853499e-01 -4.87744063e-01 -6.42031372e-01 -5.82792401e-01
2.40704566e-01 -9.81452346e-01 5.57040393e-01 -3.65705699e-01
-1.29922080e+00 1.55802816e-01 2.54629403e-01 -1.21877027e+00
-7.31524706e-01 -4.35332686e-01 -4.39952046e-01 7.32563794e-01
-1.05571020e+00 -8.23140204e-01 5.09806275e-01 4.23942387e-01
-4.51915897e-02 4.62008566e-01 7.94584513e-01 -3.50633025e-01
-5.74298859e-01 3.19493473e-01 -3.95557955e-02 -5.13549149e-01
3.68325412e-01 -1.32208848e+00 3.73965174e-01 1.03180945e+00
-4.56926376e-01 5.58424532e-01 1.32149458e+00 -7.53699243e-01
-1.88155615e+00 -1.27742064e+00 4.13026780e-01 -4.89006191e-01
1.50872731e+00 -2.85736024e-01 -5.70568621e-01 9.38069582e-01
-3.47745299e-01 3.81210260e-02 3.50710422e-01 9.25157592e-02
-6.20178580e-01 -4.39631939e-01 -1.15931249e+00 6.26837432e-01
6.50480568e-01 -6.76768541e-01 -4.36381012e-01 1.95897534e-01
1.01043344e+00 -1.98626459e-01 -1.38484633e+00 6.38784230e-01
5.68319917e-01 -7.97084689e-01 5.78794777e-01 -5.96255720e-01
-1.40794180e-03 -7.13018537e-01 -5.20319283e-01 -7.17863321e-01
1.24561533e-01 -1.28057384e+00 -8.65635812e-01 1.17054963e+00
5.62873185e-01 -9.47056055e-01 4.07415509e-01 1.11553299e+00
-9.31619182e-02 -8.21662009e-01 -1.36202776e+00 -1.22665620e+00
-1.47908762e-01 -1.24448287e+00 8.37200999e-01 3.83423895e-01
4.12188739e-01 -2.53660411e-01 -3.75947148e-01 7.47869074e-01
7.80437589e-01 5.11583567e-01 4.34168369e-01 -9.25663829e-01
-9.76833522e-01 -4.75567758e-01 -4.20151293e-01 -3.78401339e-01
2.06168935e-01 -3.50605369e-01 3.96702796e-01 -7.30572581e-01
1.61667764e-02 -2.78021485e-01 -4.93949860e-01 6.13166869e-01
3.00703645e-01 -3.54546547e-01 -2.85752624e-01 -3.84313986e-02
-7.63338387e-01 5.24477959e-01 3.95942599e-01 -1.39452800e-01
-1.18145414e-01 6.16194546e-01 -4.58675057e-01 5.42065859e-01
7.32094049e-01 -4.12575871e-01 -4.42613780e-01 4.09174442e-01
7.09805906e-01 7.72378266e-01 5.85325897e-01 -1.01154947e+00
1.77192032e-01 -7.43867934e-01 -1.86979309e-01 -6.35211051e-01
9.74384919e-02 -9.09173667e-01 5.58803618e-01 7.64499366e-01
-5.53022206e-01 1.06473364e-01 2.56042868e-01 9.94111896e-01
-1.50847966e-02 2.15176299e-01 6.68407977e-01 4.87277299e-01
-1.98993891e-01 3.07314157e-01 -6.38510823e-01 -5.94967939e-02
1.22590828e+00 2.60650218e-01 -2.46410325e-01 -4.28936571e-01
-6.88742638e-01 5.12306929e-01 5.19741476e-01 -1.70141935e-01
4.27509189e-01 -9.64274824e-01 -3.76960725e-01 -3.22755910e-02
1.20972052e-01 -1.62679300e-01 1.38463057e-03 1.16249561e+00
1.02023967e-01 2.98552305e-01 4.27969813e-01 -7.14652479e-01
-7.68361330e-01 8.74953926e-01 3.63089085e-01 -4.49620157e-01
-4.31185275e-01 2.15360269e-01 -1.89722940e-01 7.90927708e-02
2.92780787e-01 -6.82446122e-01 6.96411073e-01 -5.48949778e-01
8.00847232e-01 5.18765748e-01 1.03374675e-01 3.39749277e-01
-5.23849130e-01 -1.87380135e-03 4.27879483e-01 -5.81821620e-01
1.13645017e+00 -4.63658981e-02 -3.90592277e-01 8.79484415e-01
5.59302568e-01 2.26348132e-01 -1.65502715e+00 -2.51945369e-02
7.22427785e-01 -1.09409556e-01 -3.49582762e-01 -7.66670406e-01
-5.34591377e-01 5.79051435e-01 2.21176490e-01 4.91500020e-01
1.20736790e+00 -1.28726661e-01 4.75337952e-01 3.38012695e-01
9.13554609e-01 -9.82750654e-01 -5.48938930e-01 3.52146894e-01
5.47748446e-01 -4.28308815e-01 5.33383377e-02 -3.07236701e-01
-4.35922563e-01 9.20460582e-01 1.45252690e-01 -1.99888632e-01
7.29619443e-01 1.02539086e+00 -6.55818582e-01 2.68487304e-01
-1.47933745e+00 3.91894728e-01 -1.71977803e-01 4.70157951e-01
-1.60684884e-02 4.72032994e-01 1.10807650e-01 9.82730806e-01
7.91693479e-02 5.63940499e-03 6.37973189e-01 1.21424687e+00
-5.28333008e-01 -1.04116416e+00 -4.81282920e-01 5.60298026e-01
-4.75152314e-01 1.73777252e-01 7.17631131e-02 5.55181086e-01
-4.32498574e-01 1.18840981e+00 -2.34228715e-01 -4.13471401e-01
2.90747643e-01 1.80566952e-01 2.83877760e-01 -4.67727393e-01
-2.07160681e-01 7.15960488e-02 4.48043704e-01 -9.99381483e-01
1.73167691e-01 -8.45312476e-01 -1.20845890e+00 -4.10278380e-01
-6.05982691e-02 1.87223449e-01 4.38163131e-01 1.11019051e+00
2.65575320e-01 4.10418361e-01 9.60344672e-01 -3.43539089e-01
-1.49203944e+00 -5.54275215e-01 -7.22666800e-01 -2.14090273e-01
2.98452199e-01 -5.02356112e-01 -4.54468936e-01 -5.18287160e-02] | [4.753530979156494, 2.2891292572021484] |
cb713d67-1de3-4a15-a43f-6201fcb7a8d0 | bggan-bokeh-glass-generative-adversarial | 2011.02242 | null | https://arxiv.org/abs/2011.02242v1 | https://arxiv.org/pdf/2011.02242v1.pdf | BGGAN: Bokeh-Glass Generative Adversarial Network for Rendering Realistic Bokeh | A photo captured with bokeh effect often means objects in focus are sharp while the out-of-focus areas are all blurred. DSLR can easily render this kind of effect naturally. However, due to the limitation of sensors, smartphones cannot capture images with depth-of-field effects directly. In this paper, we propose a novel generator called Glass-Net, which generates bokeh images not relying on complex hardware. Meanwhile, the GAN-based method and perceptual loss are combined for rendering a realistic bokeh effect in the stage of finetuning the model. Moreover, Instance Normalization(IN) is reimplemented in our network, which ensures our tflite model with IN can be accelerated on smartphone GPU. Experiments show that our method is able to render a high-quality bokeh effect and process one $1024 \times 1536$ pixel image in 1.9 seconds on all smartphone chipsets. This approach ranked First in AIM 2020 Rendering Realistic Bokeh Challenge Track 1 \& Track 2. | ['Jian Cheng', 'Cong Leng', 'Chenghua Li', 'Zhenyu Guo', 'Jiamin Lin', 'Congyu Qiao', 'Ming Qian'] | 2020-11-04 | null | null | null | null | ['bokeh-effect-rendering'] | ['computer-vision'] | [ 4.97317344e-01 2.64681317e-02 5.64374804e-01 -2.28352889e-01
-5.20243883e-01 -3.69043231e-01 5.17793536e-01 -4.69168603e-01
-1.56531841e-01 4.78532284e-01 -3.22315134e-02 -2.61062622e-01
2.39118740e-01 -9.66072381e-01 -8.50660920e-01 -5.66935718e-01
2.98868895e-01 -2.04802066e-01 3.66024584e-01 -2.17723206e-01
1.31105825e-01 2.71046579e-01 -1.57564461e+00 3.11934173e-01
8.07303429e-01 1.09021854e+00 7.26192057e-01 9.99266267e-01
1.24910712e-01 9.21741426e-01 -8.04704010e-01 -3.88006270e-01
4.99632567e-01 -5.73956192e-01 -1.56958818e-01 -1.35612771e-01
9.05871272e-01 -9.16278362e-01 -3.61778408e-01 8.78378153e-01
7.90255010e-01 2.75190845e-02 1.88092932e-01 -1.09388900e+00
-1.16464779e-01 1.09836869e-01 -8.79586637e-01 1.08335435e-01
3.35273951e-01 6.66395068e-01 3.21487010e-01 -7.64514565e-01
3.85571748e-01 1.31127226e+00 6.23931587e-01 5.40555775e-01
-9.90292370e-01 -9.51432645e-01 -1.30676359e-01 6.69486597e-02
-1.19717252e+00 -5.42765915e-01 6.53443813e-01 -1.06900539e-02
8.73812675e-01 6.35632217e-01 7.18952179e-01 8.20793867e-01
4.12481129e-01 4.25707489e-01 1.26273537e+00 -2.58349717e-01
1.43575326e-01 -7.84397647e-02 -4.01567429e-01 5.07887363e-01
3.98548879e-02 2.25390613e-01 -5.92012346e-01 1.21521629e-01
1.09898019e+00 -2.26666611e-02 -4.38672602e-01 2.07309037e-01
-1.09333634e+00 9.42214206e-02 5.52189052e-01 -1.48269564e-01
-2.09893718e-01 8.04914534e-01 2.40050945e-02 -1.50566071e-01
5.10607958e-01 3.01550895e-01 2.22478490e-02 -3.23423505e-01
-1.18955994e+00 1.60857543e-01 1.82561293e-01 9.75075483e-01
7.00249434e-01 1.15541331e-01 -3.73447716e-01 7.60618746e-01
4.34323885e-02 1.04656363e+00 1.86569795e-01 -1.28466392e+00
5.04230142e-01 2.40089133e-01 3.05176586e-01 -1.00564861e+00
-2.37297773e-01 -2.80365199e-01 -1.29977775e+00 5.67178965e-01
1.40199289e-01 -2.58082420e-01 -8.60934258e-01 1.43637896e+00
3.63879144e-01 5.11535347e-01 -3.04092407e-01 1.07206368e+00
7.86512315e-01 1.11456501e+00 -3.49421650e-01 -1.23234063e-01
1.32900596e+00 -1.18924308e+00 -8.58088672e-01 -2.96726286e-01
1.96345061e-01 -1.09548342e+00 1.42974222e+00 8.15568745e-01
-1.50778949e+00 -7.95617580e-01 -1.25782216e+00 -4.39744920e-01
2.58773386e-01 8.68560746e-02 5.72883844e-01 8.18556488e-01
-1.20485711e+00 2.91590214e-01 -7.32531130e-01 7.51923248e-02
4.20574754e-01 1.20929025e-01 2.58762062e-01 -4.40306425e-01
-7.75258780e-01 5.26928782e-01 -6.71339408e-02 1.61131188e-01
-7.84114122e-01 -1.13696516e+00 -4.50895429e-01 2.02287167e-01
2.94360012e-01 -8.21521521e-01 1.14619923e+00 -8.72357190e-01
-1.96183169e+00 5.65331757e-01 -1.96425974e-01 -4.20940608e-01
6.44944549e-01 -2.40476921e-01 -4.52196926e-01 1.46300852e-01
-1.79676086e-01 7.97320962e-01 9.94008124e-01 -1.12760282e+00
-6.49609864e-01 -1.80142120e-01 2.04466656e-01 3.06424767e-01
-1.89180955e-01 6.23155497e-02 -7.37295747e-01 -5.49020827e-01
-2.22041905e-01 -7.93298423e-01 -2.35696465e-01 1.25250310e-01
-3.87102365e-01 3.64304245e-01 9.93724585e-01 -5.45928895e-01
1.19626355e+00 -2.35720229e+00 -5.80400527e-01 -1.66970938e-02
4.26764637e-01 4.18393105e-01 -6.69918358e-02 2.10149512e-01
1.50851294e-01 9.93141998e-03 -8.07833970e-02 -5.73156118e-01
-9.86249372e-02 -1.71632618e-01 -4.80984539e-01 1.99306905e-01
-2.62681454e-01 9.17842388e-01 -7.90996790e-01 -1.64575040e-01
6.84453726e-01 6.66164160e-01 -6.17494524e-01 2.52882749e-01
-1.77404866e-01 4.46395159e-01 -5.10771684e-02 1.82367906e-01
1.43061841e+00 -1.90087482e-01 4.36574034e-02 -2.78688550e-01
-2.87881732e-01 3.13448697e-01 -1.07684743e+00 1.81310916e+00
-1.08568823e+00 8.32445562e-01 2.68617421e-01 -1.77801162e-01
6.56332612e-01 -7.08946735e-02 5.71983494e-02 -1.21094990e+00
4.45803888e-02 1.96550488e-01 -2.51414686e-01 -9.59861651e-02
6.83816135e-01 7.93235004e-02 6.36704490e-02 3.38632077e-01
-5.44958532e-01 -6.56685650e-01 -1.96890831e-02 2.51413435e-01
1.15365255e+00 -3.55831310e-02 -1.40773863e-01 -9.81490687e-02
1.95292592e-01 -4.31850910e-01 3.64778161e-01 7.50228107e-01
2.45700523e-01 1.07714629e+00 2.47265935e-01 -2.51637727e-01
-1.18656981e+00 -1.06789339e+00 1.60005599e-01 6.32164717e-01
5.03560364e-01 -6.33715093e-01 -1.14386380e+00 -2.10273847e-01
-6.18337214e-01 7.28090465e-01 -2.07424775e-01 8.28321800e-02
-6.43420756e-01 -6.28018379e-01 4.83932793e-01 2.62363195e-01
1.16492724e+00 -9.61179256e-01 -1.15126872e+00 1.14377417e-01
-2.47741267e-01 -1.16090608e+00 -7.48807132e-01 -5.26232004e-01
-5.56234896e-01 -8.50918472e-01 -9.27346766e-01 -3.57193798e-01
5.12615919e-01 6.43193245e-01 1.27649915e+00 8.91543254e-02
-3.73202235e-01 1.09056093e-01 -8.38937908e-02 -2.92209238e-01
-3.42164814e-01 -3.45288724e-01 -4.25699413e-01 2.26697028e-02
-2.21714959e-01 -6.91451192e-01 -1.32674420e+00 3.89196068e-01
-1.05208433e+00 6.40829980e-01 4.69015300e-01 5.33608556e-01
5.05625308e-01 4.87245023e-01 -4.63233655e-03 -5.31828165e-01
3.27447206e-01 9.25620794e-02 -9.73724246e-01 -2.71748733e-02
-4.60154384e-01 -3.73068869e-01 7.72943616e-01 -3.27187955e-01
-1.50366747e+00 -2.05736235e-01 -1.12716027e-01 -4.18442339e-01
8.22707415e-02 -2.27889240e-01 -4.68429029e-01 7.33473971e-02
6.48724377e-01 1.29709750e-01 -2.17687920e-01 -2.76165307e-01
3.26546222e-01 6.86759531e-01 5.65017462e-01 -1.50137633e-01
6.47434473e-01 8.78096223e-01 2.21654952e-01 -8.83295476e-01
-4.20918882e-01 -1.67894900e-01 3.93036395e-01 -4.08045739e-01
7.48099029e-01 -1.13910830e+00 -8.03194702e-01 1.00286806e+00
-1.13251519e+00 -8.55057418e-01 -3.79240781e-01 1.65596217e-01
-3.18436533e-01 3.12875628e-01 -5.98457634e-01 -6.62229836e-01
-5.93608439e-01 -1.28794074e+00 1.27816224e+00 4.32668149e-01
2.43853122e-01 -3.13561410e-01 -1.85520783e-01 3.10298264e-01
7.65535057e-01 -6.74453229e-02 3.49645853e-01 8.14604282e-01
-1.09277642e+00 5.42577393e-02 -6.23473048e-01 5.39039433e-01
2.15693861e-02 -1.00081973e-01 -1.27596533e+00 -3.51043403e-01
1.79165244e-01 5.86038120e-02 6.07643366e-01 7.93890893e-01
1.59683013e+00 -2.55144089e-01 -1.48916990e-01 9.51520979e-01
1.54026473e+00 2.71694213e-01 1.40386593e+00 -1.24086730e-01
1.04252458e+00 1.63084999e-01 7.02397764e-01 4.62235123e-01
2.14105576e-01 1.10442507e+00 5.05640924e-01 -5.48167884e-01
-7.45424330e-01 -2.62722760e-01 4.81550962e-01 4.23162043e-01
1.01376668e-01 -7.65744627e-01 -5.94708800e-01 2.34526396e-01
-1.37046790e+00 -8.15062463e-01 -5.47317088e-01 2.33354640e+00
6.81374311e-01 2.79643890e-02 -2.45078772e-01 -9.09429640e-02
6.33908808e-01 3.04125458e-01 -4.05946761e-01 -4.60002393e-01
-2.60502517e-01 5.24790406e-01 8.16070318e-01 7.17988372e-01
-5.58767200e-01 7.58355200e-01 5.18046045e+00 1.46511400e+00
-1.42896593e+00 2.14141995e-01 1.00885177e+00 -3.85939986e-01
-5.42735100e-01 -1.22298658e-01 -6.77380264e-01 8.45707357e-01
9.09058869e-01 2.50059903e-01 5.68758547e-01 5.01265943e-01
6.09598398e-01 -5.11029899e-01 -7.17696071e-01 1.57081342e+00
6.80369139e-02 -1.36751401e+00 -2.83002734e-01 2.18004316e-01
8.55033159e-01 -1.02071814e-01 4.07419235e-01 -7.27285966e-02
2.48431742e-01 -1.03614807e+00 6.44100308e-01 4.76809084e-01
1.45458031e+00 -7.70396233e-01 2.39396557e-01 2.22612754e-01
-9.71755147e-01 2.74583668e-01 -2.43522316e-01 -1.18295543e-01
5.08439779e-01 1.19590414e+00 -7.57878840e-01 3.12556475e-01
9.81138110e-01 3.46198827e-01 -5.94958425e-01 8.24289441e-01
-2.76790142e-01 4.57871318e-01 -6.90798998e-01 1.89955980e-01
-3.53835672e-02 -1.46251738e-01 3.46515298e-01 1.00657022e+00
8.01544845e-01 2.35108629e-01 -3.63090158e-01 7.98261106e-01
-2.24054754e-02 -3.03849250e-01 -3.40363830e-01 4.59565699e-01
3.15905780e-01 1.34219241e+00 -6.79396451e-01 -3.60446811e-01
-1.06391132e-01 1.46978772e+00 -4.71105963e-01 3.21432143e-01
-1.26192212e+00 -2.57179290e-01 5.48067331e-01 5.65483391e-01
1.96552336e-01 -6.21914379e-02 -2.19480783e-01 -1.10622597e+00
2.70997375e-01 -8.61071527e-01 -3.81918490e-01 -1.45771766e+00
-8.05935323e-01 5.18985808e-01 -2.57396370e-01 -1.29357338e+00
-1.32527100e-02 -3.70756119e-01 -4.68221128e-01 8.69229436e-01
-1.47461128e+00 -1.03208709e+00 -8.83551061e-01 5.26402116e-01
4.82478976e-01 5.38227499e-01 5.48123181e-01 6.67200565e-01
-3.96658838e-01 5.50193667e-01 1.34121567e-01 -5.12614429e-01
8.20162952e-01 -9.91672218e-01 8.09215188e-01 1.10951078e+00
-2.50917554e-01 2.45079324e-01 6.16944134e-01 -5.19772947e-01
-1.17864573e+00 -9.62439537e-01 4.86821473e-01 -1.40422717e-01
-2.57545169e-02 -6.69850349e-01 -4.01100904e-01 2.54820436e-01
4.46686149e-01 2.01747879e-01 1.42860219e-01 -5.22302032e-01
-1.00825496e-01 -5.22412300e-01 -1.19468844e+00 7.97955334e-01
1.16006112e+00 -2.58922815e-01 4.41448629e-01 1.43685788e-01
7.83241570e-01 -9.39991355e-01 -3.14144850e-01 3.22622687e-01
4.51819271e-01 -1.62125623e+00 1.22758484e+00 5.59525132e-01
7.34913647e-01 -4.96389210e-01 -3.52879986e-02 -1.19193304e+00
8.23573321e-02 -1.09343839e+00 -3.11628599e-02 1.21796298e+00
-3.83012299e-03 -5.16273141e-01 6.64852381e-01 4.71435755e-01
-2.73311108e-01 -4.74752247e-01 -7.40346432e-01 -5.32036424e-01
-5.32052457e-01 -6.45001411e-01 8.79311740e-01 5.23819208e-01
-5.97770452e-01 3.22851926e-01 -6.20919228e-01 9.26525518e-02
7.03485429e-01 -5.94958253e-02 1.24684465e+00 -5.68391323e-01
-6.03849709e-01 -1.74705103e-01 -1.38714626e-01 -1.67926943e+00
-5.76609671e-01 -1.42084911e-01 6.54256195e-02 -1.29078448e+00
2.02819258e-02 -6.73671305e-01 1.08139291e-01 -1.04313232e-01
-1.89411625e-01 6.26283169e-01 4.62169290e-01 -1.02834232e-01
-4.36023146e-01 3.68225962e-01 1.44458473e+00 1.75699946e-02
-1.18119404e-01 -2.04409227e-01 -5.16521454e-01 7.75604844e-01
6.23510242e-01 -5.03659062e-02 -8.93266797e-01 -7.68243194e-01
4.27030176e-01 8.25928226e-02 7.89402723e-01 -1.29519880e+00
-9.64991301e-02 7.37522319e-02 4.05387014e-01 -6.50590003e-01
6.53552651e-01 -7.29210734e-01 5.84710121e-01 2.10535839e-01
1.42617732e-01 -1.41591758e-01 2.46192575e-01 2.80679077e-01
4.73049507e-02 1.59570441e-01 8.75451148e-01 1.12462640e-01
-5.19926190e-01 2.73138851e-01 -1.91054106e-01 9.21091437e-02
6.97563469e-01 -2.16521055e-01 -5.69243371e-01 -6.74651802e-01
-7.39573091e-02 -1.73118740e-01 7.54078865e-01 7.77093247e-02
6.98476970e-01 -1.13907850e+00 -4.73681748e-01 3.13747108e-01
-3.56219381e-01 4.58833665e-01 1.10099804e+00 6.39005005e-01
-1.03600097e+00 1.12252101e-01 9.34916288e-02 -6.23070180e-01
-1.19467175e+00 4.08684582e-01 3.61237854e-01 -4.76232469e-01
-7.90676594e-01 9.39067006e-01 8.56497884e-01 7.72703066e-02
-8.50305706e-02 -5.16718030e-01 3.38774115e-01 -3.45458508e-01
9.39169705e-01 4.41469759e-01 2.26828307e-01 -2.11540475e-01
-1.72126010e-01 5.29705882e-01 2.04052761e-01 -2.87856132e-01
9.53536510e-01 -4.36843365e-01 8.22820235e-03 -4.93048467e-02
1.07934248e+00 4.82246250e-01 -1.63055611e+00 1.70355439e-01
-1.08060801e+00 -8.38514209e-01 3.42859238e-01 -8.67778838e-01
-1.26260257e+00 9.79385674e-01 9.63774920e-01 9.81822796e-03
1.71427441e+00 -3.99454296e-01 1.42331123e+00 -2.42654085e-01
4.30073291e-01 -9.47750390e-01 3.86108868e-02 5.89795818e-04
7.63824940e-01 -8.09403181e-01 -4.28281277e-02 -7.66793311e-01
-3.22953135e-01 5.75688183e-01 6.42596543e-01 -1.58686325e-01
3.32247406e-01 6.15219057e-01 9.50464688e-04 -1.10576592e-01
-4.55929607e-01 2.24092603e-01 4.34689224e-02 5.30479193e-01
3.54600549e-02 5.17877191e-02 -2.52096713e-01 1.32818609e-01
-4.35672462e-01 5.96349649e-02 7.46518850e-01 4.97294366e-01
-1.41815441e-02 -8.75537515e-01 -4.50235486e-01 1.74199387e-01
-3.49427313e-01 -4.97465104e-01 2.40441740e-01 4.32859808e-01
3.01676095e-01 1.12261581e+00 3.32422405e-01 -1.90451160e-01
3.73985082e-01 -7.49126136e-01 6.71814799e-01 -2.92075336e-01
-4.53747988e-01 2.68877089e-01 -6.64885789e-02 -1.05151761e+00
-2.64707237e-01 -1.04347102e-01 -9.57948208e-01 -6.40473425e-01
-1.34797364e-01 -3.99086803e-01 7.16161966e-01 4.21069384e-01
6.99204206e-01 7.12433696e-01 7.13576138e-01 -1.03483021e+00
1.83452278e-01 -8.32270205e-01 -5.52127063e-01 2.83599973e-01
1.90635502e-01 -2.32392550e-01 -4.01124775e-01 6.66832700e-02] | [10.640488624572754, -2.286220073699951] |
1c62f8ef-3318-4d86-a206-7c47dbdf94d8 | hierarchical-multi-building-and-multi-floor | 2112.12478 | null | https://arxiv.org/abs/2112.12478v1 | https://arxiv.org/pdf/2112.12478v1.pdf | Hierarchical Multi-Building And Multi-Floor Indoor Localization Based On Recurrent Neural Networks | There has been an increasing tendency to move from outdoor to indoor lifestyle in modern cities. The emergence of big shopping malls, indoor sports complexes, factories, and warehouses is accelerating this tendency. In such an environment, indoor localization becomes one of the essential services, and the indoor localization systems to be deployed should be scalable enough to cover the expected expansion of those indoor facilities. One of the most economical and practical approaches to indoor localization is Wi-Fi fingerprinting, which exploits the widely-deployed Wi-Fi networks using mobile devices (e.g., smartphones) without any modification of the existing infrastructure. Traditional Wi-Fi fingerprinting schemes rely on complicated data pre/post-processing and time-consuming manual parameter tuning. In this paper, we propose hierarchical multi-building and multi-floor indoor localization based on a recurrent neural network (RNN) using Wi-Fi fingerprinting, eliminating the need of complicated data pre/post-processing and with less parameter tuning. The RNN in the proposed scheme estimates locations in a sequential manner from a general to a specific one (e.g., building->floor->location) in order to exploit the hierarchical nature of the localization in multi-building and multi-floor environments. The experimental results with the UJIIndoorLoc dataset demonstrate that the proposed scheme estimates building and floor with 100% and 95.24% accuracy, respectively, and provides three-dimensional positioning error of 8.62 m, which outperforms existing deep neural network-based schemes. | ['Kyeong Soo Kim', 'Abdalla Elmokhtar Ahmed Elesawi'] | 2021-12-23 | null | null | null | null | ['indoor-localization'] | ['computer-vision'] | [-8.20150673e-02 -5.03148973e-01 -5.96055463e-02 -4.22371387e-01
-5.55213749e-01 -4.45647061e-01 9.41483155e-02 -2.87585020e-01
-4.39867437e-01 7.74696112e-01 3.00860815e-02 -7.17478156e-01
-4.92270470e-01 -1.15948462e+00 -5.09475172e-01 -7.50940561e-01
-3.90311480e-02 2.74057053e-02 1.71984941e-01 -7.04226121e-02
7.15404898e-02 3.91488075e-01 -1.48688471e+00 -1.98345751e-01
9.17438030e-01 1.30190337e+00 5.98895609e-01 6.95174456e-01
-2.01866757e-02 3.75355393e-01 -6.63808703e-01 4.42017540e-02
-1.63416080e-02 2.82534137e-02 -3.65733385e-01 -4.58249331e-01
3.65681946e-02 -4.20703977e-01 -5.02462447e-01 6.34012938e-01
7.24971771e-01 3.65825236e-01 1.59594715e-01 -9.17670488e-01
-5.05415440e-01 5.82226515e-01 -4.60373670e-01 1.65520594e-01
2.86001652e-01 -5.71615577e-01 4.76110369e-01 -5.94793200e-01
-1.05472825e-01 6.86573088e-01 1.15421295e+00 -3.05757578e-02
-7.99226880e-01 -8.46512079e-01 6.79182913e-03 7.02514173e-03
-2.08438158e+00 -4.40706313e-01 6.15335584e-01 -5.33251353e-02
5.90766668e-01 3.71260107e-01 4.19307768e-01 1.09203315e+00
2.17281520e-01 3.23691249e-01 5.44775188e-01 -3.20844680e-01
1.20512262e-01 1.82402711e-02 -1.69358656e-01 6.30829275e-01
6.14030182e-01 -1.20002896e-01 -1.88519835e-01 -5.22251874e-02
8.08837116e-01 6.68125987e-01 -2.08523303e-01 -4.83315513e-02
-1.29187739e+00 3.07798475e-01 7.15428352e-01 9.05907512e-01
-3.46703291e-01 4.47501481e-01 7.54194036e-02 -2.51224607e-01
6.06864430e-02 2.26501510e-01 -5.20310163e-01 -4.37579095e-01
-1.07851529e+00 -1.32149145e-01 4.67010558e-01 1.19385731e+00
6.21520877e-01 1.08819313e-01 2.85518408e-01 9.06903803e-01
5.06683350e-01 8.73047173e-01 7.32287467e-01 -8.48197758e-01
8.36491883e-01 2.11995617e-01 4.56411213e-01 -1.43726194e+00
-8.54421854e-01 -9.96477067e-01 -1.29657388e+00 -7.39187956e-01
3.33391815e-01 -4.68871176e-01 -4.83977199e-01 1.65402544e+00
2.08574370e-01 4.58819628e-01 -2.89256513e-01 3.26366067e-01
4.28216338e-01 5.09940684e-01 -2.89585412e-01 2.39932999e-01
1.01048934e+00 -7.82828569e-01 -5.30035317e-01 -1.46881148e-01
6.19813621e-01 -5.69264829e-01 9.12768006e-01 2.75538061e-02
-5.03156424e-01 -9.21917260e-01 -1.21880782e+00 3.28425348e-01
-7.06567049e-01 4.86994028e-01 6.13857210e-01 1.29120874e+00
-9.53981876e-01 4.62628424e-01 -8.73076618e-01 -4.60672170e-01
3.23349833e-02 7.60941803e-01 -6.03118949e-02 1.19899511e-02
-1.24909341e+00 2.07026556e-01 2.72262126e-01 6.79459572e-01
1.34398052e-02 -2.64260471e-01 -6.32631660e-01 2.01722771e-01
2.33360771e-02 -5.02934039e-01 1.04822075e+00 -2.82598346e-01
-1.47163546e+00 -1.99116513e-01 -4.78037417e-01 -1.40206560e-01
1.70528129e-01 -1.21017478e-01 -1.03283274e+00 -4.88086998e-01
3.48510057e-01 1.42323047e-01 2.93778121e-01 -1.12863076e+00
-1.10506606e+00 -2.76441395e-01 8.81164521e-02 -1.89676851e-01
-5.09004653e-01 -5.28231561e-01 -3.27447832e-01 -3.66300732e-01
7.55609810e-01 -1.08476365e+00 -2.69510418e-01 -7.83671856e-01
-6.44813359e-01 1.95129782e-01 7.39334822e-01 -6.49426460e-01
1.74922240e+00 -2.02392411e+00 -7.92449653e-01 7.26621151e-01
-5.15559874e-02 -3.38213891e-02 3.26544464e-01 3.87924403e-01
2.44498000e-01 8.84895697e-02 1.81741968e-01 -5.42735219e-01
1.52944848e-01 9.23587754e-02 -4.99888621e-02 4.10956442e-01
-8.12766552e-01 6.10812604e-01 -9.59621370e-01 -2.38445193e-01
4.05072749e-01 6.97005868e-01 -3.58465523e-01 -2.63515353e-01
7.47343183e-01 6.53451800e-01 -6.65526569e-01 8.64523709e-01
7.92443633e-01 -4.20722097e-01 3.12806129e-01 -5.41665219e-02
-5.44575095e-01 3.70902956e-01 -1.49472666e+00 1.58182192e+00
-1.19074035e+00 5.14503777e-01 -2.58616030e-01 -7.45600998e-01
8.90659571e-01 4.27571654e-01 5.41434765e-01 -8.06084931e-01
1.72558665e-01 6.88832879e-01 -2.56414473e-01 -2.70746887e-01
7.17531741e-01 4.99351323e-01 -4.85577226e-01 2.94468641e-01
-4.00523454e-01 6.16783977e-01 -1.64373163e-02 -3.48848134e-01
1.08335817e+00 5.58756478e-02 1.64571509e-01 -4.58649807e-02
8.20410490e-01 -7.47491419e-01 5.91005147e-01 1.15283430e+00
-9.52411816e-03 3.78494859e-01 -5.04694402e-01 -3.44680727e-01
-5.04503787e-01 -9.00826931e-01 -2.51665056e-01 9.83923435e-01
2.94152379e-01 -2.14872688e-01 -5.49419343e-01 -2.77325094e-01
1.45746497e-02 4.57838833e-01 -2.70477235e-01 1.38366014e-01
-7.00472355e-01 -5.85392594e-01 9.04892623e-01 5.70793271e-01
1.27558041e+00 -5.39741457e-01 -4.25368518e-01 4.95038569e-01
-5.44954360e-01 -1.11683512e+00 -3.61115485e-01 2.55304784e-01
-7.09329844e-01 -4.55119610e-01 -7.84439027e-01 -8.84477615e-01
5.98596990e-01 8.77861559e-01 5.99192500e-01 1.13469735e-01
4.11606014e-01 1.60491526e-01 -4.23523299e-02 1.34217329e-02
3.44451249e-01 9.20434475e-01 4.70804811e-01 2.16311455e-01
1.70593470e-01 -9.39760566e-01 -8.23411822e-01 8.35003793e-01
-3.63632560e-01 -2.40159065e-01 5.06557822e-01 6.32700503e-01
3.80065054e-01 6.49106622e-01 6.72835052e-01 -4.66665089e-01
3.11372280e-01 -6.86187744e-01 -6.24069571e-01 2.15519205e-01
-4.57380176e-01 -4.07249987e-01 9.00821328e-01 -1.80194765e-01
-9.17294204e-01 1.06635675e-01 -5.43135881e-01 3.86442751e-01
-2.13195547e-01 5.69085002e-01 -5.02286017e-01 -4.15197372e-01
4.00672555e-01 3.76165003e-01 -9.27025259e-01 -5.29805183e-01
4.51429859e-02 1.22964704e+00 5.13773680e-01 -3.43449205e-01
9.88078892e-01 4.67849076e-01 1.28113227e-02 -9.33519006e-01
-5.23974597e-01 -6.39505088e-01 -6.16165400e-01 -1.86642885e-01
4.15798306e-01 -7.96002448e-01 -1.32379425e+00 3.16138715e-01
-1.02670217e+00 -5.69839366e-02 2.71920413e-01 6.98635578e-01
-1.89474657e-01 1.83152378e-01 -3.12577367e-01 -9.43702102e-01
-1.35567173e-01 -1.01309288e+00 8.28654051e-01 6.14408135e-01
-2.24086449e-01 -8.32965136e-01 -2.89606936e-02 3.86077374e-01
9.30542588e-01 5.63524552e-02 2.97366500e-01 -1.39522240e-01
-8.22738886e-01 -6.56437457e-01 -2.30581582e-01 -1.31612897e-01
6.69437647e-01 -4.73018318e-01 -1.03064740e+00 -1.52069854e-03
-2.00910047e-01 3.89473587e-01 2.71360695e-01 6.31460965e-01
1.47175765e+00 -1.90464914e-01 -7.04811752e-01 9.82165515e-01
1.33594012e+00 7.35071480e-01 6.94737852e-01 7.08082259e-01
7.56175458e-01 1.43166108e-03 3.27919632e-01 4.02973175e-01
7.59392321e-01 7.03981638e-01 2.74404138e-01 -1.65334284e-01
3.41554642e-01 -5.08020341e-01 -1.41193837e-01 9.16815579e-01
-3.44370037e-01 -4.60510641e-01 -8.20039630e-01 3.82784694e-01
-1.84737575e+00 -9.64598119e-01 -8.16264004e-02 2.44529390e+00
1.82175010e-01 1.19256534e-01 -1.47932723e-01 4.37764794e-01
7.47460544e-01 1.68625668e-01 -3.12659502e-01 3.78288189e-03
2.12095559e-01 3.06378305e-02 1.07988882e+00 5.46788335e-01
-1.34197736e+00 5.57081819e-01 5.20145845e+00 7.04181671e-01
-1.17479277e+00 1.79001063e-01 5.04003227e-01 2.68965036e-01
-1.52679430e-02 -5.15622199e-01 -8.89976501e-01 8.85158300e-01
1.17188847e+00 4.53220993e-01 5.96088886e-01 1.05158556e+00
3.80318791e-01 -2.76496410e-01 -4.80703652e-01 1.20261848e+00
-4.61948335e-01 -1.38238752e+00 -7.28434801e-01 3.13606083e-01
6.23664975e-01 1.05679616e-01 2.12080866e-01 4.46310937e-01
-7.58904144e-02 -6.07530117e-01 6.36741936e-01 4.76323396e-01
7.75231481e-01 -9.78567958e-01 1.07846844e+00 3.42552692e-01
-1.94779623e+00 -3.55995417e-01 -1.71075881e-01 -1.35266511e-02
3.30403984e-01 7.49720991e-01 -5.04763484e-01 6.93984270e-01
9.83858347e-01 3.95434767e-01 -3.72351587e-01 1.11594808e+00
-9.95874852e-02 5.26364684e-01 -7.10399270e-01 -2.83322155e-01
1.87986255e-01 -1.26422122e-01 -1.15477055e-01 1.00090778e+00
1.09745073e+00 -3.25470358e-01 -6.21790811e-03 2.48914152e-01
-1.35403797e-01 -3.23689371e-01 -6.35693669e-01 3.97206068e-01
8.49869013e-01 1.11929023e+00 -8.63126874e-01 -7.66234994e-02
-3.05447787e-01 9.49518800e-01 -1.69539750e-01 6.45118892e-01
-1.14861667e+00 -1.10340655e+00 5.63913405e-01 1.79607704e-01
4.31585550e-01 -8.03042591e-01 -3.03948075e-01 -7.93072164e-01
-1.63842030e-02 -1.78317294e-01 -2.24212602e-01 -3.56031656e-01
-6.19293809e-01 4.63170528e-01 -5.69618583e-01 -1.23202491e+00
-2.76693553e-01 -2.25854293e-01 -3.64344209e-01 7.13497281e-01
-1.40990674e+00 -1.07463408e+00 -5.26515603e-01 5.81003487e-01
9.50645581e-02 1.38016403e-01 9.37540710e-01 1.19276857e+00
-8.00781071e-01 9.54596579e-01 8.76433253e-01 3.08325797e-01
3.74744207e-01 -9.02581871e-01 5.62530816e-01 8.34406674e-01
5.29561713e-02 1.21768045e+00 2.62397736e-01 -5.02162278e-01
-1.23416281e+00 -1.17249835e+00 1.43171978e+00 -1.49276271e-01
4.62172031e-01 -6.13642395e-01 -1.82087377e-01 6.96074069e-01
-3.35475206e-01 -6.80591464e-02 8.00669849e-01 4.66328144e-01
1.16109975e-01 -6.41090393e-01 -1.15075648e+00 6.46912813e-01
1.31387019e+00 -5.14523029e-01 1.94798768e-01 1.76243439e-01
3.99104506e-01 -4.02642667e-01 -7.38163114e-01 1.79422304e-01
1.04235899e+00 -1.03572512e+00 1.31420934e+00 5.11687934e-01
-3.35949361e-01 -6.51573896e-01 -6.05693936e-01 -9.11406577e-01
-5.99867940e-01 -5.37027180e-01 -1.48827434e-01 1.36425924e+00
3.59337032e-01 -1.23084819e+00 8.88950586e-01 3.91285956e-01
-1.02395490e-01 -6.92968965e-01 -1.20225585e+00 -8.70278358e-01
-8.27925622e-01 -7.45509088e-01 1.28249443e+00 6.28696203e-01
-4.25812513e-01 1.01106251e-02 -4.90241826e-01 6.92801595e-01
3.81800860e-01 -1.22875325e-01 8.09053540e-01 -1.06788564e+00
-2.72736639e-01 -1.53551176e-01 -3.62758577e-01 -1.74075425e+00
-4.10503387e-01 -4.79388386e-01 1.51169032e-01 -1.74134684e+00
-6.32364690e-01 -1.02571774e+00 -6.42507315e-01 2.45709226e-01
2.98669815e-01 4.20892030e-01 -3.42684358e-01 1.08144052e-01
-7.17540383e-01 3.75791490e-01 5.98019242e-01 -1.92765549e-01
-4.90234673e-01 7.44405091e-01 -6.42024875e-01 8.05777550e-01
9.17680740e-01 -3.68677646e-01 -4.71941561e-01 -5.91965854e-01
3.54791313e-01 -1.71609133e-01 3.64205986e-02 -1.71374750e+00
4.58642453e-01 1.63224578e-01 8.10931921e-01 -7.26216495e-01
5.02529323e-01 -1.27789247e+00 3.38668019e-01 5.03599346e-01
3.21912587e-01 1.43146530e-01 -4.34791483e-02 5.94477534e-01
-1.00148190e-03 3.73616721e-03 2.25806475e-01 2.23952234e-01
-5.43240666e-01 2.41602018e-01 -6.57781482e-01 -5.60643196e-01
6.16896629e-01 -6.21090710e-01 -1.22435242e-01 -5.46243846e-01
-3.74271035e-01 4.31221128e-02 5.10257147e-02 3.03995520e-01
3.89197052e-01 -1.55780065e+00 3.00356120e-01 3.01602989e-01
-1.30942628e-01 -1.68633252e-01 5.28198361e-01 7.02613652e-01
-5.09092391e-01 1.00023031e+00 2.47560576e-01 -5.68328738e-01
-8.44625831e-01 -3.99423689e-02 4.67570961e-01 -3.72908622e-01
-3.03519100e-01 7.85477698e-01 -2.36242652e-01 -8.49157810e-01
4.13172871e-01 -7.24492431e-01 -1.24100000e-01 -2.45098218e-01
5.75367451e-01 7.94994116e-01 1.17332906e-01 -6.37042522e-01
-5.70818305e-01 8.65911663e-01 3.95527035e-01 7.93852285e-02
1.07233143e+00 -7.10810781e-01 2.13509277e-01 5.03551126e-01
1.02061939e+00 3.49577218e-01 -7.41168141e-01 -7.67211020e-02
1.34859368e-01 -4.95675802e-01 6.73971549e-02 -5.62712014e-01
-9.66192663e-01 4.39743042e-01 9.46307659e-01 4.08561200e-01
1.02981246e+00 -4.11554813e-01 1.34697700e+00 7.90936291e-01
1.19956565e+00 -1.10174143e+00 -5.70494711e-01 5.71667671e-01
1.12074964e-01 -9.56155837e-01 -2.14438826e-01 1.88126117e-02
3.04653794e-01 1.04335654e+00 3.07293922e-01 1.97613522e-01
9.72409129e-01 7.36721151e-04 -1.27901286e-01 2.88969547e-01
3.70924979e-01 -8.84261578e-02 -3.63552235e-02 5.85304916e-01
4.24784899e-01 2.77723849e-01 2.32371598e-01 8.59312177e-01
-4.51466769e-01 8.03616121e-02 4.56433818e-02 1.01609743e+00
-5.16137481e-01 -1.09444511e+00 -6.05909586e-01 5.11956394e-01
-2.47661993e-01 5.06899171e-02 5.42599201e-01 6.93855464e-01
7.03916728e-01 1.40830469e+00 -9.91147757e-02 -8.70225251e-01
3.92154038e-01 -1.73919484e-01 5.09626269e-02 -3.10786217e-02
-3.96335483e-01 -6.80444762e-02 -2.31907819e-03 -6.26540899e-01
-2.64843285e-01 -5.21046579e-01 -1.27158475e+00 -6.59220159e-01
-5.04774332e-01 1.91005915e-01 1.08580673e+00 9.71319973e-01
6.50465906e-01 8.90675783e-01 9.47309494e-01 -1.07404137e+00
1.62018865e-01 -7.98170567e-01 -7.61450887e-01 -5.11827469e-01
4.25005674e-01 -7.26500750e-01 -1.73180595e-01 -5.62363207e-01] | [6.402823448181152, 0.9138978123664856] |
bb85a5b4-3464-4fe2-b5d5-6f61aba56af1 | learning-from-the-best-contrastive | 2210.01459 | null | https://arxiv.org/abs/2210.01459v1 | https://arxiv.org/pdf/2210.01459v1.pdf | Learning from the Best: Contrastive Representations Learning Across Sensor Locations for Wearable Activity Recognition | We address the well-known wearable activity recognition problem of having to work with sensors that are non-optimal in terms of information they provide but have to be used due to wearability/usability concerns (e.g. the need to work with wrist-worn IMUs because they are embedded in most smart watches). To mitigate this problem we propose a method that facilitates the use of information from sensors that are only present during the training process and are unavailable during the later use of the system. The method transfers information from the source sensors to the latent representation of the target sensor data through contrastive loss that is combined with the classification loss during joint training. We evaluate the method on the well-known PAMAP2 and Opportunity benchmarks for different combinations of source and target sensors showing average (over all activities) F1 score improvements of between 5% and 13% with the improvement on individual activities, particularly well suited to benefit from the additional information going up to between 20% and 40%. | ['Paul Lukowicz', 'Sungho Suh', 'Vitor Fortes Rey'] | 2022-10-04 | null | null | null | null | ['wearable-activity-recognition'] | ['time-series'] | [ 7.28832006e-01 2.70678222e-01 -2.82623619e-01 -2.82061011e-01
-8.89125884e-01 -3.69921207e-01 4.40835744e-01 3.04626584e-01
-7.23569274e-01 8.59592259e-01 2.73085117e-01 1.71032742e-01
-4.93978620e-01 -5.64522386e-01 -8.27149093e-01 -6.01660073e-01
-2.10954890e-01 1.46840960e-01 1.54147640e-01 1.24517538e-01
-3.32795769e-01 3.26489151e-01 -1.90095139e+00 2.21670583e-01
6.89063549e-01 1.44889736e+00 7.88487345e-02 4.69345272e-01
2.54643023e-01 3.67826253e-01 -8.32224548e-01 1.67777594e-02
2.94675171e-01 -3.41114372e-01 -3.36457342e-01 1.17058367e-01
3.18764776e-01 -1.79565310e-01 2.29779899e-01 5.06714225e-01
6.26432538e-01 1.44634575e-01 2.95782536e-01 -1.17093873e+00
3.90763611e-01 1.08157866e-01 -1.71868175e-01 3.89862917e-02
6.88385069e-01 5.92774786e-02 6.10283315e-01 -5.01371026e-01
3.50252092e-01 6.33809626e-01 8.81961346e-01 4.96379524e-01
-1.39449382e+00 -3.37864012e-01 -3.05755064e-02 -8.20671022e-03
-1.27598011e+00 -5.53580821e-01 6.48079216e-01 -4.76576030e-01
1.01369393e+00 5.69642901e-01 6.54787958e-01 1.56093001e+00
1.76803306e-01 6.20718718e-01 1.06967425e+00 -3.04296941e-01
4.59447742e-01 3.74750316e-01 2.39607334e-01 3.87184888e-01
5.53887069e-01 -1.83633640e-01 -8.29565525e-01 -1.07563280e-01
3.72851670e-01 1.99748382e-01 -1.55960619e-01 -3.60142797e-01
-1.10798097e+00 2.97101706e-01 2.87153542e-01 5.40177047e-01
-8.63628149e-01 -1.39607061e-02 3.05527717e-01 1.29525170e-01
2.74852008e-01 4.79184061e-01 -3.56712759e-01 -5.97066820e-01
-1.13615310e+00 6.51616254e-04 8.33810627e-01 6.89873695e-01
5.79419494e-01 -3.19933891e-01 -5.04843652e-01 8.04631472e-01
8.01106170e-02 3.23275656e-01 5.31310380e-01 -7.57515252e-01
6.13207817e-01 8.45224798e-01 4.32733744e-01 -7.18415499e-01
-4.45448101e-01 -3.57059598e-01 -6.66664958e-01 7.53268972e-02
6.01143360e-01 -2.58972198e-01 -7.38514066e-01 1.67098546e+00
2.32098684e-01 7.18390942e-02 -2.64529169e-01 5.47621727e-01
3.63832265e-01 2.39783898e-01 3.28547567e-01 -4.18543518e-01
1.31916511e+00 -6.28414810e-01 -7.08192289e-01 -4.53791916e-01
4.36851501e-01 -3.17656428e-01 1.33114624e+00 7.15187311e-01
-1.06595421e+00 -6.18939579e-01 -1.50050628e+00 2.68367380e-01
-3.92726928e-01 2.56833762e-01 3.42693210e-01 1.10750282e+00
-6.30589902e-01 9.25322413e-01 -1.21922040e+00 -4.09241855e-01
3.68253529e-01 7.33202875e-01 -5.33607244e-01 -8.28155354e-02
-8.09318781e-01 7.74193287e-01 2.28706777e-01 6.39887452e-02
-6.11337304e-01 -7.71665812e-01 -5.16157329e-01 3.50275598e-02
3.71838331e-01 -3.88902187e-01 8.81153345e-01 -9.80151653e-01
-1.46607327e+00 3.97208273e-01 1.23199709e-01 -4.63476568e-01
9.17454660e-01 -6.39419496e-01 -3.95588398e-01 1.22839620e-03
-9.91115272e-02 2.75407463e-01 6.84308469e-01 -7.08616078e-01
-4.34192359e-01 -3.20973158e-01 7.71386772e-02 -1.45808179e-02
-7.21978128e-01 -3.63430053e-01 -2.76268721e-01 -6.01707399e-01
-2.70949334e-01 -1.13839483e+00 1.46947771e-01 -1.06771111e-01
-2.60822982e-01 -7.22786114e-02 7.84904242e-01 -1.03998554e+00
1.23595357e+00 -2.02419567e+00 2.57393047e-02 2.85922587e-01
-2.22626075e-01 4.05759424e-01 1.42894089e-01 2.30491981e-01
-1.93561748e-01 -1.28039658e-01 -1.01312608e-01 -4.87244338e-01
-1.60325080e-01 3.47751290e-01 3.42243165e-01 5.06237626e-01
4.68905345e-02 4.39569652e-01 -7.19656765e-01 -6.13302179e-02
3.76296669e-01 6.27116084e-01 -8.86694118e-02 3.79148513e-01
6.77790791e-02 5.67694485e-01 -3.05222839e-01 5.67757428e-01
2.48336554e-01 -1.05215624e-01 1.03102863e-01 -2.51879185e-01
1.20017983e-01 6.39336586e-01 -1.41162956e+00 1.85663950e+00
-6.50266469e-01 2.66157269e-01 -1.37998089e-02 -8.97830486e-01
6.03256464e-01 5.78120589e-01 9.00028467e-01 -6.70969605e-01
-1.36543050e-01 1.08760826e-01 -2.15479940e-01 -6.04477108e-01
9.84234214e-02 1.45210281e-01 -6.76795170e-02 3.83819729e-01
1.42704099e-02 5.80335259e-01 1.98653385e-01 -3.24112475e-01
1.51272547e+00 2.74142802e-01 2.59182930e-01 -1.68394670e-01
4.90043789e-01 -4.27250862e-01 4.82731014e-01 6.24187171e-01
6.22726157e-02 3.67683560e-01 2.89257526e-01 -1.36823848e-01
-8.40683222e-01 -1.06883252e+00 -7.83681199e-02 9.84345019e-01
-3.98798734e-01 -4.52541023e-01 -7.24172175e-01 -8.90984952e-01
-8.14867988e-02 6.57055855e-01 -6.98352098e-01 -5.24958789e-01
-4.26547438e-01 -6.31100357e-01 5.50650179e-01 7.43390858e-01
6.16986632e-01 -8.63183677e-01 -1.18576372e+00 3.74055833e-01
-1.53434590e-01 -8.44485879e-01 -1.06935926e-01 5.38320124e-01
-1.12256765e+00 -8.47634971e-01 -8.71981859e-01 1.07629076e-01
6.33595288e-01 -1.48900226e-01 7.78208375e-01 -1.98323771e-01
-1.87396884e-01 7.92271554e-01 -2.56049156e-01 -4.32529122e-01
-1.03787847e-01 2.59365916e-01 1.37730449e-01 2.81194776e-01
2.27020323e-01 -7.13290691e-01 -6.46260023e-01 4.01806146e-01
-6.95773005e-01 -2.79250026e-01 6.81344211e-01 6.78203642e-01
5.10156989e-01 1.58457586e-03 4.64738071e-01 -6.17800951e-01
5.33381999e-01 -5.81526518e-01 -1.72035933e-01 2.43663505e-01
-8.93501341e-01 2.88755000e-01 1.94179744e-01 -7.69918442e-01
-9.12523687e-01 3.92089307e-01 -7.27917030e-02 -1.19791806e-01
-1.24214277e-01 2.82488346e-01 -5.03347993e-01 -1.48885325e-03
8.43253255e-01 -3.89582925e-02 -6.72479672e-03 -8.18438709e-01
-2.26538330e-01 7.96492457e-01 3.51272196e-01 -4.79364336e-01
3.30362290e-01 2.46706292e-01 1.99036285e-01 -9.26570415e-01
-6.49383545e-01 -5.40241182e-01 -3.90854120e-01 -2.65519589e-01
7.23957837e-01 -7.90957510e-01 -6.04469419e-01 3.31020027e-01
-7.87396729e-01 -1.95427924e-01 -6.79700434e-01 4.80000377e-01
-5.10808349e-01 2.86457449e-01 -1.20287515e-01 -1.14693379e+00
-3.19206834e-01 -9.62453008e-01 1.15611565e+00 1.43089026e-01
-6.62396550e-01 -8.24426055e-01 -7.30436742e-02 4.70029026e-01
3.83706868e-01 5.65731227e-01 4.55931932e-01 -6.21395886e-01
-1.25935897e-01 -7.39653349e-01 3.73720825e-01 7.73831069e-01
4.25313830e-01 -7.01926351e-01 -1.21805882e+00 -4.97515768e-01
1.33753940e-01 -1.59844369e-01 5.49879134e-01 1.11790083e-01
1.03990185e+00 -3.87566358e-01 -4.52209324e-01 1.47281429e-02
1.17921042e+00 -3.57751176e-02 9.22189832e-01 2.33752027e-01
4.67999637e-01 6.89728975e-01 5.87327838e-01 3.39658916e-01
-2.52664816e-02 1.05122483e+00 2.82991886e-01 4.73568700e-02
-1.23633161e-01 -1.62221834e-01 6.86647415e-01 2.21146882e-01
-2.92919010e-01 3.12361214e-02 -5.84690392e-01 5.73998690e-01
-1.79354799e+00 -5.89093149e-01 7.75519002e-04 3.05362082e+00
6.47530198e-01 3.20211500e-01 6.19999826e-01 7.35520303e-01
2.78300226e-01 -1.66521475e-01 -5.41849256e-01 9.53303743e-03
2.60946602e-01 3.00340414e-01 7.93733776e-01 9.36191678e-02
-1.05235851e+00 -2.20282733e-01 6.37513542e+00 5.54957390e-01
-9.52382922e-01 4.55057830e-01 4.97356951e-01 -5.29989958e-01
2.62487113e-01 -3.25671673e-01 -6.01459086e-01 8.42914999e-01
1.52158535e+00 3.92470330e-01 3.35606039e-01 8.47521424e-01
1.75021902e-01 -7.09309936e-01 -1.47377312e+00 1.22594035e+00
-4.19951491e-02 -6.58799291e-01 -6.07903838e-01 2.32409939e-01
4.57962871e-01 -1.99819371e-01 -2.20032215e-01 6.71252981e-02
-4.18883055e-01 -8.44211519e-01 6.13898158e-01 7.89174139e-01
6.53236330e-01 -6.38977051e-01 7.79049635e-01 5.06355345e-01
-9.45675850e-01 -1.77767083e-01 8.76017436e-02 -2.75295585e-01
-1.03820346e-01 8.12116086e-01 -8.74616206e-01 4.58057016e-01
8.05123687e-01 4.29548204e-01 -7.80203938e-01 1.06435633e+00
-2.09987447e-01 6.66111112e-01 -8.41456652e-01 -3.14619914e-02
-2.35979095e-01 8.64679068e-02 5.15859127e-01 1.05552888e+00
7.17528343e-01 -5.49022198e-01 5.93839735e-02 5.14165759e-01
7.78176412e-02 -2.74600610e-02 -6.51254058e-01 5.18559739e-02
2.33410075e-01 1.12726176e+00 -4.63793606e-01 -1.95450738e-01
-4.41745490e-01 1.03138745e+00 -3.76495235e-02 1.37378871e-01
-6.13387167e-01 -2.90525526e-01 4.92284238e-01 7.70883381e-01
8.40327740e-02 -1.91836521e-01 -3.33690614e-01 -8.31060231e-01
7.31584311e-01 -6.53775811e-01 4.24154967e-01 -4.43444431e-01
-1.10511553e+00 1.97419614e-01 1.67119831e-01 -1.38922381e+00
-4.71049368e-01 -5.19116223e-01 -3.36934268e-01 9.31906998e-01
-8.14305246e-01 -9.56810772e-01 -4.66057509e-01 4.85106438e-01
1.92252636e-01 1.49088636e-01 1.01033270e+00 7.36379027e-01
-3.26384366e-01 6.61411405e-01 -2.30525523e-01 -2.51187474e-01
5.19325197e-01 -1.08091891e+00 4.15518172e-02 7.91553915e-01
1.15182064e-01 6.82349324e-01 7.38236427e-01 -5.63453317e-01
-1.36591816e+00 -8.57263625e-01 8.40966582e-01 -7.30761111e-01
9.24970433e-02 -5.92429817e-01 -6.97197318e-01 5.32209635e-01
-1.26452103e-01 -8.06202814e-02 8.24713707e-01 2.49451131e-01
-1.54310554e-01 -4.59015220e-01 -1.34263229e+00 2.19903529e-01
1.00072122e+00 -3.90865415e-01 -5.01911998e-01 2.18419060e-01
1.56229166e-02 -1.69978186e-01 -1.10868561e+00 3.79024416e-01
8.33904386e-01 -8.21881115e-01 9.24465060e-01 -2.63232231e-01
-2.34611377e-01 -1.39704108e-01 -9.11313444e-02 -1.16813171e+00
1.10512368e-01 -6.91569269e-01 -3.86755019e-01 1.36883128e+00
3.45978469e-01 -6.03628874e-01 9.73618150e-01 9.74652886e-01
2.52194762e-01 -4.83082533e-01 -1.37872648e+00 -1.10588121e+00
-7.23198473e-01 -5.52371025e-01 2.31044769e-01 4.97699708e-01
9.23452452e-02 3.17064673e-01 -7.16564238e-01 -2.06584662e-01
4.32187557e-01 -6.49054289e-01 7.18679607e-01 -1.54769993e+00
-7.03509808e-01 5.93836680e-02 -4.60260332e-01 -8.10036063e-01
-4.36510593e-01 -5.41165352e-01 1.43225491e-01 -1.40743124e+00
2.44964827e-02 -3.80242974e-01 -5.96322715e-01 7.15658665e-01
5.68781830e-02 2.70315647e-01 7.79786482e-02 -1.07284978e-01
-5.72336614e-01 1.27955064e-01 5.22706807e-01 -1.11291826e-01
-4.47262913e-01 4.62351650e-01 -6.43729210e-01 5.29210865e-01
5.29008746e-01 -5.93014717e-01 -4.49723244e-01 1.53264245e-02
3.90161425e-01 4.88348044e-02 5.02300143e-01 -1.66324937e+00
-6.11685999e-02 2.50898987e-01 6.18769348e-01 -1.39122814e-01
6.81598485e-01 -1.40671229e+00 6.24538600e-01 5.45326054e-01
-3.15875053e-01 -1.59153923e-01 1.87308505e-01 6.10540807e-01
1.54869705e-01 -3.03138971e-01 4.22973782e-01 1.53584599e-01
-2.20856234e-01 -1.33617952e-01 -2.87778467e-01 -4.48870182e-01
9.84277487e-01 -4.76886481e-01 8.05734470e-02 -4.47193325e-01
-9.95057940e-01 -1.76887065e-01 1.86641738e-01 5.33507168e-01
1.17286474e-01 -1.38485301e+00 -1.61407918e-01 2.95823187e-01
3.64865810e-01 -2.29280740e-01 6.74030632e-02 1.10184205e+00
1.33667350e-01 2.00929269e-01 -4.41480517e-01 -4.70541924e-01
-1.47586906e+00 4.38260287e-01 2.78579265e-01 -4.93898213e-01
-5.05639791e-01 3.78247857e-01 -4.08079833e-01 -6.18515769e-03
6.16942704e-01 -5.62846422e-01 -2.25936659e-02 2.45835289e-01
7.12446630e-01 9.10637558e-01 7.37138510e-01 -2.90906101e-01
-5.87627232e-01 2.85895735e-01 1.96321398e-01 -9.18302536e-02
1.37764621e+00 -2.47774981e-02 3.62396955e-01 8.03980470e-01
1.01314557e+00 3.25245373e-02 -1.47728860e+00 -3.82528082e-02
4.30842876e-01 -3.26262712e-01 -3.03723756e-02 -1.09046674e+00
-6.78103030e-01 8.15600753e-01 1.33912623e+00 2.57716298e-01
1.22125292e+00 -1.78628221e-01 6.68985069e-01 3.33404809e-01
5.18714547e-01 -1.37109613e+00 9.26765054e-02 -7.68401101e-02
6.93622708e-01 -9.42816317e-01 2.71552145e-01 -1.58560798e-01
-3.73909861e-01 6.74801886e-01 -1.20526145e-03 -1.04007445e-01
4.76300240e-01 1.76780209e-01 -3.38316292e-01 -1.76117253e-02
-3.61686140e-01 -1.28625974e-01 4.63232517e-01 7.80242741e-01
4.36012954e-01 1.71301037e-01 -5.40621519e-01 9.15494382e-01
3.36444378e-01 2.94539958e-01 6.34239614e-02 1.16539645e+00
-1.15704685e-01 -1.42057729e+00 -3.09356391e-01 7.75915563e-01
-6.38707399e-01 4.76936579e-01 -2.57222801e-01 5.45072734e-01
3.87599528e-01 1.16959262e+00 -1.83254048e-01 -4.63829726e-01
8.54801893e-01 4.89593834e-01 5.17692924e-01 -5.80734074e-01
-7.63781250e-01 1.42527133e-01 4.86370653e-01 -8.92731190e-01
-6.39190137e-01 -9.70170975e-01 -8.80586684e-01 2.03394562e-01
-2.19341479e-02 1.51080526e-02 9.67202604e-01 1.03435445e+00
4.73025829e-01 8.36323202e-01 2.58920342e-01 -9.99282062e-01
-6.16355479e-01 -1.21556866e+00 -5.80425322e-01 6.17584288e-01
3.01790625e-01 -9.38990176e-01 -8.05849582e-03 1.07832097e-01] | [7.485788822174072, 0.942331850528717] |
c41345cf-2dfc-4365-be46-353fb6e8ed81 | learning-entity-and-relation-embeddings-for | null | null | https://dl.acm.org/doi/10.5555/2886521.2886624 | https://dl.acm.org/doi/10.5555/2886521.2886624 | Learning Entity and Relation Embeddings for Knowledge Graph Completion | Knowledge graph completion aims to perform link prediction between entities. In this paper, we consider the approach of knowledge graph embeddings. Recently, models such as TransE and TransH build entity and relation embeddings by regarding a relation as translation from head entity to tail entity. We note that these models simply put both entities and relations within the same semantic space. In fact, an entity may have multiple aspects and various relations may focus on different aspects of entities, which makes a common space insufficient for modeling. In this paper, we propose TransR to build entity and relation embeddings in separate entity space and relation spaces. Afterwards, we learn embeddings by first projecting entities from entity space to corresponding relation space and then building translations between projected entities. In experiments, we evaluate our models on three tasks including link prediction, triple classification and relational fact extraction. Experimental results show significant and consistent improvements compared to state-of-the-art baselines including TransE and TransH. The source code of this paper can be obtained from https://github.com/mrlyk423/relation_extraction. | ['Xuan Zhu', 'Yang Liu', 'Maosong Sun', 'Zhiyuan Liu', 'Yankai Lin'] | 2015-01-25 | null | null | null | aaai-2015-2015-1 | ['triple-classification', 'knowledge-graph-embeddings', 'knowledge-graph-embeddings'] | ['graphs', 'graphs', 'methodology'] | [-3.88081312e-01 4.87687886e-01 -5.80374897e-01 -3.34644556e-01
-3.10984522e-01 -4.87209797e-01 5.53959250e-01 5.34817398e-01
-2.40314111e-01 6.95170522e-01 4.94270146e-01 -2.40175515e-01
-2.45578736e-01 -1.24863648e+00 -7.40841866e-01 -1.30198002e-01
-1.75393566e-01 7.31537342e-01 2.27568641e-01 -2.42315382e-01
-3.32195520e-01 2.60120660e-01 -9.32703018e-01 1.39640570e-01
7.64686227e-01 7.64679849e-01 -3.33392113e-01 1.68722644e-01
-3.57326001e-01 8.49551916e-01 -1.04538806e-01 -1.12316144e+00
9.74997282e-02 4.64263074e-02 -1.12508404e+00 -2.26764306e-01
2.73643196e-01 -6.04736712e-03 -1.01679814e+00 9.49656427e-01
2.45033965e-01 5.34719937e-02 5.59948206e-01 -1.56352019e+00
-1.44133437e+00 9.54947948e-01 -4.37111855e-01 -3.65126655e-02
6.17715538e-01 -5.85469127e-01 1.75630093e+00 -1.45967197e+00
8.94579828e-01 1.03609705e+00 6.99471533e-01 1.80988938e-01
-1.18265676e+00 -5.11462033e-01 9.57132876e-02 5.91348171e-01
-1.85147440e+00 -3.03823799e-01 6.19169414e-01 -4.29446191e-01
1.13461983e+00 2.49142513e-01 6.43536627e-01 7.65591323e-01
-1.99205667e-01 7.97962785e-01 5.36893725e-01 -2.87053138e-01
-2.66465575e-01 3.04903835e-01 4.40890878e-01 7.11736023e-01
7.41033494e-01 -2.93444872e-01 -5.57924867e-01 -1.98404804e-01
5.94132841e-01 2.38734018e-02 -5.60221136e-01 -7.16927290e-01
-1.38291061e+00 7.65067995e-01 8.16247761e-01 2.58544326e-01
-4.36473817e-01 2.69808676e-02 2.93770671e-01 2.16294602e-01
4.67237145e-01 4.33399647e-01 -5.42707860e-01 1.99432701e-01
-1.88004583e-01 1.40307903e-01 1.07704723e+00 1.39750552e+00
9.08630311e-01 -5.68276703e-01 7.82311521e-03 9.31307495e-01
4.37469184e-01 4.59924117e-02 1.98210061e-01 -5.56940317e-01
8.48294258e-01 1.05500185e+00 9.09337252e-02 -1.31637049e+00
-2.77658015e-01 -1.88875899e-01 -5.91165543e-01 -5.84393978e-01
-1.41916769e-02 -1.34016410e-01 -6.90946579e-01 1.43503678e+00
6.98577344e-01 3.79735619e-01 3.24113041e-01 7.20354676e-01
1.33749473e+00 6.75139070e-01 6.08395599e-02 1.22451380e-01
1.55659580e+00 -1.19771004e+00 -9.94324923e-01 1.15635886e-03
9.75662827e-01 -6.73554420e-01 5.51207423e-01 -4.13522214e-01
-8.79738450e-01 -3.04482698e-01 -9.05750453e-01 -5.73536575e-01
-9.76010203e-01 1.71292350e-01 9.82979298e-01 2.49735862e-01
-8.24331582e-01 5.53496778e-01 -8.22057366e-01 -5.62902987e-01
3.72810215e-01 3.02735746e-01 -9.21459913e-01 -2.02847183e-01
-1.72659326e+00 1.12932575e+00 7.52228618e-01 1.57112002e-01
-1.08564258e-01 -7.93600500e-01 -1.26807249e+00 2.94147879e-01
5.81584096e-01 -9.24951613e-01 6.84323847e-01 4.80324142e-02
-8.32842469e-01 8.50969195e-01 -2.46644080e-01 -4.05027062e-01
4.46033962e-02 -3.85147721e-01 -9.45893645e-01 -2.46727705e-01
1.46592125e-01 4.35200334e-01 1.29419699e-01 -1.34740138e+00
-6.02903545e-01 -3.26541662e-01 4.12802577e-01 2.52684712e-01
-5.75081646e-01 -6.36674240e-02 -8.64867628e-01 -3.61883909e-01
3.12094688e-01 -8.48048210e-01 1.34197935e-01 -2.50719100e-01
-8.17130208e-01 -5.94160318e-01 7.38329887e-01 -7.74768531e-01
1.59355152e+00 -1.83933699e+00 3.56960565e-01 2.75944084e-01
5.43070376e-01 2.66622752e-01 -2.46278453e-03 8.06252897e-01
-4.78816748e-01 3.35490286e-01 -6.45741820e-02 -2.85933644e-01
2.87367582e-01 3.02855700e-01 -2.73018986e-01 2.32173890e-01
2.58942127e-01 1.35710573e+00 -9.77366745e-01 -5.72637320e-01
-5.40590920e-02 7.32169092e-01 -3.27504605e-01 8.00025687e-02
8.66967253e-03 -3.60265262e-02 -6.07480705e-01 7.00726330e-01
5.71114123e-01 -4.97767776e-01 6.42780185e-01 -6.51023030e-01
2.80515283e-01 6.41570628e-01 -1.22621834e+00 1.54998839e+00
-4.53156680e-01 6.08488321e-01 -5.32833695e-01 -8.40102315e-01
8.91396046e-01 5.19506335e-01 5.92919827e-01 -2.02898964e-01
2.91644912e-02 1.08794719e-01 -1.30286038e-01 -5.15808642e-01
9.17538404e-01 1.91329196e-01 6.88006505e-02 2.05333531e-01
3.29331636e-01 3.11997414e-01 2.77687371e-01 5.37533700e-01
1.17174363e+00 2.03145668e-01 4.96678978e-01 2.16547683e-01
4.70179319e-01 -1.53290316e-01 7.51242161e-01 -1.73443109e-02
7.26097915e-03 1.71179622e-01 6.81930542e-01 -2.94227332e-01
-8.21942031e-01 -1.33916795e+00 -1.82482511e-01 7.88203061e-01
3.98632079e-01 -1.16100192e+00 -1.41959339e-01 -1.09199810e+00
3.35856080e-01 7.40866184e-01 -7.40246356e-01 -9.11419839e-02
-5.84896624e-01 -5.11458635e-01 3.11043143e-01 7.15368509e-01
1.98848620e-01 -7.53356636e-01 3.77314329e-01 1.90979749e-01
-3.16531420e-01 -1.36732960e+00 -5.17553568e-01 -1.58753365e-01
-6.29646599e-01 -1.27641118e+00 -3.32574457e-01 -1.07448721e+00
6.67079091e-01 2.46727258e-01 1.31379104e+00 6.90558702e-02
-7.80622289e-02 4.33792561e-01 -5.06284416e-01 -4.61427495e-02
1.85560495e-01 2.65654117e-01 7.93900788e-02 3.43967415e-02
8.21259558e-01 -6.45158350e-01 -4.26987529e-01 1.67878747e-01
-7.51564384e-01 -8.28471109e-02 3.55619162e-01 8.20772350e-01
7.21381366e-01 8.48085359e-02 3.91586751e-01 -1.27062213e+00
5.38635373e-01 -8.10003698e-01 -2.62080848e-01 7.43840873e-01
-7.69713402e-01 -1.03191147e-02 1.85649067e-01 3.49786729e-02
-8.79258454e-01 -1.27162680e-01 1.27693251e-01 -4.75779980e-01
1.78430155e-01 9.28530872e-01 -4.19421077e-01 1.56581923e-01
2.93482810e-01 -2.24469937e-02 -6.39550507e-01 -3.73800635e-01
9.38740313e-01 5.34097612e-01 1.58604875e-01 -4.72272933e-01
1.16411805e+00 3.05441082e-01 -1.11911960e-01 -4.71463799e-01
-9.16243970e-01 -6.26600742e-01 -7.96498060e-01 2.24907696e-01
8.13452780e-01 -1.11924088e+00 -4.14519548e-01 -1.45241842e-01
-1.22660589e+00 1.59331635e-01 -4.25151229e-01 5.77089727e-01
-1.00955479e-01 3.18180323e-01 -6.63455606e-01 -2.37501144e-01
-1.66565314e-01 -6.56706095e-01 7.49765098e-01 2.71515042e-01
-2.10086107e-01 -1.49142671e+00 2.61868387e-01 4.48888361e-01
2.64201388e-02 1.95866883e-01 8.05355430e-01 -8.87651026e-01
-7.96153188e-01 -4.49972510e-01 -4.76445317e-01 -4.71280590e-02
3.86854827e-01 4.10460010e-02 -5.83647966e-01 -6.19273521e-02
-7.18732417e-01 -1.93775550e-01 9.47998822e-01 -1.73446581e-01
6.56039655e-01 -2.85939008e-01 -8.97938609e-01 7.32251525e-01
1.50277543e+00 -1.67680651e-01 7.57299602e-01 2.86371440e-01
1.26197433e+00 6.66749895e-01 7.39832282e-01 1.25303969e-01
1.18085897e+00 7.80707717e-01 9.15113837e-02 9.23167020e-02
-2.12984830e-01 -6.81460202e-01 5.90983108e-02 1.01926267e+00
-1.89922586e-01 -3.40953290e-01 -9.42176223e-01 9.36961174e-01
-2.00241041e+00 -9.33634102e-01 -4.98384535e-01 1.98578608e+00
9.81073618e-01 -1.77384242e-01 -1.79354966e-01 -2.05113783e-01
8.02313030e-01 1.99661732e-01 -3.66032794e-02 -7.50076622e-02
-9.76548716e-02 1.36751786e-01 6.26886606e-01 6.85380042e-01
-1.14727426e+00 1.19133162e+00 4.61462355e+00 5.60228586e-01
-5.50376177e-01 3.46568912e-01 -5.18818758e-02 2.42277622e-01
-7.10877836e-01 3.96653533e-01 -9.49016750e-01 1.33383110e-01
6.32997334e-01 -6.75077856e-01 1.70573696e-01 6.07549608e-01
-4.24373597e-01 4.14974183e-01 -1.33406496e+00 8.66373003e-01
-3.64739485e-02 -1.35392010e+00 -2.14180518e-02 7.34724104e-02
8.56738925e-01 -2.64144409e-02 -1.99010104e-01 6.70036674e-01
4.70091641e-01 -1.00797939e+00 7.94229284e-02 5.06304920e-01
6.39749229e-01 -5.30809462e-01 8.26029658e-01 -2.74427474e-01
-1.75249612e+00 3.71971190e-01 -4.66272652e-01 3.22622836e-01
3.70559096e-01 6.52660668e-01 -8.82809699e-01 1.31577253e+00
4.87326026e-01 1.11156225e+00 -5.35325050e-01 9.61435258e-01
-7.20023513e-01 2.96519309e-01 -6.66135997e-02 2.21019566e-01
-1.68714896e-01 -4.13381159e-01 5.05856812e-01 1.15056145e+00
3.19448143e-01 1.84906274e-01 1.17878288e-01 8.41147602e-01
-7.07874358e-01 3.61570925e-01 -8.30566466e-01 -2.88471967e-01
9.39962566e-01 1.42727804e+00 -3.17641675e-01 -4.27320063e-01
-9.34277534e-01 1.05764925e+00 7.62043655e-01 5.56721330e-01
-1.00519073e+00 -5.73379636e-01 9.09198403e-01 -2.57680006e-02
5.76737702e-01 -1.18707776e-01 -1.02474779e-01 -1.52152455e+00
3.29606652e-01 -7.68512487e-02 6.43581927e-01 -5.71066558e-01
-1.37837780e+00 5.21122694e-01 -8.80339444e-02 -1.15249455e+00
1.68339536e-01 -4.57783163e-01 -4.96695101e-01 8.41018260e-01
-1.59651637e+00 -1.44475937e+00 -7.04850182e-02 3.83898556e-01
-1.15469165e-01 -5.81950974e-03 1.01622260e+00 5.71095049e-01
-7.42858827e-01 7.99300134e-01 4.20738459e-02 6.26309752e-01
7.61677504e-01 -1.53506494e+00 6.14209950e-01 5.92068672e-01
5.83826840e-01 1.02119815e+00 3.18095446e-01 -8.15282941e-01
-1.31602573e+00 -1.41163170e+00 1.77683234e+00 -7.79769123e-01
1.10673869e+00 -2.70533323e-01 -1.04159248e+00 1.47356629e+00
3.32433522e-01 4.34763044e-01 1.06182492e+00 8.82777095e-01
-7.43216157e-01 2.44046915e-02 -8.34547520e-01 6.88223362e-01
1.29782045e+00 -6.86386943e-01 -7.96690404e-01 5.13878822e-01
9.63744044e-01 -3.68626893e-01 -1.72912025e+00 4.55167800e-01
3.92932177e-01 -5.22642910e-01 1.07752633e+00 -9.42371666e-01
4.99344379e-01 -5.26380002e-01 -2.56227195e-01 -1.43678844e+00
-4.76775706e-01 -2.03064516e-01 -9.59013224e-01 1.53099597e+00
9.00173962e-01 -8.49314451e-01 7.79078305e-01 6.27984941e-01
6.67519197e-02 -1.07012033e+00 -7.86334336e-01 -8.36492360e-01
-1.89369526e-02 -5.74083850e-02 8.06305766e-01 1.58338368e+00
4.48634893e-01 7.22811878e-01 -1.10333398e-01 6.67793214e-01
4.47196126e-01 4.74987000e-01 7.11345434e-01 -1.25537610e+00
-1.54894412e-01 -1.12992585e-01 -7.62261689e-01 -9.04927611e-01
4.15088356e-01 -1.27325451e+00 -5.33036292e-01 -2.16102028e+00
2.69796103e-01 -5.26412785e-01 -4.19209510e-01 6.55087650e-01
-4.12624896e-01 2.46455483e-02 5.26478067e-02 1.52741686e-01
-6.75191402e-01 8.93062830e-01 1.02227557e+00 -2.74182290e-01
-8.63156989e-02 -3.72120291e-01 -7.81496406e-01 4.62439746e-01
7.18940854e-01 -3.91277611e-01 -4.34104562e-01 -6.16341949e-01
3.89221042e-01 -1.14606895e-01 2.49991685e-01 -5.65655291e-01
3.85893941e-01 9.73811373e-02 1.26311600e-01 -3.54959011e-01
5.08858800e-01 -9.68375564e-01 2.60131180e-01 -6.75061941e-02
-1.40221611e-01 -1.44734427e-01 -4.61713821e-02 7.37461388e-01
-5.78762710e-01 -2.20680740e-02 5.48105054e-02 2.75853157e-01
-9.20618832e-01 6.43258631e-01 5.01722991e-01 1.54328391e-01
1.15932703e+00 1.21746138e-01 -6.01342559e-01 -2.75856048e-01
-1.07897794e+00 6.36871159e-01 4.07665163e-01 6.86239362e-01
8.28131795e-01 -1.95101047e+00 -7.02469945e-01 -3.06588262e-01
5.77286780e-01 1.87321290e-04 2.01860532e-01 1.10158241e+00
-2.14698598e-01 5.88121116e-01 2.43161559e-01 3.82969640e-02
-1.32720673e+00 7.38157690e-01 1.91033959e-01 -4.49705094e-01
-5.57155311e-01 9.83592451e-01 1.00004472e-01 -9.22614574e-01
-1.12879230e-02 -1.71152309e-01 -3.97094190e-01 2.72110283e-01
2.50158191e-01 3.58419329e-01 -6.94669411e-02 -8.59296620e-01
-6.47235572e-01 4.96341407e-01 -2.38906384e-01 2.35231102e-01
1.32335544e+00 -1.39879018e-01 -3.71284217e-01 2.84507692e-01
1.42951345e+00 2.90890664e-01 -4.90008026e-01 -7.16220975e-01
2.74570048e-01 -6.08053684e-01 -8.26820657e-02 -4.44053710e-01
-1.16042531e+00 4.84129876e-01 -5.45372395e-03 3.65350783e-01
7.06053972e-01 4.29269195e-01 9.44628060e-01 3.19987744e-01
3.77776206e-01 -8.86350572e-01 -4.57767993e-01 4.79186267e-01
8.42727542e-01 -1.26666999e+00 1.86612785e-01 -1.02223289e+00
-7.06636608e-01 8.88126254e-01 6.91148937e-01 1.35476645e-02
9.28644657e-01 -2.02710897e-01 -1.86587691e-01 -6.28644049e-01
-8.67669106e-01 -5.88024914e-01 5.51730335e-01 6.30927265e-01
7.89056063e-01 4.35005128e-01 -4.07282442e-01 5.62685013e-01
-7.51345903e-02 -7.33115673e-02 4.30458546e-01 7.91920543e-01
-6.49205223e-02 -1.54727089e+00 7.75756836e-02 4.46116358e-01
-1.89835235e-01 -3.60918224e-01 -4.87711102e-01 9.52897429e-01
1.84117571e-01 7.99538970e-01 -9.48511958e-02 -6.97268665e-01
6.24772549e-01 2.87977785e-01 3.64915729e-01 -8.81157398e-01
-1.12883104e-02 -4.79148000e-01 6.36109591e-01 -4.76580024e-01
-3.00712109e-01 -7.20833302e-01 -1.42090225e+00 -4.34046388e-01
-5.34492135e-01 2.41180003e-01 2.14477137e-01 5.99776685e-01
5.29100597e-01 7.36696124e-01 4.05226916e-01 -8.19488019e-02
-1.26812235e-01 -9.16732609e-01 -7.19519377e-01 5.34994781e-01
-1.71207562e-01 -8.48423600e-01 1.69346724e-02 -3.03426255e-02] | [8.84736156463623, 7.994575500488281] |
33752631-1dcb-45ff-8fd2-1c4a775c590a | conditional-generation-of-medical-images-via | 2012.04764 | null | https://arxiv.org/abs/2012.04764v3 | https://arxiv.org/pdf/2012.04764v3.pdf | Conditional Generation of Medical Images via Disentangled Adversarial Inference | Synthetic medical image generation has a huge potential for improving healthcare through many applications, from data augmentation for training machine learning systems to preserving patient privacy. Conditional Adversarial Generative Networks (cGANs) use a conditioning factor to generate images and have shown great success in recent years. Intuitively, the information in an image can be divided into two parts: 1) content which is presented through the conditioning vector and 2) style which is the undiscovered information missing from the conditioning vector. Current practices in using cGANs for medical image generation, only use a single variable for image generation (i.e., content) and therefore, do not provide much flexibility nor control over the generated image. In this work we propose a methodology to learn from the image itself, disentangled representations of style and content, and use this information to impose control over the generation process. In this framework, style is learned in a fully unsupervised manner, while content is learned through both supervised learning (using the conditioning vector) and unsupervised learning (with the inference mechanism). We undergo two novel regularization steps to ensure content-style disentanglement. First, we minimize the shared information between content and style by introducing a novel application of the gradient reverse layer (GRL); second, we introduce a self-supervised regularization method to further separate information in the content and style variables. We show that in general, two latent variable models achieve better performance and give more control over the generated image. We also show that our proposed model (DRAI) achieves the best disentanglement score and has the best overall performance. | ['Qicheng Lao', 'Yiping Wang', 'Ximeng Mao', 'Mohammad Havaei'] | 2020-12-08 | null | null | null | null | ['medical-image-generation'] | ['medical'] | [ 7.44096100e-01 4.53056991e-01 -2.38526106e-01 -3.53089601e-01
-5.21439075e-01 -4.95745748e-01 7.23932385e-01 -1.34881616e-01
-3.47930819e-01 6.57431424e-01 3.47515523e-01 -1.10769063e-01
1.77936092e-01 -9.89127100e-01 -8.01666141e-01 -1.08444166e+00
3.21632087e-01 1.66966334e-01 -3.94766212e-01 -6.08975403e-02
-2.10263789e-01 3.02429736e-01 -1.14227951e+00 2.58587360e-01
1.20447886e+00 8.03046405e-01 3.42658684e-02 5.61454535e-01
1.16548697e-02 9.12066996e-01 -7.57129610e-01 -3.74890894e-01
4.21295643e-01 -9.40056801e-01 -4.70344365e-01 3.58648628e-01
1.28608465e-01 -2.69570142e-01 -2.44775951e-01 1.13000488e+00
4.72269446e-01 -2.22287998e-01 6.89005554e-01 -1.20990944e+00
-7.89605856e-01 5.10603547e-01 -3.92737448e-01 -5.29200256e-01
1.48455143e-01 2.98902333e-01 7.85936654e-01 -4.67628032e-01
7.77452171e-01 1.07452834e+00 2.34528825e-01 8.68378520e-01
-1.80586314e+00 -7.08084524e-01 -1.23126199e-02 -3.07119399e-01
-1.27219474e+00 -2.27396786e-01 1.08477271e+00 -6.25143528e-01
1.18559107e-01 4.69334394e-01 7.01001823e-01 1.35527265e+00
3.04415375e-01 8.18029881e-01 1.41007030e+00 -5.15921474e-01
3.36126745e-01 5.22588551e-01 -4.26498473e-01 7.37029791e-01
1.42730564e-01 2.71991134e-01 -3.02250057e-01 -1.22513771e-01
9.53292966e-01 1.49201125e-01 -5.02719223e-01 -6.27273977e-01
-1.10757494e+00 1.09357560e+00 5.04189670e-01 9.56691206e-02
-2.39045799e-01 3.52258198e-02 7.27078244e-02 2.75743306e-01
2.86216825e-01 6.90676570e-01 -1.59008011e-01 4.50748056e-01
-8.65102410e-01 2.81374604e-01 6.42616630e-01 8.42481852e-01
8.05222392e-01 1.61633715e-01 -5.93258798e-01 7.77934492e-01
2.30439752e-01 6.71745062e-01 7.22429991e-01 -7.51221061e-01
3.41819406e-01 5.94272316e-01 1.46699259e-02 -1.05490804e+00
-1.57539457e-01 -4.22831982e-01 -1.15696478e+00 4.87809807e-01
2.76384920e-01 -3.92868221e-01 -1.25165880e+00 2.26181602e+00
4.98687811e-02 4.83997129e-02 1.11155644e-01 6.73619092e-01
7.82373726e-01 4.40466106e-01 1.17591478e-01 -1.39846951e-01
1.29247928e+00 -7.88286626e-01 -9.05834913e-01 -2.38854304e-01
5.12040377e-01 -6.21396244e-01 9.46278811e-01 2.83048689e-01
-1.09400690e+00 -3.72785062e-01 -1.33809984e+00 -4.23203893e-02
-2.94451743e-01 2.18658969e-01 5.05160034e-01 8.01027596e-01
-8.54688585e-01 5.87479830e-01 -7.50169039e-01 1.75930172e-01
4.07249510e-01 4.87086445e-01 -3.69881183e-01 -9.86741334e-02
-1.30506837e+00 5.88376582e-01 2.18931705e-01 -8.75121355e-02
-7.80788362e-01 -6.58065915e-01 -1.09030449e+00 -5.41168638e-02
4.17888135e-01 -1.12758100e+00 7.13626087e-01 -1.30207086e+00
-1.62760079e+00 8.99555624e-01 1.74022853e-01 -3.13001603e-01
9.55394626e-01 -3.48500609e-02 -1.11586541e-01 9.04989243e-02
-3.03657893e-02 8.18992853e-01 1.15026438e+00 -1.50623262e+00
-1.28806859e-01 -2.84305215e-01 1.16047477e-02 4.67366949e-02
-3.15242887e-01 -4.55695897e-01 -4.38287526e-01 -1.09123373e+00
-1.27970362e-02 -1.17547297e+00 -4.08873856e-01 1.34726524e-01
-8.05532694e-01 4.60923702e-01 5.23166835e-01 -6.48561001e-01
1.08718693e+00 -2.07328296e+00 4.33870643e-01 2.89242089e-01
4.13248688e-01 2.63830990e-01 -2.39166498e-01 1.66200146e-01
-1.48088619e-01 2.99162924e-01 -6.60288095e-01 -5.89684069e-01
-2.39983007e-01 4.04680997e-01 -2.15720728e-01 2.98730880e-01
3.81482035e-01 1.08592534e+00 -8.92021477e-01 -4.34793770e-01
1.43962726e-01 7.00444043e-01 -8.69566619e-01 4.00524765e-01
-3.48360270e-01 9.51364279e-01 -5.13413548e-01 1.87334761e-01
7.45418012e-01 -2.22257808e-01 3.51814687e-01 -2.92494357e-01
1.47481620e-01 -6.56143352e-02 -8.89676392e-01 1.73366237e+00
-4.56624717e-01 2.92321980e-01 -6.44644946e-02 -7.28228211e-01
8.49735260e-01 4.65397596e-01 5.41133642e-01 -5.58150947e-01
1.77505344e-01 -1.16727099e-01 -3.82815604e-03 -4.99503642e-01
2.19366536e-01 -4.65425670e-01 -1.35697201e-01 3.40533137e-01
5.51452041e-02 -1.01667136e-01 -1.34851903e-01 2.52738416e-01
9.23436821e-01 1.08562127e-01 1.16502956e-01 -6.57313839e-02
4.95404869e-01 -3.29088122e-01 7.51618266e-01 7.48947263e-01
8.81894007e-02 9.26007986e-01 8.41162741e-01 2.54303943e-02
-1.18342173e+00 -1.02676880e+00 5.00410460e-02 4.97759730e-01
1.13073867e-02 -2.18256235e-01 -8.85134220e-01 -7.78906345e-01
-1.02402948e-01 6.89694405e-01 -9.46210742e-01 -4.98334497e-01
-5.57423711e-01 -8.90219450e-01 5.11338711e-01 4.38081056e-01
3.58636051e-01 -1.06597245e+00 -3.23595196e-01 -1.27788886e-01
-1.54943407e-01 -9.57589865e-01 -6.78162694e-01 5.44235483e-02
-7.04611063e-01 -7.80837953e-01 -8.62545013e-01 -4.94017899e-01
1.12359536e+00 -2.04258904e-01 8.99028540e-01 3.92339518e-03
-4.09659028e-01 1.44684583e-01 -2.11857021e-01 -3.52778316e-01
-7.37530887e-01 -5.06768450e-02 -3.30674708e-01 4.70009506e-01
-2.21846774e-01 -6.16967976e-01 -8.37503850e-01 5.31961247e-02
-1.36401916e+00 5.04557788e-01 8.83602917e-01 1.28706408e+00
5.72336435e-01 -2.05732733e-01 4.05208468e-01 -1.55440629e+00
5.22316217e-01 -3.30386311e-01 -4.16265994e-01 1.19624346e-01
-8.02374601e-01 3.68847907e-01 7.08941519e-01 -6.19225800e-01
-1.12030411e+00 2.16367319e-01 -1.42826617e-01 -5.91518342e-01
-2.76283193e-02 3.54876608e-01 -5.30357420e-01 8.39740336e-02
6.39064908e-01 2.92556673e-01 3.90748501e-01 -3.00523609e-01
6.18979871e-01 2.66089618e-01 3.50616366e-01 -4.27727640e-01
9.12811279e-01 6.38822556e-01 2.01252364e-02 -5.10207534e-01
-7.59349108e-01 6.01601005e-02 -4.55946594e-01 1.61403015e-01
1.09035134e+00 -8.31424534e-01 -4.90866393e-01 4.31115657e-01
-8.84573400e-01 -2.88295090e-01 -6.05073631e-01 4.16005075e-01
-6.65026486e-01 2.09253043e-01 -6.01960003e-01 -6.50357723e-01
-3.16137522e-01 -1.23723781e+00 8.71234357e-01 7.48226270e-02
-2.05343515e-01 -1.05713010e+00 -1.78784784e-02 3.36134017e-01
3.85678977e-01 7.51167357e-01 1.15303504e+00 -4.42941308e-01
-7.30328441e-01 -1.94768339e-01 -1.19152792e-01 5.25836229e-01
4.29046959e-01 -4.16950583e-01 -9.97499824e-01 -4.30296898e-01
4.01891738e-01 -1.80718675e-01 9.14886653e-01 2.82674909e-01
1.15897620e+00 -6.67651951e-01 -1.83225051e-01 9.21376467e-01
1.51092434e+00 1.19619377e-01 8.58793557e-01 1.28951622e-03
1.06374061e+00 6.05809867e-01 5.06520979e-02 3.18179131e-01
1.03451379e-01 6.13717318e-01 2.14177594e-01 -5.48239231e-01
-2.91678190e-01 -5.19216776e-01 2.33062267e-01 6.09033048e-01
6.16592206e-02 -2.78147161e-01 -4.51954484e-01 2.92925984e-01
-1.78706491e+00 -8.13250124e-01 3.18662167e-01 2.44146895e+00
1.19538391e+00 4.72581685e-02 -1.09658070e-01 1.14602737e-01
4.73851949e-01 1.36078671e-01 -6.62518620e-01 -1.76829994e-01
-1.47032499e-01 2.65428513e-01 5.43703377e-01 5.67688644e-01
-8.69486749e-01 6.40400112e-01 6.26898432e+00 6.58212245e-01
-1.39088738e+00 -6.69330880e-02 9.31998074e-01 -2.19269637e-02
-8.49295616e-01 -7.09719285e-02 -3.58495176e-01 5.73913574e-01
4.91815954e-01 2.10281331e-02 3.63518566e-01 6.22655809e-01
2.42236197e-01 2.60124803e-01 -1.15957677e+00 7.50687122e-01
2.61704862e-01 -1.10852396e+00 4.58851278e-01 1.95979774e-01
9.14487362e-01 -6.14768267e-01 4.02208358e-01 1.53608965e-02
2.80665070e-01 -1.21049869e+00 6.30881131e-01 6.73330784e-01
1.06993842e+00 -6.66342080e-01 6.07266903e-01 3.24292630e-01
-5.54105699e-01 4.46533784e-03 6.27229214e-02 3.22680593e-01
9.94565040e-02 6.48807406e-01 -5.74789345e-01 5.12602687e-01
-2.92754237e-04 3.88640821e-01 -4.02417868e-01 5.71138620e-01
-5.80065548e-01 3.69629800e-01 1.87146083e-01 4.05133486e-01
-8.79434422e-02 -4.61714655e-01 6.66713536e-01 9.98059332e-01
4.93050106e-02 1.12401303e-02 1.87334210e-01 1.36123252e+00
-2.06607908e-01 2.20780633e-02 -5.99573314e-01 -9.85402167e-02
8.34079310e-02 1.12131512e+00 -5.48569322e-01 -3.57169420e-01
-2.88554728e-01 1.18691730e+00 8.32329914e-02 5.33047974e-01
-7.45434761e-01 -2.63791382e-01 6.36561990e-01 1.73475683e-01
2.37802520e-01 7.48145059e-02 -6.11668706e-01 -1.32627213e+00
-1.06219403e-01 -9.71530139e-01 6.77546188e-02 -6.15532637e-01
-1.20889401e+00 7.69563854e-01 -1.18856698e-01 -1.15460086e+00
-4.59806353e-01 -4.40855116e-01 -4.31069344e-01 1.09986126e+00
-1.46095562e+00 -1.29342639e+00 -3.13402176e-01 6.35433912e-01
1.80127934e-01 -5.93931861e-02 9.66446042e-01 2.64821470e-01
-5.86215496e-01 1.03225732e+00 2.88588088e-03 3.54289740e-01
7.80729890e-01 -1.36125898e+00 -9.78155732e-02 7.18979895e-01
1.55539494e-02 7.28677392e-01 7.09952176e-01 -7.16421843e-01
-1.27933836e+00 -1.14434540e+00 6.27646923e-01 -2.90878475e-01
1.82260305e-01 -5.42239666e-01 -7.36118674e-01 6.57933831e-01
1.02887541e-01 -1.69959188e-01 8.13591957e-01 -2.47201651e-01
-4.21113044e-01 -6.29772842e-02 -1.38324773e+00 8.07483792e-01
6.54334307e-01 -4.92278486e-01 -1.72620654e-01 -1.28246169e-03
9.35909748e-01 -4.60553974e-01 -7.28914618e-01 4.31589961e-01
5.77064753e-01 -9.32608247e-01 8.94178510e-01 -3.80851716e-01
8.08513403e-01 -1.86779872e-01 1.55047690e-02 -1.31823707e+00
-2.78481960e-01 -7.06640363e-01 -8.46128352e-03 1.19073129e+00
6.11106575e-01 -7.05286741e-01 1.01242232e+00 7.84137309e-01
8.63290802e-02 -9.00763094e-01 -2.93534607e-01 -5.08730292e-01
2.23552048e-01 -2.71274596e-02 6.15625381e-01 1.05016780e+00
-2.34276101e-01 3.88503790e-01 -7.78789461e-01 8.22502095e-03
5.51229298e-01 1.22265920e-01 7.05441177e-01 -9.09843743e-01
-5.91049552e-01 -3.11088443e-01 -2.82945544e-01 -8.89340222e-01
-8.29233080e-02 -9.44895327e-01 5.07847480e-02 -1.23204732e+00
4.36322182e-01 -5.84946871e-01 -1.90368056e-01 3.70836973e-01
-3.42301637e-01 2.89268911e-01 3.95018101e-01 2.85281658e-01
1.45925656e-01 5.35812855e-01 1.72159982e+00 -1.94789827e-01
-3.97512078e-01 -3.92447002e-02 -9.89469171e-01 5.57025969e-01
6.53858781e-01 -4.60023195e-01 -6.58488691e-01 -3.81768107e-01
2.02801488e-02 2.59628147e-01 2.81454533e-01 -6.31963968e-01
-9.55384895e-02 -1.49056301e-01 5.35009205e-01 7.08281398e-02
3.75891656e-01 -6.70391977e-01 2.41460323e-01 5.15370667e-01
-6.15318596e-01 -3.79799962e-01 -8.25203061e-02 5.33009589e-01
-2.26314843e-01 -1.90433815e-01 9.03330386e-01 -1.96229234e-01
1.18053421e-01 4.10185665e-01 -1.74370646e-01 -1.17063478e-01
7.48162150e-01 5.19131944e-02 -4.62677628e-02 -4.98527795e-01
-9.27483022e-01 6.06510900e-02 6.53507292e-01 3.17806721e-01
5.45538127e-01 -1.46234775e+00 -7.32036710e-01 6.46137655e-01
5.73378354e-02 1.40508879e-02 2.25881889e-01 5.35184801e-01
-3.02310735e-01 -2.25385614e-02 -1.24925040e-01 -4.91167575e-01
-1.06383193e+00 7.49655068e-01 1.57922328e-01 -6.47496998e-01
-5.51896632e-01 5.75345218e-01 7.60372400e-01 -2.88372755e-01
6.86403811e-02 -1.03656024e-01 -1.65009186e-01 -3.74944415e-03
4.58172768e-01 -3.35610174e-02 -2.30569810e-01 -6.22248590e-01
1.36841729e-01 4.78103578e-01 -1.35516897e-01 -3.32076490e-01
1.14473617e+00 -1.47032989e-02 -3.62212621e-02 2.95666069e-01
1.37008309e+00 2.57957250e-01 -1.31292188e+00 -1.53949425e-01
-4.20775086e-01 -4.43737000e-01 -1.93235744e-02 -9.52525795e-01
-1.25216162e+00 6.57009542e-01 6.42243028e-01 1.58662915e-01
1.31671727e+00 -1.68949395e-01 6.49870872e-01 -2.69308865e-01
6.64582895e-03 -7.77399063e-01 2.59979159e-01 3.47252772e-03
9.07904387e-01 -1.22737455e+00 -1.31573873e-02 -6.18820906e-01
-7.58580863e-01 6.45773709e-01 3.94335210e-01 -1.81221291e-01
5.56346655e-01 2.37304494e-01 2.66305208e-01 -3.21165025e-02
-4.72312629e-01 8.77636373e-02 5.89897573e-01 5.58952153e-01
4.94230270e-01 2.25085363e-01 -4.68604863e-01 5.39596438e-01
-3.43221366e-01 -6.66497722e-02 3.76675367e-01 7.26101577e-01
9.14248675e-02 -1.51369679e+00 -3.33521187e-01 4.29718733e-01
-5.99537134e-01 -7.72421062e-02 -3.40672523e-01 5.07237971e-01
4.18095797e-01 7.15326965e-01 -1.19368762e-01 -3.66372049e-01
2.47921720e-01 -5.28992265e-02 4.62062836e-01 -6.81258321e-01
-4.95819241e-01 2.15963840e-01 -2.36859143e-01 -4.89995927e-01
-2.76623040e-01 -3.92613292e-01 -9.16241050e-01 -9.68606099e-02
-3.79105300e-01 1.39575349e-02 7.07409441e-01 8.47357750e-01
3.30810487e-01 8.11056435e-01 8.86399329e-01 -6.16912186e-01
-4.69493538e-01 -6.79682136e-01 -4.78087336e-01 8.18184257e-01
4.15674597e-01 -3.10904235e-01 -3.59822482e-01 2.45675012e-01] | [11.650036811828613, -0.30870458483695984] |
ca7e8edd-f95f-49f4-a0ee-62b43d95ce69 | jet-images-deep-learning-edition | 1511.05190 | null | http://arxiv.org/abs/1511.05190v3 | http://arxiv.org/pdf/1511.05190v3.pdf | Jet-Images -- Deep Learning Edition | Building on the notion of a particle physics detector as a camera and the
collimated streams of high energy particles, or jets, it measures as an image,
we investigate the potential of machine learning techniques based on deep
learning architectures to identify highly boosted W bosons. Modern deep
learning algorithms trained on jet images can out-perform standard
physically-motivated feature driven approaches to jet tagging. We develop
techniques for visualizing how these features are learned by the network and
what additional information is used to improve performance. This interplay
between physically-motivated feature driven tools and supervised learning
algorithms is general and can be used to significantly increase the sensitivity
to discover new particles and new forces, and gain a deeper understanding of
the physics within jets. | ['Lester Mackey', 'Benjamin Nachman', 'Luke de Oliveira', 'Michael Kagan', 'Ariel Schwartzman'] | 2015-11-16 | null | null | null | null | ['jet-tagging'] | ['graphs'] | [-2.38154858e-01 -2.67500967e-01 -4.61913794e-01 -4.17921156e-01
-2.85562575e-01 -6.60848081e-01 1.07149863e+00 1.85512722e-01
-4.57140595e-01 3.28108400e-01 3.15054297e-01 -3.98411393e-01
-2.42949650e-01 -1.07350719e+00 -5.97703218e-01 -7.23329127e-01
-1.32152244e-01 8.55176806e-01 5.24459124e-01 -1.48563579e-01
3.38552594e-01 9.07463491e-01 -9.71967697e-01 3.97559166e-01
-2.97151685e-01 9.01132166e-01 1.62919372e-01 1.18967676e+00
-2.85479903e-01 8.13500762e-01 -2.41932184e-01 -4.73230243e-01
2.46762484e-01 -5.40322185e-01 -4.84986931e-01 -5.16699135e-01
2.25415707e-01 -9.91228074e-02 -5.36599040e-01 1.06523752e+00
2.10075662e-01 1.69934407e-01 5.16144514e-01 -9.88699794e-01
-7.08150446e-01 4.67884660e-01 -2.13411510e-01 9.36636209e-01
-2.53056526e-01 5.10457814e-01 9.81778681e-01 -7.75507033e-01
9.44739223e-01 1.46971488e+00 6.86297178e-01 3.77537191e-01
-1.13591325e+00 -5.74262619e-01 -5.24474382e-01 -3.02661099e-02
-2.64211655e-01 -3.28306437e-01 8.20739210e-01 -8.71131301e-01
1.16195047e+00 2.90569156e-01 8.66758347e-01 1.01535511e+00
5.89735687e-01 2.56928325e-01 7.31546760e-01 -6.71744406e-01
1.01466030e-01 1.80038318e-01 2.32524052e-01 9.60409582e-01
4.71867353e-01 1.01931596e+00 -8.10108066e-01 -1.68430820e-01
1.08494925e+00 5.75244427e-04 2.94356674e-01 -7.47007132e-01
-1.21727371e+00 1.25457072e+00 5.57550132e-01 5.71757019e-01
2.24919319e-01 5.48179924e-01 4.92780566e-01 -8.25387053e-03
5.35463333e-01 1.24169290e+00 -7.09982932e-01 -3.19618165e-01
-8.67736697e-01 4.54313278e-01 4.64927346e-01 4.61590976e-01
6.20727062e-01 6.07188419e-02 -1.97932452e-01 4.38119657e-02
4.44440991e-01 6.58486664e-01 2.17181176e-01 -1.04315376e+00
-1.83611289e-02 5.26111066e-01 1.51368544e-01 -6.53669655e-01
-7.72648811e-01 -5.82973361e-01 -3.00697744e-01 7.84204066e-01
3.17037463e-01 -1.90714180e-01 -9.42951620e-01 1.08778763e+00
1.81957901e-01 -5.71590811e-02 -2.45968372e-01 1.04452729e+00
9.01070595e-01 5.36777258e-01 1.02069080e-01 5.56021035e-01
1.44432259e+00 -7.82524228e-01 4.01438624e-02 -2.57117718e-01
5.16768992e-01 -8.88492525e-01 5.02046347e-01 4.49513405e-04
-1.29483080e+00 -6.74847543e-01 -1.16942489e+00 -2.51910090e-01
-6.80260479e-01 -2.18868796e-02 1.03374243e+00 6.48587346e-01
-8.45385015e-01 1.14268148e+00 -1.13022244e+00 -2.17583120e-01
6.27786040e-01 3.16986680e-01 2.63452847e-02 6.03417516e-01
-6.62393928e-01 7.64094591e-01 5.61982393e-01 -4.24236298e-01
-6.30976915e-01 -1.23741150e+00 -4.61932182e-01 2.38670453e-01
-4.80773933e-02 -8.80978167e-01 1.50065088e+00 -3.48420590e-01
-1.38719392e+00 8.97111893e-01 4.18964736e-02 -6.21377707e-01
2.03820810e-01 -1.06612056e-01 -4.36713099e-01 3.82524371e-01
-1.15120202e-01 4.21064019e-01 8.30592275e-01 -9.72283661e-01
-8.11747968e-01 -6.67936131e-02 1.89648737e-04 -5.43015361e-01
1.45655079e-03 5.73288739e-01 -1.90678407e-02 -3.52927089e-01
1.51089514e-02 -8.27111125e-01 -2.15118721e-01 1.30754218e-01
-3.74949396e-01 6.33088052e-02 1.07909477e+00 -2.03115001e-01
7.57293310e-03 -1.98095596e+00 7.23377764e-02 1.18303142e-01
7.03591347e-01 4.12614346e-01 1.53999865e-01 3.76863837e-01
-1.42206296e-01 2.28606731e-01 4.88751292e-01 6.03478812e-02
2.50938267e-01 -9.44277421e-02 -2.82499582e-01 3.06070596e-01
4.58767772e-01 1.37861204e+00 -1.20545626e+00 1.55248091e-01
8.04524601e-01 4.64663923e-01 -4.38545048e-01 1.92286670e-01
-3.67297798e-01 8.54923844e-01 -5.12930512e-01 7.64594898e-02
6.90920115e-01 -3.55281413e-01 -8.69633108e-02 -1.63922310e-01
-4.14377153e-01 5.46990573e-01 -5.43556213e-01 1.31032908e+00
-3.04249138e-01 1.05085802e+00 1.25719383e-01 -9.78094935e-01
6.87075317e-01 -3.94153446e-02 4.82039452e-01 -1.04015434e+00
2.10476160e-01 -6.28131330e-02 3.81299824e-01 -4.11235332e-01
3.17404151e-01 -5.29774725e-01 1.90439537e-01 3.78088474e-01
5.29494941e-01 2.77051423e-03 3.69049646e-02 2.78748870e-01
1.35506928e+00 -5.13830371e-02 -1.16027147e-01 -4.69921529e-01
-1.45096987e-01 4.24822748e-01 1.33481920e-01 1.07628036e+00
8.96941796e-02 1.59789562e-01 3.50559503e-01 -9.67490435e-01
-1.75314856e+00 -1.30741930e+00 -2.20415622e-01 1.37293911e+00
-4.71173273e-03 -6.05557740e-01 2.68189088e-02 -5.08506656e-01
2.16780007e-01 5.01996040e-01 -3.59447628e-01 -4.84760165e-01
-5.61305761e-01 -6.76053286e-01 1.49104893e-01 9.33716118e-01
5.15361726e-02 -1.15295923e+00 -7.50583112e-01 4.05229181e-02
8.65693390e-01 -9.55546677e-01 3.45409513e-01 5.67167521e-01
-5.83254874e-01 -1.06572700e+00 -6.31827265e-02 -4.42804217e-01
3.28729481e-01 1.67886913e-01 1.66074967e+00 2.93611549e-02
-7.66785324e-01 3.04152966e-01 -3.62151951e-01 -7.90897250e-01
-7.97789931e-01 -3.63956898e-01 2.25708820e-02 -6.06983304e-01
4.56984878e-01 -3.49121362e-01 -7.59024441e-01 -3.12558293e-01
-4.24482822e-01 5.85166030e-02 5.86321592e-01 6.73139274e-01
1.01828091e-01 3.13120782e-01 -2.35668555e-01 -8.96841705e-01
3.16450655e-01 -3.97856772e-01 -1.06478775e+00 -2.13785231e-01
-7.80797452e-02 4.73135889e-01 5.42217970e-01 -8.70173126e-02
-9.74691153e-01 -2.33727828e-01 -3.27778280e-01 -2.68163234e-01
-1.44509524e-01 -1.36623144e-01 1.78972259e-01 -6.09243691e-01
4.09583211e-01 -3.03494811e-01 -6.39166594e-01 -6.86948121e-01
7.90579379e-01 -5.93198128e-02 6.96911514e-01 -5.24909019e-01
1.28736079e+00 6.94059849e-01 5.21788299e-01 -6.13275528e-01
-9.96458054e-01 -5.74775219e-01 -1.03149927e+00 -7.03224242e-02
1.24362421e+00 -4.76999074e-01 -7.94753909e-01 -1.30153835e-01
-1.13912714e+00 -9.12623107e-02 -8.26917291e-01 8.44699621e-01
-2.88197309e-01 -1.87854499e-01 -5.50698042e-01 -6.67229891e-01
-1.21203050e-01 -8.81653249e-01 1.30401576e+00 4.36219752e-01
7.20566884e-02 -1.29542983e+00 3.98988038e-01 -8.57476890e-02
5.76769173e-01 1.03848934e-01 1.53104532e+00 -8.18323433e-01
-7.05515146e-01 -1.30353957e-01 -5.36792517e-01 7.82616436e-02
-3.38698536e-01 2.63963211e-02 -8.40079188e-01 -1.56269804e-01
1.87110156e-01 -3.03632379e-01 1.19084084e+00 5.25087237e-01
1.03139734e+00 2.15777978e-01 -6.01591647e-01 7.74712980e-01
1.38345873e+00 3.03142190e-01 4.82362621e-02 3.97617787e-01
6.76079452e-01 2.86570657e-02 8.85657519e-02 2.42379010e-01
-4.41455215e-01 7.07855344e-01 3.50228578e-01 -5.09848475e-01
-4.96332794e-01 -1.96512043e-01 -4.09250110e-02 7.18247950e-01
-1.03729561e-01 -1.46914618e-02 -8.52662385e-01 2.21464708e-02
-1.61336672e+00 -1.21933353e+00 -3.99950951e-01 1.86976218e+00
2.44388923e-01 8.12359571e-01 8.91891420e-02 -2.23411933e-01
5.60151935e-01 2.09035158e-01 -4.20054615e-01 -7.68225431e-01
3.02557588e-01 9.74153101e-01 6.98691368e-01 2.94477731e-01
-1.23194265e+00 6.93198264e-01 7.92056274e+00 4.93585587e-01
-1.10812056e+00 4.49539810e-01 3.15411508e-01 -2.99111098e-01
-2.52436638e-01 1.68667242e-01 -8.87009859e-01 3.43374997e-01
1.07012558e+00 1.61833898e-03 3.12973142e-01 7.92832613e-01
1.16762079e-01 8.44011605e-02 -1.47321975e+00 7.66792059e-01
-4.69867080e-01 -2.25483584e+00 -1.96993455e-01 2.21026555e-01
5.14642417e-01 5.39702892e-01 3.17724757e-02 3.39797080e-01
8.73368263e-01 -8.07308912e-01 5.91147542e-01 7.83196568e-01
3.46568376e-01 -4.75580633e-01 4.86029565e-01 6.76404387e-02
-1.00662947e+00 8.93756971e-02 -8.19757342e-01 -2.96627551e-01
6.98393732e-02 9.67380822e-01 -9.01497841e-01 3.07224721e-01
5.61611176e-01 4.25348639e-01 -5.61617136e-01 1.24243772e+00
-5.44765629e-02 8.03406715e-01 -2.23947152e-01 -1.04229130e-01
3.79514813e-01 2.18974911e-02 7.57944107e-01 1.41195095e+00
2.10444629e-01 -3.05293620e-01 2.62055337e-01 1.16334331e+00
-8.67053941e-02 -4.93870884e-01 -6.92177653e-01 -4.87876147e-01
-9.49851274e-02 1.67268777e+00 -1.19400656e+00 -3.85848641e-01
-6.19658828e-01 3.64680588e-01 2.46265247e-01 8.63065720e-02
-7.33703494e-01 3.22193243e-02 8.33238304e-01 2.49609739e-01
6.37083828e-01 -7.38573134e-01 1.94171686e-02 -9.87353206e-01
-3.94214630e-01 -2.16455981e-02 -9.34373960e-02 -9.03937101e-01
-1.39744544e+00 3.22534628e-02 -3.57521325e-01 -6.78177118e-01
-6.81461468e-02 -1.32045496e+00 -1.20000958e+00 5.23492813e-01
-1.48026574e+00 -8.24791670e-01 7.72663280e-02 8.78844038e-03
2.96968788e-01 -5.33922970e-01 7.38983810e-01 -2.24726751e-01
-2.21790344e-01 -2.41426304e-01 6.67531550e-01 3.38120610e-01
2.37493411e-01 -1.61774492e+00 9.42112803e-01 6.62503898e-01
9.83290553e-01 2.34344006e-01 7.41283715e-01 -7.37055182e-01
-1.84420800e+00 -9.72386479e-01 3.65996242e-01 -8.38033974e-01
1.12217581e+00 -9.54322159e-01 -5.17592371e-01 5.65382957e-01
3.79459769e-01 5.95661879e-01 4.99833852e-01 3.97636205e-01
-3.49440783e-01 1.51966929e-01 -8.79446924e-01 5.90566136e-02
1.12926078e+00 -7.35617399e-01 -6.89091802e-01 1.02792776e+00
5.24695694e-01 -2.60288388e-01 -6.54746294e-01 2.48018369e-01
3.21784019e-01 -1.23916864e+00 1.30942583e+00 -1.40528047e+00
4.13439631e-01 2.68435180e-01 4.12265152e-01 -1.36152995e+00
-1.09257674e+00 -4.46592212e-01 -5.96768595e-02 9.56005335e-01
2.24584684e-01 -2.64802009e-01 9.94666398e-01 2.16538906e-01
1.97045039e-02 -3.27767402e-01 -1.02374506e+00 -8.94617379e-01
3.40058506e-01 -4.94658351e-01 3.30008864e-01 6.20170593e-01
-1.71331584e-01 2.30530530e-01 -1.67067777e-02 2.04104424e-01
4.71763194e-01 4.49132502e-01 3.29335243e-01 -1.66608000e+00
-5.70745766e-01 -6.83385015e-01 -8.80489886e-01 -2.97735095e-01
-6.86138570e-02 -1.31887293e+00 -2.48552218e-01 -1.19539976e+00
5.42378128e-01 -4.08235222e-01 -4.92249340e-01 6.19677901e-02
2.47002736e-01 3.72856021e-01 3.81329179e-01 -5.94682023e-02
-7.13760257e-01 1.45608634e-01 1.05323100e+00 -1.34562701e-01
4.00041819e-01 -7.00758547e-02 -2.72730887e-01 1.24985981e+00
7.17164397e-01 -7.02857852e-01 8.57655406e-02 -5.58703125e-01
6.74852610e-01 -2.03620493e-01 9.49803650e-01 -1.33432984e+00
1.67562068e-01 -1.39852762e-02 1.34608555e+00 -4.50835764e-01
4.26218182e-01 -5.41551471e-01 -3.49903315e-01 4.27741855e-01
-3.20624858e-01 -8.07591304e-02 1.21618576e-01 3.24077994e-01
1.34185612e-01 -6.79902077e-01 8.41935754e-01 -7.05524862e-01
-6.91043437e-01 9.22013894e-02 -2.84171045e-01 2.61454374e-01
9.36650932e-01 5.00934601e-01 -6.98099434e-01 -3.58207896e-02
-8.31501424e-01 -2.33452931e-01 3.32854003e-01 4.35306400e-01
2.82449186e-01 -1.51066971e+00 -3.06827903e-01 3.75908017e-01
-2.15468705e-01 -7.36558497e-01 -3.33854139e-01 2.92655349e-01
-5.65060556e-01 7.73409605e-01 -3.53883356e-01 -6.36252105e-01
-9.43848610e-01 5.98185778e-01 4.78121400e-01 -3.02451074e-01
-1.06363428e+00 1.15293121e+00 2.05305919e-01 -1.20126180e-01
-3.24561805e-01 -4.49176043e-01 2.30394363e-01 -4.52851146e-01
5.22669256e-01 6.77449405e-02 1.49712205e-01 -2.86253542e-01
-2.72315592e-01 3.46520543e-01 -7.67413294e-03 6.22680336e-02
1.45410645e+00 4.30592507e-01 5.11185788e-02 4.00599331e-01
1.07249594e+00 3.67837429e-01 -1.26547706e+00 -4.33079302e-02
2.61137635e-01 -5.20265996e-01 4.93527263e-01 -9.41628695e-01
-1.14816773e+00 1.23806977e+00 8.41499984e-01 6.19973719e-01
2.58703470e-01 6.05804145e-01 9.82693017e-01 3.51723313e-01
7.84746781e-02 -8.32094252e-01 2.59509563e-01 4.10211295e-01
2.89519846e-01 -1.33465910e+00 4.87150624e-02 -1.42909825e-01
1.18020505e-01 1.73764098e+00 3.89658928e-01 -5.71055532e-01
7.32477069e-01 9.74267542e-01 -2.65583217e-01 -8.11045170e-01
-6.38335705e-01 -3.76327604e-01 4.81578618e-01 5.32262385e-01
2.50969678e-01 6.71519190e-02 1.64548025e-01 3.71876091e-01
-4.99473095e-01 -1.79793060e-01 1.95506677e-01 9.87043679e-01
-8.19732666e-01 -1.45660102e+00 -2.24871740e-01 6.95494652e-01
-2.43792981e-01 -4.14895862e-02 -4.25005287e-01 7.08797395e-01
5.65301776e-01 5.31877100e-01 3.76968354e-01 -2.32513800e-01
-3.17536220e-02 3.32293838e-01 7.78960884e-01 -8.68406951e-01
-5.59356213e-01 -1.36682004e-01 -2.91888323e-02 -6.31288111e-01
-4.59461331e-01 -4.28284854e-01 -1.22689271e+00 -1.63938954e-01
-1.66794837e-01 2.66547561e-01 1.04514551e+00 1.30968630e+00
2.50027418e-01 8.09462786e-01 3.90716493e-01 -1.26899767e+00
-3.53059232e-01 -4.12227988e-01 -5.06208956e-01 4.13670212e-01
3.33853364e-01 -8.76629472e-01 -1.65334836e-01 -3.32865238e-01] | [15.700263977050781, 2.9188120365142822] |
91a353b5-579f-480f-a90c-93bfa6524d8c | vibe-video-inference-for-human-body-pose-and | 1912.05656 | null | https://arxiv.org/abs/1912.05656v3 | https://arxiv.org/pdf/1912.05656v3.pdf | VIBE: Video Inference for Human Body Pose and Shape Estimation | Human motion is fundamental to understanding behavior. Despite progress on single-image 3D pose and shape estimation, existing video-based state-of-the-art methods fail to produce accurate and natural motion sequences due to a lack of ground-truth 3D motion data for training. To address this problem, we propose Video Inference for Body Pose and Shape Estimation (VIBE), which makes use of an existing large-scale motion capture dataset (AMASS) together with unpaired, in-the-wild, 2D keypoint annotations. Our key novelty is an adversarial learning framework that leverages AMASS to discriminate between real human motions and those produced by our temporal pose and shape regression networks. We define a temporal network architecture and show that adversarial training, at the sequence level, produces kinematically plausible motion sequences without in-the-wild ground-truth 3D labels. We perform extensive experimentation to analyze the importance of motion and demonstrate the effectiveness of VIBE on challenging 3D pose estimation datasets, achieving state-of-the-art performance. Code and pretrained models are available at https://github.com/mkocabas/VIBE. | ['Nikos Athanasiou', 'Michael J. Black', 'Muhammed Kocabas'] | 2019-12-11 | vibe-video-inference-for-human-body-pose-and-1 | http://openaccess.thecvf.com/content_CVPR_2020/html/Kocabas_VIBE_Video_Inference_for_Human_Body_Pose_and_Shape_Estimation_CVPR_2020_paper.html | http://openaccess.thecvf.com/content_CVPR_2020/papers/Kocabas_VIBE_Video_Inference_for_Human_Body_Pose_and_Shape_Estimation_CVPR_2020_paper.pdf | cvpr-2020-6 | ['monocular-3d-human-pose-estimation'] | ['computer-vision'] | [ 3.05625647e-02 -2.92330384e-02 -3.30722153e-01 -1.41607821e-01
-8.91512692e-01 -7.57174373e-01 5.17212033e-01 -6.21471941e-01
-4.35776621e-01 4.51993287e-01 3.57381076e-01 1.60005726e-02
3.27879846e-01 -3.58420312e-01 -1.11136472e+00 -3.50942314e-01
-2.24231914e-01 5.47964036e-01 2.96965182e-01 -1.29759401e-01
-4.01463717e-01 3.17471355e-01 -1.14418399e+00 1.10989749e-01
2.91381806e-01 7.56740987e-01 -3.41914445e-01 1.21967983e+00
7.34106839e-01 7.64964581e-01 -3.03417444e-01 -4.22101647e-01
5.25653839e-01 -5.91521800e-01 -7.57239461e-01 1.51280418e-01
9.80323970e-01 -8.92475486e-01 -8.66948783e-01 7.01748788e-01
5.34691572e-01 2.85666049e-01 5.48245490e-01 -1.53088152e+00
-3.06789100e-01 7.12728202e-02 -2.72136033e-01 2.25225791e-01
6.33926332e-01 7.71727741e-01 6.92958534e-01 -6.25319839e-01
1.07644093e+00 1.19486189e+00 1.01074243e+00 9.89886880e-01
-1.24631894e+00 -4.04014081e-01 8.15117881e-02 -2.91198511e-02
-1.22230291e+00 -5.28422952e-01 7.78560519e-01 -7.26718545e-01
8.24067950e-01 1.25923380e-01 1.04448354e+00 1.84898138e+00
1.34409666e-01 9.76837337e-01 4.34374809e-01 -8.77627954e-02
1.68075319e-03 -7.28552938e-01 -4.13325280e-01 9.04235721e-01
1.40159309e-01 3.51311773e-01 -6.13013983e-01 -1.82667896e-02
1.22025681e+00 -1.33728132e-01 -4.15142745e-01 -8.57079089e-01
-1.35092545e+00 4.91658807e-01 3.13319772e-01 -4.53661568e-02
-2.90445626e-01 8.31262171e-01 3.19790274e-01 -5.08925058e-02
2.31305718e-01 8.72574672e-02 -3.18178415e-01 -4.98171717e-01
-9.66391146e-01 7.82808006e-01 6.40033185e-01 7.69376278e-01
2.04817697e-01 3.35047901e-01 -8.30138922e-02 2.61449546e-01
1.76673666e-01 7.77258456e-01 5.32525241e-01 -1.68311691e+00
4.46672380e-01 1.27007738e-01 2.89890110e-01 -1.08587646e+00
-3.93274546e-01 -1.34457320e-01 -4.92226571e-01 2.33398765e-01
8.04240346e-01 -1.45291924e-01 -1.01753926e+00 2.02197146e+00
5.25902629e-01 5.52738309e-01 -1.36843219e-01 1.29855955e+00
6.73261166e-01 2.66389757e-01 -3.26440781e-02 3.28984261e-01
9.13502991e-01 -1.11117637e+00 -2.47444183e-01 -4.61833477e-01
5.55248797e-01 -3.13147187e-01 7.78772116e-01 1.61840051e-01
-1.30750430e+00 -5.63550711e-01 -8.47245216e-01 -1.59577072e-01
2.36091703e-01 -1.16993822e-01 4.57470655e-01 4.47499931e-01
-9.54512060e-01 7.57404625e-01 -1.52445138e+00 -3.95005554e-01
3.73424768e-01 3.23060602e-01 -7.43283331e-01 8.15695599e-02
-9.73328710e-01 7.26824522e-01 3.53648886e-02 2.21729681e-01
-1.50372303e+00 -7.67751992e-01 -1.30177987e+00 -5.23585796e-01
4.71247405e-01 -1.30487919e+00 1.67710888e+00 -9.97068405e-01
-1.21805549e+00 1.03470445e+00 -1.21724773e-02 -5.15657604e-01
1.16440606e+00 -6.68363929e-01 2.41610897e-03 6.01582885e-01
7.51798004e-02 1.02073002e+00 9.12807584e-01 -1.08290470e+00
-1.51971221e-01 -2.35883445e-01 -7.39514083e-02 1.51005000e-01
1.80835128e-01 -3.47378969e-01 -8.14036489e-01 -9.32371318e-01
-1.14899032e-01 -1.34385443e+00 -2.43048549e-01 3.80056649e-01
-3.02970737e-01 3.29171836e-01 5.68634689e-01 -9.31998253e-01
8.52256715e-01 -1.87469304e+00 5.26718020e-01 -1.49616644e-01
2.24746183e-01 3.19802344e-01 -1.74725547e-01 1.66964769e-01
1.03197843e-02 -1.89928897e-02 -2.78019488e-01 -5.03234804e-01
3.61896567e-02 1.88380033e-01 -1.56076746e-02 7.49388039e-01
2.69301593e-01 1.31790364e+00 -1.07119179e+00 -4.40212429e-01
3.69137555e-01 5.46013713e-01 -8.75409484e-01 3.90534729e-01
-3.26924980e-01 9.13057268e-01 -4.43960667e-01 8.11622381e-01
1.18726879e-01 -2.71380007e-01 1.04135171e-01 -2.99264103e-01
3.61438096e-01 -1.17380664e-01 -8.69122803e-01 2.31794000e+00
7.79688507e-02 5.91085017e-01 8.39953721e-02 -7.38035321e-01
2.07853332e-01 4.29213881e-01 8.72817695e-01 -3.00793767e-01
3.15846473e-01 9.95908305e-02 -9.60079804e-02 -6.72877550e-01
4.02342409e-01 -4.44595702e-02 -1.92011222e-01 1.55050427e-01
4.18767892e-02 -1.03467830e-01 -6.79245219e-02 1.30389422e-01
1.36243880e+00 9.86091256e-01 1.25622287e-01 2.90451825e-01
2.03234270e-01 2.06692562e-01 7.29136825e-01 5.22967041e-01
-6.62676811e-01 1.03376162e+00 1.96117342e-01 -6.02697551e-01
-1.45205915e+00 -1.29166663e+00 4.31561977e-01 7.13699520e-01
9.58634540e-02 -2.80192852e-01 -8.67098987e-01 -7.02862859e-01
1.99303135e-01 1.50716901e-01 -7.33186722e-01 -1.17527835e-01
-9.81276631e-01 -2.83429682e-01 9.68516827e-01 8.22344005e-01
3.60663176e-01 -8.55441451e-01 -9.32839751e-01 1.75947338e-01
-3.66431832e-01 -1.55527246e+00 -8.11471701e-01 -5.65409899e-01
-7.83077419e-01 -1.30387318e+00 -9.74513710e-01 -4.16584611e-01
4.45086181e-01 1.47691900e-02 1.22473752e+00 3.07567921e-02
-3.71360749e-01 8.99446368e-01 -3.37063909e-01 8.22621435e-02
-4.89510804e-01 -1.41702846e-01 3.34488630e-01 -2.56051779e-01
-6.55410485e-03 -7.80091882e-01 -9.66562033e-01 3.45927387e-01
-7.46340334e-01 8.63113105e-02 2.44828716e-01 6.62127435e-01
6.25006616e-01 -5.98032296e-01 7.95009062e-02 -4.03228581e-01
-4.35019135e-02 -3.59315872e-01 -3.97571951e-01 -1.71917140e-01
1.34996995e-02 -2.57893521e-02 2.69761473e-01 -6.54229939e-01
-5.41096687e-01 3.15310806e-01 -3.07958573e-01 -1.05695045e+00
-2.75949329e-01 8.77990872e-02 -7.85668194e-03 -7.50613362e-02
6.39536202e-01 8.81858245e-02 2.89247483e-01 -2.39362761e-01
4.44635302e-01 7.05508813e-02 1.24807727e+00 -6.15239739e-01
9.10610080e-01 8.09632599e-01 2.27046564e-01 -5.62788367e-01
-7.22330809e-01 -3.59638691e-01 -9.09849942e-01 -4.85896498e-01
1.31711566e+00 -1.25320256e+00 -7.43704617e-01 7.28977501e-01
-9.06592011e-01 -8.31390262e-01 -6.40373230e-02 5.80682278e-01
-1.10778928e+00 5.13994753e-01 -7.71328032e-01 -6.97662115e-01
-2.54301995e-01 -1.04328835e+00 1.41205728e+00 -2.31108323e-01
-7.30478346e-01 -8.71336877e-01 2.76849091e-01 5.83396792e-01
-1.69834599e-01 1.00413191e+00 2.67840773e-01 -2.27141246e-01
-7.56784201e-01 -2.54707873e-01 4.77891535e-01 1.98349923e-01
-2.36590788e-01 -6.41432703e-02 -7.61324823e-01 -3.35379213e-01
-3.08315367e-01 -6.08558357e-01 7.46223092e-01 5.91741204e-01
6.87101424e-01 -2.26952806e-01 -2.55876988e-01 1.06905711e+00
1.00098205e+00 -2.10986614e-01 5.29670894e-01 3.12570661e-01
1.19425857e+00 3.80217522e-01 4.91802305e-01 2.84264386e-01
5.01177490e-01 9.66816843e-01 5.14357865e-01 8.32946002e-02
-3.32004666e-01 -6.79220855e-01 5.79217255e-01 5.47501326e-01
-2.32344106e-01 -1.50508672e-01 -9.77390826e-01 6.32843673e-01
-1.97272217e+00 -1.16257846e+00 1.05225250e-01 2.03133416e+00
6.71508908e-01 1.28204167e-01 6.24522805e-01 -8.30359682e-02
4.60389256e-01 2.43985355e-01 -7.34179497e-01 1.72637999e-01
1.47980109e-01 -9.79330316e-02 5.00069082e-01 6.12066031e-01
-1.38875258e+00 9.20054972e-01 5.87083626e+00 3.52322489e-01
-1.05587399e+00 2.86759175e-02 4.26825285e-01 -5.40261626e-01
-1.67855918e-01 -2.75363863e-01 -5.25947332e-01 3.63055497e-01
7.51994014e-01 2.89176971e-01 3.59081328e-01 7.12592483e-01
4.70773935e-01 9.59083810e-02 -1.29326248e+00 1.16259992e+00
2.08503813e-01 -1.24763799e+00 -7.82008395e-02 3.78311314e-02
7.66163826e-01 1.52139425e-01 -7.91867673e-02 1.12538390e-01
2.99125820e-01 -1.08702767e+00 1.16942704e+00 6.83921039e-01
9.01985586e-01 -4.08552974e-01 3.65728736e-01 4.31240201e-01
-1.22638118e+00 1.87971711e-01 1.50672004e-01 -7.35815167e-02
4.95140940e-01 -6.50540218e-02 -5.39515555e-01 5.95499992e-01
7.14992940e-01 8.39236617e-01 -4.30897087e-01 7.66551793e-01
-2.87365288e-01 5.00058472e-01 -4.37674046e-01 3.77020717e-01
2.65604556e-01 1.86656728e-01 8.96797955e-01 9.46070075e-01
2.75520921e-01 7.05964193e-02 4.29997921e-01 7.39311695e-01
-1.26548454e-01 -5.02272308e-01 -7.41119504e-01 -1.63629666e-01
2.91645646e-01 7.98070669e-01 -5.44490218e-01 -2.42807850e-01
-2.69233108e-01 1.32287800e+00 1.91377386e-01 3.76034647e-01
-1.00044024e+00 3.34960610e-01 9.71993148e-01 2.45711386e-01
4.01465774e-01 -7.23265409e-01 1.35838747e-01 -1.47426212e+00
1.26797691e-01 -1.01379228e+00 3.97122264e-01 -8.16110015e-01
-1.07766461e+00 1.82619110e-01 2.68866241e-01 -1.32797086e+00
-9.91882443e-01 -6.56224549e-01 -4.43836689e-01 5.00432014e-01
-6.99184835e-01 -1.41899025e+00 -4.45941687e-01 5.10949671e-01
5.72582603e-01 1.36604533e-01 4.79835808e-01 2.50962168e-01
-4.09452289e-01 6.65670156e-01 -3.33215058e-01 4.92226273e-01
7.01585293e-01 -9.72383380e-01 9.56015110e-01 9.90898311e-01
2.80746371e-01 2.53347963e-01 8.27258527e-01 -7.63477743e-01
-1.64355159e+00 -1.17237198e+00 3.01455855e-01 -1.25584078e+00
5.17912149e-01 -3.62416297e-01 -5.89101851e-01 1.02317846e+00
-3.77986670e-01 4.38970923e-01 3.30656946e-01 -4.42525804e-01
-2.76876420e-01 3.04055899e-01 -9.25125480e-01 8.05834293e-01
1.60145175e+00 -4.82730836e-01 -5.99735260e-01 -6.71527162e-02
7.00703204e-01 -8.87385607e-01 -8.59718621e-01 6.39402449e-01
1.06020725e+00 -9.17110920e-01 1.28300798e+00 -8.56296718e-01
6.71510577e-01 -3.68443966e-01 -1.60070658e-01 -1.03781569e+00
-9.24016014e-02 -7.11600304e-01 -5.79328716e-01 6.34472311e-01
-3.53719853e-02 9.48538035e-02 1.23135316e+00 6.18909121e-01
1.27626173e-02 -6.03479326e-01 -8.67070198e-01 -9.45542872e-01
1.24811247e-01 -7.69406676e-01 1.49792835e-01 7.75310814e-01
-4.31580037e-01 1.03760011e-01 -7.96470225e-01 1.44143134e-01
6.96672976e-01 -1.09601036e-01 1.34995866e+00 -6.18684769e-01
-6.94421768e-01 -2.00371385e-01 -8.96548867e-01 -1.26473057e+00
3.43860477e-01 -6.81306064e-01 2.05254808e-01 -1.25870788e+00
4.98378053e-02 2.59702832e-01 2.51323849e-01 4.56828147e-01
-1.65854678e-01 7.47702360e-01 4.81751233e-01 3.00904304e-01
-7.08972871e-01 5.67268670e-01 1.35920346e+00 1.24759157e-03
-6.37194812e-02 -1.98337451e-01 -6.23106919e-02 9.75249708e-01
5.31948745e-01 -4.00165737e-01 -4.45125043e-01 -5.38189292e-01
-1.03196807e-01 3.29518855e-01 1.15620542e+00 -1.17232597e+00
-4.77671549e-02 -4.83649820e-02 7.31195092e-01 -5.12397587e-01
6.66312456e-01 -5.39605677e-01 4.14841712e-01 6.83945298e-01
-1.97742060e-01 1.83268502e-01 2.90232956e-01 7.36842811e-01
1.28963485e-01 1.14361338e-01 4.91257489e-01 -3.63870949e-01
-9.04653132e-01 6.68679059e-01 -1.04841731e-01 5.55318594e-01
8.35351527e-01 -3.20775986e-01 -3.96997854e-02 -7.11454511e-01
-8.90185952e-01 1.20088987e-01 8.94133389e-01 4.25839543e-01
4.87079144e-01 -1.43284345e+00 -6.53736651e-01 1.45257944e-02
9.32149589e-03 1.84846595e-01 3.46603125e-01 7.12861240e-01
-8.97278965e-01 2.27260441e-01 -3.05902183e-01 -8.78072798e-01
-1.21720290e+00 4.56607461e-01 6.71209335e-01 -7.34699443e-02
-8.51893187e-01 6.87668800e-01 1.78117946e-01 -4.03765619e-01
9.86526832e-02 -2.99218476e-01 4.13174868e-01 -5.70162714e-01
3.02011371e-01 3.78152341e-01 -3.52864593e-01 -8.98328543e-01
-4.22206610e-01 7.16761947e-01 4.91783559e-01 -5.66299736e-01
1.01445198e+00 -4.34801988e-02 4.86445785e-01 4.06449467e-01
1.22426474e+00 -6.68088123e-02 -1.95720899e+00 2.90058017e-01
-2.98564315e-01 -6.15602732e-01 -4.10589635e-01 -4.50921595e-01
-8.32742631e-01 6.72148883e-01 4.27903950e-01 -4.36310142e-01
7.63645172e-01 -4.80544753e-02 1.11135042e+00 3.33731681e-01
3.52126002e-01 -7.91529179e-01 3.57572675e-01 5.05134404e-01
7.84111142e-01 -1.17679560e+00 -5.57478741e-02 -3.14242840e-01
-5.83516121e-01 8.17543268e-01 7.86101997e-01 -3.46609056e-01
3.02876174e-01 1.61641076e-01 1.78628415e-01 -2.22349521e-02
-5.32894313e-01 -1.15941674e-01 4.99731481e-01 7.87564993e-01
3.76564443e-01 -1.30827606e-01 1.10149875e-01 4.36146885e-01
-3.46345603e-01 2.32009724e-01 2.34525859e-01 1.00748003e+00
5.99508323e-02 -9.43275154e-01 -3.97423893e-01 -1.18238971e-01
-6.46247625e-01 2.19075069e-01 -3.84616107e-01 1.04547119e+00
2.81381961e-02 5.15828609e-01 2.36020721e-02 -5.42618752e-01
2.70153195e-01 2.26272941e-02 9.76570308e-01 -3.10413897e-01
-2.61862576e-01 4.08639163e-02 3.38115931e-01 -1.12607324e+00
-4.93424088e-01 -7.18834400e-01 -1.24270892e+00 -3.79231870e-01
4.17906821e-01 -2.68568486e-01 2.05769554e-01 8.28522384e-01
3.39150250e-01 4.04198796e-01 -1.94140635e-02 -1.48607731e+00
-5.24136424e-01 -7.38535523e-01 -1.07246511e-01 1.00619435e+00
6.57944441e-01 -6.75480008e-01 -1.04352497e-01 5.24716854e-01] | [7.187196731567383, -0.5875267386436462] |
ad6e9de2-81a2-4ed1-830e-561c137b797a | a-deep-learning-approach-to-clustering-visual | 2106.06234 | null | https://arxiv.org/abs/2106.06234v2 | https://arxiv.org/pdf/2106.06234v2.pdf | A deep learning approach to clustering visual arts | Clustering artworks is difficult for several reasons. On the one hand, recognizing meaningful patterns based on domain knowledge and visual perception is extremely hard. On the other hand, applying traditional clustering and feature reduction techniques to the highly dimensional pixel space can be ineffective. To address these issues, in this paper we propose DELIUS: a DEep learning approach to cLustering vIsUal artS. The method uses a pre-trained convolutional network to extract features and then feeds these features into a deep embedded clustering model, where the task of mapping the input data to a latent space is jointly optimized with the task of finding a set of cluster centroids in this latent space. Quantitative and qualitative experimental results show the effectiveness of the proposed method. DELIUS can be useful for several tasks related to art analysis, in particular visual link retrieval and historical knowledge discovery in painting datasets. | ['Gennaro Vessio', 'Giovanna Castellano'] | 2021-06-11 | null | null | null | null | ['art-analysis'] | ['computer-vision'] | [ 4.12565619e-02 -4.96658474e-01 -9.88947451e-02 -2.73308605e-01
-3.64832014e-01 -5.33404112e-01 5.59298575e-01 1.39856443e-01
-1.25947520e-01 1.92006975e-01 1.33315861e-01 1.37737751e-01
-5.76882601e-01 -8.67849827e-01 -4.87443209e-01 -8.58425438e-01
1.50167346e-01 5.41492581e-01 -4.41162810e-02 3.71847242e-01
4.94224042e-01 8.13407362e-01 -1.59995818e+00 5.88837743e-01
6.34185493e-01 1.06894898e+00 3.60587120e-01 3.14947665e-01
-3.39885443e-01 7.68722892e-01 -4.17083293e-01 -2.59625107e-01
1.84059620e-01 -2.76496887e-01 -6.11731768e-01 3.57551038e-01
3.08464855e-01 -8.01183581e-02 -4.59049940e-01 1.05951715e+00
2.95242339e-01 2.77040154e-01 8.70008171e-01 -1.21377850e+00
-7.14575350e-01 5.24936855e-01 -7.41053641e-01 -5.76476417e-02
-1.53683454e-01 -5.06799445e-02 9.87591982e-01 -8.13609958e-01
7.46951401e-01 1.33102727e+00 5.15425503e-01 2.17283025e-01
-1.25857747e+00 -6.00319922e-01 3.95822562e-02 3.26423705e-01
-1.58841646e+00 -3.80815685e-01 1.22078168e+00 -8.76744211e-01
4.02298123e-01 6.32335693e-02 6.99394464e-01 7.04716504e-01
-1.90565750e-01 8.53216171e-01 9.03408229e-01 -3.51953179e-01
5.37518263e-01 -1.52693363e-02 -1.20053403e-01 6.83611989e-01
-1.55139975e-02 -3.42897743e-01 -5.77442288e-01 1.55039355e-01
8.77046883e-01 3.99249882e-01 1.49160195e-02 -5.51164925e-01
-1.02347541e+00 9.21641588e-01 7.82470822e-01 5.20767868e-01
-5.29120743e-01 4.69643652e-01 2.83196390e-01 -1.12536825e-01
3.24067175e-01 5.31528234e-01 6.77640289e-02 1.25145569e-01
-1.20518553e+00 4.85047288e-02 4.75889117e-01 5.36490202e-01
8.35162461e-01 -1.89029425e-01 -2.02585738e-02 8.85966957e-01
4.52521473e-01 4.90791304e-03 2.33705342e-01 -1.16109288e+00
2.73617536e-01 9.88655806e-01 -2.45900244e-01 -1.44502771e+00
-2.11353078e-01 -3.25008631e-01 -8.96221697e-01 5.32165468e-01
3.90140355e-01 1.39019758e-01 -9.54247177e-01 1.48726153e+00
2.16243654e-01 -3.26931626e-02 -1.25651300e-01 1.15763271e+00
6.07671320e-01 7.06950128e-01 7.63628855e-02 -1.50092626e-02
1.23731530e+00 -7.76125491e-01 -7.98336744e-01 -2.40555350e-02
3.06803156e-02 -9.36699808e-01 1.05546296e+00 4.30410743e-01
-8.02750170e-01 -4.79939967e-01 -1.01458490e+00 -1.86533764e-01
-4.54349786e-01 5.02723575e-01 7.90587664e-01 1.94744378e-01
-8.58978450e-01 6.19997323e-01 -9.85019803e-01 -5.26603162e-01
1.05372334e+00 4.69143599e-01 -9.86714363e-02 -1.53529868e-01
-5.31538546e-01 2.12407619e-01 4.72418547e-01 1.71367332e-01
-7.13416636e-01 -3.39076191e-01 -5.09707034e-01 2.79127508e-01
3.00019950e-01 -4.88975167e-01 5.46607435e-01 -1.11526549e+00
-1.35688138e+00 7.12144732e-01 -1.27254933e-01 -1.69367522e-01
4.19666111e-01 -2.97125399e-01 -2.29347393e-01 4.37048376e-01
1.65324345e-01 6.43299222e-01 9.55603123e-01 -1.47656417e+00
-4.43891525e-01 -5.48908830e-01 -3.45920771e-01 1.39826566e-01
-8.36224854e-01 -2.76823938e-01 -9.98711646e-01 -6.80240512e-01
3.57701689e-01 -7.09303975e-01 -1.15939461e-01 2.89251179e-01
-5.78990877e-01 -4.27167892e-01 1.17067087e+00 -4.83332276e-01
1.07199836e+00 -2.24141264e+00 3.73418391e-01 6.54783070e-01
3.45145166e-01 2.60654911e-02 5.61372228e-02 2.62923509e-01
7.93496892e-03 1.40156224e-02 -2.55877584e-01 -1.54008090e-01
-3.54638882e-02 -6.19215220e-02 -3.51428360e-01 4.43969220e-01
4.19310592e-02 8.81094813e-01 -8.05287302e-01 -7.00967968e-01
5.96888244e-01 5.80915272e-01 -3.31870258e-01 1.23457804e-01
-4.45114046e-01 3.18085700e-01 -4.92387414e-01 4.70698416e-01
3.75113130e-01 -4.58021253e-01 2.59437859e-01 -3.67242485e-01
-9.98005196e-02 -2.84567047e-02 -1.14719462e+00 2.00066042e+00
-3.56459320e-01 1.14299905e+00 -1.73866913e-01 -9.57241237e-01
9.12457585e-01 5.04416302e-02 8.35252762e-01 -5.52454293e-01
2.41348103e-01 -1.00098245e-01 -1.80455670e-01 -6.02180183e-01
2.63803720e-01 -7.60393813e-02 1.41545847e-01 5.87767124e-01
-1.23235725e-01 2.82078415e-01 5.33451363e-02 3.49061430e-01
1.06005073e+00 5.75767942e-02 -3.24788123e-01 -1.33815914e-01
2.92244285e-01 2.73558944e-01 4.16694909e-01 2.01694846e-01
1.20532118e-01 4.34721678e-01 4.79416519e-01 -5.32501876e-01
-9.88680363e-01 -1.20613003e+00 9.85469148e-02 7.31803775e-01
2.35016823e-01 -4.40504074e-01 -6.91753507e-01 -5.07935762e-01
8.59470665e-02 4.81180012e-01 -7.02168941e-01 -9.99782458e-02
-4.54458565e-01 -3.44488412e-01 2.63839662e-01 5.30067086e-01
5.90059400e-01 -1.16873240e+00 -5.58601856e-01 -1.69114266e-02
-1.56793997e-01 -8.63727748e-01 -1.95469290e-01 9.71822739e-02
-8.20416391e-01 -1.05904901e+00 -3.75729799e-01 -9.55765069e-01
8.85643363e-01 3.03527325e-01 8.28410566e-01 1.61702514e-01
-6.10149264e-01 3.02339792e-01 -2.25902051e-01 -1.47846252e-01
5.16503938e-02 1.58666059e-01 -1.65249467e-01 3.85864109e-01
6.35328174e-01 -6.08256340e-01 -8.23865592e-01 1.13550253e-01
-1.02114487e+00 1.52525946e-03 6.61620617e-01 3.87091875e-01
7.28183985e-01 6.68044865e-01 3.63681555e-01 -6.69889927e-01
4.88964528e-01 -3.12671512e-01 -5.91805875e-01 3.35208476e-01
-2.25829124e-01 5.05462708e-03 4.97610897e-01 -3.56277823e-01
-8.91226172e-01 6.38020277e-01 3.22537661e-01 -8.86205554e-01
-1.00937888e-01 5.82268596e-01 -5.67608416e-01 2.70191401e-01
4.91629750e-01 -4.10168879e-02 -1.92764074e-01 -6.53396308e-01
5.69930553e-01 7.00668156e-01 5.98533750e-01 -3.78107250e-01
8.37638378e-01 1.00236368e+00 1.24664180e-01 -9.67753708e-01
-8.08237791e-01 -4.64971632e-01 -9.79273140e-01 -5.80565691e-01
1.34404051e+00 -6.75618112e-01 -8.11479390e-01 5.28951641e-03
-9.19176817e-01 -3.25277776e-01 -3.41941029e-01 4.01509792e-01
-5.67445874e-01 4.09612209e-01 -3.50783676e-01 -7.86682665e-01
-9.94060487e-02 -8.79784465e-01 1.01633167e+00 2.72375286e-01
-3.45181882e-01 -8.90968025e-01 -1.36195645e-02 4.95270878e-01
-1.61144108e-01 5.28461456e-01 1.15472627e+00 -9.34665799e-02
-8.59097064e-01 -1.47007555e-01 -3.33532542e-01 1.48346081e-01
3.09306651e-01 1.76786199e-01 -9.57568228e-01 -9.77393463e-02
-2.26405576e-01 -4.23696697e-01 1.07718968e+00 4.84879345e-01
1.66326725e+00 -1.19968712e-01 -5.31174898e-01 6.36101067e-01
1.78077042e+00 1.88705176e-01 6.89033389e-01 3.73938650e-01
9.42066729e-01 7.44820356e-01 4.65904564e-01 4.02548105e-01
1.14798725e-01 5.68531036e-01 4.12670285e-01 -8.48975554e-02
-3.42040777e-01 -2.00641036e-01 -8.12972784e-02 5.56406677e-01
-7.49263763e-02 -7.13744313e-02 -1.22701585e+00 8.17285419e-01
-2.02362561e+00 -1.03897750e+00 -5.71814328e-02 1.87520492e+00
5.18640280e-01 -1.27529919e-01 -8.18985403e-02 2.38827884e-01
8.78978431e-01 -1.21086732e-01 -5.82339585e-01 -1.43968891e-02
-9.41735581e-02 1.08049959e-01 9.21476111e-02 7.07106441e-02
-1.21667457e+00 1.10859573e+00 5.28802395e+00 8.26794565e-01
-1.14636815e+00 8.79955888e-02 5.60197055e-01 -2.05796450e-01
-2.12293677e-02 6.02683797e-02 -2.03883573e-01 4.31291401e-01
2.25344524e-01 2.69609213e-01 6.60750449e-01 7.31525898e-01
3.29917111e-02 -5.86805046e-02 -1.27564180e+00 1.35716248e+00
4.60392348e-02 -1.50478029e+00 2.25315005e-01 2.61090189e-01
9.22673702e-01 -2.73649305e-01 3.21525127e-01 -1.07784048e-01
4.28944737e-01 -1.22820139e+00 5.47547817e-01 8.29519808e-01
5.67139089e-01 -1.10103190e+00 2.72752672e-01 -1.64774060e-03
-1.16949356e+00 -7.91457519e-02 -5.15578747e-01 8.35914016e-02
-1.50475532e-01 6.52322054e-01 -5.87556422e-01 3.03870559e-01
8.31034601e-01 9.39303458e-01 -6.25316441e-01 1.14963424e+00
-2.12491900e-01 4.92854714e-01 -2.91102141e-01 1.79062963e-01
2.02977583e-01 -2.89394051e-01 1.66605547e-01 9.51830745e-01
3.16656619e-01 7.92079885e-03 1.38018489e-01 1.14678550e+00
-2.17672482e-01 -4.22141068e-02 -6.04788423e-01 -3.94867361e-01
3.87106121e-01 1.54080272e+00 -1.36819112e+00 -2.36914709e-01
1.11980095e-04 1.15312505e+00 4.63936478e-01 4.06014979e-01
-5.16678035e-01 -5.44667184e-01 5.60913146e-01 1.08072534e-01
5.66051841e-01 -3.79565090e-01 -7.65972257e-01 -9.55243349e-01
-4.80113663e-02 -3.88599485e-01 4.28008497e-01 -8.25382590e-01
-1.32266378e+00 3.69824082e-01 -4.44523871e-01 -1.26840949e+00
5.40668964e-02 -4.39095825e-01 -7.16497302e-01 4.71003383e-01
-1.14485431e+00 -1.15803659e+00 -5.25394082e-01 7.95540392e-01
5.36855996e-01 -4.00298744e-01 6.78768635e-01 3.54614049e-01
-5.78514457e-01 2.28948236e-01 4.58651453e-01 4.67862457e-01
5.63995779e-01 -1.14822066e+00 -1.06692880e-01 6.98748231e-01
4.54548687e-01 6.37511909e-01 4.12416190e-01 -7.02473164e-01
-1.32564127e+00 -1.35928571e+00 4.82325137e-01 -3.46915305e-01
5.87987363e-01 -5.39040327e-01 -8.08575213e-01 3.28853369e-01
2.43603066e-01 -2.81129330e-01 9.18594360e-01 7.12612793e-02
-2.58620024e-01 -2.06498235e-01 -7.84409106e-01 6.66207850e-01
8.74351323e-01 -6.23757124e-01 -3.19053054e-01 5.04518628e-01
2.32502848e-01 2.15243027e-01 -1.00277889e+00 -1.19934879e-01
5.74531078e-01 -6.22177780e-01 9.91599739e-01 -3.75870198e-01
7.60069132e-01 -6.07823253e-01 -3.33719909e-01 -1.11387491e+00
-4.09273773e-01 -2.10488185e-01 -7.54841464e-03 1.46866560e+00
-7.56030679e-02 1.74397260e-01 1.10392654e+00 3.84816229e-01
1.43190220e-01 -6.26527131e-01 -7.87900925e-01 -6.03162587e-01
-1.38477743e-01 -4.13933337e-01 5.74586391e-01 1.19699574e+00
-1.90398887e-01 5.90102375e-01 -1.08827680e-01 1.59737900e-01
8.41133654e-01 6.64321065e-01 7.95038223e-01 -1.58757401e+00
-2.17321720e-02 -4.56601381e-01 -5.63194871e-01 -6.43209577e-01
1.91318735e-01 -1.01416647e+00 5.06581366e-02 -1.96615946e+00
1.92804500e-01 -7.00782299e-01 -2.76689947e-01 3.86729330e-01
1.86545238e-01 4.33310449e-01 3.07930231e-01 6.03878140e-01
-8.44594419e-01 7.50091910e-01 8.92561257e-01 -5.36293566e-01
-3.26309204e-01 -3.57915908e-01 -5.87017179e-01 7.42867589e-01
8.72185409e-01 -4.93336082e-01 -3.72520030e-01 -7.76525795e-01
2.63061464e-01 -3.88317823e-01 5.71733773e-01 -1.22804201e+00
5.10623395e-01 -9.65966284e-02 8.73434365e-01 -8.77640486e-01
5.24271369e-01 -1.23402476e+00 1.07705049e-01 3.86525780e-01
-3.46827805e-01 -3.12815815e-01 7.98504725e-02 7.80111849e-01
-2.40983963e-01 -3.91739942e-02 6.37490809e-01 -7.75193349e-02
-8.65818381e-01 7.61364996e-02 -3.42995048e-01 -3.91347110e-01
1.15793741e+00 -2.58802265e-01 -1.32116660e-01 -2.50642180e-01
-7.95519233e-01 2.47298509e-01 5.31499505e-01 4.38028246e-01
8.12009692e-01 -1.64721715e+00 -3.21669757e-01 8.88053328e-02
9.44974273e-02 8.30643028e-02 2.87098080e-01 4.42520648e-01
-6.13607883e-01 2.90762454e-01 -3.30636889e-01 -9.78850663e-01
-1.17515731e+00 7.16676354e-01 5.24542741e-02 7.99858421e-02
-6.84160888e-01 7.06063688e-01 2.36550689e-01 -1.09985173e-01
4.34187204e-01 1.44682094e-01 -3.19382340e-01 2.95045495e-01
3.31105053e-01 3.59847307e-01 -9.55442265e-02 -7.07701743e-01
-3.15365970e-01 8.05741429e-01 4.18741405e-02 -7.69910142e-02
1.83305681e+00 -7.74922222e-02 -3.75097990e-01 4.33568239e-01
1.28918910e+00 -3.21909964e-01 -1.20000374e+00 -3.25969368e-01
2.30845883e-01 -7.31877029e-01 3.66826564e-01 -6.14288986e-01
-1.43478358e+00 1.13673878e+00 9.42760170e-01 3.52635682e-02
1.22282171e+00 2.13484034e-01 6.13965392e-01 5.12068450e-01
-1.14410810e-01 -1.47863483e+00 6.08382821e-01 -1.12907909e-01
7.18780935e-01 -1.03767109e+00 2.66420364e-01 -2.88845837e-01
-5.07671177e-01 1.19405687e+00 5.88502705e-01 -2.30965137e-01
7.81540096e-01 -8.81441236e-02 1.52478382e-01 -8.38270724e-01
-4.40938860e-01 -3.20702910e-01 3.54564756e-01 4.44935739e-01
2.82781839e-01 -2.02927198e-02 6.94212690e-02 2.79224008e-01
-1.44345120e-01 -1.41372979e-01 7.81956092e-02 7.99069703e-01
-3.37581635e-01 -9.55695570e-01 -5.73809206e-01 5.25594532e-01
-2.38809109e-01 3.42228830e-01 -8.42963696e-01 5.36522985e-01
4.58531737e-01 9.52541232e-01 4.10877496e-01 -4.44683611e-01
-3.56422476e-02 -2.85691582e-02 2.59718120e-01 -5.13440430e-01
-2.93517679e-01 3.65852714e-01 -5.59347749e-01 -3.99447083e-01
-6.46111369e-01 -7.86804736e-01 -1.34691107e+00 -1.48124620e-01
-1.40819550e-01 -7.20671564e-02 9.63388264e-01 8.19444120e-01
2.74554312e-01 8.19031775e-01 5.71385086e-01 -8.12501252e-01
1.71055645e-01 -5.78479886e-01 -5.73314667e-01 6.12383068e-01
-6.14763834e-02 -6.90912545e-01 2.37100989e-01 4.19082403e-01] | [9.138351440429688, 3.2024831771850586] |
8bfc5675-a5e9-438e-aa67-29bb3b06c4f2 | road-traffic-reservoir-computing | 1912.00554 | null | https://arxiv.org/abs/1912.00554v1 | https://arxiv.org/pdf/1912.00554v1.pdf | Road traffic reservoir computing | Reservoir computing derived from recurrent neural networks is more applicable to real world systems than deep learning because of its low computational cost and potential for physical implementation. Specifically, physical reservoir computing, which replaces the dynamics of reservoir units with physical phenomena, has recently received considerable attention. In this study, we propose a method of exploiting the dynamics of road traffic as a reservoir, and numerically confirm its feasibility by applying several prediction tasks based on a simple mathematical model of the traffic flow. | ['Hiroyasu Ando', 'Hanten Chang'] | 2019-12-02 | null | null | null | null | ['3d-car-instance-understanding'] | ['computer-vision'] | [-1.26627341e-01 -4.43759143e-01 -1.88417256e-01 3.94984305e-01
1.07414208e-01 6.32206798e-02 9.48864222e-01 -4.96643819e-02
-3.93113166e-01 1.12541902e+00 -8.72291401e-02 -3.99402589e-01
1.30635887e-01 -1.06747377e+00 -6.34209216e-01 -8.20246279e-01
-4.48558897e-01 7.56021887e-02 1.20680161e-01 -4.06929433e-01
4.08142090e-01 7.75065303e-01 -1.58878541e+00 -1.61095306e-01
8.03545713e-01 1.08723080e+00 2.45011270e-01 2.13836595e-01
-3.97646278e-02 1.14661312e+00 -8.01844418e-01 3.13683242e-01
7.06078187e-02 -4.96480405e-01 -1.89909488e-01 -4.16065365e-01
-1.37822360e-01 -3.75436753e-01 -9.98520494e-01 7.55411208e-01
2.98857719e-01 3.24643284e-01 5.39835572e-01 -8.36602867e-01
-7.68138230e-01 3.89885932e-01 -6.51084632e-02 5.35001099e-01
-8.23702365e-02 4.01297629e-01 6.43133700e-01 -9.02454972e-01
2.68947124e-01 9.16509569e-01 8.42804134e-01 2.39444688e-01
-1.17601275e+00 -4.26719069e-01 -1.23294279e-01 1.32316977e-01
-1.39186406e+00 -2.84636825e-01 1.05062318e+00 -4.95428771e-01
1.31624210e+00 -1.84503630e-01 1.05803549e+00 8.57955635e-01
7.46316135e-01 5.83188891e-01 1.17986739e+00 -3.01476955e-01
6.33027256e-01 1.37314558e-01 -1.55121401e-01 5.34467995e-01
5.86017430e-01 7.45957553e-01 -4.60660815e-01 -2.04428598e-01
1.13732886e+00 1.44153967e-01 3.66022475e-02 -4.24880311e-02
-7.10081518e-01 8.04839492e-01 6.05107665e-01 4.36402977e-01
-7.89456010e-01 8.07401597e-01 3.71910423e-01 6.78210258e-02
5.46629906e-01 5.42602658e-01 1.05672717e-01 -3.77963454e-01
-1.03935790e+00 3.99617553e-01 8.06048155e-01 4.58965182e-01
4.38938230e-01 1.01892531e+00 1.56422049e-01 5.71468115e-01
1.41460806e-01 8.67249429e-01 8.81088912e-01 -8.31612051e-01
1.30488366e-01 3.11589718e-01 3.11216891e-01 -9.18065608e-01
-2.09703669e-01 -3.18202764e-01 -1.32652044e+00 1.49167955e-01
1.21211536e-01 -1.40668914e-01 -7.85222471e-01 1.15124750e+00
-2.14779779e-01 7.50317574e-01 3.32482785e-01 7.22055018e-01
4.06721562e-01 1.08689356e+00 -3.36770080e-02 -1.21271186e-01
9.35742319e-01 -8.71915936e-01 -7.12202787e-01 -7.69209713e-02
2.46910989e-01 1.23635620e-01 5.45944035e-01 8.46662149e-02
-1.02934504e+00 -4.72967356e-01 -9.72357750e-01 2.69580305e-01
-8.17431986e-01 -4.69969958e-03 7.50105083e-01 3.87753338e-01
-1.04321969e+00 9.65902746e-01 -1.03613210e+00 4.20383587e-02
1.11340210e-01 1.68265149e-01 3.71676832e-01 4.17986363e-01
-1.70960140e+00 1.26637757e+00 3.72523785e-01 7.41135716e-01
-8.11903298e-01 -5.06939173e-01 -8.24228823e-01 3.62969428e-01
-2.10678931e-02 -3.81053448e-01 1.22744310e+00 -2.26691142e-01
-2.09623098e+00 -3.17094289e-02 -2.28157774e-01 -8.24483931e-01
4.62680250e-01 1.16731316e-01 -5.07251740e-01 -1.50610209e-02
-4.15994793e-01 -2.75706261e-01 8.23243380e-01 -8.18813920e-01
-1.72792390e-01 5.44562377e-02 -3.61863405e-01 -2.62581468e-01
-2.27131695e-01 -4.45057005e-01 3.54189694e-01 -4.63344872e-01
-4.25411426e-02 -9.54113424e-01 -5.77617705e-01 -4.00659233e-01
-1.63083654e-02 -7.99457505e-02 6.49683714e-01 -3.96211088e-01
1.20230007e+00 -2.02668023e+00 -8.87190029e-02 3.57023805e-01
1.60190403e-01 5.86626470e-01 1.83476239e-01 8.42504799e-01
3.37602466e-01 2.24705473e-01 -3.75161827e-01 -2.22121909e-01
-2.00161308e-01 2.98915058e-01 -8.17062438e-01 3.63350064e-01
6.13387764e-01 1.29793870e+00 -1.19032252e+00 -1.82880625e-01
4.13143754e-01 6.61116481e-01 -5.28339408e-02 8.44114553e-03
-9.03336033e-02 1.74306855e-01 -5.88298798e-01 5.84315300e-01
3.74054849e-01 -4.26713169e-01 1.93880200e-01 5.17001688e-01
-6.11185491e-01 2.58476585e-01 -8.68669391e-01 8.50044906e-01
-7.11430967e-01 1.08254468e+00 -1.03713684e-01 -1.18040061e+00
1.20176029e+00 3.22849482e-01 6.13449693e-01 -1.12926924e+00
2.92601604e-02 4.90413338e-01 -3.06134520e-04 -4.91399080e-01
8.60160589e-01 -3.89710933e-01 -9.17302258e-03 4.76402938e-01
-3.85882109e-01 -4.07430530e-01 1.45367518e-01 -1.65928021e-01
1.07739079e+00 -1.53183350e-02 1.45165831e-01 -1.90964684e-01
2.79192746e-01 2.25725219e-01 1.55504271e-01 8.35851550e-01
-1.57234862e-01 3.09448726e-02 3.14925045e-01 -5.09563923e-01
-1.26203215e+00 -1.05123746e+00 -3.53022963e-01 2.68515378e-01
3.70478362e-01 9.34851542e-02 -2.02073246e-01 5.74532449e-01
6.74108386e-01 5.11065483e-01 -8.14759135e-01 -2.47529745e-01
-1.11322284e+00 -8.53375375e-01 5.66282868e-01 6.63841188e-01
3.91666085e-01 -1.67968714e+00 -1.10803998e+00 7.95200408e-01
3.58339101e-01 -9.61655855e-01 2.93680906e-01 1.55662805e-01
-1.23845708e+00 -5.64832211e-01 -7.20574915e-01 -4.26987797e-01
2.68518150e-01 2.61559278e-01 9.93760049e-01 3.73028964e-01
-2.59930581e-01 -1.48324296e-01 8.59094188e-02 -2.24907860e-01
-5.97626567e-01 2.07083687e-01 2.68383771e-01 -1.83958560e-01
3.37976888e-02 -7.07572281e-01 -3.74039710e-01 -9.41959769e-02
-8.36730957e-01 -1.03009813e-01 6.62817717e-01 1.03755224e+00
3.16139072e-01 1.82748839e-01 7.09123969e-01 -4.89778548e-01
1.09933412e+00 -7.29170442e-01 -9.88383830e-01 -2.75832713e-02
-7.38943458e-01 1.96176648e-01 1.01841879e+00 -5.91499388e-01
-7.74587452e-01 -2.15369493e-01 2.34351814e-01 -5.67628741e-01
3.49759161e-01 1.02523887e+00 5.92269838e-01 3.18026394e-02
2.07882985e-01 8.19113195e-01 2.70237178e-01 -3.62705827e-01
7.01585971e-03 5.46079159e-01 5.69867015e-01 -5.33567488e-01
6.79382682e-01 4.64348257e-01 3.17118019e-01 -1.17396367e+00
-5.28190844e-02 1.10602733e-02 -6.52973592e-01 -3.43445808e-01
1.58349425e-01 -7.36042917e-01 -1.15377057e+00 7.11369038e-01
-1.29255688e+00 -5.14924288e-01 -6.11028552e-01 5.88057697e-01
-5.84190965e-01 -9.95618030e-02 -9.14752066e-01 -1.43886328e+00
-1.28188595e-01 -8.46216381e-01 6.16683066e-01 3.36269289e-01
1.99067697e-01 -1.22418332e+00 3.81231189e-01 -7.02843845e-01
1.07575095e+00 3.89470309e-01 6.91027939e-01 6.60946742e-02
-7.75451422e-01 -3.91729385e-01 -2.14160025e-01 2.14001477e-01
3.66835371e-02 2.79484034e-01 -7.81962156e-01 -6.20435104e-02
7.72572905e-02 1.67393759e-01 1.02015662e+00 5.05107939e-01
8.51166725e-01 -1.70281753e-01 -4.56053227e-01 2.62912333e-01
1.67745984e+00 3.41592103e-01 6.52190745e-01 3.17185700e-01
4.57285821e-01 3.62112969e-01 8.79828036e-02 6.08503163e-01
-4.02764119e-02 3.27885836e-01 2.99254984e-01 6.16182908e-02
1.52319610e-01 -4.76794749e-01 4.30740356e-01 1.19239318e+00
-2.40396321e-01 3.61344069e-02 -1.09411573e+00 6.55992746e-01
-1.91047859e+00 -1.31006980e+00 -3.13490219e-02 1.98459029e+00
5.04037797e-01 2.70704627e-01 9.18639749e-02 -8.21147263e-02
6.10879362e-01 2.30933741e-01 -9.02635932e-01 -6.03349566e-01
-1.71345279e-01 3.64324689e-01 6.12513185e-01 3.58692169e-01
-5.83127558e-01 9.10527110e-01 8.12771511e+00 1.72895983e-01
-1.72772205e+00 7.71235526e-02 3.54631096e-01 1.01394996e-01
-4.13777009e-02 -3.05291474e-01 -6.80143774e-01 8.92554760e-01
1.66429889e+00 -5.06739020e-01 5.52865028e-01 7.26663291e-01
6.13529861e-01 -3.44325691e-01 -7.30056107e-01 6.93122685e-01
-4.71718371e-01 -1.89399087e+00 -9.72396955e-02 2.69829720e-01
5.65969706e-01 3.76804590e-01 2.44799152e-01 5.30487657e-01
2.51944453e-01 -1.21412623e+00 6.83314979e-01 8.82528484e-01
6.27886236e-01 -3.39135081e-01 8.01515520e-01 4.43744183e-01
-1.27741969e+00 -3.16861123e-01 -4.16368365e-01 -8.06765676e-01
5.11141241e-01 5.17953396e-01 -5.90858638e-01 1.04270719e-01
4.47340608e-01 8.30006123e-01 -1.58300921e-01 1.01804030e+00
1.43909216e-01 7.53446698e-01 -6.16380930e-01 -6.24867320e-01
4.23802525e-01 -4.75410938e-01 3.07930619e-01 1.07573354e+00
3.93575639e-01 -8.34941268e-02 -7.36622438e-02 1.29397035e+00
1.38847515e-01 -2.20737845e-01 -1.04938781e+00 -4.17050570e-01
5.85072219e-01 6.27154112e-01 -4.68081862e-01 -5.39231062e-01
-1.47978306e-01 2.86746860e-01 1.39422804e-01 5.18207967e-01
-9.15723085e-01 -4.33495909e-01 6.66632891e-01 1.65070549e-01
4.02711004e-01 -7.24944592e-01 -4.73104835e-01 -1.17797387e+00
9.98594016e-02 7.08323717e-03 -5.37571013e-01 -7.72921801e-01
-1.09886312e+00 8.25721085e-01 9.82156675e-03 -1.39162982e+00
-6.99781239e-01 -7.10361540e-01 -8.65381956e-01 1.17091227e+00
-1.98416579e+00 -5.01492620e-01 -3.35776538e-01 2.16448903e-01
1.84080362e-01 -3.01194727e-01 7.53154337e-01 1.22174934e-01
-8.65598857e-01 5.11666462e-02 8.66412401e-01 5.61133251e-02
-3.18805009e-01 -9.78732646e-01 7.69629776e-01 6.29146457e-01
-3.53069216e-01 6.00013435e-01 9.12086308e-01 -7.21265733e-01
-1.70103037e+00 -8.43020797e-01 8.56177211e-01 -9.85316858e-02
1.10057592e+00 -3.29797894e-01 -1.28646433e+00 2.60420591e-01
-9.51647013e-02 3.09105396e-01 -5.59380613e-02 -4.03166860e-01
6.05104715e-02 -2.41638701e-02 -1.07425129e+00 4.87065524e-01
6.49706304e-01 -7.67915368e-01 -3.76105934e-01 1.29309043e-01
4.27055359e-01 -3.46749246e-01 -8.47385347e-01 4.70285267e-02
7.10665941e-01 -4.54296023e-01 9.00944948e-01 -4.87539977e-01
4.89339173e-01 -6.61973432e-02 1.93753298e-02 -1.43972385e+00
-1.94038600e-01 -5.63050508e-01 -7.68465400e-01 5.87303638e-01
2.31863871e-01 -1.22220671e+00 8.20863843e-01 7.28417635e-01
1.23855695e-01 -7.94275880e-01 -1.22531486e+00 -1.26882994e+00
5.10780096e-01 -1.63552776e-01 7.78563321e-01 6.72745466e-01
2.66576886e-01 -7.56042227e-02 -4.72711354e-01 4.50371280e-02
4.04584587e-01 2.30030909e-01 3.23129833e-01 -1.36066604e+00
-1.21795595e-01 -7.05298960e-01 -3.22452784e-01 -8.37662816e-01
4.30127889e-01 -5.41978538e-01 2.55484372e-01 -1.30133200e+00
-3.06466550e-01 -7.40970194e-01 -6.32260442e-01 9.46477950e-02
3.22652698e-01 2.28789568e-01 1.85771734e-01 7.33501434e-01
8.77836272e-02 1.04454708e+00 1.04163313e+00 -2.87736565e-01
-4.70800489e-01 -6.63491115e-02 -8.83046016e-02 1.49105981e-01
9.72309709e-01 -4.27342594e-01 8.63860175e-02 -7.87242129e-02
3.30843776e-02 3.57242376e-01 3.53452772e-01 -1.17226136e+00
3.54873359e-01 -4.00223255e-01 1.92262575e-01 -5.15837371e-01
6.92625642e-01 -6.93710148e-01 2.28127703e-01 1.04370713e+00
-1.52524799e-01 3.74967873e-01 3.59544307e-01 5.99630833e-01
-5.35623312e-01 -2.41119489e-01 5.83856523e-01 -1.55266032e-01
-7.62833118e-01 1.05855681e-01 -1.01404297e+00 -3.01908076e-01
8.09938133e-01 -3.70514959e-01 -4.38208371e-01 -1.43626586e-01
-4.31954205e-01 2.89251395e-02 2.66853690e-01 2.33857676e-01
7.31256366e-01 -1.15630066e+00 -5.00097096e-01 3.01801890e-01
-3.91106695e-01 -3.43179375e-01 -2.51332298e-02 5.83091497e-01
-7.07017124e-01 9.27189946e-01 -3.09615076e-01 -4.37952250e-01
-1.73959434e-01 5.38373530e-01 7.12286055e-01 -3.48833233e-01
-9.40226674e-01 -1.19739190e-01 -5.81713021e-01 9.77453310e-03
-1.35908827e-01 -5.51414251e-01 2.33651195e-02 -1.19172648e-01
5.22051454e-01 5.15221953e-01 -8.77226964e-02 -6.66742682e-01
-2.39402488e-01 2.73801297e-01 5.21558225e-01 -1.36949331e-01
1.63890755e+00 1.87413439e-01 -1.71527371e-01 1.01844931e+00
9.58112359e-01 -4.30823863e-01 -1.21352422e+00 -4.22080696e-01
2.32603744e-01 -2.66381890e-01 3.15517068e-01 -4.97249842e-01
-1.04771864e+00 1.08111274e+00 4.50890243e-01 8.06160688e-01
5.91528475e-01 -4.78039205e-01 9.83123064e-01 6.10641897e-01
5.54325342e-01 -1.05334735e+00 -9.19574052e-02 8.44567955e-01
8.12155485e-01 -1.03866506e+00 9.80975805e-04 6.10348806e-02
-8.56015235e-02 1.23036861e+00 3.73354316e-01 -7.45742798e-01
1.08690453e+00 3.28835070e-01 -2.34183475e-01 -9.23322886e-02
-8.76188695e-01 1.02683399e-02 -2.73348451e-01 2.79084146e-01
2.22529635e-01 3.51906300e-01 -1.06727377e-01 2.68661603e-03
2.35005561e-02 5.22020042e-01 9.55068409e-01 1.08889067e+00
-2.60477275e-01 -5.52179456e-01 -1.54160261e-01 5.22836149e-01
-3.29875909e-02 -2.60944456e-01 1.43171459e-01 9.13251340e-01
-5.78816473e-01 5.40687382e-01 5.28635323e-01 -1.85795560e-01
2.11714283e-01 -3.04703172e-02 1.30673170e-01 -4.52304810e-01
-6.83818758e-01 -4.19419140e-01 -4.14733797e-01 -3.18077892e-01
-2.04245910e-01 -7.25347817e-01 -1.38645935e+00 -6.37960494e-01
-2.36765921e-01 3.77616793e-01 1.07492530e+00 9.06358123e-01
6.51892126e-01 5.50914168e-01 7.71221101e-01 -1.31971931e+00
-7.23794639e-01 -1.04452252e+00 -1.08287346e+00 -5.93171418e-02
6.33741558e-01 -1.04772496e+00 -5.53745866e-01 -4.04467881e-01] | [6.65191650390625, 3.3306615352630615] |
1fe74197-cc9a-4b0b-87f4-eb79048cb6de | stock-movement-prediction-based-on-bi-typed | null | null | https://openreview.net/forum?id=nXyXgrVOk8k | https://openreview.net/pdf?id=nXyXgrVOk8k | Stock Movement Prediction Based on Bi-typed and Hybrid-relational Market Knowledge Graph via Dual Attention Networks | Stock Movement Prediction (SMP) aims at predicting listed company's stock future price trend, which is a challenging task due to the volatile nature of financial markets. Recent financial studies show that the momentum spillover effect plays a significant role in stock fluctuation. However, previous studies typically only learn the simple connection information among related companies, which inevitably fail to model complex relations of listed companies in real financial market. To address this issue, we first construct a more comprehensive Market Knowledge Graph (MKG) which contains bi-typed entities including listed companies and their associated executives, and hybrid-relations including the explicit relations and implicit relation. Afterward, we propose \textsc{DanSmp}, a novel Dual Attention Networks to learn the momentum spillover signals based upon the constructed MKG for stock prediction. The empirical experiments on our constructed datasets against nine SOTA baselines demonstrate that the proposed \textsc{DanSmp} is capable of improving stock prediction with the constructed MKG. | ['Anonymous'] | 2021-11-16 | null | null | null | acl-arr-november-2021-11 | ['stock-prediction'] | ['time-series'] | [-8.72432411e-01 7.33166263e-02 -6.46118045e-01 -1.31392479e-01
-1.45695940e-01 -6.49104655e-01 6.58678830e-01 -4.19468135e-01
3.42832282e-02 8.43617022e-01 5.18590152e-01 -6.23144686e-01
-1.95901498e-01 -1.26269495e+00 -7.49907196e-01 -2.86230534e-01
-1.86312765e-01 5.44387341e-01 3.25360298e-01 -5.83983302e-01
3.90589535e-01 1.54604465e-01 -6.63916051e-01 4.40550856e-02
6.93989694e-01 1.30474818e+00 -1.13355629e-01 -1.85098186e-01
-5.81453323e-01 1.75353861e+00 -3.25264603e-01 -9.35118556e-01
6.35745645e-01 -1.96377113e-01 -5.63627720e-01 -3.48443270e-01
-3.73437554e-02 -4.11690414e-01 -6.64594412e-01 1.23810601e+00
2.31376197e-02 -2.64307618e-01 4.19710368e-01 -1.33630025e+00
-1.23497546e+00 1.60365129e+00 -8.32315803e-01 9.50955331e-01
-3.53663802e-01 1.47779444e-02 1.89111114e+00 -9.27160144e-01
6.21036232e-01 8.00497293e-01 7.97811151e-01 4.73632440e-02
-8.71964037e-01 -1.36327875e+00 8.09115946e-01 4.08413351e-01
-9.86224771e-01 8.10768306e-02 9.63550627e-01 -4.01525676e-01
1.11536276e+00 2.66896077e-02 1.03518045e+00 7.49987066e-01
4.32721257e-01 7.36556947e-01 9.95901048e-01 3.24356943e-01
-2.98637092e-01 1.48194388e-01 5.03029704e-01 2.33885050e-01
7.40174711e-01 1.32978752e-01 -9.17407334e-01 1.24153100e-01
1.04643571e+00 -6.47615865e-02 -3.35757196e-01 1.71039566e-01
-1.32433426e+00 9.04264987e-01 8.01411033e-01 5.84626794e-01
-5.83806157e-01 3.59805435e-01 3.00369769e-01 5.62310636e-01
8.92911315e-01 6.00812197e-01 -9.13076341e-01 -1.22580126e-01
-7.35312164e-01 2.98098415e-01 9.93680716e-01 1.01780760e+00
4.63709831e-01 3.64411682e-01 8.99116024e-02 1.40315756e-01
4.79310006e-01 3.47011119e-01 8.19953322e-01 -4.44704205e-01
1.07635105e+00 9.12107348e-01 9.02131349e-02 -1.22911763e+00
-6.38388932e-01 -1.13885808e+00 -7.48607874e-01 -1.70481905e-01
3.61694046e-03 -3.55534673e-01 -3.86848271e-01 1.40759623e+00
-2.35564753e-01 7.80400336e-01 2.23578244e-01 6.12534642e-01
9.12627578e-01 7.12224722e-01 -2.71192998e-01 -4.10532445e-01
1.10332358e+00 -1.22989273e+00 -9.65824664e-01 -2.04766437e-01
3.87643546e-01 -2.35172868e-01 5.29215276e-01 -7.12975562e-02
-9.13231850e-01 -1.64671302e-01 -1.05233324e+00 1.01331778e-01
-4.58298892e-01 -1.21739559e-01 9.84220386e-01 1.76938087e-01
-8.25453997e-01 7.05423176e-01 -7.57403553e-01 6.02351844e-01
5.09001732e-01 5.35763323e-01 1.23028941e-01 7.80407488e-01
-1.72036421e+00 9.21579361e-01 5.51210880e-01 6.40620530e-01
-1.86392426e-01 -9.55676675e-01 -5.65297484e-01 4.71269995e-01
5.76098680e-01 -5.29367805e-01 1.11433017e+00 -6.33743227e-01
-1.29614663e+00 3.68809104e-01 4.44835991e-01 -9.14824188e-01
4.78806317e-01 -3.91961485e-01 -6.85617387e-01 -2.77812034e-01
1.46928772e-01 6.13078400e-02 2.65478104e-01 -6.44386590e-01
-7.60649860e-01 -1.54090345e-01 9.90698561e-02 1.92224547e-01
-4.66950923e-01 -1.22423358e-01 -2.20029980e-01 -1.11483991e+00
1.43748015e-01 -5.78442156e-01 -7.22027421e-02 -8.85595322e-01
-6.53286695e-01 -3.64502579e-01 5.40304959e-01 -8.12500179e-01
1.66964948e+00 -1.56000447e+00 -1.84957787e-01 3.52348149e-01
2.88343310e-01 1.69168543e-02 5.47519773e-02 4.42865372e-01
-3.55694652e-01 3.53615165e-01 1.97895333e-01 1.54070750e-01
3.21610540e-01 -7.51510561e-02 -9.74123240e-01 -1.28359675e-01
2.98798621e-01 1.65899909e+00 -6.65321648e-01 -8.47430676e-02
-4.39699292e-01 -1.01060726e-01 -2.96829462e-01 -2.44259670e-01
-5.14609039e-01 6.67478815e-02 -8.50288689e-01 6.32547677e-01
6.60781324e-01 -7.41811156e-01 2.16180786e-01 -1.47699252e-01
-1.89547583e-01 8.04516256e-01 -1.08393764e+00 8.54713380e-01
6.97654709e-02 5.01617432e-01 -5.32729506e-01 -7.44058192e-01
1.15475786e+00 3.04293573e-01 6.55601203e-01 -7.99174666e-01
6.62894398e-02 5.46644986e-01 4.17820394e-01 -5.65802402e-05
6.17341399e-01 -4.18311566e-01 9.61185023e-02 6.23562872e-01
-3.08380693e-01 2.77234375e-01 1.92713991e-01 3.14506441e-01
1.03614140e+00 3.79273966e-02 1.55879274e-01 -3.28445524e-01
4.86363828e-01 -7.09525272e-02 1.16896248e+00 2.19333664e-01
2.17544157e-02 1.24694824e-01 1.12759280e+00 -5.36949515e-01
-7.27487385e-01 -7.46917903e-01 -1.50231738e-02 4.35574442e-01
3.24281901e-02 -2.68960983e-01 -7.22703934e-02 -7.72734880e-01
4.87208098e-01 6.66379511e-01 -6.06641114e-01 3.79167944e-02
-5.82126737e-01 -1.18084288e+00 2.52349377e-01 9.74609077e-01
9.88306105e-01 -1.18730533e+00 -1.23404562e-02 3.95974904e-01
1.15330711e-01 -1.03124082e+00 -5.89753032e-01 8.30822363e-02
-7.91588247e-01 -1.30846608e+00 -6.10962570e-01 -6.42877698e-01
1.76628888e-01 -2.44628325e-01 1.44231677e+00 -1.09466106e-01
5.32576919e-01 -2.16866508e-01 -1.13938190e-01 -5.85900247e-01
1.53157696e-01 3.87980431e-01 -1.33647338e-01 1.01668283e-01
7.02112496e-01 -7.36700594e-01 -5.30822754e-01 1.94578260e-01
-4.47832882e-01 2.03470647e-01 6.47432268e-01 8.11136365e-01
3.84455413e-01 6.20460987e-01 1.26814926e+00 -1.36318028e+00
8.50104451e-01 -7.40525365e-01 -9.48854327e-01 1.94785610e-01
-1.25818396e+00 -5.03974743e-02 1.77304789e-01 -1.56513050e-01
-1.11945975e+00 -5.42224586e-01 1.16079241e-01 -3.51472527e-01
9.74640071e-01 1.38353717e+00 -6.43084245e-03 3.08016360e-01
-1.12744868e-01 4.05141771e-01 -3.61901790e-01 -3.45544577e-01
9.50722471e-02 1.55857816e-01 4.25466627e-01 -1.21276505e-01
1.14184189e+00 2.29388595e-01 -4.14251722e-02 3.43353488e-02
-1.14511228e+00 -1.85985640e-01 -5.14336765e-01 1.17522769e-01
6.49454832e-01 -1.19305408e+00 -7.96799839e-01 8.17106903e-01
-9.03214991e-01 -1.31663710e-01 -1.42857447e-01 6.43813610e-01
-2.15241909e-01 -1.59863085e-01 -1.21548653e+00 -8.49251211e-01
-5.88595688e-01 -6.19257331e-01 3.60605150e-01 1.96997285e-01
2.01084003e-01 -1.31137359e+00 1.56673282e-01 3.36129546e-01
5.20189881e-01 2.16929600e-01 9.50473905e-01 -1.10744011e+00
-1.21474802e+00 -1.48734391e-01 -4.94297743e-01 3.20627600e-01
3.72304231e-01 -3.01546961e-01 -4.43054855e-01 1.35575294e-01
2.47508497e-03 6.48050681e-02 1.22569644e+00 2.38488838e-01
3.29921633e-01 -4.58275110e-01 -1.16115145e-01 4.93989736e-01
1.30868113e+00 5.07809818e-01 4.35233802e-01 7.81651616e-01
8.84107947e-01 3.93394738e-01 4.94246572e-01 3.57732564e-01
8.24555397e-01 1.87315732e-01 2.93531120e-01 3.36227506e-01
4.85385954e-01 -4.01534170e-01 4.23673481e-01 1.33526182e+00
-3.52284402e-01 8.91634747e-02 -8.58062029e-01 2.58381724e-01
-1.99451268e+00 -1.21449935e+00 -2.85490930e-01 1.54688501e+00
9.23659205e-01 7.60326564e-01 2.92209070e-02 -1.31326661e-01
5.65054357e-01 4.86742228e-01 -9.21231329e-01 4.65266973e-01
-5.55568814e-01 -1.03856608e-01 7.49161720e-01 2.39529490e-01
-1.02058566e+00 9.42777216e-01 5.77909231e+00 3.85042310e-01
-1.11396205e+00 -2.06265718e-01 7.54943252e-01 -7.42299408e-02
-8.33634138e-01 1.14727942e-02 -1.21089625e+00 8.09137642e-01
6.68607473e-01 -8.74400496e-01 8.78277868e-02 7.37424850e-01
-3.01799718e-02 4.60499346e-01 -9.05654728e-01 7.11445510e-01
-3.99463683e-01 -1.95058906e+00 5.32205924e-02 3.24062914e-01
1.01987481e+00 1.16786607e-01 2.73307443e-01 6.26008570e-01
5.85331619e-01 -7.81505167e-01 8.58233213e-01 8.50663364e-01
5.70792444e-02 -8.32148671e-01 1.13713443e+00 3.01719129e-01
-1.59173548e+00 -2.93515205e-01 -3.17291617e-01 -1.97314382e-01
2.09757313e-01 6.06847763e-01 -4.21561509e-01 9.26319540e-01
6.50309741e-01 1.44422424e+00 -4.92719769e-01 5.82458973e-01
-3.09039533e-01 7.02082992e-01 -3.07206400e-02 -1.21205449e-01
3.90253842e-01 -6.53849661e-01 2.28905201e-01 5.60069978e-01
4.36941981e-01 1.68323040e-01 -2.10961804e-01 1.24957812e+00
-5.80040395e-01 5.27534112e-02 -4.06226546e-01 -7.21130192e-01
3.22104007e-01 1.07446790e+00 -7.78418243e-01 -3.22439164e-01
-8.39567959e-01 5.38159430e-01 3.09761971e-01 2.77205259e-01
-8.23148847e-01 -5.21800332e-02 4.78209376e-01 1.44196838e-01
5.84043860e-01 1.33433007e-02 -4.74074423e-01 -1.59617996e+00
2.79479653e-01 -6.24212682e-01 5.40105999e-01 -7.09779680e-01
-1.63481700e+00 6.40036047e-01 -3.29434931e-01 -1.17888141e+00
-1.52526319e-01 -6.39447033e-01 -9.96594250e-01 1.16749728e+00
-2.01331425e+00 -1.13329148e+00 2.55495012e-01 3.20541650e-01
1.58487782e-01 -6.80449963e-01 1.83646381e-01 3.28387886e-01
-9.10516739e-01 7.91300684e-02 1.06360346e-01 6.02901459e-01
1.97749227e-01 -1.56961226e+00 9.79048073e-01 6.60869956e-01
4.08307850e-01 7.50444055e-01 6.61802739e-02 -1.27392352e+00
-9.69120681e-01 -1.09135354e+00 1.02697790e+00 -4.14515376e-01
1.63563311e+00 3.16800885e-02 -1.17851841e+00 1.38498533e+00
4.18422788e-01 -1.39056891e-01 3.77242923e-01 2.20386684e-01
-4.80903655e-01 -2.98453987e-01 -3.34959626e-01 3.28161418e-01
1.01852274e+00 -4.19714749e-01 -9.32467639e-01 1.86083958e-01
1.02673125e+00 -3.23371977e-01 -1.02733898e+00 4.65570360e-01
2.46759504e-01 -9.54314470e-01 7.63750076e-01 -6.13687634e-01
6.88042521e-01 -1.52243152e-01 2.99635708e-01 -1.17932010e+00
-5.19811690e-01 -4.42206681e-01 -5.25214911e-01 1.55258656e+00
9.58093882e-01 -1.11877275e+00 1.04870343e+00 6.98292851e-01
2.36090589e-02 -9.54230011e-01 -7.51034200e-01 -8.66180658e-01
3.02049935e-01 -1.78923979e-01 1.04625618e+00 1.39693010e+00
6.55574128e-02 6.29582286e-01 -5.24527013e-01 3.31629701e-02
1.89328253e-01 7.09600508e-01 3.42644006e-01 -1.48024881e+00
-5.16557574e-01 -8.83506596e-01 -2.10738376e-01 -9.18719292e-01
3.63441885e-01 -1.07280183e+00 -6.88382089e-01 -1.63614237e+00
1.73669010e-01 -4.87445921e-01 -8.50672662e-01 2.62409359e-01
-2.65030205e-01 -1.65923730e-01 2.53938854e-01 6.63336098e-01
-3.34909767e-01 6.94268882e-01 1.49659288e+00 -2.59501666e-01
-1.56393707e-01 3.28452319e-01 -9.97147322e-01 6.85094118e-01
7.22759664e-01 -2.10252911e-01 -3.94092619e-01 -6.52195290e-02
1.14786541e+00 7.99902529e-02 6.11720793e-02 -4.13734019e-01
4.51174676e-01 -1.70148820e-01 3.17488551e-01 -1.10918093e+00
8.47456679e-02 -6.16224289e-01 2.86486179e-01 6.19672120e-01
-1.78412467e-01 5.14100373e-01 -3.20643513e-03 6.98659658e-01
-7.14639723e-01 2.60783564e-02 2.38118153e-02 -2.87562340e-01
-8.23855102e-01 6.51372612e-01 1.86965108e-01 1.68751225e-01
9.18601274e-01 2.59381086e-01 -6.62520528e-01 -4.47230071e-01
-6.18730187e-01 7.41545379e-01 -2.48820618e-01 5.17350078e-01
3.62415910e-01 -1.65578282e+00 -7.62311995e-01 -1.11833833e-01
-3.30126256e-01 1.71831042e-01 7.77524617e-03 1.06852436e+00
-6.10154212e-01 7.25704193e-01 -8.05912986e-02 3.29163998e-01
-5.89723587e-01 6.71051025e-01 5.40644228e-01 -9.51916099e-01
-8.08701396e-01 9.50740397e-01 2.93809682e-01 -1.30974725e-01
-1.35079324e-01 -8.03086042e-01 -6.33847415e-01 5.74691117e-01
2.27656364e-01 3.34860355e-01 -2.16669291e-01 -4.14267629e-01
-2.04465017e-01 4.79135573e-01 -3.35480154e-01 -8.48125294e-03
1.73172104e+00 -4.96453606e-02 -4.42579240e-01 7.44132876e-01
6.83778346e-01 1.91400483e-01 -1.14026344e+00 -5.18903852e-01
8.23164880e-01 -2.05090404e-01 -6.42870143e-02 -6.66582763e-01
-1.82722306e+00 3.78004730e-01 -3.10649246e-01 5.21507442e-01
9.07934368e-01 -1.83942199e-01 1.07314229e+00 6.08183563e-01
3.92823189e-01 -1.12710714e+00 4.45684157e-02 6.71274245e-01
1.06727469e+00 -1.10286546e+00 9.46785212e-02 -6.01833820e-01
-8.26099038e-01 1.03787374e+00 7.94926524e-01 -3.02274823e-01
1.27276349e+00 2.43372336e-01 2.81921208e-01 -5.01665115e-01
-1.09272170e+00 -4.43949625e-02 3.30985606e-01 -8.28857198e-02
2.40172833e-01 -1.36144429e-01 -2.25123107e-01 1.32672799e+00
-7.16880679e-01 -3.76388058e-02 6.27291143e-01 6.80344462e-01
-1.61868274e-01 -1.03170705e+00 4.00645733e-02 9.66221690e-01
-7.01456666e-01 -5.46416104e-01 -4.83819366e-01 1.06509650e+00
-3.58548090e-02 2.86202043e-01 3.61115575e-01 -5.82833827e-01
4.18149173e-01 1.44363582e-01 -1.80427864e-01 -5.78309059e-01
-9.57968771e-01 1.99824959e-01 1.17249258e-01 -2.17213780e-01
-4.29750264e-01 -8.61365616e-01 -1.46312487e+00 -4.48435247e-01
-4.81105119e-01 2.71221399e-01 -9.75852609e-02 7.74523675e-01
3.12753111e-01 9.42782640e-01 6.44847155e-01 -1.11763172e-01
-7.63195038e-01 -1.03072178e+00 -1.40649819e+00 2.42571607e-01
1.43186256e-01 -7.81990767e-01 -2.19565824e-01 -2.12068051e-01] | [4.336575508117676, 4.32593297958374] |
a4d27dca-6617-44b6-957a-8aeecc387f52 | m2ts-multi-scale-multi-modal-approach-based | 2203.09707 | null | https://arxiv.org/abs/2203.09707v2 | https://arxiv.org/pdf/2203.09707v2.pdf | M2TS: Multi-Scale Multi-Modal Approach Based on Transformer for Source Code Summarization | Source code summarization aims to generate natural language descriptions of code snippets. Many existing studies learn the syntactic and semantic knowledge of code snippets from their token sequences and Abstract Syntax Trees (ASTs). They use the learned code representations as input to code summarization models, which can accordingly generate summaries describing source code. Traditional models traverse ASTs as sequences or split ASTs into paths as input. However, the former loses the structural properties of ASTs, and the latter destroys the overall structure of ASTs. Therefore, comprehensively capturing the structural features of ASTs in learning code representations for source code summarization remains a challenging problem to be solved. In this paper, we propose M2TS, a Multi-scale Multi-modal approach based on Transformer for source code Summarization. M2TS uses a multi-scale AST feature extraction method, which can extract the structures of ASTs more completely and accurately at multiple local and global levels. To complement missing semantic information in ASTs, we also obtain code token features, and further combine them with the extracted AST features using a cross modality fusion method that not only fuses the syntactic and contextual semantic information of source code, but also highlights the key features of each modality. We conduct experiments on two Java and one Python datasets, and the experimental results demonstrate that M2TS outperforms current state-of-the-art methods. We release our code at https://github.com/TranSMS/M2TS. | ['Chen Lyu', 'Yuexiu Gao'] | 2022-03-18 | null | null | null | null | ['code-summarization'] | ['computer-code'] | [ 2.04208717e-01 5.50827496e-02 -3.75529677e-01 -3.11508536e-01
-1.10912800e+00 -5.99451661e-01 1.58081681e-01 5.56591928e-01
2.87979722e-01 1.12755872e-01 7.51193225e-01 -9.12364051e-02
1.75405845e-01 -6.32650137e-01 -6.40928209e-01 -2.97756821e-01
1.01430506e-01 -2.08406195e-01 4.55715984e-01 -1.10548370e-01
6.82833731e-01 -2.83900023e-01 -1.83208108e+00 9.89038348e-01
1.40053642e+00 6.37236118e-01 4.09450889e-01 5.39899230e-01
-1.13829386e+00 1.25711024e+00 -6.24519110e-01 -4.51574028e-01
-3.16409200e-01 -4.84699041e-01 -9.26154673e-01 -2.30742574e-01
4.63352799e-01 2.02079266e-02 -2.35954359e-01 1.34657860e+00
1.73976973e-01 -1.59678891e-01 3.14117819e-01 -1.31759512e+00
-7.00668156e-01 1.19202900e+00 -7.58052051e-01 -5.03046811e-02
7.95289218e-01 7.80064613e-02 1.24469864e+00 -8.86967123e-01
5.28831363e-01 1.09041798e+00 8.55907917e-01 4.80929226e-01
-9.12073851e-01 -6.10376179e-01 1.12166554e-01 7.37749562e-02
-1.08323073e+00 -4.04000074e-01 8.87457788e-01 -6.77720010e-01
1.24757338e+00 2.32645035e-01 4.74985272e-01 9.51202095e-01
2.79899687e-01 1.19712913e+00 5.00101864e-01 -2.76814967e-01
-1.88978412e-03 -6.53616861e-02 5.23168445e-01 1.17021155e+00
3.81694347e-01 -6.70273304e-01 -5.72641492e-01 -5.76753616e-01
1.27413228e-01 3.19045812e-01 -7.09229037e-02 -2.59178251e-01
-1.31453347e+00 8.27846467e-01 3.38284492e-01 4.01243716e-01
-7.59235471e-02 3.08589280e-01 9.31207359e-01 1.65718362e-01
2.22507298e-01 2.36519918e-01 -6.16013885e-01 -3.76169354e-01
-1.04947722e+00 3.54642004e-01 6.90467238e-01 1.42223179e+00
1.18759835e+00 1.24570802e-02 -4.27343935e-01 1.00308371e+00
5.83935142e-01 5.91507554e-01 8.94022584e-01 -8.15034270e-01
1.05140340e+00 1.49402189e+00 -4.17685121e-01 -1.04955387e+00
-1.50958568e-01 -1.74112603e-01 -4.40425575e-01 -3.90791893e-01
-2.02682301e-01 1.24036139e-02 -6.35130525e-01 1.43841183e+00
8.07553977e-02 -3.81200947e-02 3.22485626e-01 1.89849377e-01
1.45626795e+00 6.90162182e-01 5.65094054e-02 8.30267742e-02
1.45392966e+00 -1.14995730e+00 -4.73578542e-01 -4.65625972e-01
9.45838749e-01 -8.57330501e-01 1.12009549e+00 -8.11162218e-02
-8.42343390e-01 -4.01296467e-01 -8.47586811e-01 -2.33906746e-01
-1.93586305e-01 5.10992527e-01 5.44829130e-01 3.22019070e-01
-8.48427117e-01 3.73754293e-01 -9.49253261e-01 -2.58815259e-01
5.47280014e-01 -9.81787667e-02 -2.37165257e-01 -1.38495669e-01
-7.47107565e-01 2.18332335e-01 7.17933357e-01 -3.46667856e-01
-7.50757098e-01 -9.28339124e-01 -1.52585363e+00 4.08638567e-01
4.92284983e-01 -6.31815970e-01 1.46951330e+00 -9.04802203e-01
-1.07139337e+00 5.18100679e-01 -6.68411851e-01 2.46715434e-02
-1.03012636e-01 -1.75070047e-01 -3.07990164e-01 5.15008047e-02
5.85272491e-01 2.31716663e-01 6.82108819e-01 -1.28481710e+00
-7.26124048e-01 -2.99660772e-01 1.18139520e-01 -1.23378739e-01
-4.80778337e-01 2.62636274e-01 -4.23388243e-01 -7.97654331e-01
5.48414476e-02 -5.96748114e-01 -1.09115764e-01 -6.21921122e-01
-7.04063058e-01 -4.75529313e-01 8.71120453e-01 -8.30803335e-01
1.70514357e+00 -2.48518777e+00 3.48338038e-01 -4.06891033e-02
4.70448107e-01 1.53456509e-01 -3.52873564e-01 8.36578131e-01
-3.96128148e-02 2.12816283e-01 -7.40760326e-01 -4.66026008e-01
2.29004338e-01 -2.94309501e-02 -6.70880795e-01 -7.46034533e-02
2.80339476e-02 9.91218865e-01 -1.10884285e+00 -8.58349621e-01
-1.33073330e-01 6.68643788e-02 -7.94468820e-01 2.56784260e-01
-4.26891834e-01 3.18071209e-02 -9.59068298e-01 6.63913012e-01
5.93146682e-01 -3.10193300e-01 2.75115371e-02 -7.40344375e-02
-1.31640032e-01 4.50189084e-01 -8.64195347e-01 2.25581765e+00
-6.78501189e-01 3.97202641e-01 -4.13829654e-01 -9.21588778e-01
9.83101904e-01 1.48765102e-01 4.37994808e-01 -4.10169095e-01
-2.59261459e-01 5.01109600e-01 -5.58957040e-01 -1.06336129e+00
5.67956328e-01 2.52717555e-01 -7.27292538e-01 6.13434970e-01
1.15404733e-01 -2.10688487e-01 4.03198630e-01 7.69324780e-01
1.30124688e+00 3.93919468e-01 4.61948186e-01 -1.55204743e-01
7.37434268e-01 2.51343697e-01 7.03895867e-01 5.36507010e-01
2.97105640e-01 4.60456938e-01 8.97774518e-01 -2.48775959e-01
-7.43807018e-01 -9.44431722e-01 2.98181444e-01 9.92681384e-01
1.82947516e-01 -1.33493137e+00 -9.58570957e-01 -1.19532275e+00
5.08648623e-03 7.38923848e-01 -5.05692542e-01 -3.61602575e-01
-7.42003500e-01 -5.56049764e-01 8.19811761e-01 6.48295403e-01
5.31610191e-01 -1.00389671e+00 -6.96752667e-01 1.41887784e-01
-6.92195714e-01 -8.65232587e-01 -6.29617870e-01 -2.77288735e-01
-8.24479699e-01 -1.29821944e+00 -3.03682834e-01 -8.14392745e-01
7.35173285e-01 3.27596217e-01 9.93938625e-01 5.33186734e-01
-1.43887371e-01 4.02862310e-01 -7.30672479e-01 -1.79413438e-01
-8.23439419e-01 3.01204890e-01 -6.20808959e-01 -2.48345271e-01
4.24107194e-01 -5.60075819e-01 -2.61901617e-01 -1.75781667e-01
-1.12136972e+00 2.58291066e-01 8.22451055e-01 4.60210413e-01
2.94750303e-01 -2.08773781e-02 4.46636051e-01 -1.11045039e+00
5.61023593e-01 -9.40309227e-01 -2.96830863e-01 4.05862451e-01
-1.61204338e-01 3.61335516e-01 1.02463257e+00 4.60193902e-02
-1.47366035e+00 -2.89595313e-02 -1.17567159e-01 -1.21456109e-01
-1.16134070e-01 1.13321400e+00 -1.28588304e-01 2.36490980e-01
5.49748838e-01 7.30516076e-01 -1.90581426e-01 -6.90204024e-01
2.32692763e-01 8.93217981e-01 4.28033262e-01 -1.01687813e+00
8.52168858e-01 3.25947911e-01 -4.09403652e-01 -5.95661283e-01
-6.99725509e-01 -5.11550784e-01 -5.21113873e-01 1.33314207e-01
5.10589063e-01 -8.44862461e-01 -3.16742472e-02 5.71777821e-01
-1.39517844e+00 1.45315621e-02 -1.95249423e-01 5.89945801e-02
-5.94509900e-01 9.26699579e-01 -5.55604458e-01 -2.55312592e-01
-3.20676804e-01 -1.37074053e+00 1.50030291e+00 2.38497689e-01
-3.30287635e-01 -8.25619578e-01 2.79779673e-01 4.13781732e-01
3.11672539e-01 3.20638567e-01 1.29392850e+00 -7.53203630e-01
-6.31331563e-01 -6.05492480e-02 -2.60932952e-01 1.00245275e-01
3.54777932e-01 3.60690355e-01 -7.56464183e-01 -1.98317334e-01
-3.72387506e-02 -2.56780058e-01 1.00272012e+00 -7.32262526e-03
1.25467420e+00 -7.10154831e-01 -4.88799453e-01 6.28243804e-01
1.59852862e+00 2.60603428e-02 5.20309269e-01 2.66645581e-01
1.09329712e+00 6.35140359e-01 4.02709872e-01 6.54757321e-01
8.72105420e-01 4.49473172e-01 4.30038393e-01 4.21111107e-01
-2.72320837e-01 -4.87724751e-01 8.94304097e-01 1.33686101e+00
3.32912356e-01 2.68590033e-01 -1.17765009e+00 7.75675595e-01
-2.08508611e+00 -9.83988106e-01 -4.92272615e-01 1.72731662e+00
1.04192734e+00 -3.81891608e-01 -3.96756008e-02 -2.03307509e-01
6.66210473e-01 1.91434398e-01 -4.53612387e-01 -2.54706293e-01
2.10265666e-01 -1.21006250e-01 6.97386041e-02 6.68588579e-02
-8.22110474e-01 7.75826275e-01 5.03776121e+00 1.10786796e+00
-7.96188235e-01 2.03898489e-01 -1.77749202e-01 2.13132709e-01
-9.19715285e-01 4.53385711e-01 -6.07357562e-01 6.78129375e-01
8.40014219e-01 -7.08173692e-01 3.62514853e-01 1.16650569e+00
-1.97462708e-01 9.19602346e-03 -1.11664701e+00 8.57963085e-01
2.31606230e-01 -1.37486851e+00 3.01883668e-01 -3.91573429e-01
9.51644897e-01 1.61707729e-01 -2.55248666e-01 6.34042621e-01
4.58301842e-01 -6.20923221e-01 1.05736125e+00 3.72485727e-01
6.43652141e-01 -5.20506799e-01 6.94785237e-01 1.87987685e-01
-1.92977262e+00 -4.72707540e-01 -2.03666970e-01 3.25145692e-01
-3.04121822e-01 5.83974361e-01 -3.81034523e-01 1.13252282e+00
6.62873089e-01 1.35705245e+00 -1.19870210e+00 1.12256801e+00
-3.07675570e-01 3.98810506e-01 1.47839427e-01 -1.23423273e-02
2.08648607e-01 1.82687566e-01 5.92856526e-01 1.64967620e+00
6.08076394e-01 -3.52279186e-01 2.99291670e-01 1.33580875e+00
-8.77108201e-02 2.36889645e-01 -6.18244886e-01 -2.56538063e-01
6.04936600e-01 1.21654308e+00 -6.55879200e-01 -7.35440135e-01
-7.44198143e-01 5.91894686e-01 2.15713352e-01 3.94247115e-01
-8.28180790e-01 -1.01417100e+00 5.91789842e-01 -3.75907928e-01
3.36066633e-01 1.13464193e-02 -2.76529223e-01 -1.80720174e+00
4.85358149e-01 -1.02446878e+00 6.63515985e-01 -6.63546026e-01
-8.61001074e-01 5.04368484e-01 2.91866541e-01 -1.36376476e+00
-1.12844244e-01 -1.59526631e-01 -9.16908979e-01 4.67172027e-01
-1.34664035e+00 -1.29099870e+00 -4.83117998e-01 5.03405988e-01
1.00060856e+00 -3.45703512e-01 6.20533407e-01 1.73207462e-01
-6.14604950e-01 5.27346492e-01 8.76771659e-02 3.90402555e-01
2.40156800e-01 -1.27554941e+00 6.54248774e-01 1.25423658e+00
9.71129816e-03 1.21392679e+00 4.66500729e-01 -8.31958592e-01
-1.61003268e+00 -1.38610113e+00 8.35053623e-01 -5.24541736e-01
6.20694995e-01 -2.54682809e-01 -1.22388685e+00 9.19026434e-01
1.89436316e-01 -2.37865001e-01 7.07927406e-01 -1.91572979e-01
-8.04490447e-01 9.38022360e-02 -8.43749583e-01 3.68884951e-01
1.03532195e+00 -7.03133643e-01 -1.07738912e+00 6.75311014e-02
1.00703406e+00 -3.17240626e-01 -8.25072885e-01 2.52489924e-01
3.11853528e-01 -1.18148017e+00 7.22351551e-01 -3.87188941e-01
9.89660740e-01 -5.49133360e-01 -2.51993418e-01 -1.30555201e+00
1.20637052e-01 -4.20716941e-01 -2.29858324e-01 1.67681479e+00
2.70799667e-01 -4.74871397e-01 3.53574187e-01 1.36488751e-01
-7.22452104e-01 -5.14736891e-01 -6.06155694e-01 -5.95463455e-01
-1.66582931e-02 -4.53878015e-01 1.00106013e+00 9.39618945e-01
6.15654767e-01 9.53799188e-02 1.83759570e-01 1.35398621e-03
6.87783718e-01 6.84882820e-01 7.19350755e-01 -1.05995476e+00
-9.20258835e-02 -7.36853659e-01 -2.84106255e-01 -7.24593639e-01
6.54808879e-01 -1.47561932e+00 6.22931272e-02 -1.81068575e+00
7.71290243e-01 -1.33560658e-01 1.14165843e-01 8.33588839e-01
-3.93059075e-01 -4.45813984e-01 4.19539586e-02 4.06606108e-01
-8.42628658e-01 7.34268486e-01 6.67719483e-01 -5.01674175e-01
-6.02037795e-02 -2.69230485e-01 -9.88423288e-01 7.83592522e-01
7.52760947e-01 -7.99697101e-01 -6.40439212e-01 -7.37608790e-01
4.20379490e-01 1.07603788e-01 2.83725798e-01 -9.51124191e-01
2.89791405e-01 -2.86199003e-01 -9.43771154e-02 -4.04818773e-01
-3.17828387e-01 -6.11921728e-01 1.07337432e-02 6.11650646e-01
-3.56850564e-01 9.31452960e-03 3.99044275e-01 4.60487366e-01
-4.29226935e-01 -6.52022004e-01 4.79645163e-01 -3.94882590e-01
-9.56718922e-01 2.22514808e-01 -3.62896532e-01 3.94911796e-01
8.52574587e-01 -9.73371863e-02 -7.36188591e-01 3.08848560e-01
-6.01453148e-02 3.05583388e-01 7.94226468e-01 8.48197162e-01
8.88539553e-01 -1.31785059e+00 -7.18970954e-01 4.00870234e-01
7.71227658e-01 1.45159572e-01 4.85328823e-01 7.25244522e-01
-4.69774395e-01 2.08611608e-01 -7.68080726e-02 -5.01980484e-01
-1.30752349e+00 5.09189546e-01 1.51565550e-02 -1.13792308e-01
-8.28372836e-01 5.04153013e-01 2.52053976e-01 -6.13066196e-01
-2.21758336e-01 -5.19847393e-01 -3.85617107e-01 6.10166462e-04
7.62900710e-01 3.11123282e-01 6.63280860e-02 -6.73940778e-01
-5.04176497e-01 8.84920895e-01 -1.54138923e-01 4.46004152e-01
1.46036792e+00 -9.08504874e-02 -7.91289151e-01 3.84399503e-01
1.39384997e+00 4.70478117e-01 -8.30303967e-01 -5.01945019e-01
4.86419290e-01 -6.37753665e-01 -4.52621400e-01 -4.59756047e-01
-1.09802365e+00 6.94238484e-01 -2.32408375e-01 1.72125250e-01
9.72884834e-01 4.19937402e-01 1.10038531e+00 3.29246819e-01
4.05983031e-01 -6.09797418e-01 3.05094153e-01 5.64872921e-01
7.62535036e-01 -8.51921797e-01 -2.89509058e-01 -4.34629083e-01
-6.21485770e-01 1.45487773e+00 7.88283169e-01 9.84395668e-02
1.29311606e-01 3.83551508e-01 -1.97379246e-01 -4.77506697e-01
-7.67914832e-01 3.28932516e-02 1.87731147e-01 5.04966915e-01
5.32165647e-01 -2.00031728e-01 -3.39055508e-02 9.44285333e-01
-1.15638003e-01 -1.11413680e-01 9.44272280e-01 1.23560226e+00
-6.34571970e-01 -1.20134938e+00 -2.54113287e-01 6.00016832e-01
-3.18240970e-01 -4.00853753e-01 -3.88278008e-01 3.10475498e-01
1.00040793e-01 8.33446622e-01 -3.50707144e-01 -5.14311254e-01
3.94702911e-01 1.75991341e-01 6.73666149e-02 -1.23141062e+00
-6.36256695e-01 -3.55777770e-01 -2.49994203e-01 -6.69013500e-01
-3.99822831e-01 -8.99116218e-01 -1.79706419e+00 -1.66051283e-01
1.88368373e-02 4.03812259e-01 5.43600917e-01 8.69783878e-01
7.43947804e-01 8.40333879e-01 5.84308028e-01 -5.62138200e-01
-4.90676075e-01 -7.76994467e-01 -1.34775385e-01 5.01034200e-01
5.94189584e-01 -5.38039744e-01 -3.51306349e-01 3.82106870e-01] | [7.602161884307861, 7.997286796569824] |
a2a4df8b-64c5-4b95-bc7f-8191c4670b33 | automated-story-generation-as-question | 2112.03808 | null | https://arxiv.org/abs/2112.03808v1 | https://arxiv.org/pdf/2112.03808v1.pdf | Automated Story Generation as Question-Answering | Neural language model-based approaches to automated story generation suffer from two important limitations. First, language model-based story generators generally do not work toward a given goal or ending. Second, they often lose coherence as the story gets longer. We propose a novel approach to automated story generation that treats the problem as one of generative question-answering. Our proposed story generation system starts with sentences encapsulating the final event of the story. The system then iteratively (1) analyzes the text describing the most recent event, (2) generates a question about "why" a character is doing the thing they are doing in the event, and then (3) attempts to generate another, preceding event that answers this question. | ['Mark Riedl', 'Nitya Tarakad', 'Jonathan Balloch', 'Spencer Frazier', 'Louis Castricato'] | 2021-12-07 | null | null | null | null | ['generative-question-answering'] | ['natural-language-processing'] | [ 4.74640161e-01 4.96097475e-01 7.27569684e-02 -2.05324635e-01
-9.80471551e-01 -5.69531500e-01 9.48690355e-01 6.03120148e-01
4.75990213e-02 1.07088423e+00 9.01403904e-01 -1.59774214e-01
2.49275371e-01 -1.22515547e+00 -4.40440893e-01 -3.33846420e-01
4.40197974e-01 7.00685084e-01 2.45133430e-01 -4.25807923e-01
5.11395514e-01 -1.04764007e-01 -1.28965175e+00 4.68040586e-01
5.96166015e-01 1.92621142e-01 4.16786343e-01 9.04961884e-01
-5.66312134e-01 1.66349435e+00 -9.14100885e-01 -4.08037990e-01
-3.92243236e-01 -1.50093496e+00 -1.22616792e+00 2.78323025e-01
-1.85260057e-01 -4.05028999e-01 -3.55350584e-01 7.44349480e-01
2.01973811e-01 3.26360703e-01 7.66751170e-01 -1.10635424e+00
-5.93912005e-01 1.29079664e+00 -1.74260095e-01 9.23425332e-02
8.56944442e-01 7.64022544e-02 9.09910500e-01 -5.47286034e-01
9.64851558e-01 1.07856727e+00 4.38337535e-01 8.54938865e-01
-9.19537544e-01 -9.83942151e-02 3.54794264e-02 5.20674884e-02
-1.26660895e+00 -2.87022620e-01 1.00985956e+00 -5.07649660e-01
1.09594429e+00 2.20011845e-01 9.90861058e-01 1.07537901e+00
3.77184182e-01 8.25027049e-01 6.66022480e-01 -4.89559263e-01
5.15153110e-01 -9.19910446e-02 3.14311862e-01 3.65716845e-01
-1.39252409e-01 -3.96534711e-01 -5.84243476e-01 -3.25768113e-01
3.89967471e-01 -4.96140152e-01 1.44005209e-01 4.68201220e-01
-1.27568161e+00 1.03157139e+00 -3.51767570e-01 5.90370238e-01
-7.15834796e-01 5.43523908e-01 3.97511452e-01 1.80710912e-01
2.01430500e-01 6.43167257e-01 3.17478180e-02 -5.43507695e-01
-1.13759780e+00 9.44226444e-01 1.12492406e+00 7.82787740e-01
4.29744333e-01 2.16498479e-01 -4.27280515e-01 3.70491415e-01
2.62977064e-01 -3.10261007e-02 4.50070471e-01 -7.41582811e-01
3.73096406e-01 4.41538870e-01 3.58473837e-01 -8.84211302e-01
-1.59814894e-01 -1.38400882e-01 -3.39989126e-01 -7.61399493e-02
3.61276329e-01 -3.75827760e-01 -6.10927403e-01 1.76156378e+00
1.84178308e-01 -1.40692547e-01 4.09072548e-01 4.63695973e-01
1.30070937e+00 1.16593277e+00 1.69778839e-01 -4.94095922e-01
1.34318650e+00 -8.11060607e-01 -9.31519389e-01 -4.11027402e-01
4.92784768e-01 -7.43955970e-01 8.29943240e-01 1.22022025e-01
-1.45554650e+00 -4.22755092e-01 -1.10631967e+00 -6.82215393e-02
-5.57165220e-02 3.30829620e-03 4.85585898e-01 3.27491283e-01
-9.87421930e-01 2.08947986e-01 -3.67066175e-01 -6.42217636e-01
-2.05478799e-02 -8.38031247e-03 1.23399854e-01 2.95153916e-01
-1.11338949e+00 7.38453388e-01 8.50152016e-01 -3.77431273e-01
-1.13535619e+00 -2.85686582e-01 -8.87440622e-01 1.63265437e-01
5.31785786e-01 -9.90994036e-01 1.68011844e+00 -8.27314258e-01
-1.56950116e+00 7.45296717e-01 -5.81538796e-01 -5.66287398e-01
2.19089702e-01 -1.62421361e-01 -3.31844985e-01 -3.82193998e-02
3.40448380e-01 6.48282290e-01 5.17202795e-01 -1.29473555e+00
-6.06781304e-01 -1.41612396e-01 2.78815150e-01 3.78169119e-01
1.86104864e-01 3.38643372e-01 -1.60161078e-01 -8.32795799e-01
5.51209450e-02 -5.50257206e-01 -2.25470185e-01 -7.36854613e-01
-7.21939504e-01 -4.50221241e-01 6.87310815e-01 -6.07272148e-01
1.47335756e+00 -1.74260139e+00 -1.27476025e-02 -2.52588779e-01
1.00217521e-01 -9.57559720e-02 -1.46787092e-01 1.21217752e+00
-7.01639280e-02 2.57368207e-01 -1.30003229e-01 -1.33944348e-01
-2.13403553e-01 -4.39243801e-02 -8.20376635e-01 -2.36099675e-01
2.84844130e-01 9.03215766e-01 -1.11816657e+00 -6.54812455e-01
-9.24286768e-02 4.70765643e-02 -4.64547545e-01 3.54279518e-01
-8.55651975e-01 3.50763053e-01 -5.68739057e-01 1.50078490e-01
-1.27799720e-01 -1.95132345e-01 2.92698815e-02 2.73017198e-01
-2.26700783e-01 7.64945507e-01 -1.06314266e+00 1.54599679e+00
7.20672905e-02 8.54845822e-01 -6.79712057e-01 -6.30356371e-01
1.13339746e+00 6.72508419e-01 1.91930264e-01 -5.54093383e-02
2.35760078e-01 -1.99166685e-01 -1.22411780e-01 -7.58656025e-01
8.64386499e-01 -5.08480906e-01 -4.16121036e-01 1.08417678e+00
-4.49573137e-02 -7.16695487e-01 7.14973629e-01 3.85229319e-01
1.21431589e+00 4.73330885e-01 8.60413551e-01 2.59089828e-01
5.78494847e-01 4.77695078e-01 3.85424525e-01 8.99313033e-01
1.40514299e-01 7.19642878e-01 8.56992602e-01 -4.80740219e-01
-1.34160674e+00 -1.07291305e+00 6.63160324e-01 9.05113161e-01
1.00437766e-02 -6.62536979e-01 -9.41663206e-01 -4.33994055e-01
-6.28503680e-01 1.69994879e+00 -6.69306636e-01 -1.19743347e-01
-7.20232785e-01 -4.19980049e-01 6.23334825e-01 3.58563185e-01
4.97113407e-01 -1.54322886e+00 -7.44116306e-01 7.44040370e-01
-7.13491142e-01 -7.00051904e-01 -4.46274757e-01 -1.82806432e-01
-7.14514554e-01 -6.21664166e-01 -4.75204855e-01 -7.18804657e-01
5.30035675e-01 3.08522452e-02 1.18610966e+00 5.71703725e-03
1.99023888e-01 1.68343544e-01 -5.30580282e-01 -4.81379330e-01
-1.00199473e+00 -2.45759506e-02 -3.54995728e-01 -5.47745004e-02
2.05846384e-01 -4.99735147e-01 9.81718972e-02 -2.61700064e-01
-9.08158958e-01 5.63819051e-01 1.47870898e-01 5.47544360e-01
5.00760615e-01 4.23497409e-01 9.18542087e-01 -9.03899729e-01
1.21801555e+00 -6.96345508e-01 1.04691535e-01 3.88460755e-01
-2.49751836e-01 1.99634627e-01 6.72261715e-01 -3.74656916e-01
-1.41899419e+00 -8.02042112e-02 -3.39183360e-01 3.41051579e-01
-3.98290783e-01 7.49671400e-01 -7.31922388e-02 1.02852035e+00
8.39245021e-01 8.44290316e-01 -3.57006162e-01 5.32813370e-02
3.03484946e-01 2.07571283e-01 7.01067328e-01 -4.43535745e-01
8.07942212e-01 3.69843811e-01 -3.64201486e-01 -8.15027595e-01
-8.74672592e-01 -2.06589308e-02 -2.51957864e-01 -7.36527264e-01
1.06211162e+00 -5.45634449e-01 -5.24595320e-01 4.16935086e-01
-1.68029165e+00 -2.91916102e-01 -8.15004230e-01 2.55803257e-01
-1.02672577e+00 2.29514435e-01 -5.27615607e-01 -1.17533946e+00
-4.75375563e-01 -4.28068370e-01 7.34913290e-01 5.83403051e-01
-1.14291096e+00 -7.43311584e-01 3.79128784e-01 2.52960324e-01
6.92577884e-02 5.20942748e-01 1.20496809e+00 -7.97548413e-01
-2.81202525e-01 -3.00939649e-01 3.25050950e-01 -2.01109380e-01
9.83677432e-02 1.27856836e-01 -4.17813480e-01 3.19042951e-01
2.62664407e-01 -3.27184439e-01 2.49673530e-01 3.63978237e-01
4.44466114e-01 -4.88478869e-01 -1.68611005e-01 -1.65450409e-01
1.22824252e+00 7.33273089e-01 8.68662119e-01 1.64252877e-01
2.16048151e-01 7.82517672e-01 4.81153905e-01 5.36214650e-01
6.99838161e-01 4.58221048e-01 6.23655654e-02 2.60657579e-01
-1.74496010e-01 -9.59758639e-01 6.15759730e-01 6.33456469e-01
8.85460004e-02 -7.50818968e-01 -8.51664960e-01 9.27548945e-01
-2.14473605e+00 -1.49291122e+00 -5.41097283e-01 1.62549186e+00
9.08399045e-01 3.18526268e-01 2.25616485e-01 1.99908257e-01
5.66788137e-01 3.31461906e-01 -4.30521190e-01 -5.62147975e-01
-2.58662820e-01 -5.56940660e-02 -3.21201146e-01 4.69358742e-01
-4.45131749e-01 1.34479165e+00 6.81030416e+00 6.39516115e-01
-5.84087729e-01 1.39608100e-01 5.65589964e-01 -6.83998987e-02
-6.58459425e-01 4.25381362e-01 -7.68112123e-01 2.46534020e-01
8.36752892e-01 -9.55347896e-01 2.77308583e-01 6.28216684e-01
5.81242144e-01 -5.28682709e-01 -1.16092134e+00 5.62473834e-01
4.34813589e-01 -1.49931598e+00 5.73386848e-01 -3.30778986e-01
6.60311401e-01 -7.44924963e-01 -4.62932229e-01 2.02689111e-01
6.21118724e-01 -9.02820826e-01 1.26833403e+00 7.71455765e-01
2.87512720e-01 -9.50637400e-01 4.77471352e-01 8.88422668e-01
-1.02211607e+00 1.12855576e-01 -2.24147085e-02 -4.47481483e-01
7.40987599e-01 4.01114166e-01 -1.12664032e+00 2.62776077e-01
1.72824413e-01 1.20145120e-01 -2.31874213e-01 7.14516759e-01
-8.73239696e-01 8.27148557e-01 7.72607476e-02 -5.63853502e-01
2.14705825e-01 -1.06300242e-01 9.11281347e-01 1.08087313e+00
4.96948987e-01 5.90898395e-01 -8.66386741e-02 1.25235176e+00
1.22107446e-01 2.22804949e-01 -8.91169608e-01 -6.43063068e-01
3.09313357e-01 8.74264836e-01 -9.61870968e-01 -6.89653754e-01
1.50269615e-02 1.11800456e+00 -1.23769613e-02 3.11625630e-01
-7.75098026e-01 -3.06554884e-01 9.53335911e-02 3.47779214e-01
-5.97180724e-02 -2.23429203e-01 -3.14957589e-01 -8.89141619e-01
-1.77503362e-01 -7.65480399e-01 3.29757422e-01 -1.52902877e+00
-5.83241343e-01 7.90154815e-01 5.78173809e-02 -7.29703367e-01
-9.98665571e-01 4.22093332e-01 -1.31600487e+00 6.64944470e-01
-4.47805017e-01 -1.10370803e+00 -1.00956820e-01 2.23195478e-01
1.10677981e+00 -1.02413528e-01 9.72779274e-01 -2.28092536e-01
-1.68838888e-01 1.04623675e-01 -5.69728315e-01 1.00517333e-01
5.99740371e-02 -1.04702568e+00 5.81809103e-01 1.16227186e+00
2.07133546e-01 3.95924777e-01 1.18308401e+00 -1.14187825e+00
-9.95322049e-01 -8.40577662e-01 1.79254389e+00 -3.99854988e-01
5.43905973e-01 -2.78014243e-01 -6.93004668e-01 6.81990504e-01
4.86870706e-01 -1.25628173e+00 7.85556376e-01 -3.60659420e-01
4.94538806e-02 4.04788494e-01 -1.08661652e+00 1.11482823e+00
7.82469213e-01 -3.65046531e-01 -1.20279539e+00 4.98779178e-01
9.27671313e-01 -2.09048390e-01 -2.91923046e-01 -2.04379976e-01
2.71373093e-01 -8.23732197e-01 4.25966710e-01 -9.43281174e-01
1.07296193e+00 -4.21865106e-01 1.94124151e-02 -1.13955629e+00
-3.02986354e-01 -1.07788789e+00 -8.72281492e-02 1.48703301e+00
6.00343764e-01 -1.54413790e-01 7.28568792e-01 4.46976930e-01
-8.89063999e-02 -7.16486633e-01 -5.29243231e-01 -3.57301295e-01
-1.66792482e-01 -6.89706147e-01 7.05701709e-01 7.01275468e-01
3.39328170e-01 7.48155355e-01 -5.63812971e-01 -2.83170640e-01
2.36124426e-01 1.73712194e-01 8.16973090e-01 -9.27476585e-01
-3.80133241e-01 -4.24196571e-01 5.65187596e-02 -9.80831563e-01
-1.40185878e-01 -6.88888013e-01 3.48526061e-01 -2.15968370e+00
4.15926039e-01 2.82353371e-01 5.55937648e-01 3.61161917e-01
-1.84778780e-01 -2.45488033e-01 1.89913884e-01 1.40500799e-01
-4.56992686e-01 4.70944911e-01 1.00842154e+00 -6.75761104e-02
-5.26605129e-01 1.57782268e-02 -1.02264380e+00 8.06281984e-01
8.80429327e-01 -6.48538649e-01 -5.27750909e-01 -7.25858882e-02
8.49155009e-01 8.31338942e-01 1.62510604e-01 -9.79055226e-01
3.98223609e-01 -5.04168272e-01 6.54719546e-02 -9.66993213e-01
1.77147999e-01 -1.58287644e-01 6.22012913e-01 5.94746649e-01
-6.76905036e-01 3.16777557e-01 -3.69262137e-02 2.61382520e-01
-3.09335887e-01 -7.93113947e-01 4.84266371e-01 -2.86953568e-01
-3.84927362e-01 -9.66765806e-02 -1.25160122e+00 6.83767274e-02
1.22714889e+00 -4.44933593e-01 -8.94008670e-03 -1.00047565e+00
-8.46975982e-01 1.81508541e-01 2.78451890e-01 5.10246515e-01
7.44204223e-01 -1.36954498e+00 -1.18549800e+00 -3.93925399e-01
6.98504737e-03 -1.81716099e-01 1.10663503e-01 2.10184500e-01
-5.88195264e-01 2.99156338e-01 5.71208373e-02 7.37429559e-02
-1.17945397e+00 3.93668830e-01 6.31057918e-02 -6.14999115e-01
-7.57388532e-01 8.82623792e-01 -3.35891843e-02 1.72559828e-01
-1.13299966e-01 4.15529497e-02 -6.35452867e-01 2.53603697e-01
8.51960301e-01 1.75798446e-01 -4.87756252e-01 -6.36870205e-01
3.29983793e-02 9.47353691e-02 -3.67003083e-02 -7.84891963e-01
1.36277544e+00 -8.65174532e-02 -2.35792994e-01 8.90243828e-01
5.35219967e-01 -1.96191430e-01 -7.61931121e-01 -4.33598831e-02
3.80278379e-02 2.28648093e-02 -4.64728296e-01 -6.65272653e-01
-3.74539554e-01 4.64879930e-01 -2.66390502e-01 6.42637908e-01
1.03994918e+00 3.88650447e-01 1.10548902e+00 3.14992636e-01
2.73744106e-01 -1.15359449e+00 2.82635123e-01 8.64826620e-01
1.16192842e+00 -4.55299437e-01 -1.19088978e-01 -1.46311119e-01
-9.01911676e-01 1.19640374e+00 4.85973746e-01 -4.74240556e-02
1.91793293e-01 3.27319473e-01 -1.83820590e-01 -5.10826945e-01
-1.13900697e+00 -1.12038620e-01 -4.93903160e-02 4.31906343e-01
5.84833324e-01 -8.82253889e-03 -7.97399104e-01 6.49970531e-01
-8.64822447e-01 1.66300818e-01 9.33863401e-01 9.45291579e-01
-7.20381379e-01 -1.02923214e+00 -3.59123766e-01 4.04570460e-01
-3.18196058e-01 -1.86449736e-01 -9.95910406e-01 5.15184402e-01
2.09792987e-01 1.34591162e+00 -7.56582767e-02 -4.46722925e-01
2.01942191e-01 3.20190549e-01 3.95823866e-01 -1.04696655e+00
-7.35705495e-01 -2.70963274e-02 4.33701575e-01 -2.07213029e-01
-5.78501374e-02 -1.05290329e+00 -1.75762594e+00 -2.98949301e-01
6.85453415e-02 3.13928962e-01 4.19113785e-01 1.22258520e+00
-2.11154800e-02 5.90921700e-01 3.87820542e-01 -2.34083936e-01
-1.08181871e-01 -1.04317641e+00 -2.65335709e-01 2.97616601e-01
-1.14704534e-01 -4.93352562e-02 -1.14340305e-01 5.32238007e-01] | [11.672494888305664, 8.846604347229004] |
99de7658-5f3a-4664-9615-dc36d415dfc0 | safer-situation-aware-facial-emotion | 2306.09372 | null | https://arxiv.org/abs/2306.09372v1 | https://arxiv.org/pdf/2306.09372v1.pdf | SAFER: Situation Aware Facial Emotion Recognition | In this paper, we present SAFER, a novel system for emotion recognition from facial expressions. It employs state-of-the-art deep learning techniques to extract various features from facial images and incorporates contextual information, such as background and location type, to enhance its performance. The system has been designed to operate in an open-world setting, meaning it can adapt to unseen and varied facial expressions, making it suitable for real-world applications. An extensive evaluation of SAFER against existing works in the field demonstrates improved performance, achieving an accuracy of 91.4% on the CAER-S dataset. Additionally, the study investigates the effect of novelty such as face masks during the Covid-19 pandemic on facial emotion recognition and critically examines the limitations of mainstream facial expressions datasets. To address these limitations, a novel dataset for facial emotion recognition is proposed. The proposed dataset and the system are expected to be useful for various applications such as human-computer interaction, security, and surveillance. | ['Bharat Bhargava', 'Mijanur Palash'] | 2023-06-14 | null | null | null | null | ['facial-emotion-recognition'] | ['computer-vision'] | [ 3.33822295e-02 -3.13532948e-01 -1.40780434e-01 -8.01343679e-01
-1.30558804e-01 -2.52386987e-01 3.31159085e-01 -4.07671839e-01
-4.95555788e-01 4.12807375e-01 -2.64476147e-02 1.44855112e-01
1.80260211e-01 -4.18348521e-01 -2.67549008e-01 -8.98987889e-01
-4.40154344e-01 -1.71869367e-01 -2.72085696e-01 -3.85333210e-01
-6.82001635e-02 1.11483777e+00 -1.96480846e+00 4.38577205e-01
1.12312123e-01 1.33830988e+00 -4.19241250e-01 4.22084659e-01
1.27905503e-01 6.16590917e-01 -8.54837179e-01 -5.73695123e-01
2.45500296e-01 -1.86755478e-01 -4.00706142e-01 5.15830964e-02
3.52543235e-01 -5.55639803e-01 -1.88583791e-01 6.64789200e-01
8.62324119e-01 9.02148709e-02 4.14418101e-01 -1.47402120e+00
-4.48639154e-01 -2.38038704e-01 -6.52699888e-01 3.39733928e-01
3.24515939e-01 2.63064653e-02 3.00600618e-01 -8.33482623e-01
6.28218770e-01 1.18030870e+00 7.81286895e-01 8.93637121e-01
-5.90121567e-01 -1.20386493e+00 1.32508501e-01 1.45470858e-01
-1.64696074e+00 -8.25034857e-01 7.39731550e-01 -3.73209804e-01
7.90365517e-01 1.89795732e-01 4.40638870e-01 1.36426926e+00
1.11335516e-01 6.29858136e-01 1.21857333e+00 -3.88142437e-01
1.33305475e-01 3.14220607e-01 -3.49427491e-01 7.54759431e-01
-2.54799515e-01 1.28569022e-01 -5.86669445e-01 -3.19791794e-01
4.49675351e-01 3.71900797e-02 -1.16817765e-01 8.86000171e-02
-4.85938400e-01 7.78272390e-01 3.23155254e-01 4.19683665e-01
-6.34113729e-01 -2.22951055e-01 6.78983569e-01 1.61796764e-01
8.67089629e-01 -1.19041577e-01 -4.89603192e-01 -3.20019245e-01
-8.79510045e-01 9.27207991e-02 5.82983494e-01 5.27851701e-01
4.58880186e-01 2.62322545e-01 5.67905838e-03 8.85033011e-01
3.68804544e-01 5.32981098e-01 3.17869335e-01 -7.37360179e-01
-1.55950919e-01 3.70119780e-01 -1.10601120e-01 -1.40924609e+00
-7.39997029e-01 1.05221815e-01 -6.83753610e-01 1.19293272e-01
3.60395573e-03 -5.56291580e-01 -9.06596303e-01 1.91687810e+00
5.22202849e-01 3.74178499e-01 9.62158963e-02 8.33004892e-01
1.13554370e+00 4.97095823e-01 3.13997120e-01 -2.98119158e-01
1.38067377e+00 -4.12820220e-01 -9.81199324e-01 1.42109513e-01
5.48509955e-01 -6.63429201e-01 6.07593179e-01 5.40166140e-01
-6.15737796e-01 -3.61995727e-01 -6.35069966e-01 4.33000475e-01
-6.91142559e-01 2.38854632e-01 7.48198211e-01 1.25448096e+00
-1.22924364e+00 -3.69669348e-02 -5.14549196e-01 -8.20239305e-01
6.55307412e-01 6.02043748e-01 -6.89648092e-01 2.01886952e-01
-1.19324768e+00 7.96520174e-01 -4.32950705e-02 3.82469356e-01
-5.98102987e-01 -2.41140440e-01 -8.85987103e-01 -9.87271890e-02
1.73823208e-01 -1.29043525e-02 1.09617651e+00 -1.61910045e+00
-1.56599224e+00 1.16488087e+00 -4.41106498e-01 -7.05226138e-02
7.18721598e-02 -1.62538052e-01 -1.09795773e+00 3.37387770e-01
-3.49827766e-01 5.28276920e-01 9.50162649e-01 -9.07723188e-01
-3.21409702e-01 -6.26463294e-01 -1.65833518e-01 -1.58395648e-01
-5.37332773e-01 7.84874678e-01 -3.57360274e-01 -6.57146394e-01
-6.03835225e-01 -1.09965241e+00 1.14319123e-01 8.63234401e-02
2.37278253e-01 -1.66626707e-01 1.35691690e+00 -6.58802986e-01
1.02607453e+00 -2.44711900e+00 -4.53568429e-01 3.90493155e-01
-8.90973732e-02 6.54994011e-01 -3.09462577e-01 2.62155265e-01
-1.85015097e-01 1.08585946e-01 2.40551338e-01 -2.45931283e-01
-2.83256024e-01 1.33889824e-01 6.47040382e-02 6.38304889e-01
6.64185286e-01 5.52689612e-01 -5.76728642e-01 -3.34750563e-01
1.11995310e-01 8.12549412e-01 -4.09749031e-01 2.49983072e-01
1.89106971e-01 3.95729452e-01 -4.27976370e-01 1.15185750e+00
9.19426680e-01 3.63824159e-01 6.33476526e-02 -7.15720430e-02
7.51100779e-02 -5.49277127e-01 -7.64064610e-01 1.16893959e+00
-3.86357546e-01 7.59524047e-01 4.98929024e-01 -7.55472660e-01
1.16761553e+00 6.11849606e-01 6.94723248e-01 -6.75136626e-01
6.32520318e-01 -6.78106770e-02 -6.58525005e-02 -9.57096756e-01
3.45401525e-01 -1.17909722e-01 8.78485590e-02 2.84775764e-01
1.43062472e-01 3.63369346e-01 -1.95299208e-01 -2.38799855e-01
8.69112432e-01 -1.19051367e-01 2.85137206e-01 -7.20037445e-02
6.08636081e-01 -5.67983985e-01 6.66906774e-01 2.60559499e-01
-8.35386276e-01 1.73294738e-01 3.01685065e-01 -7.65260041e-01
-2.76948065e-01 -7.31430590e-01 -2.03011110e-01 1.36002815e+00
-2.50574797e-01 -1.66645363e-01 -9.15751755e-01 -8.32905471e-01
-9.10656154e-02 1.77830160e-01 -9.43553090e-01 -1.49800479e-01
-1.48142099e-01 -8.39616477e-01 8.32232177e-01 4.66459990e-01
6.43973351e-01 -1.43450892e+00 -8.73022676e-01 -3.60390097e-02
-3.48173976e-02 -1.33202791e+00 -8.31569359e-02 -3.59588236e-01
-2.58256763e-01 -1.14725769e+00 -5.84777594e-01 -6.51218712e-01
6.03500724e-01 1.09236743e-02 8.30063343e-01 2.07260773e-01
-6.05193257e-01 6.19997978e-01 -5.10812283e-01 -8.40739250e-01
-4.21212949e-02 -1.35612994e-01 3.52457613e-01 6.89110160e-01
9.70449448e-01 -9.77174044e-02 -4.78686363e-01 4.12086248e-01
-1.04847360e+00 -6.95430040e-01 4.00598109e-01 7.01776505e-01
7.72530064e-02 1.92738980e-01 8.51956069e-01 -6.99642897e-01
7.41623104e-01 -7.01920390e-01 -2.69584775e-01 1.22075900e-01
-1.43910438e-01 -6.06693566e-01 1.83138609e-01 -4.65965182e-01
-1.35821307e+00 3.95451300e-02 -4.51194286e-01 -3.53352159e-01
-5.22801578e-01 3.98283213e-01 -2.05985948e-01 -4.01294351e-01
4.82776821e-01 1.30640399e-02 2.34446481e-01 -2.43182987e-01
-9.93640572e-02 1.32498729e+00 3.48149657e-01 -3.71784776e-01
1.96964353e-01 6.64984345e-01 -5.25689982e-02 -1.15384007e+00
-5.38886011e-01 -4.25087750e-01 -5.85018754e-01 -5.78608334e-01
6.84605122e-01 -1.03510821e+00 -8.42065811e-01 9.80526686e-01
-1.00888681e+00 -7.01940730e-02 4.20298338e-01 2.70337343e-01
-2.46822298e-01 -7.53033608e-02 -4.37451184e-01 -1.19890761e+00
-4.74358916e-01 -1.04397035e+00 1.23518836e+00 5.32277822e-01
-3.57774526e-01 -8.56135428e-01 -3.73657298e-04 2.88815856e-01
6.81301534e-01 6.82355583e-01 3.36426467e-01 -5.91210663e-01
1.48557395e-01 -3.11595351e-01 -1.55002356e-01 5.16170144e-01
4.83517736e-01 3.85021955e-01 -1.46865630e+00 -3.56814057e-01
-9.04352143e-02 -5.82042754e-01 3.87626797e-01 2.01948155e-02
1.41977632e+00 -2.85225987e-01 -1.46867424e-01 6.14358127e-01
1.06451654e+00 5.53383172e-01 7.33913362e-01 2.72422403e-01
2.11657405e-01 9.08138573e-01 7.11932659e-01 8.85332823e-01
2.02799395e-01 6.98158801e-01 4.90105242e-01 -4.00911182e-01
4.56990242e-01 2.46219516e-01 3.61684859e-01 1.12982064e-01
-1.47571668e-01 -5.46607301e-02 -8.66139412e-01 4.13726181e-01
-1.48278761e+00 -1.10759306e+00 4.46947068e-01 1.72411835e+00
4.10461813e-01 -5.18886209e-01 2.17060238e-01 -1.20631360e-01
5.70048332e-01 2.43071407e-01 -4.35215354e-01 -1.01731277e+00
-1.73483700e-01 5.21402359e-01 -3.34832072e-02 -1.44002631e-01
-1.19984424e+00 1.00148022e+00 6.96509647e+00 5.02409220e-01
-1.82526934e+00 1.65376574e-01 9.26914275e-01 -3.58821839e-01
3.92526269e-01 -7.44113326e-01 -5.04771292e-01 3.89064968e-01
1.04641712e+00 2.19009221e-01 4.04542476e-01 8.13842297e-01
4.14789021e-01 -6.83199987e-02 -5.27002454e-01 1.19421458e+00
6.21908486e-01 -8.94556463e-01 -2.07661971e-01 -1.61855787e-01
4.69147414e-01 3.75110917e-02 3.59574199e-01 3.65121037e-01
-4.02639300e-01 -1.22391069e+00 3.11426669e-01 3.63920569e-01
9.29248095e-01 -1.09082568e+00 1.10278761e+00 6.14148658e-03
-8.26662183e-01 -1.18115917e-01 1.30014494e-01 -1.52321354e-01
-4.80467565e-02 1.16517603e-01 -7.74566591e-01 2.54639447e-01
1.20273054e+00 5.82735598e-01 -5.12005866e-01 4.36851978e-01
1.63418233e-01 4.32705343e-01 -2.72480696e-01 -8.24982151e-02
1.49724975e-01 7.89704397e-02 1.44915566e-01 1.54370677e+00
3.61055404e-01 3.72681558e-01 -1.89804003e-01 2.40402564e-01
-9.63580012e-02 3.47761840e-01 -9.24640715e-01 -2.69528538e-01
1.79352582e-01 1.44995368e+00 -3.49678159e-01 1.32444818e-02
-6.31948650e-01 8.82593036e-01 1.48368821e-01 4.66680825e-01
-9.13456023e-01 -3.86797130e-01 1.19953585e+00 -2.54669003e-02
2.36527234e-01 6.57670721e-02 2.19982296e-01 -8.66587698e-01
7.81908073e-03 -1.14613259e+00 4.85585302e-01 -8.10955942e-01
-1.09834349e+00 1.01161361e+00 6.01827390e-02 -7.48554587e-01
-2.15418175e-01 -8.63223433e-01 -6.38228774e-01 6.94531977e-01
-1.42967975e+00 -1.24087059e+00 -4.67587173e-01 8.53406429e-01
1.69944167e-01 -6.24582767e-01 1.13366950e+00 3.26898754e-01
-8.89258504e-01 9.93465543e-01 -1.13017246e-01 2.82942891e-01
8.26884270e-01 -4.15220708e-01 -5.15008792e-02 6.72510147e-01
-2.59597272e-01 7.00982273e-01 3.71852368e-01 -3.13853115e-01
-1.30335629e+00 -1.16748524e+00 7.29082525e-01 -2.27807909e-01
3.33147675e-01 -5.08218765e-01 -6.92027867e-01 5.23151696e-01
2.95239985e-01 3.18364203e-01 1.24376130e+00 1.16015367e-01
-4.36391801e-01 -3.63216370e-01 -1.80872476e+00 6.39018059e-01
8.13005149e-01 -5.20627618e-01 1.25138508e-02 -2.42743338e-03
6.78789243e-03 -2.46886268e-01 -8.83526444e-01 8.09116244e-01
9.28032696e-01 -1.14042222e+00 5.69261312e-01 -6.76073790e-01
-6.45950809e-02 -7.12855672e-03 -2.19905511e-01 -1.10861158e+00
-7.36535788e-02 -8.29115450e-01 -1.08405747e-01 1.38290834e+00
8.76081828e-03 -8.20696115e-01 7.11215556e-01 7.25415707e-01
4.59197164e-01 -9.90013182e-01 -1.04880321e+00 -4.99062628e-01
-5.09257555e-01 -3.24671417e-01 9.24068570e-01 9.89601672e-01
-1.66743696e-01 -1.47034228e-01 -5.59256136e-01 1.70822546e-01
-5.00477664e-03 -2.44437590e-01 9.47440922e-01 -9.63022888e-01
3.30204040e-01 -2.48636767e-01 -8.51709366e-01 -1.08369172e-01
6.83110237e-01 -3.42953622e-01 -2.20663413e-01 -6.98508620e-01
2.31384058e-02 -2.56408840e-01 -5.19403875e-01 7.39138842e-01
1.00976065e-01 6.24971092e-01 1.96316496e-01 -3.95809203e-01
-3.74959916e-01 6.85407221e-01 7.10422873e-01 -2.46066209e-02
-7.77737275e-02 5.92640555e-03 -8.44674945e-01 7.02966630e-01
1.12158895e+00 -2.26058826e-01 -2.34811321e-01 -5.90112768e-02
-2.48642623e-01 -3.82212967e-01 1.33739740e-01 -6.90236092e-01
-2.74373591e-01 -2.56565273e-01 7.02464104e-01 -2.81704098e-01
5.80637395e-01 -1.07733905e+00 1.12097420e-01 6.85112774e-02
1.02855153e-01 3.73390555e-01 8.25425506e-01 1.90831140e-01
-2.91254044e-01 8.78660977e-02 8.73205304e-01 1.75478891e-01
-9.66101587e-01 4.61520880e-01 -5.97351015e-01 -2.41377488e-01
1.47028077e+00 -3.00412565e-01 -2.93840915e-01 -6.04780316e-01
-5.09088218e-01 -2.30025947e-02 4.69831437e-01 7.90562034e-01
7.19121277e-01 -1.34977615e+00 -6.34006202e-01 6.28698707e-01
4.77836877e-01 -6.57463253e-01 4.03434902e-01 7.87798345e-01
-4.79588300e-01 1.82490185e-01 -5.90865791e-01 -5.28696954e-01
-1.86520123e+00 3.20288301e-01 3.53139818e-01 2.45516151e-01
-9.93759483e-02 6.99290037e-01 -6.32060645e-03 -4.25203413e-01
3.13065439e-01 1.30780563e-01 -3.59999269e-01 1.19248942e-01
9.19082642e-01 7.37900361e-02 1.21889286e-01 -1.31676292e+00
-6.76617563e-01 4.76968616e-01 -2.29350291e-02 1.33033052e-01
1.32067358e+00 -8.43565837e-02 -2.02562004e-01 1.41201109e-01
1.39973152e+00 -1.51671581e-02 -9.42075431e-01 -1.14245750e-01
-1.17011130e-01 -6.22280300e-01 1.06422706e-02 -8.43163311e-01
-1.29142070e+00 7.14920521e-01 1.12529635e+00 -1.68879732e-01
1.81430197e+00 -2.62094378e-01 5.78918993e-01 2.96383709e-01
3.63534033e-01 -1.12962592e+00 2.39489181e-03 5.10359645e-01
8.51719737e-01 -1.32514441e+00 -2.35359341e-01 -8.67426321e-02
-8.64356160e-01 1.09816337e+00 6.74677908e-01 4.33027208e-01
9.58356678e-01 3.79823774e-01 7.96226859e-01 -3.16838473e-01
-6.49458885e-01 -7.28162453e-02 1.95865095e-01 8.47309470e-01
5.55514574e-01 4.87230197e-02 7.50506744e-02 5.26342452e-01
-1.33486735e-02 3.28160733e-01 1.97584212e-01 1.04859734e+00
1.54842306e-02 -8.51925910e-01 -4.60745066e-01 3.96175653e-01
-9.56521869e-01 3.56198788e-01 -7.28627920e-01 9.23438966e-01
3.87279868e-01 1.02963459e+00 1.23954795e-01 -4.13572073e-01
2.61261016e-01 2.37557232e-01 3.49603742e-01 -3.43284130e-01
-7.32972145e-01 -1.99959591e-01 8.42538103e-02 -7.01974273e-01
-9.02482927e-01 -6.69848800e-01 -9.24782932e-01 -3.74443620e-01
-6.85797855e-02 -1.12305708e-01 8.08787823e-01 7.45018303e-01
6.58151329e-01 1.21621430e-01 9.10598159e-01 -8.66611540e-01
-6.54402152e-02 -9.33646321e-01 -6.64376318e-01 5.67666411e-01
4.38049704e-01 -9.44109201e-01 -1.54161096e-01 -1.12574836e-02] | [13.484009742736816, 1.854284644126892] |
7893c7f5-3b65-4fc7-a3cc-294ae45fe6df | glipv2-unifying-localization-and-vision | 2206.05836 | null | https://arxiv.org/abs/2206.05836v2 | https://arxiv.org/pdf/2206.05836v2.pdf | GLIPv2: Unifying Localization and Vision-Language Understanding | We present GLIPv2, a grounded VL understanding model, that serves both localization tasks (e.g., object detection, instance segmentation) and Vision-Language (VL) understanding tasks (e.g., VQA, image captioning). GLIPv2 elegantly unifies localization pre-training and Vision-Language Pre-training (VLP) with three pre-training tasks: phrase grounding as a VL reformulation of the detection task, region-word contrastive learning as a novel region-word level contrastive learning task, and the masked language modeling. This unification not only simplifies the previous multi-stage VLP procedure but also achieves mutual benefits between localization and understanding tasks. Experimental results show that a single GLIPv2 model (all model weights are shared) achieves near SoTA performance on various localization and understanding tasks. The model also shows (1) strong zero-shot and few-shot adaption performance on open-vocabulary object detection tasks and (2) superior grounding capability on VL understanding tasks. Code will be released at https://github.com/microsoft/GLIP. | ['Jianfeng Gao', 'Jenq-Neng Hwang', 'Lu Yuan', 'Lijuan Wang', 'Xiyang Dai', 'Liunian Harold Li', 'Yen-Chun Chen', 'Xiaowei Hu', 'Pengchuan Zhang', 'Haotian Zhang'] | 2022-06-12 | null | null | null | null | ['open-vocabulary-object-detection', 'referring-expression-segmentation', 'phrase-grounding'] | ['computer-vision', 'computer-vision', 'natural-language-processing'] | [ 1.48331240e-01 2.73072660e-01 -3.88517171e-01 -4.30640429e-01
-1.30671728e+00 -5.96038401e-01 6.92854404e-01 -8.75426084e-03
-5.55276573e-01 3.29711586e-01 1.56744450e-01 -5.50962389e-01
5.48556507e-01 -4.04944718e-01 -8.87032032e-01 -3.66480440e-01
2.83592224e-01 4.52008545e-01 4.23088312e-01 -5.21224700e-02
1.13601968e-01 2.25359470e-01 -1.55301738e+00 5.04098535e-01
7.76875734e-01 9.76503670e-01 8.28989506e-01 9.97155547e-01
-5.35959244e-01 9.75706458e-01 -2.61102498e-01 -2.36897528e-01
9.30102766e-02 -1.86631590e-01 -7.92454422e-01 7.63549954e-02
9.89824772e-01 -2.63115317e-01 -1.93048298e-01 1.00022805e+00
3.22698981e-01 2.25539669e-01 6.29625738e-01 -1.39383268e+00
-1.29173136e+00 4.05386955e-01 -8.55970025e-01 3.69289994e-01
2.94395715e-01 5.53618252e-01 1.39505231e+00 -1.49795318e+00
5.21296263e-01 1.49714410e+00 6.12089515e-01 7.09790349e-01
-1.30505109e+00 -5.94759226e-01 4.36914831e-01 1.17077149e-01
-1.67428291e+00 -4.73170102e-01 1.26673147e-01 -6.03872895e-01
1.32685184e+00 -1.04897439e-01 4.44443047e-01 1.07922113e+00
1.61538690e-01 1.29067600e+00 9.07725155e-01 -3.75452876e-01
5.87635450e-02 2.59574234e-01 5.09053588e-01 8.89066458e-01
1.40583396e-01 -1.51744727e-02 -5.77181041e-01 3.02255958e-01
5.54277778e-01 -2.97042549e-01 -2.26688281e-01 -3.66602778e-01
-1.15290976e+00 9.93712127e-01 6.18720949e-01 1.01475947e-01
-1.56815231e-01 7.69937634e-01 2.66041219e-01 6.14914782e-02
4.44479465e-01 3.66126388e-01 -5.08031368e-01 3.11636835e-01
-9.63554978e-01 -4.52236012e-02 4.66304034e-01 1.16834247e+00
1.06821060e+00 1.46142498e-01 -6.97948337e-01 6.05077028e-01
7.72553921e-01 9.28061843e-01 1.40142590e-01 -7.00995982e-01
4.24717516e-01 2.33978301e-01 -1.00782849e-01 -4.66299325e-01
-3.03240120e-01 -3.21481675e-01 -3.86342883e-01 2.22665966e-02
1.01642609e-01 -4.93979966e-03 -1.53314853e+00 1.91168606e+00
-6.62014931e-02 5.38358927e-01 1.76614031e-01 7.32764900e-01
1.53360927e+00 9.47207749e-01 6.23980880e-01 5.84245324e-02
1.68753421e+00 -1.40158975e+00 -6.18954241e-01 -9.51828539e-01
5.48611820e-01 -5.75430870e-01 1.43644321e+00 1.58840548e-02
-1.03428471e+00 -9.74804997e-01 -9.23513293e-01 -6.73105717e-01
-6.11903369e-01 1.44683599e-01 8.01808000e-01 4.22563612e-01
-1.68244576e+00 -4.35902536e-01 -6.74914837e-01 -6.15946591e-01
5.71964383e-01 1.47204280e-01 -2.57106453e-01 -2.30019554e-01
-1.03734159e+00 8.64144444e-01 6.34929299e-01 -2.03468487e-01
-1.31854653e+00 -9.16127026e-01 -1.38223958e+00 -5.90039343e-02
5.91728330e-01 -9.09160852e-01 1.43195939e+00 -8.86552215e-01
-8.94498467e-01 1.34197080e+00 -3.99910897e-01 -7.13760316e-01
6.20925054e-02 -3.98254424e-01 -1.37871370e-01 3.93217430e-02
5.28124809e-01 1.51465631e+00 9.39469278e-01 -1.48103714e+00
-8.93231213e-01 -9.72904488e-02 4.16096821e-02 3.91704023e-01
2.42681324e-01 -1.87328413e-01 -1.09989345e+00 -3.33308429e-01
2.97786724e-02 -5.68077207e-01 -5.06428108e-02 2.62611449e-01
-2.45209068e-01 -4.24918801e-01 7.78127730e-01 -8.17920923e-01
6.77574635e-01 -2.04791522e+00 -4.61795107e-02 -2.39608109e-01
4.71690565e-01 2.43269801e-01 -7.72341430e-01 1.62692145e-01
5.06862439e-02 1.29375592e-01 -1.11794636e-01 -6.50945544e-01
3.88506688e-02 3.26076031e-01 -5.16442597e-01 3.30842495e-01
4.07216579e-01 1.71980524e+00 -1.02902758e+00 -5.56453764e-01
5.02588391e-01 4.18209702e-01 -4.58616197e-01 1.72922403e-01
-7.05516040e-01 3.40205759e-01 -2.70658761e-01 7.86830127e-01
5.77955186e-01 -5.92331588e-01 -3.76905739e-01 -3.92057240e-01
-1.58423573e-01 1.31152179e-02 -7.35066473e-01 2.05494905e+00
-6.50175929e-01 1.12795043e+00 -1.06115602e-02 -6.12102926e-01
6.10276580e-01 2.91234907e-03 1.27851263e-01 -8.75296235e-01
-2.54559871e-02 -1.51566565e-01 -2.35582665e-01 -4.54094976e-01
5.75046659e-01 1.03656605e-01 -1.48916811e-01 2.39059493e-01
6.76051974e-01 -3.90919358e-01 -3.80338877e-02 5.91109037e-01
6.84854507e-01 1.71500221e-01 5.27277827e-01 -3.05630803e-01
4.70744401e-01 1.34811997e-01 1.27651006e-01 1.28539574e+00
-3.73371840e-01 7.03462541e-01 5.64056300e-02 4.95393062e-04
-7.47273266e-01 -1.44369566e+00 4.53390740e-02 1.54310632e+00
4.02424753e-01 -3.70213836e-01 -6.36410952e-01 -3.96453261e-01
7.20787048e-02 1.15212870e+00 -6.94876850e-01 -8.05856884e-02
-2.55215559e-02 -3.10006499e-01 6.81724310e-01 7.78257966e-01
5.72151840e-01 -1.31222951e+00 -4.04202700e-01 -1.94851309e-01
-3.22461247e-01 -1.50766540e+00 -6.18587673e-01 1.83605224e-01
-2.08595201e-01 -7.22572148e-01 -5.26458502e-01 -1.25008214e+00
4.05832797e-01 7.13347077e-01 1.45218372e+00 6.51178285e-02
-5.63417733e-01 1.08636761e+00 -2.86656529e-01 -6.42611444e-01
-4.93198782e-02 -2.06393510e-01 3.73347811e-02 -1.59943968e-01
6.80304110e-01 2.42938727e-01 -3.72640818e-01 -1.78853080e-01
-6.67251050e-01 3.20094705e-01 6.81884170e-01 7.78171062e-01
1.09270310e+00 -5.77305019e-01 3.21882904e-01 -4.82518882e-01
5.55348337e-01 -2.44963169e-01 -6.67200863e-01 6.65823758e-01
-3.24741721e-01 -1.31704703e-01 -1.81601927e-01 -4.65026081e-01
-1.08676577e+00 1.94680784e-02 -3.82617980e-01 -5.62789798e-01
-2.75306314e-01 2.72504359e-01 -8.21440965e-02 -3.49014014e-01
6.19938493e-01 6.49440527e-01 -9.09016430e-02 -1.65894389e-01
1.30250168e+00 4.73937660e-01 8.55289876e-01 -4.31343913e-01
9.42997098e-01 4.00331885e-01 -7.14544415e-01 -1.10355723e+00
-1.17350519e+00 -8.87075126e-01 -6.30175591e-01 -2.11739782e-02
1.51763785e+00 -1.45136631e+00 -5.25258541e-01 1.83238953e-01
-1.44982481e+00 -6.53438747e-01 -4.83518749e-01 2.24910557e-01
-6.95201397e-01 2.54482031e-01 -4.08990443e-01 -8.36996496e-01
-4.36992168e-01 -1.17301512e+00 1.41037345e+00 4.43882138e-01
9.12795439e-02 -9.77808475e-01 -1.16439998e-01 4.58834589e-01
1.60620362e-01 -1.73878014e-01 5.09134948e-01 -6.70314312e-01
-8.63103628e-01 1.19377077e-01 -8.42353642e-01 3.69254529e-01
-2.24557906e-01 -4.15961266e-01 -1.30739212e+00 -2.84041584e-01
-2.09761009e-01 -7.52127945e-01 1.28020239e+00 8.03904355e-01
1.12618494e+00 1.56466246e-01 -3.64109963e-01 9.70472634e-01
1.54069221e+00 1.72505826e-02 5.22998571e-01 -1.04466103e-01
9.37617898e-01 4.40118641e-01 6.37164056e-01 -1.72409520e-01
6.05585456e-01 5.93912244e-01 4.36978519e-01 -6.57962739e-01
-7.72437215e-01 -5.23368299e-01 3.93273979e-01 4.65244234e-01
3.81141394e-01 -4.29476529e-01 -1.19876230e+00 8.49230468e-01
-1.95950055e+00 -7.54991472e-01 -7.24695250e-02 1.79306448e+00
4.41571087e-01 1.77000277e-03 -3.16736192e-01 -9.77116644e-01
3.93962950e-01 3.54721218e-01 -7.31780350e-01 -2.77905881e-01
-2.57810801e-01 1.12432979e-01 5.89249611e-01 1.02733994e+00
-1.12398088e+00 1.82820940e+00 6.24460268e+00 9.27454174e-01
-7.52163768e-01 7.60488868e-01 5.73451102e-01 1.66284427e-01
-3.84942412e-01 -1.63595796e-01 -1.04850721e+00 -7.32375085e-02
4.15975749e-01 -1.98219001e-01 2.49811813e-01 8.41577113e-01
1.55326566e-02 -1.44576624e-01 -1.16494536e+00 1.38580561e+00
5.34487009e-01 -1.45199370e+00 4.22926515e-01 -2.44591057e-01
6.29561186e-01 8.05032074e-01 2.48987243e-01 7.57294774e-01
3.19989085e-01 -1.30790591e+00 8.59282017e-01 3.06565821e-01
1.09089196e+00 -2.77062625e-01 6.67466104e-01 1.96734741e-01
-1.58379006e+00 1.39885053e-01 -3.76846999e-01 2.09221303e-01
4.01837796e-01 -6.50604144e-02 -7.69016802e-01 3.04904908e-01
6.06556296e-01 6.03453398e-01 -6.20085478e-01 8.04870129e-01
-6.59932971e-01 4.60457563e-01 -3.82712185e-02 1.83809459e-01
6.89325511e-01 -9.88380760e-02 4.51902211e-01 1.54931390e+00
-2.06174850e-01 1.65807381e-01 4.98488009e-01 1.24865317e+00
-1.09954411e-02 2.08350141e-02 -7.25867927e-01 5.53572364e-03
3.19891036e-01 1.13527286e+00 -7.71391034e-01 -5.09761393e-01
-7.75571406e-01 1.15421665e+00 2.43951485e-01 8.31550479e-01
-1.04039526e+00 -7.82100558e-02 8.77277136e-01 -2.02311963e-01
5.62711656e-01 -2.82107711e-01 -2.54252493e-01 -1.17754018e+00
-5.22253573e-01 -6.02708817e-01 2.85088837e-01 -1.23305130e+00
-1.12550712e+00 5.24539292e-01 7.54089430e-02 -5.51712275e-01
-6.36560619e-02 -8.43380868e-01 -4.84675795e-01 9.75197077e-01
-1.75030613e+00 -1.66496634e+00 -4.65948373e-01 7.38599479e-01
1.21861100e+00 -5.61491437e-02 5.34883618e-01 -7.52754062e-02
-3.29890728e-01 5.35244942e-01 -4.79336202e-01 2.09858477e-01
4.94669348e-01 -1.19245470e+00 6.62992060e-01 1.16123128e+00
6.95737362e-01 6.23912930e-01 5.19613385e-01 -6.69280887e-01
-1.31278551e+00 -1.38944614e+00 8.14751565e-01 -8.44863951e-01
8.29326391e-01 -7.72816300e-01 -8.68878663e-01 1.09816909e+00
3.17291200e-01 8.32300931e-02 4.92406666e-01 1.61327139e-01
-5.12128592e-01 2.67605305e-01 -9.17506397e-01 7.22947359e-01
1.14763355e+00 -9.48181033e-01 -7.78430283e-01 7.59259105e-01
1.32126236e+00 -4.87973869e-01 -3.60787898e-01 2.33623773e-01
2.23589972e-01 -4.21746671e-01 1.43003070e+00 -5.37879765e-01
2.59156283e-02 -5.11329532e-01 -4.09279168e-01 -7.31860816e-01
-4.68785942e-01 -2.22134948e-01 -2.47976646e-01 1.02221000e+00
5.68928421e-01 -5.48883140e-01 4.00321066e-01 2.50346452e-01
-1.57527819e-01 -4.82018113e-01 -5.97415864e-01 -8.00417900e-01
-2.63019830e-01 -8.93799365e-01 -6.56648129e-02 6.23318791e-01
-6.01306081e-01 7.23477662e-01 -1.62080780e-01 5.52063644e-01
7.66943634e-01 -5.14702722e-02 7.48505950e-01 -8.08643639e-01
-3.52719545e-01 -3.52250308e-01 -3.10398847e-01 -1.53616142e+00
2.99830228e-01 -1.13146055e+00 4.95905817e-01 -2.08088255e+00
3.72984588e-01 -4.90306877e-03 -2.85620272e-01 9.02486801e-01
-2.31573299e-01 4.84627455e-01 4.06206995e-01 1.05282255e-01
-1.14142394e+00 4.41347778e-01 9.22590137e-01 -4.38379735e-01
-2.71592349e-01 -3.35007429e-01 -8.89222682e-01 7.35732496e-01
5.92253745e-01 -8.20193961e-02 -8.54105413e-01 -7.55229175e-01
-1.41822919e-01 -2.75672317e-01 6.52418375e-01 -8.22226226e-01
2.80355692e-01 9.76517424e-03 1.75975591e-01 -7.54533708e-01
3.55875403e-01 -3.17248464e-01 -3.73240262e-01 5.41774213e-01
-1.97165594e-01 -1.91884086e-01 6.65108323e-01 6.55223906e-01
-1.80113018e-01 -4.55145985e-02 6.91541791e-01 -2.38658413e-01
-1.98013830e+00 4.98500466e-01 -1.03267074e-01 2.81161964e-01
9.55431223e-01 -4.13614929e-01 -4.61215377e-01 -4.51842487e-01
-8.43325078e-01 7.83647597e-01 4.29750502e-01 7.74715722e-01
9.42084372e-01 -9.33543921e-01 -8.18491757e-01 1.23922668e-01
6.88784242e-01 9.97169688e-02 2.00810477e-01 8.63320470e-01
-4.84236091e-01 7.49112606e-01 1.06551200e-01 -1.12716031e+00
-1.18208432e+00 6.21992946e-01 3.95644665e-01 2.77823675e-02
-7.88785756e-01 1.67576408e+00 1.15878189e+00 -3.50539744e-01
5.01642525e-01 -4.50930864e-01 -8.99489373e-02 -4.21149544e-02
8.93087924e-01 -3.32802653e-01 -3.33066434e-01 -6.58744097e-01
-5.48242807e-01 7.70877779e-01 -2.50130653e-01 -1.80928707e-01
7.37731397e-01 -4.11006570e-01 1.13010824e-01 5.90699852e-01
9.15818930e-01 -4.08034712e-01 -1.37837124e+00 -3.95997941e-01
-9.41888541e-02 1.62932891e-02 3.17614943e-01 -9.13594961e-01
-7.31043518e-01 1.05699635e+00 6.45000100e-01 -2.91537493e-01
9.92538273e-01 5.65504134e-01 4.09141213e-01 3.69129866e-01
4.70903873e-01 -8.24344099e-01 3.46709460e-01 6.89024508e-01
8.78232002e-01 -1.51214290e+00 -1.59025595e-01 -4.40621853e-01
-9.34263647e-01 4.74903852e-01 9.53994215e-01 5.44705503e-02
4.06848669e-01 2.12530583e-01 2.51422346e-01 -3.66308153e-01
-6.73944652e-01 -1.03878915e+00 7.04355896e-01 9.33582783e-01
2.69420087e-01 9.10502151e-02 4.06010225e-02 4.35744554e-01
2.29477450e-01 -2.10327879e-01 1.51863126e-02 4.57814544e-01
-7.86753476e-01 -3.92118335e-01 -1.12456284e-01 2.56696701e-01
8.43731835e-02 -6.85647011e-01 -3.61691266e-01 1.07819486e+00
3.72015864e-01 9.58488286e-01 3.45100313e-01 -1.33553520e-01
-6.02610223e-02 7.45649114e-02 4.51947093e-01 -1.14268100e+00
-2.33534530e-01 -8.28637704e-02 -1.54198393e-01 -8.62703621e-01
-2.10403621e-01 -2.48602316e-01 -1.44237399e+00 1.68385521e-01
-2.98305035e-01 -2.21166790e-01 5.98554671e-01 1.00584793e+00
2.01344609e-01 6.79758549e-01 -2.76516914e-01 -7.51473248e-01
-1.36565715e-01 -7.89439321e-01 -3.98875713e-01 2.51576364e-01
5.78973889e-01 -6.49755120e-01 -1.13866448e-01 7.41735548e-02] | [10.505387306213379, 1.6264230012893677] |
3e7ea0d5-8064-4f38-b739-f840c33079ac | the-repere-corpus-a-multimodal-corpus-for | null | null | https://aclanthology.org/L12-1410 | https://aclanthology.org/L12-1410.pdf | The REPERE Corpus : a multimodal corpus for person recognition | The REPERE Challenge aims to support research on people recognition in multimodal conditions. To assess the technology progression, annual evaluation campaigns will be organized from 2012 to 2014. In this context, the REPERE corpus, a French videos corpus with multimodal annotation, has been developed. This paper presents datasets collected for the dry run test that took place at the beginning of 2012. Specific annotation tools and guidelines are mainly described. At the time being, 6 hours of data have been collected and annotated. Last section presents analyses of annotation distribution and interaction between modalities in the corpus. | ["Val{\\'e}rie Mapelli", "Matthieu Carr{\\'e}", 'Ludovic Quintard', 'Juliette Kahn', 'Olivier Galibert', 'Aude Giraudel'] | 2012-05-01 | null | null | null | lrec-2012-5 | ['person-recognition'] | ['computer-vision'] | [ 6.52734488e-02 2.14603227e-02 -4.88733426e-02 -5.63588381e-01
-3.97004157e-01 -5.66332579e-01 9.90470767e-01 1.33581445e-01
-6.31630123e-01 8.24488521e-01 7.73497701e-01 4.84945059e-01
1.32476792e-01 -1.85031816e-01 -1.15119338e-01 -1.74870178e-01
-2.70959675e-01 2.83173472e-01 -2.35158741e-01 -2.18386605e-01
7.01372027e-02 2.89046079e-01 -1.80495429e+00 9.18693364e-01
2.44408593e-01 8.00004125e-01 -2.08347291e-01 1.07682312e+00
6.89529553e-02 9.02238071e-01 -7.86166310e-01 -6.23706937e-01
-1.14474438e-01 -4.50436145e-01 -1.13714612e+00 4.53649729e-01
8.80450845e-01 -4.14058089e-01 -7.57736266e-01 7.26317406e-01
1.01263976e+00 1.39001101e-01 3.57378632e-01 -1.17364907e+00
1.99100189e-02 5.99739969e-01 1.53402060e-01 2.05673590e-01
1.33751345e+00 2.66409606e-01 6.34630799e-01 -1.03892958e+00
1.17120492e+00 1.12091315e+00 7.52690792e-01 8.55562270e-01
-7.08238006e-01 -1.89764678e-01 -2.65207112e-01 4.26302016e-01
-1.43808532e+00 -9.09369826e-01 4.34971452e-01 -8.18067372e-01
1.08869541e+00 5.45539856e-01 9.04308975e-01 1.74091005e+00
-4.86023277e-01 1.18983448e+00 1.08653843e+00 -9.13155794e-01
-5.15240692e-02 3.23996037e-01 4.67675239e-01 4.58462805e-01
3.44631560e-02 -2.59779662e-01 -8.98770750e-01 9.01702866e-02
2.03452542e-01 -6.63766861e-01 -3.95484179e-01 -9.64983776e-02
-1.28912485e+00 4.36381787e-01 -2.73743749e-01 7.75779545e-01
-3.61804098e-01 -1.47971377e-01 1.19508684e+00 2.22033799e-01
1.92336850e-02 1.38061317e-02 -8.21363032e-02 -9.82319534e-01
-7.50840664e-01 3.55511576e-01 1.09245777e+00 9.93471146e-01
-8.76504704e-02 -2.63046622e-01 -5.40546954e-01 8.26375782e-01
5.31139731e-01 3.82467091e-01 1.75443634e-01 -9.00092423e-01
9.80930090e-01 5.99078000e-01 2.63501197e-01 -8.62973392e-01
-6.09772027e-01 3.43268692e-01 -6.42394185e-01 -3.23599696e-01
6.15809679e-01 -6.74888015e-01 -5.43027818e-01 1.23365104e+00
2.04387493e-02 -3.46362144e-01 2.65268683e-01 8.59217405e-01
1.48532856e+00 3.98602039e-01 3.55202734e-01 -1.67029455e-01
1.43875265e+00 -8.27789247e-01 -1.30162680e+00 2.75563508e-01
7.58966267e-01 -8.55816126e-01 8.28482985e-01 4.37616318e-01
-1.13051951e+00 -6.98422194e-01 -8.81671071e-01 2.03078002e-01
-6.21505857e-01 4.86670613e-01 3.48782450e-01 1.34081388e+00
-8.67075682e-01 5.35790846e-02 -4.29781348e-01 -1.01801896e+00
3.34403425e-01 1.21692061e-01 -7.72488713e-01 -6.29329234e-02
-1.29575849e+00 8.41876566e-01 6.31798327e-01 4.51617926e-01
-7.00324893e-01 8.42993259e-02 -7.77805030e-01 -5.82265556e-01
1.09092794e-01 -1.29143730e-01 1.41424263e+00 -1.04508448e+00
-1.26931441e+00 1.23508751e+00 -7.06140324e-02 -3.85973185e-01
6.88442826e-01 -5.38147449e-01 -1.04107034e+00 3.39612037e-01
-2.72940814e-01 7.74027109e-01 1.48967341e-01 -8.93541574e-01
-8.16214919e-01 -1.99496031e-01 3.09713185e-01 3.79604518e-01
-5.22592604e-01 6.44607544e-01 -8.98932457e-01 -5.46138525e-01
-8.02792311e-01 -9.98312533e-01 4.09505010e-01 -6.77067518e-01
-4.05902267e-01 -2.16760918e-01 7.72687495e-01 -1.22414458e+00
1.52019775e+00 -2.13639164e+00 3.29733461e-01 2.11618140e-01
1.00963153e-01 3.99766624e-01 -7.84748197e-02 6.71003103e-01
1.07873209e-01 1.24411523e-01 7.22138137e-02 -5.30218720e-01
3.34305882e-01 7.01456219e-02 1.14180759e-01 4.57374305e-01
-2.98119694e-01 7.02319205e-01 -5.25347352e-01 -5.40421486e-01
4.10165757e-01 5.45720220e-01 -9.57370698e-02 3.19615424e-01
6.51640743e-02 6.33391678e-01 -1.45121992e-01 8.98538351e-01
5.07268488e-01 3.98752332e-01 3.14684093e-01 -4.51071829e-01
-5.17539024e-01 -2.02700138e-01 -1.11987150e+00 1.79521167e+00
-5.21323411e-03 1.34908724e+00 1.49830356e-01 -7.72437513e-01
6.01397216e-01 1.00706303e+00 6.52877927e-01 -7.43470311e-01
4.43021327e-01 9.03071985e-02 -2.29899645e-01 -1.24236357e+00
1.01121485e+00 5.48145592e-01 -4.42918628e-01 -1.71971604e-01
4.15163487e-01 5.06506562e-01 1.06253827e+00 7.15616867e-02
1.03621554e+00 1.92880988e-01 9.47238877e-02 2.05014721e-01
1.02661431e+00 1.26590759e-01 -3.57139483e-02 5.13347089e-01
-6.85035110e-01 3.85904223e-01 4.22360778e-01 -4.43863064e-01
-1.12997067e+00 -4.96279746e-01 -1.31308779e-01 8.26189160e-01
-3.43392015e-01 -9.62598681e-01 -1.00871694e+00 -8.12093258e-01
-5.28655589e-01 2.94345677e-01 -5.42528152e-01 4.37576592e-01
-5.36847770e-01 -2.72293955e-01 1.12115777e+00 4.10074294e-01
1.00956523e+00 -1.26268518e+00 -8.94955516e-01 -1.75200880e-01
-6.62904441e-01 -1.47972298e+00 -9.99278501e-02 -4.44527417e-01
-3.94715667e-01 -1.42847145e+00 -9.04305637e-01 -6.05070651e-01
2.29679376e-01 -3.10856581e-01 1.09746921e+00 -6.73400089e-02
-2.78429329e-01 1.35144138e+00 -7.64883637e-01 -1.25165910e-01
-1.13762617e-01 2.01513767e-01 -1.32336840e-01 -3.03715318e-02
6.67406380e-01 3.30071181e-01 -2.40775704e-01 5.33127785e-01
-5.55155396e-01 -1.73663065e-01 1.66942850e-01 6.28442824e-01
2.09089387e-02 -1.39572054e-01 3.27682018e-01 -5.70291221e-01
5.30505538e-01 -2.14157686e-01 -1.25236690e-01 3.10229480e-01
1.95673570e-01 -6.49127364e-01 -2.54478872e-01 -2.29164764e-01
-1.27779686e+00 2.37285540e-01 -2.47472435e-01 -1.54245216e-02
-9.24003720e-01 5.50218523e-01 -4.14855748e-01 1.11696005e-01
5.63830495e-01 -3.12433571e-01 -1.79702416e-01 -4.65492457e-01
2.91567445e-01 1.19777441e+00 4.99724805e-01 -4.94542629e-01
1.34456707e-02 1.28929392e-01 -3.72529566e-01 -1.37180424e+00
-4.25276518e-01 -4.51428503e-01 -8.36062253e-01 -1.15886247e+00
1.22910905e+00 -1.12812662e+00 -7.05258071e-01 8.53505492e-01
-1.07192028e+00 -3.67990702e-01 -2.69142479e-01 6.16893828e-01
-3.89361650e-01 2.50130832e-01 -5.80052197e-01 -1.18020868e+00
-4.30906489e-02 -8.49397063e-01 7.96040475e-01 3.03606689e-01
-6.01071000e-01 -9.19602454e-01 2.96925932e-01 8.74072552e-01
2.00436145e-01 3.88832241e-01 -4.10050631e-01 -6.00495338e-01
-2.81866677e-02 -5.01368999e-01 -3.45720313e-02 3.57822150e-01
-2.78692603e-01 2.49592047e-02 -9.57782686e-01 -4.41689014e-01
-5.14451802e-01 -7.25324512e-01 3.72208953e-01 1.09251030e-02
5.81079960e-01 -5.78097329e-02 -2.82513410e-01 -2.77613670e-01
8.29950452e-01 2.28934780e-01 9.43827868e-01 4.39943254e-01
5.40478826e-01 1.03210723e+00 7.46897519e-01 4.07407075e-01
3.24305981e-01 1.01676750e+00 1.53088570e-02 4.77480799e-01
-2.90144116e-01 2.81056855e-02 7.18510866e-01 8.68231714e-01
-5.14030457e-01 -4.63191539e-01 -1.16353178e+00 2.78732359e-01
-1.74934423e+00 -1.29796040e+00 -5.76499522e-01 2.04430389e+00
3.32464099e-01 -1.55571416e-01 7.19310880e-01 2.23728314e-01
8.93748999e-01 -6.67774258e-03 3.27099621e-01 -1.86486647e-01
-4.74226564e-01 -4.84859377e-01 -2.67147692e-03 3.06705058e-01
-1.49068546e+00 5.08245051e-01 6.90781546e+00 6.53640866e-01
-6.72495961e-01 4.30433750e-01 2.96477020e-01 -9.86653641e-02
4.27673250e-01 -3.75243813e-01 -9.97562528e-01 5.70994139e-01
1.45288622e+00 4.43492800e-01 1.60028800e-01 3.95327151e-01
1.77477747e-01 -4.18457180e-01 -9.23466206e-01 1.38051772e+00
6.05028391e-01 -1.09759045e+00 -4.39429641e-01 1.08215027e-03
5.12573600e-01 6.56012967e-02 -2.63589174e-01 5.95001221e-01
-5.02627194e-01 -9.21364963e-01 1.07645559e+00 1.17290151e+00
8.27945650e-01 -6.87166631e-01 1.24106729e+00 -5.65935113e-02
-1.24662983e+00 -2.08401218e-01 2.71612406e-01 4.06393073e-02
6.63724184e-01 1.01677828e-01 -5.42975307e-01 7.63155222e-01
7.63228834e-01 8.04350853e-01 -1.08202648e+00 1.23228657e+00
-1.70250520e-01 5.81336081e-01 -3.17919433e-01 -1.10935263e-01
-1.73974603e-01 9.87455249e-03 8.48373830e-01 1.96170616e+00
3.56734484e-01 -1.33775800e-01 2.23142073e-01 -1.41357377e-01
-6.27411008e-02 1.78537086e-01 -6.56743944e-01 -3.49299371e-01
3.41328323e-01 1.24538648e+00 -5.02543688e-01 -4.17838961e-01
-5.56990206e-01 1.12474656e+00 -7.87349492e-02 2.15958267e-01
-8.39037776e-01 -3.10838878e-01 -1.04864813e-01 3.96694764e-02
-5.62874181e-03 -3.11709613e-01 9.80435237e-02 -1.05907524e+00
-2.37904186e-03 -1.03978336e+00 6.68387711e-01 -8.91672850e-01
-9.73347366e-01 7.03637183e-01 4.93218690e-01 -1.15245020e+00
-2.64805019e-01 -6.24379396e-01 -9.38363671e-02 3.93222362e-01
-3.54092628e-01 -1.25989139e+00 -5.86406052e-01 4.92363781e-01
7.11755812e-01 -8.73365581e-01 1.13741994e+00 1.01657808e+00
-9.11239564e-01 5.22860050e-01 -3.53662312e-01 2.90949404e-01
6.45536900e-01 -1.00457108e+00 -3.10162663e-01 5.57390928e-01
7.17157498e-02 3.44850779e-01 8.05364788e-01 -7.04957664e-01
-1.30261016e+00 -5.66737235e-01 1.15223360e+00 -6.68852031e-01
5.79671562e-01 -1.63207293e-01 -3.41227502e-01 7.21165419e-01
9.52539563e-01 -3.57241392e-01 9.28334951e-01 4.83832434e-02
-2.17991788e-02 3.15256685e-01 -1.07224250e+00 2.89774120e-01
9.06567514e-01 -7.93842554e-01 -6.27742946e-01 2.31607720e-01
-2.67583489e-01 -7.25107670e-01 -1.11821604e+00 4.10164952e-01
8.27266514e-01 -8.69962394e-01 6.70424044e-01 -3.91440332e-01
1.14928104e-01 -2.62893200e-01 -5.79392374e-01 -7.46698558e-01
4.97864783e-01 -7.58329391e-01 -5.97224295e-01 1.78739464e+00
3.38341773e-01 4.78827991e-02 6.03472412e-01 6.27345979e-01
-2.18107570e-02 -1.36369929e-01 -1.08050644e+00 -4.65047657e-01
-6.29424632e-01 -7.63286412e-01 4.25982885e-02 8.35423946e-01
3.69528472e-01 5.41776836e-01 -6.44466102e-01 -3.44665378e-01
2.25015864e-01 -9.67020631e-01 1.12261772e+00 -1.03344989e+00
1.57744825e-01 -5.15648305e-01 -7.17577100e-01 -5.51019549e-01
-1.96471158e-03 -5.64021409e-01 -1.98150769e-01 -1.50977874e+00
4.07219410e-01 2.58494407e-01 2.73959368e-01 4.64570433e-01
3.19532603e-01 6.36084795e-01 4.91403729e-01 1.47174541e-02
-1.04989493e+00 1.89010710e-01 5.59805036e-01 -2.99620003e-01
-1.40703842e-02 -2.29504600e-01 9.57896486e-02 5.58784604e-01
1.20917368e+00 2.68172622e-01 8.77374690e-03 -2.70363111e-02
2.63909578e-01 -3.80548388e-01 4.07036096e-01 -1.25925279e+00
1.07605845e-01 5.30390203e-01 3.80884230e-01 -1.03342402e+00
6.43417001e-01 -9.60575759e-01 5.70484281e-01 2.60034949e-01
-4.76672649e-01 2.07256615e-01 3.08129400e-01 1.50390387e-01
-4.47388858e-01 -6.26590669e-01 2.18802378e-01 2.32034460e-01
-1.04743743e+00 -3.09455872e-01 -8.97357762e-01 -1.70014918e-01
1.24787498e+00 -2.07795218e-01 -2.48382464e-01 -2.69856244e-01
-1.27120042e+00 3.83411944e-01 1.33481264e-01 5.14213622e-01
4.05002952e-01 -1.72993886e+00 -6.37624562e-01 -3.36952060e-01
3.26390147e-01 -7.45815754e-01 7.62213945e-01 1.01189339e+00
-6.37444377e-01 7.35052049e-01 -5.82892060e-01 -5.04057825e-01
-1.95027065e+00 1.99746698e-01 1.89772680e-01 -2.96120346e-01
-5.16375184e-01 3.69215637e-01 -1.03932714e+00 -2.97001243e-01
6.52147412e-01 3.88319165e-01 -1.04480600e+00 7.81839311e-01
9.67668772e-01 9.54589903e-01 -3.07036024e-02 -1.38175654e+00
-3.03721875e-01 1.50325879e-01 3.86935234e-01 -6.88105881e-01
9.07958448e-01 -5.36431432e-01 2.94538215e-02 7.58292794e-01
9.68244135e-01 7.84701630e-02 -8.37497771e-01 3.06226403e-01
3.02082509e-01 -3.87016624e-01 -2.19149664e-01 -1.04366624e+00
-7.06647038e-01 7.19511807e-01 1.30065238e+00 5.14567256e-01
9.04967189e-01 -1.12534158e-01 3.33384544e-01 4.51539576e-01
1.25668585e-01 -1.65148520e+00 -3.83041762e-02 7.14805543e-01
1.29704678e+00 -1.00356913e+00 -1.46943077e-01 -2.91472077e-01
-8.22529912e-01 1.31870830e+00 5.98744750e-01 6.84844971e-01
2.06937879e-01 4.23447937e-02 -1.21914245e-01 -3.79833668e-01
-2.14206487e-01 -3.58926028e-01 5.00621676e-01 7.34186172e-01
7.81698942e-01 1.23985492e-01 -6.45034909e-01 6.12739980e-01
-3.67983207e-02 3.56717408e-01 3.39360029e-01 1.22515666e+00
-1.46926165e-01 -1.25935543e+00 -6.63993657e-01 1.22560516e-01
-4.50185180e-01 3.98062319e-01 -7.93063700e-01 1.14178896e+00
4.27746177e-01 1.18080080e+00 -4.12736908e-02 -4.84563231e-01
7.70415068e-01 3.54853153e-01 8.46880138e-01 -2.19847649e-01
-7.41352320e-01 -9.21066776e-02 1.22818577e+00 -4.77699012e-01
-1.25317550e+00 -1.14393115e+00 -8.27759206e-01 -3.21563393e-01
8.15746486e-02 2.43503988e-01 6.07724428e-01 8.38573515e-01
3.37943137e-02 5.48166752e-01 7.06716999e-02 -1.17381549e+00
1.81503773e-01 -1.39595449e+00 -2.53446966e-01 3.89044166e-01
-1.46243706e-01 -5.95238626e-01 1.46875195e-02 3.98489356e-01] | [10.64124584197998, 0.7847983837127686] |
792db795-7023-498d-8adf-bef9cefdfa40 | nonsmooth-analysis-and-subgradient-methods | 1701.06393 | null | http://arxiv.org/abs/1701.06393v1 | http://arxiv.org/pdf/1701.06393v1.pdf | Nonsmooth Analysis and Subgradient Methods for Averaging in Dynamic Time Warping Spaces | Time series averaging in dynamic time warping (DTW) spaces has been
successfully applied to improve pattern recognition systems. This article
proposes and analyzes subgradient methods for the problem of finding a sample
mean in DTW spaces. The class of subgradient methods generalizes existing
sample mean algorithms such as DTW Barycenter Averaging (DBA). We show that DBA
is a majorize-minimize algorithm that converges to necessary conditions of
optimality after finitely many iterations. Empirical results show that for
increasing sample sizes the proposed stochastic subgradient (SSG) algorithm is
more stable and finds better solutions in shorter time than the DBA algorithm
on average. Therefore, SSG is useful in online settings and for non-small
sample sizes. The theoretical and empirical results open new paths for devising
sample mean algorithms: nonsmooth optimization methods and modified variants of
pairwise averaging methods. | ['Brijnesh Jain', 'David Schultz'] | 2017-01-23 | null | null | null | null | ['time-series-averaging'] | ['time-series'] | [ 2.97244132e-01 -3.48791331e-01 -2.27170229e-01 -3.89900297e-01
-1.20998573e+00 -4.98047054e-01 2.24253476e-01 -1.69037774e-01
-4.01668608e-01 9.44342732e-01 1.64683759e-01 -2.32929468e-01
-7.64091432e-01 -5.02396107e-01 -6.58319771e-01 -1.22662139e+00
-7.22492576e-01 3.05102766e-01 -1.15552060e-01 4.70326021e-02
6.54101729e-01 6.15658164e-01 -1.12920880e+00 -2.78925121e-01
1.02080476e+00 8.88477325e-01 -1.59058198e-02 9.33063805e-01
-8.06105807e-02 8.24086741e-02 -3.53652537e-01 -1.39717842e-02
7.73984194e-01 -8.08968961e-01 -5.52220523e-01 1.57038897e-01
7.24008620e-01 -1.72279343e-01 4.13331538e-02 1.01435077e+00
8.20439458e-01 7.49932528e-01 5.17211676e-01 -1.51111627e+00
-5.13843656e-01 6.88489079e-01 -1.17322314e+00 5.21139026e-01
9.41926762e-02 -1.00391060e-01 9.92583871e-01 -9.38569963e-01
5.39772511e-01 1.33318067e+00 8.94838512e-01 4.59927052e-01
-1.35368645e+00 -3.31608146e-01 3.63666236e-01 3.15775961e-01
-1.26862919e+00 -3.19418937e-01 7.19223559e-01 3.19371489e-03
7.73818552e-01 9.24746811e-01 6.82832181e-01 7.86507666e-01
3.08949620e-01 1.16786742e+00 1.08892775e+00 -3.47561955e-01
6.12760663e-01 -6.63739264e-01 2.87099600e-01 5.29667377e-01
5.30777097e-01 -1.44947141e-01 -5.55745304e-01 -4.52149004e-01
6.22917891e-01 -1.79815502e-03 -2.77686387e-01 -5.19910038e-01
-1.50246930e+00 7.58770466e-01 2.85043940e-02 4.49573755e-01
-6.77998602e-01 2.95456856e-01 2.18143791e-01 5.35304368e-01
9.18743730e-01 4.58640039e-01 -3.36442560e-01 -5.43465257e-01
-1.29908252e+00 7.69776583e-01 9.21514988e-01 9.24697280e-01
4.49636489e-01 6.76052645e-02 -3.81154686e-01 8.07774901e-01
-2.96809435e-01 1.11947775e+00 2.98859656e-01 -1.36708534e+00
5.40523171e-01 8.12193826e-02 3.29141855e-01 -1.16556418e+00
-4.18317467e-01 -3.14007252e-01 -7.61400819e-01 -1.64134949e-01
6.44737005e-01 -2.28785768e-01 -4.84188139e-01 1.74247980e+00
5.83594918e-01 5.93322992e-01 -4.31956470e-01 9.01275992e-01
-2.23431617e-01 7.01075375e-01 -5.24605989e-01 -1.00324428e+00
7.22277761e-01 -7.28438735e-01 -1.00913405e+00 1.02480479e-01
5.98548830e-01 -9.01501000e-01 9.74571288e-01 5.02217770e-01
-1.11633813e+00 -3.91534902e-02 -9.29541528e-01 1.64220631e-01
-1.47310331e-01 -1.39211327e-01 6.32418752e-01 8.16009283e-01
-1.08483064e+00 1.09677064e+00 -1.01720691e+00 -5.16593337e-01
4.61449772e-01 2.63504326e-01 1.46561801e-01 -9.82327461e-02
-6.18975937e-01 5.12260675e-01 -1.09091491e-01 3.20882946e-01
-7.05703139e-01 -1.16751051e+00 -5.39006293e-01 -4.03538108e-01
4.20668542e-01 -5.33977032e-01 1.08244491e+00 -6.74410403e-01
-1.55334139e+00 5.14054418e-01 -7.93905616e-01 -8.01766157e-01
9.94857371e-01 -3.14729393e-01 -1.58490997e-03 -1.17628640e-02
-5.47173470e-02 4.92859557e-02 1.15554571e+00 -6.77385151e-01
-4.71258372e-01 -6.00113809e-01 -2.68002599e-01 -1.31005747e-02
-3.33531559e-01 -2.91021198e-01 1.95421159e-01 -7.54938066e-01
3.78967285e-01 -9.37072754e-01 -6.06482804e-01 -7.14906603e-02
-1.91930279e-01 -2.71664292e-01 8.98291290e-01 -6.22769952e-01
1.50560367e+00 -1.77437651e+00 2.72656679e-01 4.39611912e-01
1.00641087e-01 1.13299368e-02 -3.85676056e-01 5.85768163e-01
-1.36357054e-01 -2.59626627e-01 -5.49032032e-01 -3.09268802e-01
-2.51249578e-02 3.16060185e-01 -5.20741582e-01 8.89733076e-01
-3.72726232e-01 7.06462383e-01 -1.13410747e+00 -3.39546651e-01
1.14472665e-01 7.04765171e-02 -6.28611088e-01 -2.48975068e-01
1.83027759e-01 1.80227861e-01 -4.52896774e-01 5.12871623e-01
1.02364874e+00 8.90087858e-02 -1.63953543e-01 -1.44689962e-01
-3.31562161e-01 -3.94405872e-01 -1.45883417e+00 1.61535716e+00
-5.39694488e-01 8.06011379e-01 6.86004236e-02 -1.29859078e+00
7.97496855e-01 -6.70037866e-02 1.06405544e+00 -1.41552821e-01
3.58492911e-01 3.81524056e-01 -3.08685571e-01 -5.22431672e-01
4.66693372e-01 -1.13522947e-01 1.91836372e-01 7.22570419e-01
-2.47417927e-01 -1.09683096e-01 5.44618666e-01 1.65129334e-01
1.10796702e+00 -2.08360821e-01 3.30899060e-01 -7.73619056e-01
5.68917036e-01 -1.75670817e-01 5.58185637e-01 9.86279666e-01
-5.81182182e-01 5.16396761e-01 2.70950645e-01 -4.34102565e-01
-1.15767336e+00 -1.04212773e+00 -1.25094011e-01 9.99681592e-01
3.17038372e-02 -2.27146208e-01 -1.06319356e+00 -2.82920480e-01
1.80436447e-01 6.73660040e-01 -7.63609111e-01 -1.50223911e-01
-7.76385844e-01 -9.34594214e-01 3.03910091e-03 3.98586392e-01
5.58117211e-01 -4.23688084e-01 -5.02515435e-01 3.76937121e-01
-3.77769172e-01 -8.20938528e-01 -1.27648294e+00 -4.40127134e-01
-1.46371830e+00 -9.97607231e-01 -1.21904957e+00 -3.24057609e-01
8.11924696e-01 6.96416557e-01 6.41531408e-01 -5.42560458e-01
-4.19139445e-01 7.55968690e-01 -2.88484573e-01 -5.34352124e-01
1.76944226e-01 -9.35107991e-02 3.61551881e-01 4.23390597e-01
4.36495006e-01 -7.06880629e-01 -8.30067635e-01 4.53703493e-01
-6.85031116e-01 -4.47470456e-01 2.77712405e-01 7.74600148e-01
9.16848302e-01 -4.59606163e-02 4.82604861e-01 -6.37860358e-01
1.08216357e+00 -2.62812644e-01 -8.45254779e-01 3.62811178e-01
-7.51487613e-01 2.34277785e-01 5.31223774e-01 -6.29633665e-01
-7.70511150e-01 -2.35616758e-01 3.93019199e-01 -6.42155707e-01
6.38275385e-01 2.04180777e-01 2.24698439e-01 -2.00933188e-01
7.22719193e-01 2.97594815e-01 3.82704914e-01 -2.31296971e-01
4.29083198e-01 4.01238889e-01 2.31009677e-01 -8.75920773e-01
6.21064782e-01 9.51994956e-01 5.24135888e-01 -1.10207796e+00
-6.99966669e-01 -7.33998179e-01 -4.99885857e-01 -5.25148988e-01
5.04110992e-01 -2.22590819e-01 -7.71553099e-01 4.82093930e-01
-7.62577236e-01 -3.51255804e-01 -5.85778177e-01 6.44930303e-01
-1.02610910e+00 5.09580970e-01 -2.29925707e-01 -1.16840422e+00
-5.01473188e-01 -6.73845708e-01 1.05659521e+00 6.80835694e-02
-3.43425959e-01 -1.07301843e+00 2.87773907e-01 -3.18131298e-02
5.53535581e-01 6.25053346e-01 2.71051884e-01 -2.93868780e-01
-3.24211687e-01 -4.63327646e-01 1.93242803e-01 2.69191384e-01
1.78305581e-01 1.19490042e-01 -6.12996280e-01 -4.88289267e-01
3.54194671e-01 3.86710644e-01 6.77101314e-01 1.24617183e+00
1.22011828e+00 -5.06879032e-01 -3.25723380e-01 6.15350723e-01
1.38828671e+00 2.06252992e-01 3.05179089e-01 2.14145109e-02
3.94019127e-01 4.07385647e-01 9.11752939e-01 7.96837330e-01
-7.38139898e-02 2.91332304e-01 -4.63337228e-02 4.03005332e-01
2.95952320e-01 1.79562002e-01 3.43243092e-01 9.94322538e-01
-4.05544996e-01 2.76195556e-01 -6.70114934e-01 8.18784773e-01
-2.12185264e+00 -1.41467500e+00 -2.30805680e-01 2.53873825e+00
7.11529732e-01 -3.48899484e-01 4.64505315e-01 2.50329554e-01
9.56336141e-01 4.32008803e-01 -7.94795632e-01 -5.44160008e-01
-1.85405970e-01 4.26575482e-01 9.90346134e-01 7.28248954e-01
-9.74669218e-01 4.29160148e-01 7.06195450e+00 1.20158660e+00
-6.85009062e-01 1.01811655e-01 3.92876625e-01 -6.13038421e-01
-1.84090227e-01 -1.62206516e-01 -5.49116313e-01 3.24804902e-01
9.68338609e-01 -9.63853240e-01 8.98652792e-01 9.54704583e-01
7.56779253e-01 -2.19800591e-01 -1.01826203e+00 1.23976648e+00
1.10677250e-01 -1.15744460e+00 -2.23848492e-01 1.96372092e-01
1.21197772e+00 -9.66640562e-02 9.70264748e-02 -7.65388459e-02
1.51527062e-01 -4.07351166e-01 3.05422664e-01 5.38383961e-01
1.51040807e-01 -9.96559262e-01 4.34114337e-01 1.81385741e-01
-1.26963603e+00 -2.61568218e-01 -6.73598349e-01 4.30112183e-02
4.80166376e-01 1.22138572e+00 -2.11423010e-01 5.57483137e-01
4.40871596e-01 9.06700611e-01 -9.24110636e-02 1.32447207e+00
2.61493087e-01 5.13801813e-01 -5.50408125e-01 -4.95083481e-01
4.00218368e-01 -7.31937945e-01 1.16050935e+00 9.10530746e-01
6.92029595e-01 1.53611735e-01 6.04974106e-02 4.57609266e-01
3.20859790e-01 3.81384373e-01 -5.34833074e-01 -6.65760487e-02
5.39079070e-01 1.02074265e+00 -8.59118104e-01 -4.15101588e-01
-4.16764384e-03 8.50772440e-01 -2.12408036e-01 5.27405858e-01
-8.70055616e-01 -7.05906451e-01 8.84891689e-01 -8.38682875e-02
3.42084557e-01 -5.53811669e-01 -5.02401710e-01 -1.12375939e+00
3.06223005e-01 -7.33096600e-01 7.07278728e-01 -4.28278714e-01
-1.34155715e+00 1.44441023e-01 2.64517218e-01 -1.35473454e+00
-5.71094714e-02 -4.62794155e-01 -7.03230143e-01 5.18014967e-01
-8.82875800e-01 -4.16117817e-01 -1.04763120e-01 6.95596933e-01
7.46874571e-01 1.41904801e-01 4.19842064e-01 7.13556036e-02
-5.41545391e-01 5.53510606e-01 6.78021431e-01 -3.17652673e-01
5.41236460e-01 -1.34393036e+00 1.54537171e-01 1.25302231e+00
-6.49861023e-02 6.67898238e-01 1.32144618e+00 -3.85079175e-01
-1.74915409e+00 -8.97479713e-01 7.54715204e-01 -1.73269615e-01
9.17638123e-01 8.54506120e-02 -5.96602857e-01 4.79776531e-01
2.03063697e-01 -7.67027363e-02 5.15949786e-01 3.22248787e-01
-4.05563898e-02 -5.94075739e-01 -1.39074528e+00 7.72178888e-01
1.36880755e+00 -2.61628151e-01 -3.56702000e-01 6.28238976e-01
2.79160738e-01 -2.75225699e-01 -9.12382662e-01 2.31240720e-01
5.88513792e-01 -6.75151169e-01 9.86058891e-01 -5.36367953e-01
1.48118988e-01 -7.53504410e-02 -1.96398661e-01 -1.56442177e+00
-5.78038990e-02 -1.35310900e+00 -2.77301073e-01 9.01419997e-01
5.41901402e-02 -1.00765574e+00 6.16489768e-01 7.07501411e-01
1.18264124e-01 -8.85126948e-01 -1.23739290e+00 -1.52134907e+00
1.84434026e-01 -5.40223241e-01 5.50932884e-01 7.78285265e-01
8.21657255e-02 -2.84631580e-01 -3.18883896e-01 -9.39149112e-02
1.29543185e+00 4.79649574e-01 6.45641625e-01 -8.46655071e-01
-1.20653056e-01 -6.84051692e-01 -3.87421429e-01 -9.69153166e-01
-4.47023436e-02 -7.18067825e-01 1.50674190e-02 -1.19057822e+00
-5.98430969e-02 -1.30674079e-01 -2.58678347e-01 -1.25515819e-01
-2.58546263e-01 -1.82604626e-01 -4.24153507e-02 -1.08472243e-01
-4.59970385e-01 5.46029806e-01 1.42717469e+00 -1.14313200e-01
-3.57120305e-01 2.27775633e-01 -3.08226049e-01 4.00824398e-01
8.52862179e-01 -4.18041945e-01 -6.11706614e-01 -1.44049481e-01
1.73104536e-02 -1.55344397e-01 -1.86579332e-01 -8.92285705e-01
1.48528829e-01 -4.19446945e-01 -1.61788806e-01 -8.45036983e-01
-5.73225766e-02 -6.12717271e-01 -2.34959740e-02 5.81443548e-01
-4.35841322e-01 1.41404971e-01 -1.26447022e-01 8.38529050e-01
-6.65349364e-02 -2.40953788e-01 8.66838992e-01 6.23762943e-02
-2.79405445e-01 4.75630015e-01 -2.76274532e-01 3.59635621e-01
1.24058926e+00 -4.21547204e-01 -9.52356402e-03 -6.00177288e-01
-7.19125986e-01 4.44744080e-01 -1.53801948e-01 1.58441365e-01
4.24250484e-01 -1.58687449e+00 -7.05442429e-01 -8.68681818e-02
-2.41774768e-01 -3.83698553e-01 4.38398123e-01 1.54193151e+00
-4.12671626e-01 4.37444121e-01 4.23030183e-02 -6.83897614e-01
-1.35627806e+00 5.21383762e-01 1.43734440e-01 -9.80093479e-02
-5.06095290e-01 1.06754267e+00 -3.46120924e-01 -1.39888749e-01
1.86266527e-01 -7.38744497e-01 3.84020984e-01 3.55295241e-01
9.32110429e-01 1.31207025e+00 4.67916876e-02 -1.58119813e-01
-1.91248998e-01 7.83419013e-01 -1.60484314e-02 -2.72909552e-01
1.68329680e+00 -2.85245895e-01 -3.64292592e-01 6.25211239e-01
1.56440616e+00 -1.77726001e-01 -1.25614977e+00 -2.23861262e-01
1.69746414e-01 -9.64337707e-01 -8.63441378e-02 1.01797488e-02
-1.13197327e+00 5.91076553e-01 6.37266517e-01 4.98992920e-01
1.39282346e+00 -4.48843479e-01 9.38540220e-01 6.46427214e-01
7.58557677e-01 -1.57019007e+00 -1.66225940e-01 3.95449132e-01
1.11332083e+00 -7.59563863e-01 2.32682496e-01 -2.01099381e-01
-3.65651011e-01 1.18588400e+00 6.17101556e-03 -7.14577138e-01
9.90941167e-01 2.00711980e-01 -2.10626662e-01 2.10974336e-01
-5.77487171e-01 -1.29152194e-01 1.32726267e-01 3.00554693e-01
1.05726823e-01 1.89688325e-01 -1.12553835e+00 1.44232392e-01
-3.13520104e-01 8.78592059e-02 4.57331419e-01 9.29945409e-01
-4.40570235e-01 -9.59098220e-01 -5.21940470e-01 5.09844780e-01
-2.58894086e-01 2.22522467e-01 -1.61518559e-01 5.16330838e-01
-5.87703705e-01 7.93323636e-01 -2.02345550e-02 -1.61473513e-01
3.25900823e-01 7.44734555e-02 8.71111035e-01 1.88984662e-01
-1.10814020e-01 1.30084425e-01 -1.87705338e-01 -9.68208730e-01
-8.78987312e-01 -1.24146724e+00 -8.49662185e-01 -7.76948929e-01
-3.39545608e-01 1.77782312e-01 5.70329726e-01 9.80385423e-01
3.51640552e-01 2.00404570e-01 1.02597225e+00 -8.55391264e-01
-9.05767143e-01 -8.18769872e-01 -7.89183617e-01 1.69932768e-01
2.70354539e-01 -4.24601793e-01 -6.62060380e-01 -1.00762704e-02] | [6.959786891937256, 4.345320701599121] |
4766353d-9e51-4bb0-9da6-fe7a93164526 | wsfe-wasserstein-sub-graph-feature-encoder | 2305.04410 | null | https://arxiv.org/abs/2305.04410v1 | https://arxiv.org/pdf/2305.04410v1.pdf | WSFE: Wasserstein Sub-graph Feature Encoder for Effective User Segmentation in Collaborative Filtering | Maximizing the user-item engagement based on vectorized embeddings is a standard procedure of recent recommender models. Despite the superior performance for item recommendations, these methods however implicitly deprioritize the modeling of user-wise similarity in the embedding space; consequently, identifying similar users is underperforming, and additional processing schemes are usually required otherwise. To avoid thorough model re-training, we propose WSFE, a model-agnostic and training-free representation encoder, to be flexibly employed on the fly for effective user segmentation. Underpinned by the optimal transport theory, the encoded representations from WSFE present a matched user-wise similarity/distance measurement between the realistic and embedding space. We incorporate WSFE into six state-of-the-art recommender models and conduct extensive experiments on six real-world datasets. The empirical analyses well demonstrate the superiority and generality of WSFE to fuel multiple downstream tasks with diverse underlying targets in recommendation. | ['Irwin King', 'Chen Ma', 'Zixing Song', 'Menglin Yang', 'Yifei Zhang', 'Yankai Chen'] | 2023-05-08 | null | null | null | null | ['collaborative-filtering'] | ['miscellaneous'] | [ 3.20238657e-02 -2.53351867e-01 -5.75404823e-01 -5.23651302e-01
-4.36350733e-01 -8.06120336e-01 5.85133851e-01 -1.90994099e-01
-2.47369528e-01 1.95999190e-01 4.35542643e-01 -5.54388821e-01
-6.83280349e-01 -7.32911289e-01 -4.24256116e-01 -5.46209395e-01
-1.58519089e-01 3.84451598e-01 -2.13414043e-01 -2.87661016e-01
2.56098896e-01 3.78899932e-01 -1.35616660e+00 2.92160898e-01
9.08075690e-01 1.18622065e+00 2.32441187e-01 6.89533293e-01
-1.01736650e-01 1.60359398e-01 -1.18830353e-01 -5.56744158e-01
5.85529089e-01 -2.76608109e-01 -6.30388856e-01 -3.29091810e-02
4.37722802e-01 -4.09691066e-01 -7.53123045e-01 7.74276137e-01
5.21454811e-01 6.74025476e-01 9.42673206e-01 -1.09839499e+00
-1.51241338e+00 7.55114198e-01 -5.74401803e-02 3.26999307e-01
5.34428880e-02 -1.75350979e-01 1.71897340e+00 -1.14910591e+00
1.45054445e-01 8.74245644e-01 5.07966101e-01 5.33152342e-01
-1.14283085e+00 -4.41855013e-01 7.08151579e-01 -1.19793033e-02
-1.25329804e+00 -3.78565699e-01 4.71433878e-01 -5.30673504e-01
8.84598076e-01 5.35005152e-01 6.09135985e-01 1.25933921e+00
-2.08566785e-01 7.75736690e-01 4.92625147e-01 2.57738720e-04
1.12407520e-01 2.59264916e-01 3.49055886e-01 5.55219293e-01
2.81285644e-01 3.38201493e-01 -3.77310157e-01 -1.89045295e-01
9.35332179e-01 5.12663901e-01 -2.97098190e-01 -6.27996027e-01
-1.08739984e+00 1.05498242e+00 5.13243675e-01 6.21849624e-03
-3.50792140e-01 1.82335690e-01 1.44276097e-01 2.89322257e-01
6.42760932e-01 8.37763548e-01 -4.28820997e-01 -1.49853677e-02
-1.00166059e+00 3.47896993e-01 5.84092081e-01 1.00036836e+00
2.86591351e-01 3.68899740e-02 -4.42782789e-01 8.93116891e-01
7.56363392e-01 1.27852291e-01 3.26204151e-01 -9.27320778e-01
2.68077999e-01 5.35053194e-01 3.34544957e-01 -9.84798491e-01
-1.35984689e-01 -9.25168931e-01 -5.59361398e-01 -2.47366890e-01
3.80501777e-01 -9.67183337e-02 -5.26261747e-01 1.51612437e+00
3.43876749e-01 6.24985278e-01 -1.19901046e-01 1.12505221e+00
5.82717299e-01 6.60881460e-01 5.15411310e-02 6.14118204e-02
1.17807829e+00 -1.36629438e+00 -3.44416916e-01 4.49645892e-03
6.18236244e-01 -5.67101121e-01 1.22965765e+00 4.08013493e-01
-8.64673555e-01 -7.05519497e-01 -9.72391248e-01 -7.99127761e-03
-3.47583324e-01 2.12313250e-01 1.02941751e+00 7.91152120e-01
-5.76040149e-01 8.26303899e-01 -5.31214476e-01 -1.53059229e-01
4.97494161e-01 5.97114861e-01 6.52085096e-02 -8.19611698e-02
-1.30416751e+00 6.39393449e-01 -2.92894617e-02 3.47127706e-01
-7.46233284e-01 -1.12912714e+00 -5.16477942e-01 4.04711992e-01
1.44139126e-01 -6.96717441e-01 1.19371831e+00 -7.14862406e-01
-1.62451208e+00 2.35636279e-01 1.61080360e-01 -2.76431143e-01
1.33387625e-01 -3.56355458e-01 -6.00004017e-01 -3.60223651e-01
-4.81346130e-01 3.44783098e-01 7.53460348e-01 -1.03048420e+00
-6.25505269e-01 -1.12658225e-01 4.75565612e-01 4.11752820e-01
-9.36425805e-01 -1.92568768e-02 -4.93798345e-01 -7.25201964e-01
-2.58711666e-01 -8.42458367e-01 -4.14454430e-01 1.55879855e-02
-1.92039922e-01 -2.47597575e-01 4.28640902e-01 -3.18565458e-01
1.67040491e+00 -1.96810555e+00 2.34413609e-01 3.88515383e-01
3.97264630e-01 3.26474309e-01 -4.26432014e-01 3.12414169e-01
1.94746733e-01 8.45063403e-02 1.05997287e-01 -2.25777611e-01
4.23783392e-01 1.18527770e-01 -5.08797348e-01 6.28896534e-01
-1.47050560e-01 9.46808457e-01 -1.04559147e+00 -7.51068369e-02
2.60775477e-01 5.51517963e-01 -1.19240832e+00 4.99874860e-01
-1.76667064e-01 1.39242858e-01 -7.32960463e-01 4.03057873e-01
5.31927109e-01 -3.68669957e-01 8.84964317e-02 -4.31843668e-01
-7.05490038e-02 4.78559792e-01 -1.07770026e+00 1.69819117e+00
-6.27204716e-01 2.50359684e-01 -8.43265131e-02 -1.03304708e+00
6.77611172e-01 3.28404494e-02 6.98547661e-01 -5.16479313e-01
2.13157892e-01 -1.90246120e-01 2.62490362e-01 -3.27625424e-01
8.17930162e-01 2.96581030e-01 -3.66843268e-02 8.48208666e-01
1.00105546e-01 5.58054686e-01 -1.20749004e-01 2.21568614e-01
8.46469939e-01 2.43159011e-01 -1.28742963e-01 -3.62515271e-01
2.08489627e-01 -3.59022498e-01 3.17636460e-01 7.69331753e-01
-1.39566526e-01 6.29975200e-01 -2.03126092e-02 -2.36565217e-01
-9.01856124e-01 -1.12036276e+00 -2.86730886e-01 1.63685930e+00
2.41771236e-01 -5.09817064e-01 -5.29551148e-01 -1.00899494e+00
2.76985496e-01 8.26991081e-01 -6.54644608e-01 -3.77118140e-01
-3.15013051e-01 -9.37139332e-01 2.55026639e-01 7.11742461e-01
-2.33202949e-01 -5.83301306e-01 -4.82768826e-02 3.76146197e-01
1.95063263e-01 -7.38134027e-01 -1.08584642e+00 -1.51816886e-02
-7.06936955e-01 -9.01246846e-01 -6.79240346e-01 -5.14166832e-01
6.97567046e-01 7.23142803e-01 9.38242793e-01 9.00742486e-02
-2.04230733e-02 5.71430624e-01 -4.50363487e-01 4.38876860e-02
2.17133150e-01 2.04980001e-01 3.45780462e-01 1.22627445e-01
4.31695163e-01 -2.97912985e-01 -1.14022350e+00 6.41960084e-01
-6.10120177e-01 -2.59503841e-01 3.42904836e-01 9.07223821e-01
5.31551600e-01 -2.52025396e-01 7.42768943e-01 -1.12228119e+00
9.29887950e-01 -8.35964203e-01 -4.20392692e-01 2.51645058e-01
-1.04188597e+00 -8.95361081e-02 8.87851775e-01 -8.01602900e-01
-9.96368170e-01 -3.64827305e-01 -4.19241488e-01 -3.63871068e-01
9.31124762e-02 3.88022482e-01 -2.09171027e-02 6.40324131e-02
4.87490922e-01 4.66058068e-02 -2.98459440e-01 -7.25360811e-01
9.84623849e-01 9.21401024e-01 1.62468970e-01 -5.41499138e-01
5.85005224e-01 1.76308930e-01 -6.98496401e-01 -3.81437391e-01
-1.08821952e+00 -7.31664598e-01 -5.06993711e-01 -2.40124529e-03
5.95211089e-01 -8.54599953e-01 -6.12876832e-01 -2.38598153e-01
-6.96620464e-01 -2.18675852e-01 -4.23028469e-01 7.39085555e-01
-4.23720747e-01 1.62476644e-01 -5.53681195e-01 -7.71641433e-01
-5.32132030e-01 -1.16696846e+00 6.20407343e-01 9.85881835e-02
-3.21691602e-01 -1.27698255e+00 1.91186786e-01 4.37403023e-01
6.64908707e-01 -5.71799874e-01 9.24450219e-01 -1.14978123e+00
-3.79348010e-01 -2.12342158e-01 -3.28824580e-01 3.18410099e-01
1.73763931e-01 9.03053880e-02 -9.96681452e-01 -3.04182738e-01
-5.76417983e-01 1.41680002e-01 7.45114446e-01 3.38151932e-01
1.50832832e+00 -2.53528357e-01 -3.14213514e-01 7.90574193e-01
1.09961462e+00 -5.75039014e-02 2.10943446e-01 3.28896418e-02
8.88218343e-01 3.63973320e-01 7.25980222e-01 4.18099105e-01
4.48317915e-01 7.78360963e-01 4.79072005e-01 -7.99954236e-02
-3.23304981e-02 -5.25043428e-01 5.30412972e-01 1.14772022e+00
4.46474813e-02 -6.46567822e-01 -1.23706952e-01 4.16388929e-01
-1.92944467e+00 -1.09285676e+00 -2.10211631e-02 2.20903730e+00
5.00956297e-01 -1.24873549e-01 3.69062901e-01 -3.10160309e-01
4.66138631e-01 2.28926733e-01 -7.20880091e-01 -5.84737420e-01
4.00670469e-01 1.63236246e-01 6.60675347e-01 3.09596062e-01
-9.12967384e-01 8.43325078e-01 7.13623095e+00 7.48228490e-01
-7.99408615e-01 2.39409998e-01 3.62817198e-01 -5.94958425e-01
-7.28445590e-01 -2.36229822e-01 -8.86850178e-01 6.99499309e-01
1.18554068e+00 -3.81204002e-02 7.56476462e-01 8.98098469e-01
2.89483905e-01 6.08464360e-01 -1.28237545e+00 7.19977736e-01
-2.94728484e-02 -1.41845167e+00 2.17200536e-03 2.95805097e-01
7.17273474e-01 6.87911138e-02 5.66482306e-01 7.72376299e-01
4.63321269e-01 -1.09456015e+00 4.47655469e-01 6.44747734e-01
7.35062957e-01 -7.05153346e-01 4.32049125e-01 1.55785993e-01
-1.21984601e+00 -3.90417874e-01 -8.37275982e-01 8.23074207e-02
3.80735099e-01 3.34235400e-01 -3.53051275e-01 3.98626536e-01
4.18261439e-01 8.44842970e-01 -1.77131861e-01 1.06437612e+00
1.49390414e-01 8.94244015e-01 -8.09472129e-02 2.48434413e-02
3.95050049e-01 -6.38104320e-01 1.86341867e-01 1.29266357e+00
6.25692904e-01 8.36322382e-02 1.27092361e-01 7.63523519e-01
-3.05256635e-01 1.92793444e-01 -3.62256467e-01 -4.09148425e-01
5.55100381e-01 1.47739851e+00 -2.52252281e-01 -1.30440831e-01
-7.58113801e-01 9.55540657e-01 3.08028936e-01 4.34056550e-01
-1.16065133e+00 -2.01328933e-01 1.45849979e+00 1.69158697e-01
5.52175403e-01 -2.10277960e-01 -1.89288348e-01 -1.11362255e+00
-5.56216002e-01 -6.78009450e-01 2.81811357e-01 -2.18386248e-01
-1.67771387e+00 4.23014045e-01 -1.67287946e-01 -1.31470299e+00
7.87284151e-02 -8.85782659e-01 -8.62565815e-01 7.93407440e-01
-1.61509573e+00 -9.23451602e-01 1.10287100e-01 4.76211309e-01
6.60866618e-01 -3.56122434e-01 9.40497220e-01 8.27797234e-01
-1.02554262e+00 1.14929652e+00 5.00098705e-01 -1.80288985e-01
6.65173292e-01 -1.22350454e+00 5.60678542e-01 4.85187232e-01
4.00480509e-01 1.06723392e+00 4.44817722e-01 -2.06500888e-01
-1.68413842e+00 -1.34344018e+00 5.36744297e-01 -6.84146821e-01
7.87694573e-01 -4.99279171e-01 -7.04398215e-01 6.84009314e-01
6.66166246e-02 1.57495171e-01 1.37065899e+00 6.20124817e-01
-4.23861414e-01 -4.47642542e-02 -8.26116025e-01 6.73185706e-01
1.53510988e+00 -5.73274791e-01 -3.80842209e-01 3.28690886e-01
7.77602732e-01 -9.27755162e-02 -1.24675786e+00 -2.16703154e-02
9.05581057e-01 -5.55221558e-01 1.40129030e+00 -1.27417815e+00
8.91653448e-02 -1.94819406e-01 -3.82134020e-01 -1.44758618e+00
-9.30836439e-01 -5.03652871e-01 -8.09413016e-01 1.10315299e+00
6.16738617e-01 -4.06406313e-01 7.81516552e-01 7.41856575e-01
-3.47621024e-01 -1.10240436e+00 -3.58269632e-01 -6.48276985e-01
1.87525123e-01 -4.44224089e-01 9.15873289e-01 8.46356630e-01
8.16206038e-02 3.20414662e-01 -6.01952255e-01 2.83415645e-01
4.39650804e-01 3.55703890e-01 4.10896778e-01 -1.36485779e+00
-6.33381009e-01 -6.00616395e-01 1.24352060e-01 -1.78330600e+00
1.19239748e-01 -1.20162284e+00 -1.34667635e-01 -1.62509918e+00
-1.52044436e-02 -7.87702739e-01 -1.09777963e+00 4.59147096e-02
-8.77466723e-02 2.13039890e-01 1.37075931e-02 1.54243693e-01
-8.47925246e-01 6.87959313e-01 1.35576332e+00 3.20008025e-02
-1.09788328e-01 2.85733879e-01 -1.36012435e+00 5.19284368e-01
5.90901971e-01 -4.60399956e-01 -8.44419181e-01 -4.98858601e-01
3.82749259e-01 -4.19860303e-01 -8.18479359e-02 -3.17618668e-01
7.12825730e-02 -4.70385760e-01 4.01001096e-01 -1.85985774e-01
3.86499941e-01 -8.62691581e-01 -5.26245385e-02 -1.15609303e-01
-7.94922173e-01 -2.20370553e-02 -3.71814162e-01 8.67004335e-01
3.48397225e-01 -2.77401328e-01 5.16823947e-01 2.57532358e-01
-6.25484884e-01 1.04348302e+00 -2.17860848e-01 -1.26413172e-02
8.92787218e-01 -8.71926397e-02 -1.33096114e-01 -1.40127748e-01
-5.32812476e-01 2.35367343e-01 2.12968037e-01 5.67961574e-01
6.26878560e-01 -1.52640581e+00 -4.06452745e-01 2.62010932e-01
3.37067455e-01 -5.86508095e-01 3.42953950e-01 7.88785875e-01
1.85931846e-02 3.97734910e-01 1.15881592e-01 -2.64286846e-02
-7.38419890e-01 6.74579859e-01 3.48010004e-01 -1.79453894e-01
-4.80774075e-01 1.06940508e+00 1.82754979e-01 -7.57707179e-01
2.43985638e-01 -3.19237560e-01 -4.42563713e-01 1.81772918e-01
5.94416916e-01 5.30042231e-01 -1.18292786e-01 -5.54464042e-01
-2.66837422e-02 3.01621050e-01 -3.24815542e-01 2.26681337e-01
1.29300559e+00 -3.52662683e-01 5.88141084e-01 8.45570564e-02
1.29826522e+00 2.46512070e-02 -1.47283399e+00 -3.80982012e-01
-2.67389327e-01 -6.93882704e-01 5.06161153e-01 -5.87176144e-01
-1.08438241e+00 9.64423835e-01 6.19375408e-01 3.42431247e-01
5.60596287e-01 -2.36891955e-01 1.14846432e+00 2.44971663e-01
8.71090665e-02 -1.17026114e+00 -8.93972442e-02 4.26590204e-01
4.68887568e-01 -1.09314239e+00 -3.50935869e-02 -8.21703523e-02
-8.19631219e-01 8.65782142e-01 6.20605290e-01 -2.72897601e-01
9.13120508e-01 -6.97259307e-02 -1.58354759e-01 -7.72445947e-02
-8.84392202e-01 -2.46435508e-01 7.42760599e-01 4.63332832e-01
7.10812747e-01 1.67331725e-01 -1.41997889e-01 1.23589551e+00
8.80673006e-02 -3.91944647e-01 1.22349545e-01 2.95868725e-01
-5.02696693e-01 -1.24115574e+00 2.57426798e-01 9.19529140e-01
-2.80309886e-01 -1.92071795e-01 4.33543399e-02 3.82526487e-01
4.45614271e-02 9.48945582e-01 1.05119765e-01 -6.52488232e-01
4.30315882e-01 -5.57649396e-02 4.70105141e-01 -8.39289904e-01
-7.38168120e-01 -4.54866067e-02 3.61383371e-02 -5.47071278e-01
4.15986031e-02 -4.07390684e-01 -1.22831821e+00 -4.47675377e-01
-6.56568587e-01 2.31001958e-01 5.47241390e-01 8.48676980e-01
7.08624005e-01 6.88862443e-01 8.99727046e-01 -8.80219817e-01
-1.15989923e+00 -6.66700840e-01 -6.99288785e-01 3.71800512e-01
8.68263021e-02 -9.42939103e-01 -2.74034083e-01 -4.41749036e-01] | [10.084259033203125, 5.622222423553467] |
84db4700-bcb7-4065-b0ba-84a475118988 | transformer-based-entity-typing-in-knowledge | 2210.11151 | null | https://arxiv.org/abs/2210.11151v1 | https://arxiv.org/pdf/2210.11151v1.pdf | Transformer-based Entity Typing in Knowledge Graphs | We investigate the knowledge graph entity typing task which aims at inferring plausible entity types. In this paper, we propose a novel Transformer-based Entity Typing (TET) approach, effectively encoding the content of neighbors of an entity. More precisely, TET is composed of three different mechanisms: a local transformer allowing to infer missing types of an entity by independently encoding the information provided by each of its neighbors; a global transformer aggregating the information of all neighbors of an entity into a single long sequence to reason about more complex entity types; and a context transformer integrating neighbors content based on their contribution to the type inference through information exchange between neighbor pairs. Furthermore, TET uses information about class membership of types to semantically strengthen the representation of an entity. Experiments on two real-world datasets demonstrate the superior performance of TET compared to the state-of-the-art. | ['Jeff Z. Pan', 'Ru Li', 'Zhiliang Xiang', 'Víctor Gutiérrez-Basulto', 'Zhiwei Hu'] | 2022-10-20 | null | null | null | null | ['entity-typing'] | ['natural-language-processing'] | [ 2.54778732e-02 4.87502337e-01 -2.62441963e-01 -4.23899055e-01
-3.92254740e-01 -7.59170711e-01 5.07206678e-01 8.90951931e-01
-3.37892711e-01 9.55881596e-01 2.17798784e-01 -3.17685217e-01
-9.42756236e-02 -1.61647797e+00 -1.11619437e+00 -4.53099668e-01
-8.96399096e-02 5.11217356e-01 5.92222631e-01 -2.10592240e-01
-1.40670657e-01 3.01933754e-02 -1.71466637e+00 5.07730603e-01
9.50638533e-01 1.05009449e+00 2.56513432e-03 4.20501120e-02
-6.24333441e-01 8.37363720e-01 -5.37300766e-01 -9.15175796e-01
-1.29846543e-01 -7.15693831e-03 -9.78713989e-01 -6.28730714e-01
3.09003919e-01 5.88920154e-02 7.34162405e-02 9.93380249e-01
1.70251638e-01 -3.50239612e-02 4.89354759e-01 -1.31177831e+00
-6.21603906e-01 1.00776803e+00 -1.62048012e-01 -1.77962910e-02
8.22701991e-01 -4.13090676e-01 1.36725080e+00 -7.11764514e-01
9.75868464e-01 1.12292588e+00 9.29774404e-01 3.86691183e-01
-1.13950992e+00 -3.44492286e-01 1.07914090e-01 4.40618247e-01
-1.47846198e+00 -1.92175865e-01 5.55823505e-01 -4.54540938e-01
8.54790688e-01 3.76165569e-01 5.23559093e-01 5.95918000e-01
-3.66717696e-01 5.11733651e-01 7.93148100e-01 -3.59305263e-01
3.30221206e-01 2.43670776e-01 5.96725643e-01 9.48601544e-01
7.46207595e-01 -2.55255908e-01 -3.42364490e-01 -6.33110106e-01
2.84029126e-01 -4.12687302e-01 -3.37355882e-02 -3.53371114e-01
-1.12688589e+00 4.08759952e-01 7.11548448e-01 3.53750437e-01
-3.74014199e-01 6.62118867e-02 2.94359177e-01 3.04892659e-02
2.01744795e-01 3.64943445e-01 -6.94873273e-01 1.21118158e-01
-2.47764811e-01 3.95890445e-01 1.01190555e+00 1.36119163e+00
1.20176029e+00 -5.49789369e-01 -2.22061247e-01 6.98535502e-01
3.20296407e-01 2.32370526e-01 -3.74424793e-02 -5.59873164e-01
3.83520246e-01 1.29704106e+00 2.21486047e-01 -8.84491324e-01
-2.85654157e-01 -2.48621583e-01 -6.64727390e-01 -2.02111810e-01
5.46113908e-01 -2.27192923e-01 -3.06781888e-01 2.08687830e+00
9.43148434e-01 6.82695150e-01 1.83939531e-01 4.46228921e-01
9.81527925e-01 3.30967009e-01 4.71341871e-02 5.02229668e-02
1.49996805e+00 -3.76011282e-01 -5.76038301e-01 6.76961541e-02
8.44358563e-01 -1.16647243e-01 3.94483328e-01 -1.39536232e-01
-8.71930242e-01 -4.15555686e-01 -8.44080508e-01 -1.71592295e-01
-8.90457392e-01 -2.03754663e-01 6.03530824e-01 6.30153358e-01
-6.84163928e-01 4.17506069e-01 -4.69829768e-01 -3.75117548e-02
4.43656370e-02 1.88195705e-01 -3.89398634e-01 -3.16507071e-02
-1.62896693e+00 4.76124853e-01 5.92603207e-01 1.90261573e-01
-1.74008086e-01 -1.14720380e+00 -1.11497605e+00 1.38591915e-01
8.59368324e-01 -1.30568910e+00 9.20331895e-01 -5.76081812e-01
-6.50282383e-01 5.56226432e-01 -7.29850650e-01 -3.78983706e-01
1.54359609e-01 1.61713697e-02 -5.51522493e-01 -1.94901958e-01
2.13643789e-01 1.23242028e-01 1.80522650e-01 -1.31968713e+00
-1.04189932e+00 -6.03849471e-01 6.02018654e-01 -7.25530460e-02
6.29998073e-02 -2.51085430e-01 -5.55946469e-01 -5.53654253e-01
-4.42544930e-02 -7.48009026e-01 9.96560231e-02 -1.41929820e-01
-5.75124800e-01 -7.11121082e-01 4.89102811e-01 -5.44893801e-01
1.66553915e+00 -1.94522047e+00 3.81505072e-01 4.77452427e-01
5.16405761e-01 1.00342587e-01 2.16335714e-01 5.59324086e-01
1.06835261e-01 4.73077238e-01 -3.69779229e-01 -5.48103452e-03
2.99582690e-01 5.63847840e-01 -5.30221350e-02 -4.01796736e-02
9.38883796e-02 8.78280997e-01 -1.17956972e+00 -3.59858334e-01
-2.72912264e-01 3.53549957e-01 -7.05245197e-01 1.82982042e-01
-5.87212741e-01 -3.19684595e-02 -6.70961261e-01 4.74953353e-01
7.11168349e-01 -1.30281135e-01 6.08121455e-01 -5.73112905e-01
-1.06797546e-01 6.04149938e-01 -1.57805359e+00 1.33221817e+00
-3.65745664e-01 -2.42648885e-01 3.97204459e-02 -7.58002400e-01
6.97504342e-01 2.04768896e-01 2.05958322e-01 -3.71278554e-01
-3.84161085e-01 2.03180388e-01 -3.68647724e-01 -4.95008647e-01
6.53804004e-01 1.14028975e-01 -4.18986142e-01 1.84236437e-01
-1.42447889e-01 5.43411255e-01 3.97425145e-01 4.21111792e-01
1.01859331e+00 9.00412202e-02 2.65757233e-01 -1.53395906e-01
7.79867053e-01 -2.71196514e-01 1.24877357e+00 8.69275570e-01
5.03457427e-01 -1.28703028e-01 7.60908246e-01 -4.00919408e-01
-8.31932187e-01 -1.05849028e+00 -4.10511158e-02 8.41654480e-01
5.83189309e-01 -9.48790073e-01 -6.57173395e-01 -1.04950595e+00
3.44303697e-01 6.76190913e-01 -8.49350214e-01 1.06092565e-01
-6.48054540e-01 -5.40780485e-01 8.66650581e-01 7.63158083e-01
5.48957586e-01 -6.48935080e-01 -2.76497960e-01 1.41831264e-01
-8.02798271e-01 -1.20195270e+00 -1.91918150e-01 -2.32921496e-01
-2.93645680e-01 -1.29104042e+00 -8.34273454e-03 -6.54922485e-01
7.21035004e-01 -3.95336390e-01 1.11442459e+00 5.36491334e-01
2.60600507e-01 -5.07083982e-02 -4.76400197e-01 -7.11350590e-02
-4.25768524e-01 1.12803407e-01 -1.56697541e-01 2.08241999e-01
2.20013127e-01 -3.89011145e-01 -1.24419443e-01 3.74128073e-01
-9.38710332e-01 1.23359300e-01 4.92362320e-01 7.18934596e-01
5.95346570e-01 3.84505510e-01 2.63528109e-01 -1.62246752e+00
1.21802948e-01 -9.02317345e-01 -5.41431010e-01 7.22808719e-01
-6.10685170e-01 5.52129805e-01 5.86176813e-01 -1.62988141e-01
-1.46464264e+00 -2.41790175e-01 -9.02882144e-02 3.31011504e-01
-1.98167771e-01 1.04207611e+00 -7.06700087e-01 3.57680887e-01
2.68777817e-01 1.26810953e-01 -6.32682741e-01 -6.76534891e-01
4.27251726e-01 3.70107859e-01 9.15221274e-01 -1.31043160e+00
7.58542180e-01 1.51966497e-01 2.27948233e-01 -6.06524110e-01
-6.56097174e-01 -3.93992990e-01 -7.51488566e-01 3.26251723e-02
6.67622447e-01 -6.63629115e-01 -6.33869529e-01 6.93607867e-01
-1.15444624e+00 -2.70544559e-01 -1.32446751e-01 1.52998595e-02
-1.01448871e-01 4.06828582e-01 -5.37756085e-01 -4.74385053e-01
6.64014295e-02 -8.82011652e-01 1.07242155e+00 -1.20245842e-02
-7.13123828e-02 -1.14169466e+00 7.93321654e-02 1.68197304e-01
-1.94656774e-02 3.36997241e-01 1.61097240e+00 -7.84652114e-01
-7.06657827e-01 -4.11019288e-02 -3.64000738e-01 -2.88131058e-01
1.29517661e-02 -1.53818466e-02 -6.08083665e-01 2.05675319e-01
-7.00942039e-01 2.77808011e-01 3.78968090e-01 -2.95149148e-01
6.72657669e-01 -8.58572423e-01 -7.13463724e-01 5.72888494e-01
1.61349010e+00 -2.53240187e-02 9.15387630e-01 2.62596190e-01
9.76294994e-01 4.54188377e-01 5.62151074e-01 3.68434399e-01
9.39595520e-01 1.04063809e+00 1.83521897e-01 7.23704994e-02
-4.32532951e-02 -6.09201491e-01 -1.74490437e-01 3.44323903e-01
-1.49082631e-01 -3.78417075e-01 -8.20658922e-01 4.69248623e-01
-1.96754611e+00 -1.19065344e+00 -5.30204535e-01 2.41640067e+00
1.05251575e+00 -1.06935553e-01 2.77341276e-01 7.59461820e-02
8.50754023e-01 -1.83292717e-01 -2.59922147e-01 -3.94464694e-02
-4.21923772e-02 -1.09375693e-01 4.74791825e-01 6.43480241e-01
-7.04135716e-01 7.02042401e-01 5.43239737e+00 6.33360922e-01
-4.36195731e-01 -1.33541390e-01 1.16265036e-01 8.31907690e-01
-1.03615212e+00 4.91389215e-01 -1.08477628e+00 7.15388358e-01
3.92915010e-01 -4.46958035e-01 3.72632712e-01 4.35828239e-01
-4.11147892e-01 -2.16190279e-01 -1.20405436e+00 3.25335711e-01
-2.93440700e-01 -1.44711328e+00 3.09740216e-01 1.07770950e-01
3.12662870e-01 -3.11548471e-01 -5.05927980e-01 3.24635923e-01
3.77780974e-01 -3.87811750e-01 1.05318880e+00 7.99363136e-01
6.40919089e-01 -5.76575458e-01 8.41050684e-01 3.13548326e-01
-1.69953859e+00 1.01432707e-02 -1.37013540e-01 1.20824747e-01
-1.16536804e-01 8.27117026e-01 -6.13558471e-01 1.34559512e+00
6.36083961e-01 6.22463822e-01 -4.96653587e-01 9.38550830e-01
-4.31184113e-01 2.94083983e-01 -3.76454622e-01 1.75215900e-02
-2.36696318e-01 -1.88477337e-01 6.63617849e-01 1.40958834e+00
1.27525106e-01 4.58351523e-01 1.45864591e-01 1.07222855e+00
-2.62791693e-01 -3.38108808e-01 -4.67801720e-01 3.65336716e-01
1.17725158e+00 1.25248027e+00 -1.18739255e-01 -7.90411711e-01
-5.75201929e-01 6.58368230e-01 6.72677398e-01 3.83650482e-01
-7.65459001e-01 -5.17411888e-01 8.46704483e-01 6.04341030e-02
6.32229507e-01 -4.03618775e-02 -1.57344878e-01 -1.19748700e+00
4.78315294e-01 -5.09730577e-01 8.62815976e-01 -8.24640989e-01
-1.21910071e+00 4.56204504e-01 2.99558401e-01 -7.21463680e-01
-3.23504269e-01 -2.73607075e-01 -3.42613131e-01 8.69116008e-01
-1.48033774e+00 -1.22373629e+00 -3.30860525e-01 4.46740896e-01
-1.27495795e-01 4.58763659e-01 9.38451290e-01 3.80506963e-01
-6.10286713e-01 9.02308524e-01 -3.13400239e-01 5.21569431e-01
3.64354163e-01 -1.56516767e+00 4.40528959e-01 1.04949379e+00
-1.83766484e-01 1.10093713e+00 4.99248922e-01 -9.33487356e-01
-1.62053764e+00 -1.16860664e+00 1.41153145e+00 -3.04739535e-01
6.17034554e-01 -3.53751302e-01 -1.30495024e+00 9.17247057e-01
-2.38046169e-01 -9.28378999e-02 6.04429126e-01 4.18791920e-01
-1.12205410e+00 -2.49107555e-01 -1.36463404e+00 3.18952292e-01
1.33445108e+00 -4.97312665e-01 -8.29543471e-01 -4.14436817e-01
7.26409137e-01 -4.22485620e-01 -1.29263926e+00 5.64994574e-01
7.85354555e-01 -9.39637840e-01 8.95470262e-01 -2.82700211e-01
2.09985033e-01 -5.96782327e-01 -1.90345988e-01 -1.33193028e+00
-3.29006851e-01 -3.13871145e-01 -1.96459368e-01 1.81645119e+00
7.28583992e-01 -9.73215580e-01 5.07354498e-01 8.71048689e-01
-2.41400376e-01 -4.72939730e-01 -6.57480896e-01 -7.40086317e-01
-3.21263075e-01 -3.43729019e-01 1.23757720e+00 1.17166853e+00
3.58054340e-01 3.22543323e-01 1.24360852e-01 6.97975814e-01
7.70206571e-01 3.03583473e-01 6.48799241e-01 -1.37221467e+00
-5.14106631e-01 -2.75170654e-01 -5.95002592e-01 -7.84917891e-01
3.09180766e-01 -1.18598568e+00 -1.70511261e-01 -1.58625054e+00
1.77667052e-01 -1.15624964e+00 -1.93902403e-01 8.95179093e-01
-4.54495788e-01 -5.13997003e-02 -1.67843580e-01 -2.52346750e-02
-4.77908075e-01 4.86414321e-02 7.15308189e-01 -2.27543920e-01
1.52891859e-01 -1.99370846e-01 -8.11471343e-01 6.89859629e-01
3.58381540e-01 -4.84453171e-01 -1.61810696e-01 -5.28312027e-01
8.29887807e-01 3.43873441e-01 4.77886319e-01 -6.69127703e-01
4.28191215e-01 -4.70110960e-02 8.80011916e-03 -4.41824883e-01
1.64381623e-01 -7.99271762e-01 7.23189890e-01 2.41080120e-01
-3.38891715e-01 -2.17604078e-02 7.10929930e-02 6.01731539e-01
-1.55485436e-01 -2.22225562e-01 2.63854176e-01 5.80004863e-02
-1.17420542e+00 1.86844200e-01 8.83200541e-02 2.63682932e-01
6.79059744e-01 6.30756989e-02 -7.73163438e-01 1.31941438e-01
-6.77259684e-01 1.09311067e-01 8.23895574e-01 3.61081153e-01
2.97622263e-01 -1.30345523e+00 -5.36851168e-01 3.35775912e-01
4.37831223e-01 5.45645095e-02 -4.25139479e-02 7.14923859e-01
-1.38612390e-02 -9.71509889e-03 1.25732318e-01 -2.33951166e-01
-1.46103883e+00 6.47980750e-01 2.95554191e-01 -2.63629258e-01
-3.72835636e-01 5.99174380e-01 2.92979509e-01 -6.05572283e-01
3.61697488e-02 -3.77083749e-01 -2.51970559e-01 -1.04156239e-02
5.88473141e-01 5.36952257e-01 3.24754804e-01 -7.30205238e-01
-5.90000749e-01 5.99714756e-01 1.07350782e-01 8.48471597e-02
1.12321055e+00 2.36351285e-02 -6.30237043e-01 4.37113702e-01
1.01847529e+00 4.19108927e-01 -5.76792836e-01 -4.95580405e-01
2.74048090e-01 -3.77883941e-01 -4.17111099e-01 -1.11116850e+00
-8.57229412e-01 8.84618536e-02 -1.89865842e-01 2.79367983e-01
9.18338537e-01 2.58099258e-01 7.81303108e-01 3.78879786e-01
7.78966010e-01 -5.30951500e-01 -7.54529893e-01 5.53941786e-01
4.12747383e-01 -5.62952518e-01 -3.26050997e-01 -1.11558986e+00
-2.59208888e-01 7.88349509e-01 3.57002974e-01 4.02918667e-01
3.95265579e-01 4.88189399e-01 -3.66664529e-01 -1.52706861e-01
-8.36942613e-01 -3.33697140e-01 5.02726853e-01 6.24813616e-01
2.08552420e-01 1.87781036e-01 -5.15455544e-01 7.94328809e-01
-9.15354770e-03 1.80785954e-01 3.55379373e-01 8.35666418e-01
-7.78191090e-02 -1.50012922e+00 -6.08546734e-02 3.86324346e-01
-3.83509845e-01 -1.97238520e-01 -4.58281845e-01 7.21317649e-01
6.71467006e-01 8.60524476e-01 7.67877996e-02 -5.16261935e-01
4.42841560e-01 4.10203561e-02 5.76151490e-01 -5.88841736e-01
-6.85915828e-01 -4.39119130e-01 6.89271867e-01 -4.35156137e-01
-3.77026111e-01 -7.11802185e-01 -1.46944213e+00 -3.36218536e-01
-3.15609306e-01 4.72357631e-01 3.62076372e-01 1.08018804e+00
6.62072659e-01 3.56775254e-01 5.15092373e-01 -3.07407789e-02
-2.11291779e-02 -5.09683907e-01 -5.19752204e-01 8.61364424e-01
2.52585799e-01 -9.17293429e-01 -2.52124131e-01 2.79685799e-02] | [9.4928617477417, 8.623994827270508] |
cb4dcf90-af61-408b-8df7-c54ce69af01b | few-shot-3d-multi-modal-medical-image | 1810.12241 | null | http://arxiv.org/abs/1810.12241v1 | http://arxiv.org/pdf/1810.12241v1.pdf | Few-shot 3D Multi-modal Medical Image Segmentation using Generative Adversarial Learning | We address the problem of segmenting 3D multi-modal medical images in
scenarios where very few labeled examples are available for training.
Leveraging the recent success of adversarial learning for semi-supervised
segmentation, we propose a novel method based on Generative Adversarial
Networks (GANs) to train a segmentation model with both labeled and unlabeled
images. The proposed method prevents over-fitting by learning to discriminate
between true and fake patches obtained by a generator network. Our work extends
current adversarial learning approaches, which focus on 2D single-modality
images, to the more challenging context of 3D volumes of multiple modalities.
The proposed method is evaluated on the problem of segmenting brain MRI from
the iSEG-2017 and MRBrainS 2013 datasets. Significant performance improvement
is reported, compared to state-of-art segmentation networks trained in a
fully-supervised manner. In addition, our work presents a comprehensive
analysis of different GAN architectures for semi-supervised segmentation,
showing recent techniques like feature matching to yield a higher performance
than conventional adversarial training approaches. Our code is publicly
available at https://github.com/arnab39/FewShot_GAN-Unet3D | ['Christian Desrosiers', 'Arnab Kumar Mondal', 'Jose Dolz'] | 2018-10-29 | null | null | null | null | ['3d-medical-imaging-segmentation', 'brain-image-segmentation'] | ['medical', 'medical'] | [ 6.46286666e-01 7.20876276e-01 -6.18702266e-03 -5.18070638e-01
-1.35422480e+00 -6.43554032e-01 4.54901278e-01 -3.91869098e-01
-3.51892769e-01 7.08770573e-01 -9.03345719e-02 -3.36945057e-01
4.87350851e-01 -6.83756232e-01 -9.60390627e-01 -6.79603338e-01
1.13113865e-01 8.61093283e-01 -2.30348334e-02 1.28966440e-02
-3.21273178e-01 4.61072594e-01 -7.11486459e-01 2.60428309e-01
8.37835670e-01 9.59466159e-01 -2.70569712e-01 6.76955163e-01
1.48607507e-01 6.88618958e-01 -6.85567141e-01 -5.15219033e-01
6.21189952e-01 -7.55615830e-01 -1.09288418e+00 2.28815109e-01
6.23730302e-01 -4.90326613e-01 -3.76103908e-01 1.10063004e+00
6.51041865e-01 -3.01973403e-01 7.50137568e-01 -1.31426358e+00
-7.98622310e-01 6.11760199e-01 -4.00733620e-01 1.19336598e-01
1.00910656e-01 1.97626531e-01 4.73467827e-01 -4.71503943e-01
7.54018247e-01 9.00062799e-01 7.41056442e-01 1.05866456e+00
-1.34569001e+00 -6.55663908e-01 -1.85788915e-01 -2.16228262e-01
-1.14653981e+00 -1.99292853e-01 8.93309355e-01 -6.56083286e-01
6.07600808e-01 1.44876331e-01 5.14069617e-01 1.35562885e+00
1.79296821e-01 9.98491585e-01 1.63256383e+00 -2.75385261e-01
1.32156968e-01 -3.50930132e-02 -1.29604548e-01 7.11429060e-01
-9.95762795e-02 3.79525959e-01 5.50077185e-02 -1.83597147e-01
1.04407370e+00 -2.24227607e-02 -2.13547096e-01 -4.80641186e-01
-1.32464981e+00 1.10514343e+00 7.88685083e-01 3.33611488e-01
-4.95285302e-01 1.26404747e-01 3.97083312e-01 3.16069275e-01
7.99458206e-01 3.16715300e-01 -1.04496524e-01 2.42874846e-01
-1.32701159e+00 8.56382176e-02 6.26095414e-01 9.16013479e-01
4.30649072e-01 2.90096223e-01 -3.44660431e-01 6.36584938e-01
1.56894788e-01 5.68398178e-01 5.92209101e-01 -8.31414521e-01
2.61043191e-01 3.07814151e-01 -3.31390381e-01 -4.57569808e-01
-3.99682701e-01 -2.88938433e-01 -1.01743877e+00 3.67489219e-01
5.56956708e-01 -3.64086151e-01 -1.56682003e+00 1.67501807e+00
4.07212526e-01 2.83492267e-01 3.21756564e-02 8.11374128e-01
9.97047067e-01 1.27766699e-01 1.13924304e-02 1.61041811e-01
1.08730173e+00 -1.18811345e+00 -5.22125125e-01 -3.89991432e-01
2.57228374e-01 -7.31667340e-01 6.43145859e-01 1.86749592e-01
-1.15597165e+00 -3.50523621e-01 -9.21104193e-01 3.60248946e-02
-3.70171577e-01 -2.46475533e-01 5.31871021e-01 9.96110022e-01
-1.19453382e+00 5.07882714e-01 -1.17506337e+00 -1.25053842e-02
1.23292005e+00 3.58237594e-01 -4.00130212e-01 -1.64923981e-01
-1.00902379e+00 8.59826028e-01 2.73882866e-01 2.59381905e-02
-1.27443755e+00 -7.30962753e-01 -8.67700994e-01 -6.44328415e-01
5.95133901e-02 -8.36790800e-01 1.19670296e+00 -1.27400279e+00
-1.25610197e+00 1.49398732e+00 3.20757300e-01 -8.27912688e-01
1.11592436e+00 -2.36489973e-03 -1.89624444e-01 4.27628547e-01
1.29695386e-01 9.10331607e-01 1.06883216e+00 -1.34485555e+00
1.43509254e-01 -4.19675499e-01 -1.20741852e-01 -1.08714722e-01
3.10542613e-01 -1.20612524e-01 -1.60145238e-02 -1.07713366e+00
-5.18528484e-02 -1.13194418e+00 -4.04242694e-01 -1.36386007e-01
-8.88596833e-01 3.44678760e-01 6.79137230e-01 -8.75785232e-01
2.91174382e-01 -1.78781819e+00 2.62160450e-01 1.65131852e-01
3.42538953e-01 3.27146530e-01 -5.98262213e-02 -9.54730138e-02
-3.46561551e-01 1.43505827e-01 -1.02539134e+00 -5.69378555e-01
-1.06094904e-01 3.61753970e-01 -1.05514623e-01 6.69068515e-01
3.57265025e-01 1.43041122e+00 -8.40754867e-01 -5.62089562e-01
2.01107606e-01 6.00375593e-01 -3.88167948e-01 5.25046647e-01
-4.55922373e-02 1.25877047e+00 -3.52829039e-01 9.56828654e-01
8.59752238e-01 -2.39862308e-01 -2.23696649e-01 6.51285201e-02
6.60668075e-01 -2.01662108e-01 -6.19222879e-01 1.97846723e+00
-2.91633010e-01 2.63317555e-01 8.96598324e-02 -1.35699189e+00
7.08608031e-01 4.72138405e-01 6.81057334e-01 -5.17916799e-01
3.72715950e-01 3.63407582e-01 -1.72149986e-01 -1.89624280e-01
-4.30246666e-02 -5.52932382e-01 -1.53194383e-01 5.91115415e-01
4.24709588e-01 -6.75210297e-01 -9.47390422e-02 1.68302029e-01
1.22359216e+00 3.23257506e-01 5.03144264e-02 -1.63490139e-02
3.59830469e-01 8.18297192e-02 3.13993186e-01 8.34111273e-01
-4.99276221e-01 1.08443558e+00 3.84212881e-01 -2.72168845e-01
-1.19780076e+00 -1.22198141e+00 -4.25978571e-01 4.84089792e-01
-1.71109438e-01 2.42744014e-01 -1.29178178e+00 -1.28676534e+00
-3.73071246e-02 5.58526635e-01 -1.05153143e+00 -9.73313525e-02
-6.14000320e-01 -8.00388038e-01 9.20876563e-01 4.79443550e-01
5.37483037e-01 -1.34320247e+00 -4.64976847e-01 7.80813172e-02
-8.29561874e-02 -1.28758252e+00 -4.52930719e-01 1.47947654e-01
-1.03559852e+00 -1.16903222e+00 -1.38250017e+00 -8.50724518e-01
1.04174876e+00 -4.35356587e-01 1.37011981e+00 3.57304402e-02
-4.26933855e-01 6.28823459e-01 -3.59081477e-01 -4.70667273e-01
-1.06185770e+00 1.34587035e-01 -3.25100690e-01 -8.92509744e-02
1.15761824e-01 -5.73280752e-01 -8.01472068e-01 2.38361195e-01
-1.11934638e+00 1.01104893e-01 5.32112062e-01 1.20032465e+00
1.05520999e+00 -5.68185806e-01 6.89543605e-01 -1.63363242e+00
2.49393806e-01 -6.83852077e-01 -2.84165412e-01 1.41042039e-01
-5.98779917e-01 -2.13621885e-01 5.19729555e-01 -3.93371075e-01
-7.28580534e-01 2.50291288e-01 -6.16663277e-01 -6.22768164e-01
-6.24558628e-01 1.24064675e-02 1.70297235e-01 -4.60456908e-01
7.00130701e-01 2.07602888e-01 2.84562886e-01 -3.07872951e-01
4.40589428e-01 3.75685304e-01 6.94933832e-01 -3.58600706e-01
9.84996080e-01 7.39517570e-01 -1.18045546e-02 -2.13354811e-01
-7.85816967e-01 -9.67262983e-02 -9.39728558e-01 -1.18300796e-01
1.13878870e+00 -8.20913255e-01 1.28099620e-01 8.31571043e-01
-9.30524766e-01 -7.02632308e-01 -7.20139623e-01 2.16186658e-01
-1.08305109e+00 2.51942784e-01 -7.80372858e-01 -2.24390209e-01
-7.45364606e-01 -1.44041955e+00 1.16448772e+00 6.31343797e-02
4.54757828e-03 -1.22689879e+00 2.30493575e-01 7.54649460e-01
5.85009515e-01 1.08301222e+00 5.93439758e-01 -1.08186126e+00
-4.90795314e-01 -3.90733778e-01 9.11519825e-02 7.75809348e-01
1.33963853e-01 -6.21149480e-01 -1.09503484e+00 -4.31666434e-01
2.14188844e-01 -7.23111272e-01 8.35967898e-01 4.99974519e-01
1.09010386e+00 -1.80658430e-01 -1.74829766e-01 9.68383789e-01
1.33683026e+00 3.67330573e-02 7.24882424e-01 2.52153575e-01
9.50376272e-01 2.28975266e-01 3.07133943e-01 -4.10912223e-02
1.17075607e-01 3.69771838e-01 6.72597468e-01 -6.73475623e-01
-4.87228304e-01 -1.32121459e-01 -7.82880485e-02 5.51720560e-01
3.97589803e-03 5.19918604e-03 -9.58399653e-01 6.17065489e-01
-1.46583200e+00 -6.95625842e-01 2.09784672e-01 1.92060578e+00
9.84190524e-01 1.17428750e-01 2.71020055e-01 -1.50493374e-02
7.44167268e-01 1.13442741e-01 -8.93651247e-01 -3.53726774e-01
-1.15830466e-01 7.13644624e-01 6.87766969e-01 3.31774771e-01
-1.40874445e+00 8.27090919e-01 6.37239695e+00 6.88600302e-01
-1.12898326e+00 7.72180915e-01 1.06848347e+00 8.83096531e-02
-2.03778297e-01 -3.51279587e-01 -1.45917922e-01 4.22952801e-01
9.68386829e-01 2.19521120e-01 3.51361603e-01 7.06874490e-01
-3.33231270e-01 1.88546717e-01 -1.02103293e+00 7.90884137e-01
2.56616056e-01 -1.30618918e+00 -7.89640918e-02 3.63419615e-02
1.10705316e+00 4.46948171e-01 2.52290159e-01 1.58234909e-01
2.82966822e-01 -1.38794577e+00 6.92576528e-01 5.32630801e-01
1.01977456e+00 -6.42157912e-01 8.07018518e-01 2.16761127e-01
-5.61509728e-01 4.68576312e-01 7.46292621e-02 6.46930337e-01
3.73588443e-01 5.61458170e-01 -9.82138813e-01 7.26550221e-01
4.77905184e-01 6.73257172e-01 -6.30643845e-01 8.17583680e-01
-4.68875259e-01 8.25890481e-01 -1.92981303e-01 6.78619027e-01
2.68902242e-01 -1.63904458e-01 6.48634672e-01 1.08797765e+00
7.02318177e-02 -6.88405856e-02 4.38435413e-02 1.18933678e+00
-3.41813534e-01 -4.66844700e-02 -7.34046876e-01 3.37634496e-02
-3.36387567e-02 1.14548242e+00 -1.00699210e+00 -3.25540900e-01
-3.96905124e-01 1.25531387e+00 1.42549947e-01 1.57494456e-01
-9.63902593e-01 -2.99629252e-02 -1.66914128e-02 1.17012583e-01
3.39131653e-01 6.93490356e-02 -4.56726909e-01 -1.12828434e+00
-1.18718654e-01 -9.70951617e-01 4.78133082e-01 -6.09662652e-01
-1.53387594e+00 1.04290152e+00 -5.77597432e-02 -1.11534190e+00
-4.69653457e-01 -4.06572670e-01 -6.14915311e-01 8.28180969e-01
-1.53640974e+00 -1.72066629e+00 -1.50715694e-01 8.67347956e-01
3.49478751e-01 -3.17606091e-01 1.06176817e+00 3.02058518e-01
-2.53936380e-01 7.68484473e-01 8.62136707e-02 5.04095197e-01
6.95813000e-01 -1.50251031e+00 6.96671665e-01 9.39485550e-01
2.25987226e-01 3.58293243e-02 4.18057770e-01 -6.85040474e-01
-9.43199158e-01 -1.13137734e+00 2.23431394e-01 -6.20051503e-01
4.60652143e-01 -3.38840991e-01 -9.12038624e-01 1.06545758e+00
3.95180702e-01 4.52136934e-01 7.78243124e-01 -5.67709327e-01
-2.92952418e-01 3.78577113e-01 -1.87927878e+00 6.19698130e-02
7.91023433e-01 -4.91714120e-01 -6.84694529e-01 6.33668721e-01
6.73180640e-01 -9.77969587e-01 -1.24325871e+00 5.14762819e-01
1.46971256e-01 -8.16452801e-01 1.14254892e+00 -4.93000090e-01
5.72135687e-01 3.03863548e-02 1.17111191e-01 -1.47760570e+00
2.53879130e-01 -6.58535242e-01 -1.18006021e-01 9.14574265e-01
3.08678508e-01 -7.53047347e-01 9.42685127e-01 5.76772571e-01
-2.16483891e-01 -7.96779811e-01 -1.33104074e+00 -6.22487068e-01
7.30564892e-01 -2.09509850e-01 5.75916469e-01 1.02174973e+00
-6.85490668e-01 -1.29102528e-01 -4.04499233e-01 -1.16874553e-01
1.08235323e+00 1.52104706e-01 5.95125318e-01 -7.26772189e-01
-4.14316237e-01 -2.72214830e-01 -5.58395624e-01 -5.79075098e-01
3.44713956e-01 -1.36021447e+00 -8.00025016e-02 -1.30178905e+00
1.34463832e-01 -4.08247113e-01 -2.66553611e-01 7.22376525e-01
3.81036289e-02 1.00270021e+00 1.52282491e-01 2.39793330e-01
-2.75196522e-01 2.87405044e-01 1.70160151e+00 -3.80912662e-01
2.33233735e-01 3.12473863e-01 -5.20047188e-01 6.32860601e-01
9.04301465e-01 -7.83355713e-01 -1.84661150e-01 -3.85041803e-01
-4.89095747e-01 1.06892034e-01 8.20717514e-01 -8.73355508e-01
-6.06959052e-02 3.44910890e-01 4.77317125e-01 -4.66419786e-01
1.85736790e-01 -8.10093164e-01 1.91656038e-01 5.60554445e-01
-2.80563086e-01 -1.69242665e-01 3.18239629e-01 4.67026979e-01
-2.42035523e-01 -2.14522600e-01 1.12373888e+00 -4.66539323e-01
-3.71408522e-01 5.63533127e-01 -3.96101875e-03 5.70729077e-01
1.12447000e+00 -5.60038537e-02 -2.35926434e-01 -2.05117121e-01
-1.21292043e+00 -2.30577420e-02 5.66659153e-01 2.12614208e-01
6.51213467e-01 -1.32428527e+00 -8.37166250e-01 3.10590744e-01
-1.04530230e-01 3.20430070e-01 4.09513861e-01 9.53151405e-01
-6.19772911e-01 9.17508900e-02 -6.32565916e-01 -7.75415361e-01
-9.11793888e-01 4.84372348e-01 6.88107550e-01 -4.96144712e-01
-7.37581670e-01 8.43550861e-01 1.24149881e-02 -7.15419650e-01
-6.26541674e-02 -7.44903535e-02 2.06348985e-01 -2.30579332e-01
1.84958041e-01 1.08280778e-02 2.97671705e-01 -8.87177885e-01
-3.62758964e-01 3.25216204e-01 -1.38755664e-01 -4.88357432e-02
1.46319532e+00 1.74121186e-01 -3.03371623e-02 1.81081697e-01
1.35465860e+00 -3.05563271e-01 -1.25747144e+00 -3.00523609e-01
-4.21566576e-01 -2.47438475e-01 -1.58441830e-02 -1.08005106e+00
-1.62142694e+00 8.07005942e-01 9.37420964e-01 1.31234437e-01
1.14326465e+00 2.20782980e-01 1.05853188e+00 -2.28962556e-01
3.90002131e-01 -6.96751297e-01 1.67972576e-02 1.57467425e-01
8.75823677e-01 -1.54331017e+00 -2.09060863e-01 -2.17988297e-01
-7.55889714e-01 7.59540319e-01 4.63312417e-01 -5.16033769e-01
6.89251959e-01 3.11480105e-01 3.11037004e-01 -3.12238127e-01
1.83129489e-01 -3.38232443e-02 4.25427794e-01 9.03435707e-01
2.31618345e-01 3.28504652e-01 3.54430191e-02 4.79095519e-01
-2.04403147e-01 3.74128763e-03 4.68528301e-01 1.08575368e+00
3.68529141e-01 -1.37976861e+00 -3.86212766e-01 5.68314493e-01
-9.29998398e-01 -1.52183756e-01 -3.77075166e-01 8.60487759e-01
1.46436557e-01 4.85195309e-01 -6.48190305e-02 -3.33534554e-02
1.81947142e-01 2.00044543e-01 7.46417761e-01 -6.57041311e-01
-8.07051957e-01 9.31137148e-03 -2.43230507e-01 -5.55869281e-01
-8.38812232e-01 -7.10051477e-01 -1.04835320e+00 3.01151890e-02
1.13359634e-02 -1.18885532e-01 6.14995360e-01 8.54974031e-01
1.59839213e-01 6.81458592e-01 5.79329133e-01 -9.97632742e-01
-5.27544796e-01 -9.54352796e-01 -5.39816797e-01 7.97324896e-01
3.31149459e-01 -5.08579135e-01 -3.50392848e-01 1.48728475e-01] | [14.455228805541992, -2.0779223442077637] |
4d0e3946-df3c-4f76-a3fb-8d44d18723e4 | openmask3d-open-vocabulary-3d-instance | 2306.13631 | null | https://arxiv.org/abs/2306.13631v1 | https://arxiv.org/pdf/2306.13631v1.pdf | OpenMask3D: Open-Vocabulary 3D Instance Segmentation | We introduce the task of open-vocabulary 3D instance segmentation. Traditional approaches for 3D instance segmentation largely rely on existing 3D annotated datasets, which are restricted to a closed-set of object categories. This is an important limitation for real-life applications where one might need to perform tasks guided by novel, open-vocabulary queries related to objects from a wide variety. Recently, open-vocabulary 3D scene understanding methods have emerged to address this problem by learning queryable features per each point in the scene. While such a representation can be directly employed to perform semantic segmentation, existing methods have limitations in their ability to identify object instances. In this work, we address this limitation, and propose OpenMask3D, which is a zero-shot approach for open-vocabulary 3D instance segmentation. Guided by predicted class-agnostic 3D instance masks, our model aggregates per-mask features via multi-view fusion of CLIP-based image embeddings. We conduct experiments and ablation studies on the ScanNet200 dataset to evaluate the performance of OpenMask3D, and provide insights about the open-vocabulary 3D instance segmentation task. We show that our approach outperforms other open-vocabulary counterparts, particularly on the long-tail distribution. Furthermore, OpenMask3D goes beyond the limitations of close-vocabulary approaches, and enables the segmentation of object instances based on free-form queries describing object properties such as semantics, geometry, affordances, and material properties. | ['Francis Engelmann', 'Federico Tombari', 'Marc Pollefeys', 'Robert W. Sumner', 'Elisabetta Fedele', 'Ayça Takmaz'] | 2023-06-23 | null | null | null | null | ['3d-instance-segmentation-1', 'instance-segmentation', 'scene-understanding'] | ['computer-vision', 'computer-vision', 'computer-vision'] | [ 2.16263518e-01 2.53278583e-01 -2.28820547e-01 -4.61636305e-01
-1.09458387e+00 -9.49891269e-01 6.29223168e-01 1.95758104e-01
-9.04430822e-02 -1.96270202e-03 -6.42054901e-02 -1.41008928e-01
-2.18024954e-01 -7.02046633e-01 -8.83551717e-01 -4.15188760e-01
-3.56321521e-02 1.05471694e+00 5.94843388e-01 -1.89043313e-01
1.68182746e-01 7.21206963e-01 -1.83153343e+00 5.41811772e-02
7.18956113e-01 1.20562088e+00 1.76087841e-01 2.93475688e-01
-5.44273198e-01 -2.26625264e-01 -3.33277524e-01 -1.65114030e-01
7.10665345e-01 1.63101718e-01 -9.14566219e-01 3.61851662e-01
7.39488363e-01 -1.61967427e-01 -5.49125597e-02 9.37096894e-01
3.08591455e-01 3.47435862e-01 1.05203438e+00 -1.17972827e+00
-5.95630705e-01 3.10706031e-02 -4.94390130e-01 8.44286978e-02
4.46430832e-01 2.70612776e-01 1.29967403e+00 -9.34214532e-01
7.20440328e-01 1.32500207e+00 4.55976129e-01 4.84782100e-01
-1.40175176e+00 -4.83054817e-01 5.64564228e-01 -1.14035077e-01
-1.51048517e+00 -6.62390813e-02 8.35547507e-01 -6.73492908e-01
1.04478085e+00 3.00331622e-01 7.92885780e-01 8.47690284e-01
-3.71154875e-01 8.82661223e-01 9.65323567e-01 -1.32245079e-01
3.09054434e-01 -6.36951551e-02 3.23637784e-01 5.39146304e-01
3.21080655e-01 -6.81050047e-02 -3.51777256e-01 -1.19249858e-01
1.02592218e+00 -7.68256374e-04 -4.55331914e-02 -1.09216046e+00
-1.27290261e+00 1.00667679e+00 5.44655383e-01 -5.66463731e-02
-4.37363461e-02 -6.82756677e-02 4.02855426e-01 3.77582759e-02
9.86168921e-01 6.36454403e-01 -7.44271874e-01 9.03666317e-02
-7.58894444e-01 5.08933008e-01 7.32574105e-01 1.28454614e+00
1.07930422e+00 -4.53844994e-01 -1.13352850e-01 8.75556529e-01
3.16208363e-01 6.60363257e-01 1.35607589e-02 -9.22391295e-01
3.06322396e-01 9.59816158e-01 -1.11836977e-01 -5.85763872e-01
-4.00419652e-01 -5.48342429e-02 -3.42439890e-01 5.01853153e-02
3.43302578e-01 5.04464924e-01 -1.40617526e+00 1.24950147e+00
8.17222714e-01 1.54063851e-01 -8.84584486e-02 1.17892087e+00
1.16055155e+00 4.46142137e-01 -1.51425712e-02 3.30881596e-01
1.34715867e+00 -7.14910507e-01 -2.49632690e-02 -1.82807341e-01
6.45025909e-01 -5.34510732e-01 1.33953881e+00 1.15021374e-02
-6.34213090e-01 -4.50863600e-01 -7.54049063e-01 -3.29632580e-01
-6.90093815e-01 -5.90009570e-01 7.74877250e-01 6.37880504e-01
-5.70942760e-01 1.28574044e-01 -6.28879249e-01 -4.12986189e-01
8.71347249e-01 4.19360101e-01 -4.35751528e-01 -1.76713377e-01
-8.27342927e-01 6.35115862e-01 4.81411695e-01 -3.03690463e-01
-9.53013837e-01 -1.02584589e+00 -1.35723567e+00 -1.38507813e-01
8.35142195e-01 -6.60537601e-01 1.14567494e+00 -5.20523727e-01
-9.53273058e-01 1.34530139e+00 3.16843875e-02 -2.05982819e-01
2.02321276e-01 -1.82498410e-01 6.00701682e-02 2.13008612e-01
4.32692856e-01 1.03649592e+00 8.61820519e-01 -1.39104009e+00
-5.18081844e-01 -6.91687942e-01 5.74950993e-01 2.88477421e-01
1.41090706e-01 -3.88439208e-01 -7.57061005e-01 -6.02313578e-01
4.84735757e-01 -8.49466562e-01 -4.97456878e-01 1.12803474e-01
-4.86723661e-01 -4.61386710e-01 8.36194038e-01 -3.52890007e-02
6.28790498e-01 -2.23372126e+00 2.96486206e-02 2.91609377e-01
2.48092696e-01 -3.78062129e-02 -1.20036438e-01 6.76975865e-03
1.86122924e-01 4.01932359e-01 -4.89083380e-01 4.77129258e-02
2.99279392e-01 4.06436712e-01 -3.61387312e-01 4.56205100e-01
4.45635706e-01 1.11519575e+00 -7.27400959e-01 -5.96683562e-01
5.72960258e-01 2.71358281e-01 -8.23004603e-01 1.69344828e-01
-9.50786769e-01 3.85887802e-01 -8.12824726e-01 8.04772437e-01
6.96183085e-01 -3.80105197e-01 -4.25085545e-01 -2.28529111e-01
6.67357668e-02 6.68034405e-02 -1.15420616e+00 2.11823821e+00
-4.47136968e-01 2.67859191e-01 -2.68515289e-01 -1.16332364e+00
9.12278652e-01 -1.32240988e-02 7.59420037e-01 -3.73306781e-01
1.18424393e-01 2.13174373e-01 -4.11318988e-01 -4.89994615e-01
3.50771248e-01 -9.88913402e-02 -5.01372337e-01 3.92822444e-01
6.49114698e-02 -1.00861871e+00 3.44558209e-02 2.04810068e-01
8.17979157e-01 3.27251792e-01 2.82228440e-01 -3.03232580e-01
1.77682340e-01 3.47297966e-01 1.75403506e-01 7.93651283e-01
-7.60346949e-02 9.66167748e-01 3.23132068e-01 -5.08840919e-01
-9.68494534e-01 -1.34534228e+00 -4.64739710e-01 7.91958272e-01
8.81250143e-01 -2.18709424e-01 -6.42665684e-01 -9.96764958e-01
2.80641913e-01 4.02692318e-01 -6.75837815e-01 1.83077708e-01
-2.66926169e-01 -3.79106581e-01 1.12998836e-01 4.72643673e-01
2.51976818e-01 -7.18164861e-01 -7.37729847e-01 -2.85248309e-02
-5.92830554e-02 -1.48487306e+00 -6.23145223e-01 2.69425333e-01
-7.30991781e-01 -1.34511518e+00 -6.53467774e-01 -8.48872364e-01
6.87186599e-01 4.23061997e-01 1.15978205e+00 -1.77885354e-01
-6.60493255e-01 8.93594980e-01 -6.47893548e-01 -5.00813127e-01
-1.10174075e-01 2.33207703e-01 -1.15349859e-01 -8.36135587e-04
6.40633345e-01 -4.36884195e-01 -5.77189267e-01 4.65774953e-01
-8.68907928e-01 7.68539682e-02 4.12658423e-01 5.24523437e-01
1.20543027e+00 -1.85399085e-01 3.22995991e-01 -7.96423852e-01
4.89390977e-02 -3.62146139e-01 -7.21266985e-01 9.00758952e-02
-1.94762260e-01 1.13630965e-01 -1.68155981e-04 -5.65298259e-01
-7.31247127e-01 3.15363467e-01 -5.69587909e-02 -7.90602267e-01
-6.26834571e-01 1.62643224e-01 -4.27837193e-01 -7.87394121e-02
4.50498998e-01 1.49021104e-01 2.95975991e-02 -5.43548048e-01
7.11532772e-01 5.71814835e-01 1.94659367e-01 -8.30674887e-01
9.45042968e-01 7.62343347e-01 -3.05841491e-02 -1.06792653e+00
-1.25726330e+00 -9.06224608e-01 -8.82490933e-01 -1.09994248e-01
1.23033881e+00 -1.04895127e+00 -3.84237081e-01 1.19444251e-01
-9.21634078e-01 -4.27059263e-01 -7.72589207e-01 2.76015759e-01
-8.37581396e-01 2.13161603e-01 -3.91971506e-02 -6.10879540e-01
2.68971864e-02 -1.26365411e+00 1.80772936e+00 -3.72034945e-02
-2.07943454e-01 -8.15718830e-01 -2.39623278e-01 6.78140998e-01
-1.36787847e-01 2.37110734e-01 9.87076521e-01 -1.04736805e+00
-9.66397166e-01 -1.56319767e-01 -3.58368158e-01 3.08259368e-01
7.34323859e-02 -4.35471207e-01 -1.04709566e+00 4.68540899e-02
-2.91429937e-01 -5.09161592e-01 7.59175777e-01 4.79649782e-01
1.47050202e+00 2.95654893e-01 -3.96903545e-01 7.48755097e-01
1.23061430e+00 -2.50047475e-01 1.56882048e-01 9.10618231e-02
8.99709105e-01 7.22485721e-01 8.48365784e-01 3.45788032e-01
4.53530759e-01 6.26205564e-01 6.34997129e-01 -1.07914075e-01
-3.55208069e-02 -3.54625046e-01 -2.52579212e-01 3.16438466e-01
1.91069007e-01 -2.34966278e-01 -9.96315181e-01 7.23792195e-01
-1.45849621e+00 -5.24666369e-01 -7.00723305e-02 1.93781054e+00
5.50551236e-01 3.98119450e-01 1.21550046e-01 -6.79433122e-02
5.27149916e-01 3.34013611e-01 -9.52133775e-01 -5.44402935e-02
-8.95966813e-02 4.02243704e-01 6.05399132e-01 2.80703634e-01
-1.31779063e+00 1.17455220e+00 5.80437469e+00 1.13454127e+00
-7.99893200e-01 9.07804668e-02 5.39855957e-01 -1.12945676e-01
-7.34717786e-01 1.07251659e-01 -9.82333541e-01 1.00982964e-01
1.44147560e-01 1.15893401e-01 1.38613924e-01 9.12196934e-01
-3.87428887e-02 -1.26741812e-01 -1.36044145e+00 1.22617996e+00
2.51724392e-01 -1.40319586e+00 4.34315652e-01 2.51505107e-01
6.99766397e-01 1.03288390e-01 -1.29432991e-01 3.60025793e-01
1.15804933e-01 -1.04290080e+00 6.41287088e-01 1.16640024e-01
1.02145445e+00 -4.55968589e-01 1.87077686e-01 3.71881634e-01
-1.37142766e+00 1.70449123e-01 -2.17480272e-01 6.35602176e-02
2.08060592e-01 5.34920990e-01 -6.12085164e-01 4.73085344e-01
8.20416808e-01 6.99840426e-01 -3.77533257e-01 8.90980780e-01
4.28198949e-02 4.34012026e-01 -7.59923458e-01 7.79729057e-03
3.79216343e-01 -1.45859540e-01 6.68607533e-01 6.86443150e-01
3.45338248e-02 2.40002707e-01 6.44613147e-01 1.11590886e+00
-5.52825555e-02 1.77518055e-01 -7.87706196e-01 8.23692679e-02
3.10561687e-01 1.00389361e+00 -1.11752069e+00 -2.02451527e-01
-5.68830371e-01 7.22026289e-01 8.11175629e-02 2.30236903e-01
-7.73882806e-01 -1.81295872e-01 1.04679728e+00 4.04091150e-01
7.98661113e-01 -3.67115408e-01 -4.57705110e-01 -1.11614764e+00
1.39206290e-01 -4.78173554e-01 3.73795539e-01 -7.03302622e-01
-1.45065463e+00 1.70054182e-01 5.01031935e-01 -1.28545940e+00
1.40471846e-01 -8.22592080e-01 -9.82134566e-02 5.07060587e-01
-1.50738192e+00 -1.46559203e+00 -2.30415881e-01 5.49163699e-01
1.08046043e+00 2.26995051e-01 6.68065846e-01 9.02544856e-02
-3.40221375e-02 2.16301486e-01 -2.36240715e-01 8.41324404e-02
3.12246710e-01 -1.22225821e+00 6.13807738e-01 3.22916538e-01
6.00385725e-01 2.79966652e-01 5.20724833e-01 -6.29717231e-01
-1.26943135e+00 -1.22654605e+00 2.54031569e-01 -9.26030159e-01
4.40466583e-01 -8.46975505e-01 -8.21517229e-01 5.48105896e-01
-6.26576006e-01 5.22325933e-01 6.12002134e-01 1.88868657e-01
-5.47184289e-01 1.69609949e-01 -1.07162702e+00 5.10044336e-01
1.50386000e+00 -4.99474049e-01 -7.53864646e-01 4.26785439e-01
1.31394982e+00 -6.71511590e-01 -1.06478858e+00 6.47479653e-01
3.65335941e-01 -6.09856069e-01 1.25289571e+00 -5.60053766e-01
1.93128392e-01 -1.95194036e-01 -5.36392093e-01 -1.01143289e+00
1.63141683e-01 -2.51197070e-01 1.38414249e-01 9.01098073e-01
3.50447565e-01 -3.82307351e-01 9.21622694e-01 6.03672624e-01
-3.09703857e-01 -9.15139258e-01 -1.07965910e+00 -9.14211571e-01
2.02244282e-01 -1.10226095e+00 8.49950969e-01 6.86544001e-01
-5.89582384e-01 2.86070496e-01 2.26757228e-01 3.67854297e-01
7.37742424e-01 7.38204658e-01 1.03862643e+00 -1.48058665e+00
5.94724379e-02 -3.58921230e-01 -8.00706089e-01 -1.71304715e+00
3.83942425e-01 -1.08239079e+00 5.62960468e-02 -1.46803343e+00
1.61270037e-01 -8.57261658e-01 1.30540565e-01 3.09388995e-01
1.41885042e-01 4.98440504e-01 -1.35051797e-03 9.49120373e-02
-8.36729944e-01 6.64104164e-01 1.50097394e+00 -4.18287635e-01
-3.05176854e-01 -6.84276894e-02 -5.66578388e-01 7.69340098e-01
4.32257891e-01 -2.24494800e-01 -5.79038262e-01 -4.09687698e-01
-1.00359982e-02 -2.74918377e-01 7.65547156e-01 -6.31087005e-01
-2.23292768e-01 -2.73803055e-01 4.65909131e-02 -8.96772981e-01
5.90361714e-01 -9.74200070e-01 -2.34389275e-01 -1.41034842e-01
-1.65398955e-01 -6.87905669e-01 1.38060465e-01 7.49992371e-01
-1.22901890e-02 9.95188020e-03 5.45973420e-01 -3.49762172e-01
-1.08006883e+00 8.52220058e-01 3.79796810e-02 5.88702977e-01
1.29961956e+00 -7.84252465e-01 1.53694481e-01 2.48687025e-02
-9.22635853e-01 4.45850730e-01 7.10850716e-01 6.84807658e-01
7.25426018e-01 -1.03281510e+00 -3.89002502e-01 4.89422888e-01
7.60255337e-01 1.03659523e+00 3.93316776e-01 4.36764270e-01
-3.43352944e-01 3.73446375e-01 2.41037577e-01 -1.24359202e+00
-1.02748847e+00 6.46344662e-01 3.48648667e-01 1.52332380e-01
-9.84486938e-01 9.29919481e-01 8.54744434e-01 -8.82364333e-01
2.05299407e-01 -6.71742797e-01 -2.00357754e-02 -4.18816283e-02
-2.49950085e-02 -1.51202694e-01 2.11562086e-02 -7.76975930e-01
-3.76612067e-01 1.03831792e+00 9.97731835e-02 1.03158213e-01
1.26232564e+00 -1.51261196e-01 2.88454801e-01 5.98600209e-01
1.18186307e+00 -3.44819516e-01 -1.30999672e+00 -3.22116375e-01
-1.16627358e-01 -6.02179110e-01 -4.30497602e-02 -4.84714568e-01
-8.15087199e-01 8.50585520e-01 5.07802069e-01 1.73367858e-01
6.47662997e-01 9.32444572e-01 7.84526646e-01 3.48376751e-01
6.24261439e-01 -8.70282173e-01 9.81625319e-02 5.58371127e-01
7.31954515e-01 -1.68341148e+00 -6.23592809e-02 -7.96352565e-01
-3.90554935e-01 8.04935277e-01 8.80227685e-01 -1.13502167e-01
9.78603065e-01 -1.04202732e-01 5.96496686e-02 -8.01902592e-01
-3.38405550e-01 -6.24573052e-01 5.79576135e-01 7.22434878e-01
-9.74519253e-02 6.02861755e-02 3.20321560e-01 6.38272524e-01
-2.54883319e-01 -5.24049282e-01 9.87153351e-02 7.61107862e-01
-5.12126744e-01 -8.15652549e-01 -4.16305453e-01 7.72573471e-01
-9.62604582e-03 1.03541732e-01 -2.64972419e-01 1.04319668e+00
2.60219067e-01 5.66592753e-01 4.13527191e-01 -1.55840904e-01
5.12991369e-01 -5.61035387e-02 5.86709440e-01 -1.12890506e+00
5.53088300e-02 -7.79330283e-02 -1.46053627e-01 -6.40444756e-01
-5.94206810e-01 -7.60184109e-01 -1.25746024e+00 3.52757961e-01
-5.43953240e-01 -7.04240575e-02 6.39973998e-01 1.03008235e+00
2.61261284e-01 2.00987488e-01 4.02576596e-01 -1.07481658e+00
-4.12975520e-01 -5.41291356e-01 -7.41743326e-01 6.50571287e-01
3.45123559e-01 -1.12085164e+00 -4.31469887e-01 -7.98524022e-02] | [8.00770378112793, -3.144414186477661] |
a35ebfdf-3f7d-463e-8189-f088ab8a7617 | dfpenet-geology-a-deep-learning-framework-for | 1908.10907 | null | https://arxiv.org/abs/1908.10907v2 | https://arxiv.org/pdf/1908.10907v2.pdf | DFPENet-geology: A Deep Learning Framework for High Precision Recognition and Segmentation of Co-seismic Landslides | The following lists two main reasons for withdrawal for the public. 1. There are some problems in the method and results, and there is a lot of room for improvement. In terms of method, "Pre-trained Datasets (PD)" represents selecting a small amount from the online test set, which easily causes the model to overfit the online test set and could not obtain robust performance. More importantly, the proposed DFPENet has a high redundancy by combining the Attention Gate Mechanism and Gate Convolution Networks, and we need to revisit the section of geological feature fusion, in terms of results, we need to further improve and refine. 2. arXiv is an open-access repository of electronic preprints without peer reviews. However, for our own research, we need experts to provide comments on my work whether negative or positive. I then would use their comments to significantly improve this manuscript. Therefore, we finally decided to withdraw this manuscript in arXiv, and we will update to arXiv with the final accepted manuscript to facilitate more researchers to use our proposed comprehensive and general scheme to recognize and segment seismic landslides more efficiently. | ['Tianhai Jiang', 'Duoxiang Cheng', 'Chaojun Ouyang', 'Qingsong Xu', 'Xuanmei Fan'] | 2019-08-28 | null | null | null | null | ['scene-recognition'] | ['computer-vision'] | [-1.32912129e-01 9.82686877e-02 2.54239738e-01 -6.73182726e-01
-8.83877337e-01 -4.27603543e-01 -1.05980970e-01 2.17033383e-02
-3.82778466e-01 7.88511813e-01 8.87822583e-02 -6.58187687e-01
-3.87948066e-01 -1.00621665e+00 -8.08410645e-01 -7.77797401e-01
-1.85926765e-01 1.09636739e-01 4.32048589e-01 -4.11430597e-01
7.53377259e-01 4.27085668e-01 -1.49091589e+00 1.79002732e-01
9.58626568e-01 9.33955908e-01 3.76587659e-01 1.87276244e-01
8.39960761e-03 3.59548062e-01 -4.63013291e-01 -6.04287922e-01
2.40152568e-01 -2.70384699e-01 -9.21405852e-01 -3.87236297e-01
1.25278845e-01 -4.66282666e-01 -3.68575782e-01 1.28973544e+00
6.71045303e-01 1.86258987e-01 5.59817016e-01 -8.79280806e-01
-7.71327734e-01 8.12596023e-01 -3.45671654e-01 3.62339079e-01
1.10399192e-02 1.20043658e-01 8.54937732e-01 -1.04215634e+00
3.04997206e-01 1.00480378e+00 6.39719903e-01 2.85271466e-01
-4.66947317e-01 -8.27102840e-01 3.63365144e-01 4.36284125e-01
-1.31008136e+00 -5.28283775e-01 6.99996471e-01 -3.86903733e-01
9.02750790e-01 3.13742340e-01 6.43109262e-01 8.98688674e-01
9.69948694e-02 6.58251643e-01 8.66684973e-01 -1.64622411e-01
1.30533785e-01 -4.24532145e-02 2.39437938e-01 5.74812055e-01
4.13381010e-01 7.91132152e-02 -1.66551575e-01 -1.42996371e-01
7.63562441e-01 -1.47916719e-01 -2.03290030e-01 5.74383289e-02
-1.05112576e+00 9.16416168e-01 3.89716506e-01 4.63956445e-01
-4.73161414e-02 -8.70959610e-02 1.56253859e-01 3.46743822e-01
4.31144416e-01 3.08318645e-01 -4.23429251e-01 -3.06992054e-01
-7.76414573e-01 3.15088004e-01 6.67887688e-01 5.60190916e-01
7.50652134e-01 -1.27489910e-01 3.23015481e-01 1.08260465e+00
3.47252131e-01 5.68625629e-01 3.19044799e-01 -8.30746591e-01
3.98272455e-01 4.14414674e-01 -9.44866613e-02 -1.25918901e+00
-3.12490314e-01 -3.74367744e-01 -8.68396699e-01 -5.87663352e-02
1.90832958e-01 -1.60749406e-01 -7.52166808e-01 1.36632752e+00
-1.45272374e-01 -4.22979221e-02 -2.91953564e-01 1.06385446e+00
9.71126735e-01 5.49185872e-01 -8.25280398e-02 1.89819992e-01
1.24513817e+00 -7.22240984e-01 -6.78253174e-01 -1.63826972e-01
6.34151638e-01 -7.31089234e-01 9.56240356e-01 4.38319027e-01
-1.25229692e+00 -4.37092304e-01 -1.34931564e+00 1.70343161e-01
-6.43695891e-01 3.49500120e-01 8.36569905e-01 3.85264814e-01
-8.40727985e-01 9.89979506e-01 -1.11971605e+00 -4.89964336e-01
2.13402003e-01 2.77187794e-01 -3.83285791e-01 -1.00315303e-01
-1.58788490e+00 9.19985056e-01 2.82267123e-01 6.09464705e-01
-3.69742304e-01 -1.94451585e-01 -7.74912000e-01 5.02076633e-02
3.45712543e-01 -2.74155587e-01 1.06949663e+00 -4.12913501e-01
-1.09461200e+00 7.48858929e-01 -6.36467412e-02 -1.95184886e-01
4.53887880e-01 -2.51978159e-01 -2.89576650e-01 6.38520420e-02
1.99337631e-01 3.67499650e-01 2.15252608e-01 -7.66920149e-01
-5.63924193e-01 -3.82044613e-01 -4.60090525e-02 -1.51811525e-01
-3.76688540e-01 1.43848136e-01 -5.43030083e-01 -7.04203010e-01
5.41286409e-01 -6.76202476e-01 -6.91444427e-02 -3.21144015e-01
-1.41298361e-02 -4.07209784e-01 5.80973446e-01 -7.28973687e-01
1.40406632e+00 -2.18717742e+00 -5.33944428e-01 1.39357209e-01
-3.72400209e-02 2.64259100e-01 2.05768675e-01 4.42710936e-01
-2.63040692e-01 6.16086900e-01 -6.23863280e-01 -3.45900543e-02
-1.86650585e-02 1.33282647e-01 -2.57266074e-01 5.27160048e-01
3.01844448e-01 7.26239264e-01 -8.26781392e-01 -2.37328425e-01
-5.98587878e-02 7.84701779e-02 -3.76575589e-01 -8.80629569e-02
2.71425396e-01 9.65948179e-02 -6.61745012e-01 8.85159254e-01
1.12552381e+00 -1.71041399e-01 -2.74231434e-01 -1.72112793e-01
-3.56140375e-01 4.32393461e-01 -1.42091036e+00 1.39265382e+00
1.11990675e-01 4.69480127e-01 -2.31642816e-02 -1.30697322e+00
1.02480221e+00 1.36036783e-01 3.61069411e-01 -6.00358129e-01
-1.93170328e-02 5.72583139e-01 9.81635377e-02 -7.86961615e-01
6.77459180e-01 -1.38598569e-02 -1.58767588e-02 4.06794369e-01
-2.47888625e-01 9.50912163e-02 1.92011401e-01 2.19055906e-01
1.02579129e+00 3.03424925e-01 -1.62224501e-01 -4.27500963e-01
2.52114594e-01 -5.63743114e-02 5.06977022e-01 1.01513410e+00
-1.72235101e-01 7.42503643e-01 5.20723999e-01 -3.84587795e-01
-9.56012189e-01 -6.67431235e-01 -4.63990688e-01 8.29989970e-01
-8.46377909e-02 -2.02844843e-01 -4.60066378e-01 -5.73673189e-01
3.38174552e-02 4.93829846e-01 -6.54509008e-01 -4.72744852e-02
-6.37910545e-01 -8.95294130e-01 7.54758358e-01 7.37509787e-01
6.84922934e-01 -1.03942454e+00 -3.72473866e-01 1.67185500e-01
-3.37620854e-01 -3.59458894e-01 -1.76474378e-01 3.26969266e-01
-9.62575614e-01 -1.02977991e+00 -9.79401648e-01 -7.25338459e-01
6.02934003e-01 5.62874973e-01 6.75791800e-01 6.36945546e-01
2.00392902e-01 1.24377711e-02 -6.44595921e-01 -4.33639050e-01
-1.47898167e-01 2.65506953e-01 -1.83599278e-01 -3.96710515e-01
4.10199821e-01 -4.26803529e-01 -6.84518993e-01 2.28377685e-01
-1.04070103e+00 -1.07646920e-01 5.96068025e-01 8.59141648e-01
5.35492636e-02 8.19526538e-02 7.80762553e-01 -7.51682818e-01
6.30822539e-01 -5.69496989e-01 -5.27416945e-01 2.20160961e-01
-6.63118839e-01 -5.35053164e-02 2.10957706e-01 1.08741947e-01
-8.77987087e-01 -2.90184438e-01 -5.64074278e-01 -1.45508260e-01
-5.37821203e-02 9.92477238e-01 -7.74228647e-02 -1.06111318e-01
3.00377578e-01 2.20421776e-02 -1.62215248e-01 -7.71898746e-01
-2.64299601e-01 8.44471037e-01 5.47146618e-01 -5.36213577e-01
5.33969164e-01 3.07716072e-01 -3.02681625e-01 -5.26435256e-01
-6.30413532e-01 -4.05588984e-01 -4.76228982e-01 -1.55991346e-01
5.91685951e-01 -9.00843680e-01 -6.84642136e-01 8.13034713e-01
-1.08909607e+00 -3.77755046e-01 1.59443557e-01 8.13056350e-01
-9.15921852e-02 6.79025292e-01 -6.47919357e-01 -6.57660902e-01
8.19195714e-03 -1.10764432e+00 8.24462414e-01 3.72113585e-01
1.43571571e-01 -8.12329590e-01 -2.59200245e-01 1.49667293e-01
6.98454976e-01 4.47162241e-02 6.59203172e-01 -6.29485726e-01
-4.55763191e-01 -3.56447786e-01 -1.49373129e-01 2.79967934e-01
-8.48552510e-02 4.53254849e-01 -1.15137780e+00 -3.04447085e-01
6.66643158e-02 -2.87911057e-01 1.37350428e+00 4.07677889e-01
1.69962180e+00 -7.14798570e-02 -4.40487951e-01 3.48340183e-01
1.19099545e+00 3.93210530e-01 1.08046710e+00 7.60053396e-01
3.16576481e-01 8.00692976e-01 8.69857013e-01 5.85890591e-01
3.82995009e-01 2.90436983e-01 3.33153278e-01 -1.80051506e-01
2.42031708e-01 -1.81568772e-01 3.18998009e-01 1.02316296e+00
-2.76983142e-01 -3.57186347e-01 -1.00261676e+00 5.80793619e-01
-1.92386222e+00 -8.72549891e-01 -6.31466866e-01 2.48043966e+00
4.84439760e-01 2.23227203e-01 -1.02216370e-01 2.07676828e-01
7.42671549e-01 9.00500938e-02 -2.60849893e-01 -1.38377056e-01
-4.43632215e-01 1.78843975e-01 7.06547797e-01 4.18485790e-01
-1.08496666e+00 7.57122397e-01 6.57516336e+00 7.20704317e-01
-1.46329403e+00 -6.53757453e-02 7.87562132e-01 3.42239320e-01
-6.69308960e-01 5.00118136e-01 -6.75965130e-01 6.22484922e-01
9.98100042e-01 -4.66078408e-02 1.06419757e-01 6.52636826e-01
3.15284431e-01 -3.31743151e-01 -6.09318733e-01 7.11783588e-01
-1.17639922e-01 -1.27605796e+00 -3.85856539e-01 1.84726298e-01
2.89436787e-01 2.14816824e-01 -3.27694416e-02 5.24983346e-01
1.91288650e-01 -9.48220491e-01 5.70383370e-01 6.87625468e-01
5.34350455e-01 -6.61934018e-01 1.03658235e+00 2.36136496e-01
-1.02252269e+00 2.84101460e-02 -7.88132131e-01 -2.88456082e-01
1.13755994e-01 7.36534297e-01 -2.59898603e-01 9.82159972e-01
9.97878730e-01 6.23134792e-01 -8.09709191e-01 1.47688341e+00
2.37458223e-03 8.25220227e-01 -4.08436209e-01 -4.52596694e-02
2.02217087e-01 -6.47251904e-02 2.58563936e-01 1.01788557e+00
7.56764174e-01 1.83411673e-01 -8.12247917e-02 8.13876688e-01
1.33159414e-01 6.34284243e-02 -6.21137142e-01 -9.53623652e-03
4.35837597e-01 1.04090965e+00 -8.69681120e-01 -3.52686256e-01
-5.60194254e-01 6.00265384e-01 1.06324166e-01 2.37969130e-01
-7.90657878e-01 -8.58685017e-01 -1.27078369e-01 2.44707838e-01
3.30722183e-01 -2.29167923e-01 -4.82518643e-01 -1.23165917e+00
1.12009607e-01 -5.70797026e-01 3.08232695e-01 -9.20217037e-01
-1.27360392e+00 4.99526680e-01 1.80331737e-01 -1.24171162e+00
1.27658918e-01 -4.87638503e-01 -6.21616900e-01 8.70331168e-01
-1.24537516e+00 -8.25329661e-01 -3.02430838e-01 -3.82501297e-02
2.00098932e-01 5.96324913e-03 5.78242838e-01 6.72426581e-01
-4.29459512e-01 6.03108227e-01 3.74261469e-01 2.96592683e-01
9.93429184e-01 -8.33944321e-01 4.32054907e-01 7.40025163e-01
-2.98483342e-01 6.80279016e-01 5.56498408e-01 -7.91235089e-01
-1.03714943e+00 -8.83192122e-01 1.24629426e+00 -1.76136807e-01
5.73352456e-01 -1.66800559e-01 -1.29048121e+00 6.31209731e-01
9.02133510e-02 -3.81179571e-01 6.74664438e-01 4.06185351e-02
2.95299083e-01 -1.83244180e-02 -1.02811038e+00 3.89611363e-01
8.96970332e-01 -2.44570956e-01 -5.97940445e-01 3.39225948e-01
4.99033123e-01 -2.31571093e-01 -9.45009351e-01 6.15011394e-01
4.22860652e-01 -9.29955244e-01 4.73392874e-01 -2.31109217e-01
4.17651504e-01 -4.40826893e-01 -1.10194907e-01 -1.00196850e+00
-5.94798148e-01 -1.50999174e-01 4.71798480e-01 1.29082084e+00
4.42765921e-01 -1.18555212e+00 6.71675742e-01 6.62857056e-01
-5.91198981e-01 -9.03521299e-01 -8.62824678e-01 -6.67527795e-01
4.44904506e-01 -7.17282832e-01 8.44130337e-01 8.73783350e-01
1.44504875e-01 -1.42786711e-01 -3.67059857e-01 4.68965583e-02
3.13437462e-01 -2.24929303e-02 5.31795204e-01 -1.18671906e+00
-1.26825467e-01 -4.22908008e-01 -4.18649465e-02 -1.25239372e+00
-2.82052845e-01 -7.81435132e-01 3.25134061e-02 -1.37819421e+00
1.86331585e-01 -7.10216224e-01 -3.45035821e-01 6.32594526e-01
-2.58315802e-01 2.49805734e-01 -1.18427068e-01 5.71992099e-01
-3.03845853e-01 7.20210373e-01 1.20766199e+00 -5.17906621e-02
1.77996948e-01 1.38351053e-01 -1.16274178e+00 4.43395585e-01
8.45235229e-01 -5.44803858e-01 1.53147951e-01 -4.69989955e-01
2.46917248e-01 1.54771388e-01 4.77965623e-01 -9.41121280e-01
1.80085421e-01 -2.97738910e-01 4.39942777e-01 -9.23457682e-01
-1.66345127e-02 -3.57795328e-01 -7.92425722e-02 3.85199666e-01
-6.44920841e-02 1.85894638e-01 2.84808934e-01 2.18630224e-01
-2.92998374e-01 -7.26392806e-01 5.39427757e-01 -3.83254498e-01
-5.53841591e-01 1.92197204e-01 -5.34308791e-01 -4.79590654e-01
5.99188030e-01 -3.24323803e-01 -3.94879699e-01 -6.03220105e-01
-7.13117659e-01 4.08319712e-01 5.94224155e-01 2.73152888e-01
5.70337057e-01 -1.27629030e+00 -8.02306712e-01 2.24468306e-01
-7.14745224e-02 6.30814135e-02 5.85816622e-01 1.08873880e+00
-6.61233306e-01 4.99151349e-01 -1.01247489e-01 -2.29552075e-01
-7.81962752e-01 1.94438607e-01 1.41672522e-01 7.64229894e-02
-5.44874072e-01 8.02893639e-01 2.30337176e-02 -4.94370520e-01
2.64834940e-01 -2.74366498e-01 -2.51876056e-01 1.55590503e-02
4.07635152e-01 4.23873603e-01 3.81193250e-01 -4.54253465e-01
-4.50966656e-01 1.98128432e-01 -9.09341797e-02 -1.93855371e-02
1.64727521e+00 -6.72985613e-02 -2.08219383e-02 4.94294524e-01
1.08128440e+00 3.42443436e-02 -9.93341744e-01 8.97089615e-02
-5.20309210e-02 -5.46991229e-01 -1.26353592e-01 -6.00818872e-01
-1.17152381e+00 1.02360797e+00 7.33508885e-01 4.33759063e-01
1.00341594e+00 -3.82354259e-02 7.96659946e-01 5.88162839e-01
2.90987909e-01 -1.40473282e+00 -1.45725578e-01 1.04470396e+00
1.01333535e+00 -1.27729380e+00 2.49054715e-01 -1.17282480e-01
-4.23279881e-01 1.39602399e+00 6.29560232e-01 -1.36026219e-01
8.42119157e-01 4.95370924e-02 -1.04222335e-01 -4.14049953e-01
-4.95551497e-01 -6.45029619e-02 -1.23403251e-01 4.08459306e-01
7.66096652e-01 -1.71387494e-01 -6.91536486e-01 9.18090940e-01
-2.48588160e-01 -2.45015253e-03 6.01512432e-01 8.55686188e-01
-6.13726139e-01 -1.32410944e+00 -5.48509121e-01 6.67767048e-01
-4.72188175e-01 -1.81358948e-01 -2.09724188e-01 7.68084049e-01
8.59772936e-02 8.63354445e-01 -4.82091568e-02 -4.94859695e-01
2.22231343e-01 5.87971881e-02 1.00740038e-01 -2.36511573e-01
-5.08897007e-01 5.51132746e-02 -9.06238519e-03 -6.08445629e-02
-3.05301636e-01 -7.39162862e-01 -1.22472084e+00 -6.06491566e-01
-5.80490232e-01 4.71018732e-01 7.14158952e-01 8.61989081e-01
3.92380506e-01 2.76496142e-01 4.47287828e-01 -6.95173681e-01
-6.01895094e-01 -1.33199644e+00 -6.76561117e-01 1.40140116e-01
-3.00578117e-01 -8.24399471e-01 -5.12053788e-01 -2.78693646e-01] | [7.114596843719482, 2.3044590950012207] |
f9b0fa5b-b4cf-42be-85da-00d6e1c057f1 | bartpho-pre-trained-sequence-to-sequence | 2109.09701 | null | https://arxiv.org/abs/2109.09701v3 | https://arxiv.org/pdf/2109.09701v3.pdf | BARTpho: Pre-trained Sequence-to-Sequence Models for Vietnamese | We present BARTpho with two versions, BARTpho-syllable and BARTpho-word, which are the first public large-scale monolingual sequence-to-sequence models pre-trained for Vietnamese. BARTpho uses the "large" architecture and the pre-training scheme of the sequence-to-sequence denoising autoencoder BART, thus it is especially suitable for generative NLP tasks. We conduct experiments to compare our BARTpho with its competitor mBART on a downstream task of Vietnamese text summarization and show that: in both automatic and human evaluations, BARTpho outperforms the strong baseline mBART and improves the state-of-the-art. We further evaluate and compare BARTpho and mBART on the Vietnamese capitalization and punctuation restoration tasks and also find that BARTpho is more effective than mBART on these two tasks. We publicly release BARTpho to facilitate future research and applications of generative Vietnamese NLP tasks. Our BARTpho models are available at https://github.com/VinAIResearch/BARTpho | ['Dat Quoc Nguyen', 'Duong Minh Le', 'Nguyen Luong Tran'] | 2021-09-20 | null | null | null | null | ['punctuation-restoration'] | ['natural-language-processing'] | [-6.14140853e-02 2.26376146e-01 -1.00846432e-01 -2.51814336e-01
-1.26822615e+00 -5.68431973e-01 6.23155653e-01 -4.70293909e-02
-6.42240524e-01 1.11594415e+00 1.05332708e+00 -2.09183246e-01
5.68718016e-01 -5.58247924e-01 -8.82339001e-01 -6.37845576e-01
2.53283054e-01 8.22331071e-01 -1.23863980e-01 -5.12318313e-01
-4.16500773e-03 -2.68848464e-02 -9.03407156e-01 5.30589402e-01
1.10458207e+00 1.08150184e-01 5.14105618e-01 9.51798797e-01
2.58628130e-01 7.76428640e-01 -9.47065353e-01 -8.08022857e-01
1.27511218e-01 -8.33639383e-01 -7.56279290e-01 -4.03173357e-01
3.10600579e-01 -3.90236288e-01 -5.52573323e-01 8.84191871e-01
1.13150823e+00 2.57822663e-01 6.63745105e-01 -8.66360068e-01
-1.26368761e+00 1.36035359e+00 -2.79421449e-01 3.71888250e-01
3.31702501e-01 3.52532178e-01 1.16750073e+00 -1.19222820e+00
1.10670662e+00 1.36748457e+00 9.69314635e-01 6.88198268e-01
-1.08107805e+00 -3.95537376e-01 -2.66861141e-01 4.29144859e-01
-1.37409556e+00 -7.41244555e-01 4.10541862e-01 -6.57843575e-02
1.65305507e+00 2.66167432e-01 6.83313489e-01 1.63914180e+00
4.14904296e-01 1.36061251e+00 7.14210689e-01 -4.34435219e-01
1.47557007e-02 -2.84815371e-01 8.19872096e-02 5.50597370e-01
-8.18519108e-03 6.06539920e-02 -6.02578342e-01 1.43714309e-01
5.78592241e-01 -7.46626377e-01 -3.72863919e-01 6.05201066e-01
-1.41232586e+00 1.05717206e+00 2.34252080e-01 4.57212061e-01
-7.09469557e-01 1.85184896e-01 7.28160322e-01 3.23000848e-01
5.64717233e-01 4.99351412e-01 -2.77865797e-01 -5.68080902e-01
-9.70005631e-01 4.30246502e-01 1.02091181e+00 1.18860412e+00
4.30600047e-01 5.03477037e-01 -5.74530303e-01 1.15746355e+00
-2.30016947e-01 5.18302023e-01 7.19882488e-01 -1.02432919e+00
6.25505805e-01 -2.64731348e-01 -1.59088790e-01 -4.27436739e-01
-7.58881867e-02 -2.09061846e-01 -9.58871722e-01 -6.53115869e-01
9.52307601e-03 -5.37557840e-01 -9.68564987e-01 1.50019908e+00
7.34274834e-02 -3.96891236e-01 3.14498425e-01 6.17219210e-01
1.21410060e+00 1.68438387e+00 -1.02911443e-01 -3.81285548e-01
9.99786317e-01 -1.57554090e+00 -1.18765652e+00 -3.24083924e-01
7.02463806e-01 -8.80941927e-01 1.18044817e+00 2.73090363e-01
-1.31292856e+00 -6.83456421e-01 -8.71213257e-01 -6.15837872e-01
-2.53891289e-01 3.29939216e-01 3.52665007e-01 3.20182025e-01
-1.24030006e+00 6.57077849e-01 -9.72341299e-01 -5.04790843e-01
-7.64103681e-02 -2.04455465e-01 -1.62171692e-01 -1.23304039e-01
-1.45381415e+00 1.00186718e+00 1.13348234e+00 1.96033329e-01
-1.13475013e+00 -7.88960516e-01 -1.06975281e+00 2.00829506e-01
2.34578177e-01 -7.59804666e-01 1.63232851e+00 -5.65854788e-01
-1.54382706e+00 5.50271153e-01 -3.53480816e-01 -7.80345142e-01
5.82588911e-01 -5.09399652e-01 -4.30877745e-01 -3.06688640e-02
3.75765979e-01 1.01648891e+00 5.80702603e-01 -1.01945710e+00
-5.30006647e-01 2.55133301e-01 -6.20120645e-01 3.89645725e-01
-3.51215824e-02 1.41376361e-01 -5.57036638e-01 -9.22770023e-01
-6.68579280e-01 -7.31646001e-01 3.53841595e-02 -9.60706174e-01
-7.79194176e-01 -5.96144676e-01 5.83088636e-01 -1.50073206e+00
1.40608394e+00 -1.78085566e+00 2.19550356e-01 -3.59034419e-01
-2.97201365e-01 6.07864976e-01 -5.25982082e-01 1.08008468e+00
1.33792475e-01 1.91163123e-01 -3.54004979e-01 -5.66835761e-01
1.90885261e-01 5.41991949e-01 -2.55802631e-01 1.87092096e-01
2.89440453e-01 1.46125221e+00 -8.95566940e-01 -5.69404960e-01
-6.47004992e-02 6.64626658e-01 -5.87551653e-01 3.16639334e-01
-3.31459224e-01 4.19164658e-01 2.38302946e-01 5.82764685e-01
3.61165643e-01 2.88044184e-01 2.99908400e-01 2.07612440e-01
-2.43773133e-01 6.94734573e-01 -3.19237590e-01 1.78235149e+00
-3.27514619e-01 8.12006712e-01 -1.63810015e-01 -7.47474790e-01
7.63518214e-01 5.48159540e-01 -3.98408771e-02 -7.35243618e-01
1.36933655e-01 3.82315755e-01 -1.28600504e-02 -4.48263407e-01
1.13949025e+00 -3.40935998e-02 -2.52133995e-01 2.88233757e-01
4.93111044e-01 -3.04902017e-01 7.59677470e-01 4.57374066e-01
8.39120030e-01 3.17954600e-01 8.48598838e-01 -3.17204177e-01
2.47447997e-01 1.17995195e-01 8.62241209e-01 9.59996700e-01
-1.37620538e-01 7.92717695e-01 5.25692761e-01 -7.02023581e-02
-1.49954844e+00 -1.10831583e+00 2.23272517e-01 1.23964930e+00
-5.57244003e-01 -5.45145154e-01 -1.22413135e+00 -6.24115288e-01
-2.48922750e-01 1.24601161e+00 -6.71220005e-01 2.50460953e-01
-1.17128468e+00 -8.75219882e-01 9.80955660e-01 7.36508131e-01
5.64527392e-01 -1.52121603e+00 1.05238505e-01 4.93631244e-01
-7.07216144e-01 -1.13846922e+00 -1.10591900e+00 -8.18333402e-02
-4.22810435e-01 -4.34754997e-01 -1.14734912e+00 -1.15774250e+00
6.54960498e-02 -1.25377655e-01 1.30081677e+00 -2.05379277e-01
7.45715350e-02 -9.77900028e-02 -7.90683568e-01 -5.70446372e-01
-1.10786569e+00 6.47287071e-01 -1.52835429e-01 -4.36333835e-01
1.04889296e-01 -4.62935805e-01 -2.34177172e-01 -3.35103214e-01
-7.74340510e-01 1.92466110e-01 7.45469809e-01 1.14960384e+00
5.06371617e-01 -3.68501723e-01 8.35072994e-01 -9.96136069e-01
8.91307116e-01 -4.63005960e-01 -3.12078178e-01 2.50448197e-01
-2.85266191e-01 -1.59336045e-01 8.02950382e-01 -1.90451190e-01
-1.31711245e+00 -3.69468153e-01 -1.02923369e+00 -1.72197800e-02
4.98274490e-02 7.80055225e-01 -2.99409449e-01 7.40837276e-01
6.03145659e-01 5.12561738e-01 -3.93021792e-01 -6.10371172e-01
6.97448254e-01 7.72299528e-01 1.03846014e+00 -3.52000266e-01
4.29440707e-01 -4.74791117e-02 -8.32428098e-01 -1.18329096e+00
-8.78099084e-01 -3.36063266e-01 -5.49492478e-01 8.60475674e-02
1.09114742e+00 -1.02739370e+00 -2.86607236e-01 4.48911309e-01
-1.60053670e+00 -7.38915563e-01 -4.66243833e-01 2.28983566e-01
-6.17731988e-01 5.34187019e-01 -1.30139267e+00 -4.35185611e-01
-9.88911569e-01 -9.10990417e-01 1.11073339e+00 9.18286294e-02
-3.25397909e-01 -1.17095971e+00 4.57718223e-01 4.22139823e-01
2.65178382e-01 1.42485663e-01 8.21075916e-01 -7.75183380e-01
-1.50030151e-01 4.02400196e-02 3.66262048e-02 6.12296224e-01
-2.92156368e-01 -2.21181735e-02 -7.83454239e-01 -5.19395709e-01
-1.00363210e-01 -3.93273115e-01 1.24020314e+00 5.19710064e-01
5.70320010e-01 -8.40184450e-01 7.85631463e-02 6.17153168e-01
1.16547501e+00 2.30795845e-01 8.88384759e-01 2.08670333e-01
7.11847723e-01 4.12854373e-01 5.31131983e-01 3.08478922e-01
5.97119808e-01 4.02858406e-01 -6.46573529e-02 -2.11522765e-02
-5.39723039e-01 -7.79069543e-01 9.14515913e-01 1.99893641e+00
-7.14471042e-02 -8.74347925e-01 -8.38788748e-01 8.62672508e-01
-1.93422520e+00 -1.09043396e+00 -1.54849842e-01 1.73899114e+00
1.15866399e+00 -2.15617478e-01 2.11405203e-01 -3.36320728e-01
8.00358891e-01 4.93494987e-01 -6.40434921e-02 -8.66043448e-01
-5.35457134e-01 3.42163295e-01 2.92433798e-01 8.63255560e-01
-9.95907605e-01 1.43989134e+00 6.15907001e+00 1.25928235e+00
-7.27434278e-01 6.55228853e-01 3.95232946e-01 5.03370836e-02
-2.50226587e-01 -3.64578553e-02 -9.84904945e-01 3.77229005e-01
1.36583257e+00 -2.47302145e-01 4.55750793e-01 6.28073514e-01
2.33609691e-01 4.83394414e-02 -1.08034980e+00 6.13716364e-01
2.86329210e-01 -1.51662648e+00 8.66406113e-02 -1.19028546e-01
1.10781670e+00 4.87462670e-01 -3.20978940e-01 8.07647884e-01
6.47605360e-01 -1.10556185e+00 8.78375351e-01 1.32691368e-01
6.52998388e-01 -7.68893898e-01 9.08332944e-01 4.02754068e-01
-9.11100686e-01 2.92868912e-01 -5.58380127e-01 2.50802785e-01
7.74997532e-01 4.86864001e-01 -1.26056123e+00 8.06953549e-01
3.88275802e-01 9.23338711e-01 -4.14503574e-01 7.42240489e-01
-8.02405477e-01 1.23838758e+00 -3.74288261e-02 -2.48684317e-01
5.46282768e-01 -9.34458822e-02 8.11391056e-01 1.97069895e+00
3.40854526e-01 -1.18789874e-01 1.74131587e-01 7.82816947e-01
-3.25394958e-01 2.72119612e-01 -3.88365746e-01 -4.68055964e-01
4.87283498e-01 9.52140272e-01 -3.63746315e-01 -7.82060027e-01
-1.30398333e-01 1.38583744e+00 4.26233739e-01 4.69070852e-01
-8.67416203e-01 -7.35658288e-01 1.61312804e-01 -1.91996396e-01
5.48623085e-01 -3.81161839e-01 -2.87901517e-02 -1.34767985e+00
-2.81277537e-01 -1.12637091e+00 3.59447926e-01 -8.52387011e-01
-1.21794629e+00 7.82627642e-01 -6.93085417e-02 -6.19695961e-01
-6.82166636e-01 -2.40340278e-01 -6.56730711e-01 7.78881550e-01
-1.32299566e+00 -1.41495574e+00 7.69230425e-02 3.27844262e-01
1.29399681e+00 -6.79136813e-02 8.03674340e-01 1.68651491e-01
-7.63191760e-01 5.02240956e-01 4.87254202e-01 4.29634362e-01
7.70393491e-01 -1.37468410e+00 1.17453432e+00 1.25022304e+00
2.14755133e-01 3.71517390e-01 8.68583322e-01 -1.01941550e+00
-1.06901312e+00 -1.26276362e+00 1.49749398e+00 -2.77290016e-01
6.19229317e-01 -6.02595747e-01 -9.57017303e-01 1.17926061e+00
1.27872181e+00 -7.24825025e-01 4.00127083e-01 -1.06722070e-02
-8.68965685e-02 2.91196436e-01 -6.56208158e-01 8.02524269e-01
1.04377043e+00 -3.65747333e-01 -9.64828432e-01 6.10518515e-01
1.15719962e+00 -4.67020273e-01 -9.89933074e-01 1.55153319e-01
3.37783337e-01 -9.58734691e-01 5.85909009e-01 -4.83482152e-01
6.40002429e-01 2.09507719e-01 -5.29188290e-02 -1.96739531e+00
-4.63469386e-01 -1.01461554e+00 4.52173613e-02 1.58441782e+00
6.12943649e-01 -6.67943358e-01 3.19183141e-01 -4.00549442e-01
-7.55366087e-01 -4.29424644e-01 -8.77686381e-01 -1.05649865e+00
4.63282466e-01 -1.35230690e-01 4.61745530e-01 8.56647432e-01
-2.45412648e-01 7.46411026e-01 -7.64678657e-01 -2.51564324e-01
3.60068858e-01 -2.50478387e-01 7.03348875e-01 -4.57673132e-01
-5.15477777e-01 -2.52993405e-01 2.96895206e-01 -9.37705576e-01
4.84372646e-01 -1.15524256e+00 5.30342758e-01 -1.73402154e+00
2.53534585e-01 4.44765568e-01 2.97748834e-01 6.61440611e-01
-3.77268463e-01 2.84681976e-01 4.52807277e-01 1.84548765e-01
-3.18575174e-01 1.06115615e+00 1.14418137e+00 -1.14706427e-01
-4.02903408e-01 -5.84797919e-01 -5.41607380e-01 3.07348698e-01
1.00797117e+00 -4.51140791e-01 2.21570637e-02 -8.29571128e-01
-1.27953202e-01 1.80762976e-01 -7.00446367e-02 -7.48710036e-01
-1.85630228e-02 1.15679286e-01 2.76475579e-01 -8.78979921e-01
3.45735878e-01 3.89905721e-01 7.11141452e-02 5.65231383e-01
-3.09296310e-01 4.82207447e-01 1.11633897e-01 5.38946092e-02
-3.65085125e-01 -3.73734266e-01 6.18390858e-01 -3.71132851e-01
-7.70946801e-01 -1.85943898e-02 -7.06534564e-01 4.39351887e-01
4.03953463e-01 1.57183588e-01 -6.80278540e-01 -8.02127838e-01
-5.08808196e-01 2.49564767e-01 2.64225543e-01 3.18119019e-01
5.59729040e-01 -1.17618001e+00 -1.44709659e+00 2.49833800e-02
-1.59441054e-01 -1.16035379e-01 2.39796788e-01 8.78588736e-01
-9.65323806e-01 8.24163973e-01 -1.19140208e-01 -1.95401311e-01
-1.10894716e+00 5.40989876e-01 -1.20900825e-01 -5.23530424e-01
-6.67138517e-01 7.72129059e-01 1.65228352e-01 -7.37170875e-01
5.34252338e-02 -1.68213118e-02 9.75752994e-02 1.42515942e-01
5.31610131e-01 6.04206145e-01 6.29663765e-02 -7.49464035e-01
1.98001973e-03 -1.10738188e-01 -2.47194380e-01 -5.05576968e-01
1.46011770e+00 -2.24854976e-01 -2.14407563e-01 3.60931486e-01
1.14207375e+00 7.99690336e-02 -1.00296509e+00 4.21474949e-02
-4.35168631e-02 3.49141397e-02 -3.09759825e-01 -9.19398606e-01
-6.71131670e-01 8.36712897e-01 -2.20660910e-01 -2.98713557e-02
9.37161863e-01 -1.02643266e-01 1.34722281e+00 4.66897756e-01
2.41788533e-02 -1.31763458e+00 -2.52900925e-02 1.05138099e+00
1.35916853e+00 -9.27388787e-01 -1.40553117e-01 -2.84318894e-01
-1.14550292e+00 7.60651171e-01 3.47856104e-01 -2.02946097e-01
5.45843281e-02 1.55962378e-01 2.05472678e-01 3.24451506e-01
-8.69575739e-01 -1.64081231e-01 1.54444695e-01 5.66337645e-01
5.90152264e-01 1.80211619e-01 -7.89987206e-01 5.09654939e-01
-9.55222428e-01 -1.98862776e-01 7.99137235e-01 7.14924932e-01
-3.46027434e-01 -1.12651396e+00 -2.99837232e-01 5.89398593e-02
-5.63594103e-01 -7.19561517e-01 -4.85395521e-01 9.56726730e-01
-2.22672969e-01 8.63327563e-01 9.19558667e-03 -1.45061105e-01
1.19824886e-01 2.37413809e-01 3.70378643e-01 -6.38177752e-01
-8.65901947e-01 5.76765835e-01 6.26255572e-01 -3.93965125e-01
-1.30063713e-01 -7.62880981e-01 -1.02494299e+00 -5.32678485e-01
-1.67111307e-02 4.03527856e-01 5.17441869e-01 6.79273665e-01
2.31039837e-01 5.44128060e-01 3.51109684e-01 -1.09605098e+00
-5.70374727e-01 -1.52599788e+00 -4.45666015e-01 1.93398461e-01
2.60984629e-01 1.56300500e-01 3.55655923e-02 1.08518213e-01] | [12.149945259094238, 9.518951416015625] |
94cba812-a8ca-4168-9a44-cafcfdcb2920 | malibo-meta-learning-for-likelihood-free | 2307.03565 | null | https://arxiv.org/abs/2307.03565v1 | https://arxiv.org/pdf/2307.03565v1.pdf | MALIBO: Meta-learning for Likelihood-free Bayesian Optimization | Bayesian optimization (BO) is a popular method to optimize costly black-box functions. While traditional BO optimizes each new target task from scratch, meta-learning has emerged as a way to leverage knowledge from related tasks to optimize new tasks faster. However, existing meta-learning BO methods rely on surrogate models that suffer from scalability issues and are sensitive to observations with different scales and noise types across tasks. Moreover, they often overlook the uncertainty associated with task similarity. This leads to unreliable task adaptation when only limited observations are obtained or when the new tasks differ significantly from the related tasks. To address these limitations, we propose a novel meta-learning BO approach that bypasses the surrogate model and directly learns the utility of queries across tasks. Our method explicitly models task uncertainty and includes an auxiliary model to enable robust adaptation to new tasks. Extensive experiments show that our method demonstrates strong anytime performance and outperforms state-of-the-art meta-learning BO methods in various benchmarks. | ['Joaquin Vanschoren', 'Felix Berkenkamp', 'Stefan Falkner', 'Jiarong Pan'] | 2023-07-07 | null | null | null | null | ['meta-learning', 'bayesian-optimization'] | ['methodology', 'methodology'] | [ 4.77774478e-02 -4.19505000e-01 -2.81817555e-01 -5.66914797e-01
-1.24743688e+00 -3.98602486e-01 5.04556000e-01 1.78641260e-01
-7.02893615e-01 8.39246154e-01 8.56622607e-02 1.41564474e-01
-3.58601809e-01 -2.92692631e-01 -8.23784590e-01 -5.33305943e-01
1.55986130e-01 7.03230023e-01 3.01289320e-01 -3.58387292e-03
2.29369074e-01 -7.17835426e-02 -1.46858394e+00 3.30911160e-01
1.28885686e+00 9.98571098e-01 5.73267996e-01 5.14780045e-01
-2.10030302e-01 5.14106274e-01 -8.56900394e-01 -4.02856588e-01
2.44032249e-01 2.22444266e-01 -7.38439441e-01 -2.99016833e-01
3.33422124e-01 -5.03077060e-02 -2.31075734e-01 1.05090618e+00
3.52732599e-01 4.87961203e-01 6.65779412e-01 -1.42790806e+00
-6.55546486e-01 4.05277997e-01 -3.57499123e-01 2.49082968e-01
1.53146639e-01 5.99101409e-02 9.90288496e-01 -9.36291456e-01
2.02351987e-01 1.31712973e+00 8.25009048e-01 5.26323259e-01
-1.52180195e+00 -5.98876357e-01 7.44886935e-01 3.83780301e-01
-1.15287149e+00 -3.87667209e-01 5.59237421e-01 -4.16329384e-01
9.79583800e-01 5.01466207e-02 1.58021122e-01 1.37661552e+00
2.33058527e-01 1.06820881e+00 1.16775990e+00 -1.86666265e-01
3.91778469e-01 3.34939718e-01 2.71888375e-01 2.92246908e-01
2.54676342e-01 8.27494264e-02 -8.61557007e-01 -4.11468416e-01
3.20695993e-03 3.59175593e-01 -9.12791118e-02 -5.18420756e-01
-1.37413502e+00 6.58245742e-01 3.48901570e-01 5.06132245e-02
-3.62355649e-01 3.45724463e-01 4.53711122e-01 2.03388408e-01
6.23887718e-01 7.13608801e-01 -8.28227878e-01 -2.49239847e-01
-8.19618821e-01 5.86899042e-01 8.15776706e-01 1.11750054e+00
8.33472848e-01 -1.34191811e-01 -6.37782037e-01 9.31932390e-01
3.03868294e-01 4.11702543e-01 5.63233674e-01 -1.07611716e+00
7.38468528e-01 3.90220642e-01 2.29739547e-01 -5.71227789e-01
-3.04709733e-01 -5.41338444e-01 -4.87468183e-01 -6.33617267e-02
3.55242431e-01 -1.37078390e-01 -9.44253743e-01 1.77633631e+00
3.25621128e-01 -7.40351668e-03 -9.28486288e-02 5.38123369e-01
4.77755368e-01 4.94858086e-01 1.72784731e-01 -1.50772139e-01
1.09568715e+00 -1.31032336e+00 -8.08656812e-01 -6.64225876e-01
4.44638520e-01 -6.79652572e-01 1.52627051e+00 6.13699019e-01
-9.43708062e-01 -5.58665335e-01 -9.32296991e-01 -5.18228225e-02
-3.56454223e-01 -2.22802252e-01 6.51803792e-01 7.42542148e-01
-6.63940728e-01 4.22968507e-01 -8.36699665e-01 -3.07452399e-02
5.67547500e-01 3.25672239e-01 -1.88301150e-02 -4.40054715e-01
-1.01237702e+00 9.25251663e-01 5.66188693e-01 1.74108986e-02
-1.32949555e+00 -9.58826542e-01 -5.83197176e-01 9.18057039e-02
9.39210832e-01 -9.28323686e-01 1.65783226e+00 -6.65584803e-01
-1.44211709e+00 1.49840996e-01 -5.27130306e-01 -3.61624628e-01
5.36051691e-01 -7.96592295e-01 -1.87895626e-01 -4.57230359e-01
3.85944247e-02 4.06231493e-01 1.22130454e+00 -1.28050995e+00
-9.22027290e-01 -5.28666675e-01 1.24746062e-01 2.71171242e-01
-4.81216460e-01 -2.41668895e-01 -4.25648183e-01 -7.20243037e-01
-7.69552886e-02 -9.01284754e-01 -2.18969047e-01 -3.58224124e-01
-3.12252671e-01 -4.76187766e-01 7.17435122e-01 -3.70349973e-01
1.25475812e+00 -2.04880548e+00 2.91690707e-01 -1.50329918e-01
3.17523032e-01 1.27784654e-01 -1.56100571e-01 1.11202277e-01
5.30551195e-01 1.26645327e-01 -9.85866040e-02 -7.28505850e-01
4.66303304e-02 2.56919354e-01 -2.25392118e-01 1.79413483e-02
-7.39534125e-02 8.37609589e-01 -1.22699320e+00 -4.93486226e-01
-1.12390094e-01 1.90025255e-01 -5.49379170e-01 4.24777299e-01
-6.49937034e-01 5.08424401e-01 -6.69096529e-01 7.54098177e-01
4.86169726e-01 -3.31940770e-01 2.05848347e-02 -7.53232986e-02
2.98375487e-01 4.57644850e-01 -9.73026395e-01 2.06995249e+00
-8.50529790e-01 3.84260446e-01 -7.16960132e-02 -1.06591594e+00
7.77099788e-01 2.43575901e-01 3.92875522e-01 -2.72246331e-01
-1.56979337e-01 2.00939238e-01 -1.38490021e-01 -4.74742830e-01
4.07864481e-01 7.12040365e-02 -5.56474328e-02 3.72349799e-01
1.54199287e-01 -3.38254213e-01 8.99653882e-02 -3.13356556e-02
1.02640355e+00 3.11264217e-01 1.68327764e-01 -9.51716024e-03
2.01482311e-01 4.69781123e-02 8.27876806e-01 1.30021524e+00
-3.43496561e-01 4.33295310e-01 1.33832738e-01 -6.05388165e-01
-6.10650539e-01 -1.13872182e+00 -1.30960708e-02 1.77380574e+00
4.64545563e-02 -4.94928151e-01 -4.84271318e-01 -1.18018472e+00
3.09612453e-01 8.80059421e-01 -6.31487548e-01 -3.69094044e-01
-1.69936895e-01 -1.03440535e+00 1.27900630e-01 6.39795840e-01
4.19048101e-01 -7.44778991e-01 -4.76830333e-01 4.24496174e-01
-5.28683603e-01 -1.02835417e+00 -5.59494317e-01 3.07793707e-01
-1.23389435e+00 -8.78098011e-01 -5.87630689e-01 -3.85098189e-01
3.57514441e-01 5.25251031e-01 1.27992213e+00 -5.30180871e-01
-6.87936321e-02 4.51564759e-01 -1.04407705e-01 -8.71862888e-01
-1.82222202e-01 3.99067998e-01 1.32538125e-01 -3.50919403e-02
5.85269749e-01 -4.66781884e-01 -4.21834409e-01 4.42051530e-01
-6.59877777e-01 -2.24442452e-01 6.62517905e-01 1.16728497e+00
7.43301630e-01 4.80291154e-03 7.59783030e-01 -1.18059134e+00
8.84193361e-01 -6.51098132e-01 -7.19960392e-01 8.55887532e-01
-1.15827751e+00 4.84034419e-01 4.73006696e-01 -8.06764543e-01
-1.48606133e+00 1.48393055e-02 5.73007822e-01 -3.78220171e-01
-9.05283820e-03 6.11212850e-01 -8.69504288e-02 2.55841225e-01
8.47401559e-01 4.44042943e-02 -2.60297000e-01 -7.28613973e-01
3.81539345e-01 7.62920558e-01 3.95271957e-01 -9.74203348e-01
4.65187579e-01 4.65758264e-01 -9.94146615e-02 -3.00046653e-01
-1.42734444e+00 -6.34060919e-01 -4.97084767e-01 3.21624018e-02
5.38106918e-01 -1.01821744e+00 -6.40665054e-01 4.86486286e-01
-1.13723814e+00 -4.07360762e-01 -1.53020054e-01 5.74514925e-01
-5.45737922e-01 1.32588059e-01 -1.59637958e-01 -8.28825653e-01
-2.84316659e-01 -1.34862316e+00 9.53330100e-01 1.19445726e-01
-1.03862077e-01 -1.11538172e+00 7.42395297e-02 6.63284659e-01
5.83851755e-01 -1.50464252e-01 8.83045256e-01 -8.23856413e-01
-6.68903589e-01 -2.88952410e-01 -1.26464278e-01 4.68225777e-01
1.98800698e-01 -4.35926527e-01 -1.13655388e+00 -4.24653113e-01
2.91143805e-01 -6.29088283e-01 1.11455917e+00 4.46305752e-01
1.46615374e+00 -2.14739501e-01 -3.91577870e-01 6.69032812e-01
1.08422446e+00 -7.73527250e-02 8.50010216e-02 5.22896171e-01
6.74502313e-01 5.46886683e-01 8.77749383e-01 5.15570223e-01
5.92244625e-01 7.44671643e-01 4.48924273e-01 5.17426431e-01
1.07119694e-01 -2.10053727e-01 5.73479712e-01 5.91943979e-01
6.83347359e-02 -7.06873974e-03 -9.30230439e-01 5.94754994e-01
-2.18962908e+00 -4.84734505e-01 1.32305339e-01 2.61768103e+00
1.05237067e+00 2.51653165e-01 7.59039819e-02 -2.84381926e-01
4.65299338e-01 6.49780110e-02 -9.56790507e-01 -7.69050494e-02
1.55395091e-01 -8.49200711e-02 3.66728216e-01 3.03673953e-01
-1.07935047e+00 8.64225566e-01 6.99732351e+00 7.63658881e-01
-7.04055548e-01 6.54233217e-01 3.57316315e-01 -5.26649356e-01
-2.42988124e-01 6.42856676e-03 -1.21403539e+00 3.90213668e-01
7.14694738e-01 -2.52408445e-01 6.36374772e-01 1.09653807e+00
-2.21847266e-01 -1.65908739e-01 -1.46282041e+00 9.17011380e-01
9.64802355e-02 -8.53109360e-01 -1.67377934e-01 -1.79093137e-01
1.05041862e+00 3.56418192e-01 3.08416367e-01 7.07479715e-01
6.92496359e-01 -9.67255175e-01 5.99561334e-01 6.27478242e-01
3.33713949e-01 -5.14787495e-01 7.21112728e-01 6.05050981e-01
-8.84232879e-01 -5.19652545e-01 -5.76446235e-01 1.50598615e-01
-8.77048746e-02 6.86370850e-01 -9.89725828e-01 3.04236114e-01
1.02759135e+00 5.98016620e-01 -8.56377006e-01 1.02878535e+00
-3.39663386e-01 5.83369315e-01 -1.66543663e-01 -2.43801810e-03
4.29431014e-02 2.06825331e-01 7.49318361e-01 9.58074331e-01
2.78180480e-01 -6.09011114e-01 3.73764515e-01 1.07385421e+00
-6.33742139e-02 -5.95206730e-02 -6.24857187e-01 2.80616432e-02
8.51747632e-01 1.07012475e+00 1.00772560e-01 -2.04442888e-01
-5.73832929e-01 7.92853951e-01 6.96703136e-01 6.84319496e-01
-5.93904078e-01 -2.59815782e-01 7.17828870e-01 -2.07567528e-01
1.55200556e-01 -2.55655140e-01 -4.02903110e-01 -1.22370708e+00
1.37748569e-01 -9.57742155e-01 7.56698668e-01 -5.32087564e-01
-1.61663163e+00 3.39314491e-01 3.59132499e-01 -8.33413064e-01
-4.36290532e-01 -4.39564496e-01 -2.96761394e-01 8.40627909e-01
-1.80949032e+00 -1.04395342e+00 -3.80445749e-01 6.39014840e-01
8.39857936e-01 -4.36958730e-01 6.58017516e-01 -1.78951129e-01
-5.47951698e-01 7.09803700e-01 6.02576494e-01 -4.97748584e-01
1.16226292e+00 -1.44220686e+00 2.87478656e-01 5.43145478e-01
2.24224299e-01 7.13632286e-01 5.66271067e-01 -5.86050868e-01
-1.28069687e+00 -1.00529468e+00 8.84791315e-01 -8.69080961e-01
4.27571476e-01 -5.00215352e-01 -1.06392801e+00 7.64387548e-01
-1.64060518e-01 1.19436078e-01 6.47342563e-01 7.81130075e-01
-6.75389409e-01 -5.88860393e-01 -8.88553202e-01 5.10321915e-01
9.06831980e-01 -5.43996990e-01 -7.76917994e-01 5.80920577e-01
9.53010857e-01 -3.00143391e-01 -8.51527870e-01 3.89925659e-01
6.11302912e-01 -9.13723767e-01 1.09473503e+00 -1.01177776e+00
-3.21992696e-03 -9.10667703e-02 -2.63551056e-01 -1.67767429e+00
-2.99730301e-01 -7.25644886e-01 -5.23128867e-01 9.97878194e-01
5.20437717e-01 -6.94376230e-01 6.02680564e-01 9.97161865e-01
-1.84328184e-01 -5.94666779e-01 -8.56790006e-01 -1.25856030e+00
-2.13125739e-02 -7.07190096e-01 7.70645320e-01 6.21286988e-01
-1.41079292e-01 2.93607354e-01 -4.86483335e-01 6.47721514e-02
8.69135499e-01 -8.87388363e-03 8.89848709e-01 -1.49011111e+00
-5.78693032e-01 -2.71805733e-01 1.91942170e-01 -1.09446967e+00
4.08196777e-01 -6.79227948e-01 4.71214265e-01 -1.22687328e+00
3.46749127e-01 -6.34795070e-01 -8.87940228e-01 4.17401373e-01
-7.66324759e-01 -3.95933002e-01 7.38537982e-02 5.32457888e-01
-9.40151334e-01 5.89207172e-01 1.01417053e+00 -2.05400258e-01
-5.04936814e-01 4.62483287e-01 -7.18565524e-01 7.56897151e-01
7.93109119e-01 -9.54570770e-01 -6.34674370e-01 -9.24240351e-01
4.68307614e-01 -3.94712269e-01 1.56314746e-01 -8.64983261e-01
4.48105484e-01 -4.05960619e-01 2.02138528e-01 -4.04004157e-01
3.83779705e-01 -8.13507080e-01 -9.08262953e-02 1.75609574e-01
-5.78169823e-01 -2.45357696e-02 7.75004402e-02 1.17370248e+00
-1.52847350e-01 -5.14607012e-01 4.43607032e-01 -2.18608797e-01
-4.98621911e-01 2.86148757e-01 -5.54607660e-02 4.98041421e-01
7.78798699e-01 1.17900491e-01 -3.00224006e-01 -3.20049077e-01
-6.22713566e-01 5.75370193e-01 1.52730763e-01 5.81740320e-01
5.01846254e-01 -1.20442295e+00 -7.56341159e-01 -6.94614276e-02
5.32531977e-01 4.45113510e-01 5.37872501e-02 8.89196098e-01
2.87641436e-01 5.35962820e-01 1.01709306e-01 -7.78159201e-01
-9.86076236e-01 6.19080305e-01 1.25136048e-01 -6.33997083e-01
-3.72775160e-02 1.03682184e+00 2.69341558e-01 -4.96357888e-01
6.07128918e-01 -2.61373013e-01 9.66905877e-02 -1.46056917e-02
3.65258694e-01 6.75972104e-01 1.65415153e-01 2.54913997e-02
-1.75128102e-01 8.04729685e-02 -5.31453550e-01 -2.29048342e-01
1.18472338e+00 -2.04525128e-01 1.31352320e-01 1.01645935e+00
8.23227465e-01 -3.44688237e-01 -1.69146454e+00 -1.00183916e+00
5.03032923e-01 -8.76074672e-01 2.81269670e-01 -1.02651989e+00
-6.85110033e-01 1.13643384e+00 3.91691267e-01 -1.74953982e-01
9.96762514e-01 -1.35322556e-01 7.71433413e-01 9.75617826e-01
4.11102980e-01 -1.46073079e+00 6.13174379e-01 6.28010213e-01
8.79642129e-01 -1.63777971e+00 1.12329327e-01 -4.67036739e-02
-7.68274903e-01 8.88390064e-01 7.78761983e-01 3.91650468e-01
6.99333131e-01 -9.19786170e-02 -1.12277076e-01 8.44219625e-02
-1.07329977e+00 -3.30743082e-02 6.79204106e-01 9.71255243e-01
2.28749752e-01 -6.73034415e-02 2.36668095e-01 8.11853230e-01
2.42477417e-01 8.00580904e-02 7.29341898e-03 1.06889355e+00
-3.52785677e-01 -1.32294381e+00 -3.24744225e-01 5.33624113e-01
-4.10016984e-01 -1.99101016e-01 -1.88782707e-01 6.01707518e-01
-5.64605407e-02 9.49519455e-01 -2.38745004e-01 -2.31431097e-01
3.29305351e-01 5.04158676e-01 5.58620274e-01 -9.62938488e-01
-6.76347911e-01 -8.01180750e-02 -2.14883983e-01 -6.18093669e-01
-2.06411019e-01 -7.41540730e-01 -6.79994822e-01 2.92451419e-02
-4.98142064e-01 9.46590006e-02 8.59256327e-01 1.02233446e+00
6.79882407e-01 5.26437759e-01 5.34743130e-01 -5.99071503e-01
-1.35390055e+00 -1.12737691e+00 -4.01354939e-01 3.39366704e-01
4.34108794e-01 -9.47365224e-01 -2.41726786e-01 1.59827858e-01] | [9.207237243652344, 3.6400439739227295] |
f210671f-6fed-42ac-ad54-519cc6b74534 | improved-twitter-sentiment-analysis-using | 1711.11081 | null | http://arxiv.org/abs/1711.11081v1 | http://arxiv.org/pdf/1711.11081v1.pdf | Improved Twitter Sentiment Analysis Using Naive Bayes and Custom Language Model | In the last couple decades, social network services like Twitter have
generated large volumes of data about users and their interests, providing
meaningful business intelligence so organizations can better understand and
engage their customers. All businesses want to know who is promoting their
products, who is complaining about them, and how are these opinions bringing or
diminishing value to a company. Companies want to be able to identify their
high-value customers and quantify the value each user brings. Many businesses
use social media metrics to calculate the user contribution score, which
enables them to quantify the value that influential users bring on social
media, so the businesses can offer them more differentiated services. However,
the score calculation can be refined to provide a better illustration of a
user's contribution. Using Microsoft Azure as a case study, we conducted
Twitter sentiment analysis to develop a machine learning classification model
that identifies tweet contents and sentiments most illustrative of
positive-value user contribution. Using data mining and AI-powered cognitive
tools, we analyzed factors of social influence and specifically, promotional
language in the developer community. Our predictive model was a combination of
a traditional supervised machine learning algorithm and a custom-developed
natural language model for identifying promotional tweets, that identifies a
product-specific promotion on Twitter with a 90% accuracy rate. | ['Angela Lin'] | 2017-11-10 | null | null | null | null | ['twitter-sentiment-analysis'] | ['natural-language-processing'] | [-2.12377742e-01 1.60036355e-01 -7.76659250e-01 -2.50882030e-01
-3.78589451e-01 -4.18118924e-01 5.47799706e-01 9.29031670e-01
-4.51384395e-01 1.31501824e-01 6.97856426e-01 -3.71059179e-01
1.80005319e-02 -1.11267543e+00 -1.38825029e-01 -3.81606966e-01
2.68229961e-01 2.62467742e-01 8.27561915e-02 -7.62712836e-01
7.74917603e-01 1.28353506e-01 -1.47335029e+00 4.14055794e-01
7.51442432e-01 1.14810991e+00 3.56693596e-01 1.51368961e-01
-7.04616964e-01 1.14212859e+00 -6.17671132e-01 -6.66451633e-01
-1.01987481e-01 -6.43066987e-02 -3.71992379e-01 -5.52595034e-02
-2.51101285e-01 -3.07728816e-03 5.40725708e-01 9.00861382e-01
1.01870298e-01 -3.86806250e-01 3.21035087e-01 -1.19266462e+00
-6.11571550e-01 1.02148056e+00 -6.14165425e-01 3.42767060e-01
5.16477942e-01 -2.51573473e-01 1.19444156e+00 -4.19174701e-01
7.47110605e-01 9.19087350e-01 4.79658246e-01 -1.89581037e-01
-9.55224872e-01 -7.41494894e-01 -4.11016569e-02 -1.91110596e-01
-8.77734542e-01 1.08316861e-01 8.63149583e-01 -8.32555413e-01
6.95053458e-01 3.31654847e-01 1.09157920e+00 5.90717316e-01
1.36930823e-01 5.26888490e-01 1.01960135e+00 -2.90158629e-01
1.39127225e-01 8.75772953e-01 1.86709210e-01 1.18659809e-01
6.23064153e-02 -6.36977792e-01 -6.17855668e-01 -2.43289396e-01
-4.48812135e-02 7.28703320e-01 2.81676352e-01 4.13628638e-01
-9.20444727e-01 1.38443971e+00 5.91607213e-01 8.07645559e-01
-6.73503995e-01 -4.48109329e-01 2.43413270e-01 2.31643975e-01
1.03052533e+00 7.81187415e-01 -3.19371372e-01 -4.72589552e-01
-7.01178551e-01 1.48661047e-01 1.04002166e+00 4.25186485e-01
1.00918233e+00 -3.75279248e-01 2.02464789e-01 8.27101350e-01
4.49250877e-01 3.27734947e-01 7.56046653e-01 -6.55614495e-01
1.16713718e-01 1.43574703e+00 4.21501696e-02 -1.69978809e+00
-2.34333903e-01 -6.06374502e-01 -2.14851737e-01 -1.89331979e-01
1.63677722e-01 -1.53888389e-01 -6.95061386e-02 1.13189173e+00
2.20448058e-03 -4.90710557e-01 -2.13603944e-01 7.43541479e-01
7.49609828e-01 7.10151434e-01 8.02729577e-02 -3.16950858e-01
1.32541478e+00 -4.52937335e-01 -6.65729046e-01 -2.23966181e-01
8.94711733e-01 -1.00320947e+00 1.27246630e+00 4.17319894e-01
-7.52423644e-01 -2.41447657e-01 -8.87974381e-01 2.37971857e-01
-7.65394568e-01 -4.59796369e-01 1.01271737e+00 6.45714998e-01
-6.32855237e-01 7.68424273e-01 -3.03682357e-01 -3.16563904e-01
4.13055956e-01 7.15314671e-02 1.27610236e-01 2.98505068e-01
-1.27207160e+00 8.92665744e-01 -1.18416280e-01 -4.81802464e-01
7.88914692e-03 -5.93625426e-01 -3.55043709e-01 -1.36965945e-01
2.96000540e-01 -1.88583732e-01 1.08360577e+00 -1.44608200e+00
-7.49139905e-01 9.65634286e-01 -2.66298890e-01 -2.19954118e-01
-1.12024993e-02 6.57309592e-03 -8.02531600e-01 -2.46710330e-01
5.24820089e-01 7.50363916e-02 3.89906704e-01 -8.49125624e-01
-1.15196192e+00 -6.47530198e-01 1.42460734e-01 -2.88262367e-02
-1.14829206e+00 5.42386115e-01 -1.60204306e-01 -2.95035452e-01
1.24163777e-01 -6.60370231e-01 -2.49326333e-01 -9.33219850e-01
-9.12505388e-02 -4.82068539e-01 7.94347227e-01 -4.59348559e-01
1.56224203e+00 -2.12937188e+00 -4.07303929e-01 5.03991246e-01
6.00409806e-01 -8.69020820e-02 4.26614761e-01 6.93254292e-01
5.28169870e-01 6.42820120e-01 5.91359496e-01 2.07863927e-01
-1.96329549e-01 -1.68769121e-01 -1.84206635e-01 -1.34386301e-01
5.22770844e-02 5.53250790e-01 -1.04052687e+00 -2.83116311e-01
-2.40307361e-01 4.75423396e-01 -4.27113295e-01 -2.06213117e-01
-7.09051341e-02 4.70048413e-02 -8.84549916e-01 6.72684073e-01
1.37603909e-01 -6.41957760e-01 -3.38876992e-02 -2.04927325e-02
-5.81336915e-01 4.53587830e-01 -5.59748650e-01 6.04987919e-01
-5.98159075e-01 8.69821131e-01 -8.62670392e-02 -5.34075379e-01
1.40961063e+00 -2.71626115e-01 8.53632152e-01 -8.42913210e-01
4.24650013e-01 1.16606176e-01 -1.63882263e-02 -4.02219832e-01
8.72818828e-01 -2.28961840e-01 -2.42126420e-01 7.65519142e-01
-6.78058505e-01 1.37968510e-01 3.84588122e-01 5.27426600e-01
9.04214978e-01 -5.77071071e-01 3.26986134e-01 -1.27494246e-01
2.51192451e-01 3.07095855e-01 2.10635871e-01 1.31485462e-01
6.14623651e-02 4.08148728e-02 7.78441310e-01 -3.47183168e-01
-7.25360274e-01 -2.48003334e-01 -1.87038183e-02 1.52001595e+00
1.56605065e-01 -6.63198829e-01 -5.51065683e-01 -1.96961150e-01
2.64869690e-01 7.26177216e-01 -4.53952253e-01 8.44492167e-02
1.14614360e-01 -7.46605098e-01 -1.76463261e-01 1.49295986e-01
2.28805825e-01 -9.87951696e-01 -2.63825387e-01 2.56639302e-01
-3.39672565e-01 -7.70758390e-01 -2.68740177e-01 -6.02637343e-02
-6.08879209e-01 -9.41683471e-01 -3.34357023e-01 -6.74601555e-01
6.23564184e-01 5.56678593e-01 1.07128537e+00 1.27821654e-01
1.60389066e-01 5.98667376e-03 -5.30057669e-01 -1.00600314e+00
-5.55311680e-01 2.68425226e-01 -1.07512981e-01 1.90548360e-01
9.55807924e-01 -3.01724762e-01 -3.83458942e-01 6.39342725e-01
-6.91408873e-01 -9.82206240e-02 3.68288100e-01 -2.14969572e-02
2.33024567e-01 2.41951182e-01 8.75405252e-01 -1.24479532e+00
1.23290038e+00 -1.07747686e+00 -1.94008723e-01 -3.36182714e-01
-1.26872444e+00 -3.76526713e-01 3.22220206e-01 -3.54836315e-01
-7.93732166e-01 -3.01444799e-01 -6.25458956e-02 4.11360860e-01
4.32892442e-01 1.24263000e+00 3.25374246e-01 2.32232496e-01
8.48726630e-01 -2.57095039e-01 3.58506233e-01 -2.50271499e-01
1.32415742e-01 1.13446379e+00 -1.52677864e-01 9.26625952e-02
3.10766816e-01 4.55195367e-01 -5.57765245e-01 -7.16719508e-01
-1.07663023e+00 -9.77664292e-01 -1.14189014e-01 -8.23881865e-01
5.95273137e-01 -6.99121714e-01 -1.12271142e+00 2.18211412e-01
-7.46021986e-01 2.77864873e-01 -1.16334081e-01 4.77458686e-01
1.11587688e-01 -2.22079873e-01 -4.45547163e-01 -7.93861926e-01
-2.07444400e-01 -9.28595603e-01 2.83019453e-01 4.01058674e-01
-8.52530360e-01 -1.02040064e+00 -1.36027128e-01 9.67046201e-01
9.01281714e-01 4.11529869e-01 5.95532656e-01 -9.60152745e-01
-1.90486982e-01 -6.63886607e-01 -2.06306249e-01 3.13030958e-01
2.58115292e-01 6.40496239e-02 -6.64770782e-01 1.45884633e-01
1.04065821e-01 1.23896338e-02 4.94210541e-01 3.88219416e-01
7.97125101e-01 -6.34997785e-01 -4.36643213e-01 -2.14103341e-01
1.16703260e+00 2.31223851e-01 4.67831939e-01 9.08072352e-01
5.37699640e-01 9.62698281e-01 9.14176464e-01 7.15031743e-01
6.31775141e-01 3.54397088e-01 6.49474740e-01 -2.36763641e-01
7.72356570e-01 -2.53075480e-01 5.47739148e-01 9.73281860e-01
-4.72743809e-01 1.86446965e-01 -9.33013082e-01 3.30330312e-01
-1.63531709e+00 -9.62726116e-01 -4.72441852e-01 1.82759798e+00
7.43115664e-01 6.83467150e-01 5.45255125e-01 2.83800840e-01
7.87315667e-01 1.43263370e-01 -2.80116469e-01 -6.12297416e-01
-9.69408527e-02 -2.41027832e-01 7.48093724e-01 2.19475567e-01
-4.70708877e-01 6.64401531e-01 5.53912926e+00 6.40546918e-01
-1.42656231e+00 2.21116588e-01 1.08490872e+00 -2.49036521e-01
-6.52307689e-01 -8.92034173e-02 -9.54155564e-01 6.22152567e-01
9.98288810e-01 -6.64382994e-01 2.19516486e-01 1.28985977e+00
7.80783415e-01 -2.99312443e-01 -5.82335114e-01 5.64285100e-01
2.74237961e-01 -1.68074489e+00 -3.90339553e-01 4.53915298e-01
7.09096193e-01 6.73024654e-02 1.88952118e-01 1.09100819e-01
2.42670640e-01 -5.36151767e-01 5.66461325e-01 5.31372607e-01
1.14090860e-01 -8.23690295e-01 8.69640529e-01 5.01546919e-01
-6.89120293e-01 -2.74929881e-01 -5.07149100e-02 -7.57770777e-01
2.38019973e-01 9.95258689e-01 -1.14722943e+00 -2.65590817e-01
7.27185965e-01 9.39928651e-01 -4.94148493e-01 3.77146304e-01
4.25914153e-02 8.30118716e-01 6.60730302e-02 -9.20809150e-01
1.73511088e-01 -3.75832766e-01 1.78680331e-01 1.00686181e+00
3.26265126e-01 -1.40565366e-01 -1.01307020e-01 7.35147357e-01
-1.03213787e-01 6.60588384e-01 -5.49072504e-01 -8.91737938e-01
3.50108206e-01 1.67103934e+00 -1.18920588e+00 -6.64008111e-02
-5.48923850e-01 1.62720576e-01 -6.31363466e-02 -1.84716478e-01
-4.52499330e-01 -3.44932288e-01 5.63327610e-01 1.13928938e+00
-2.02425331e-01 6.56144693e-02 -4.59841222e-01 -5.61243892e-01
-2.47434095e-01 -5.83073497e-01 -1.27365617e-02 -6.63834155e-01
-1.25353658e+00 3.45353842e-01 -6.48537755e-01 -1.02066898e+00
-1.93445832e-01 -3.34192246e-01 -5.93115211e-01 7.19160676e-01
-1.20268083e+00 -9.15085077e-01 -3.36059391e-01 1.90691054e-01
8.66260380e-04 -4.17723835e-01 4.41555440e-01 4.13541317e-01
-1.12145849e-01 9.61725228e-03 -1.68120507e-02 -7.70703852e-02
6.66100860e-01 -9.05653119e-01 -2.95825880e-02 2.01100588e-01
-1.99144304e-01 7.09129989e-01 6.18774891e-01 -7.50727117e-01
-1.21182382e+00 -7.10790753e-01 1.61706853e+00 -6.21336460e-01
1.15656066e+00 5.99963814e-02 -3.81206930e-01 3.12463045e-01
3.49534154e-02 -7.49781132e-01 1.39866424e+00 7.07597256e-01
8.74032360e-03 -4.18700039e-01 -9.62039471e-01 4.63530004e-01
5.52466154e-01 -4.13709283e-01 -3.01055789e-01 4.11850035e-01
7.10097849e-01 2.78815120e-01 -1.13153434e+00 -3.61022502e-01
6.65603399e-01 -1.13860011e+00 6.48649752e-01 -4.70753551e-01
1.02777362e+00 -6.81232214e-02 1.22008314e-02 -1.21187556e+00
-4.36168820e-01 -5.30794561e-01 4.73570883e-01 1.53980041e+00
8.78786147e-01 -7.74164379e-01 8.07947636e-01 1.00087059e+00
1.07032135e-01 -8.66893709e-01 3.09235714e-02 -2.30783105e-01
-4.29521561e-01 -8.26167703e-01 7.47870445e-01 1.40723562e+00
5.29608965e-01 5.42481303e-01 -8.03116560e-02 -4.41764355e-01
2.57111657e-02 1.81192487e-01 8.22275460e-01 -1.47940528e+00
6.84296265e-02 -7.95214176e-01 -4.20448303e-01 -4.62898076e-01
-2.97700882e-01 -9.18837488e-01 -7.53478944e-01 -1.58989656e+00
4.03841496e-01 -6.23549283e-01 -2.48907268e-01 2.90939391e-01
2.18427360e-01 3.95582795e-01 2.11819917e-01 5.78730702e-01
-4.99664128e-01 -4.05628741e-01 1.32659364e+00 -1.40053496e-01
-6.04315281e-01 3.72742563e-01 -1.91630411e+00 9.02549148e-01
7.22952187e-01 -4.00633425e-01 -7.71713927e-02 1.31134301e-01
1.59172964e+00 -2.73519576e-01 -5.37063964e-02 -3.85671318e-01
2.71006525e-01 -3.71045619e-01 4.94955368e-02 -5.12958169e-01
-8.20780769e-02 -6.57511830e-01 8.83949697e-02 3.93169910e-01
-5.11297464e-01 -1.10388994e-01 -2.93373436e-01 1.61027551e-01
-5.81790805e-01 -2.95502573e-01 2.46697456e-01 1.79924741e-02
-6.02351785e-01 -2.00898591e-02 -6.27516806e-01 -2.44543776e-01
9.92128849e-01 -3.81730944e-01 -3.63042980e-01 -8.55002642e-01
-7.02107131e-01 7.83173591e-02 4.85245049e-01 7.02058792e-01
2.58696318e-01 -1.18203139e+00 -8.26711297e-01 -1.12865314e-01
2.81601548e-01 -4.76082563e-01 -1.83939427e-01 8.19594264e-01
-2.39823520e-01 5.79229407e-02 -8.82121474e-02 -4.26150620e-01
-1.22580230e+00 1.50988862e-01 -3.56029183e-01 -1.38201803e-01
1.23252556e-01 6.63185418e-01 -2.81611085e-01 2.46234611e-03
-1.55698538e-01 -1.40526503e-01 -8.81117582e-01 1.00192475e+00
6.78571343e-01 6.59332693e-01 -1.30032077e-02 -8.84898067e-01
-1.96633279e-01 1.41288728e-01 -2.83915382e-02 1.30600676e-01
1.63133562e+00 -1.47019789e-01 -3.82151842e-01 7.15542793e-01
1.29215193e+00 3.59681875e-01 -4.87662673e-01 -7.80777037e-02
2.64436454e-01 -6.53290153e-01 4.33546871e-01 -7.21464276e-01
-1.28486681e+00 3.98514867e-01 6.20494224e-02 1.23795009e+00
9.93306518e-01 3.79677355e-01 8.22298527e-01 5.85190952e-02
1.88746274e-01 -1.44269204e+00 1.47172242e-01 3.33278984e-01
7.88843989e-01 -1.39572644e+00 7.30865598e-02 -3.48815501e-01
-1.09125018e+00 1.10949528e+00 1.29758477e-01 2.14689553e-01
1.04996634e+00 3.66811186e-01 1.55285358e-01 -3.93894821e-01
-3.52794707e-01 -1.54922813e-01 1.60950005e-01 3.63946147e-02
8.22344780e-01 3.66126984e-01 -7.51208186e-01 1.18694699e+00
-6.83400393e-01 5.74989989e-02 5.90833068e-01 7.07422972e-01
-8.53180289e-01 -1.15418291e+00 -1.17656708e-01 9.71660912e-01
-8.88267577e-01 1.80867892e-02 -7.64976919e-01 4.47609067e-01
2.90579855e-01 1.29308820e+00 1.90705597e-01 -9.42817867e-01
1.29483193e-01 -3.19253981e-01 -3.48008007e-01 -5.94297707e-01
-9.60632563e-01 1.48886174e-01 4.24964935e-01 -3.34509671e-01
-7.93757558e-01 -9.88549113e-01 -1.35121107e+00 -7.57683992e-01
-5.18257976e-01 3.65683079e-01 1.32637906e+00 8.55031252e-01
4.50808704e-01 3.88434857e-01 1.15138054e+00 -4.44066644e-01
1.11283213e-01 -9.34324265e-01 -7.28754878e-01 6.35334849e-01
-2.55804151e-01 -2.78281927e-01 -6.35879397e-01 5.13088610e-03] | [10.688695907592773, 6.854066848754883] |
4b828218-d545-4fe3-bc25-51a5ee4027aa | neural-entropic-estimation-a-faster-path-to | 1905.12957 | null | https://arxiv.org/abs/1905.12957v2 | https://arxiv.org/pdf/1905.12957v2.pdf | Neural Entropic Estimation: A faster path to mutual information estimation | We point out a limitation of the mutual information neural estimation (MINE) where the network fails to learn at the initial training phase, leading to slow convergence in the number of training iterations. To solve this problem, we propose a faster method called the mutual information neural entropic estimation (MI-NEE). Our solution first generalizes MINE to estimate the entropy using a custom reference distribution. The entropy estimate can then be used to estimate the mutual information. We argue that the seemingly redundant intermediate step of entropy estimation allows one to improve the convergence by an appropriate reference distribution. In particular, we show that MI-NEE reduces to MINE in the special case when the reference distribution is the product of marginal distributions, but faster convergence is possible by choosing the uniform distribution as the reference distribution instead. Compared to the product of marginals, the uniform distribution introduces more samples in low-density regions and fewer samples in high-density regions, which appear to lead to an overall larger gradient for faster convergence. | ['Ali Al-Bashabsheh', 'Da Sun Handason Tam', 'Hing Pang Huang', 'Chung Chan', 'Chao Zhao', 'Michael Lim'] | 2019-05-30 | null | null | null | null | ['mutual-information-estimation'] | ['methodology'] | [-3.50308069e-03 3.60201508e-01 -7.08789751e-02 -2.69516140e-01
-5.72673619e-01 -3.26668292e-01 4.19632405e-01 -3.55750918e-02
-8.78705859e-01 1.18235278e+00 1.06935248e-01 -2.41281196e-01
-1.67391717e-01 -7.73040414e-01 -6.11829937e-01 -8.20226669e-01
-6.37813658e-02 3.83017987e-01 7.37562999e-02 2.84186900e-02
2.53928274e-01 1.83784530e-01 -1.38137639e+00 -1.38625845e-01
1.01037610e+00 7.70294130e-01 2.06543818e-01 6.46656096e-01
-5.07849492e-02 5.66548884e-01 -5.63925087e-01 -2.91002393e-01
3.66938353e-01 -8.59455228e-01 -8.71415496e-01 -5.10517001e-01
-2.97883544e-02 -3.05525184e-01 -6.26337994e-03 1.35068154e+00
3.23872209e-01 5.08304060e-01 1.04688728e+00 -9.97150779e-01
-2.71890223e-01 7.09787190e-01 -6.36063516e-01 -5.17956680e-04
-9.87063721e-02 -1.83331162e-01 1.15080678e+00 -5.62315702e-01
4.93248135e-01 9.81669307e-01 9.30617630e-01 5.67731261e-01
-1.36065888e+00 -4.09374177e-01 -1.58458158e-01 -1.49541900e-01
-1.49583507e+00 -1.09579191e-01 7.15593040e-01 -8.61138031e-02
8.65705371e-01 2.45633006e-01 6.35198832e-01 7.03771710e-01
1.97383106e-01 7.90128171e-01 1.00971973e+00 -4.93464679e-01
4.31496531e-01 5.27777970e-01 -4.61121351e-02 5.41227579e-01
4.96003896e-01 1.91474095e-01 -8.09413120e-02 -1.86590925e-01
9.43057179e-01 -2.15852574e-01 -4.25483555e-01 -3.83275360e-01
-8.21966529e-01 1.04731548e+00 5.57371199e-01 6.35159314e-01
-2.68600166e-01 9.90983918e-02 2.71128625e-01 4.48253661e-01
5.27434170e-01 8.26264799e-01 -5.90701759e-01 -2.96879619e-01
-1.22974920e+00 2.41858795e-01 1.16783655e+00 3.52196902e-01
1.09144092e+00 -2.14013696e-01 1.57347679e-01 8.65727842e-01
2.80306399e-01 1.41409650e-01 3.89475018e-01 -1.12475157e+00
4.22429830e-01 2.43988201e-01 1.91040397e-01 -8.59685659e-01
-3.61282229e-01 -5.12093246e-01 -8.80052090e-01 1.42672732e-01
8.67428124e-01 -5.19981742e-01 -4.73966688e-01 2.20943403e+00
2.93589421e-02 -3.48396331e-01 1.71899989e-01 7.33551621e-01
2.11121719e-02 6.28692806e-01 -1.06388375e-01 -2.79131711e-01
8.34209144e-01 -6.68575704e-01 -5.29650688e-01 -1.25221565e-01
8.09605896e-01 -3.85963857e-01 7.97935843e-01 1.54242575e-01
-1.06666911e+00 -1.17162950e-01 -1.17687416e+00 8.23982805e-02
-4.43627626e-01 -3.79180051e-02 5.10174930e-01 5.12152314e-01
-1.15947831e+00 1.18115819e+00 -8.18368495e-01 -1.91557392e-01
1.39811411e-01 4.85058784e-01 -3.45118672e-01 5.21899104e-01
-1.34967530e+00 1.10409319e+00 8.39439332e-01 -1.19746201e-01
-9.68805179e-02 -5.35197258e-01 -9.07078087e-01 2.22920969e-01
5.88039635e-03 -5.63092649e-01 1.03078640e+00 -1.36018753e+00
-1.59381378e+00 4.64661717e-01 -1.19112991e-01 -4.64528561e-01
5.98876357e-01 5.61309978e-02 3.31909776e-01 -5.90935983e-02
-1.78769469e-01 8.75211716e-01 4.29697573e-01 -1.14138043e+00
-1.44479841e-01 -9.76864249e-02 -4.26592156e-02 3.18989933e-01
-3.15474391e-01 -4.09080803e-01 -1.18493229e-01 -3.91085416e-01
1.21820182e-01 -8.76237869e-01 -2.52923965e-01 -1.47671729e-01
-2.44211510e-01 -8.70753601e-02 4.64743018e-01 -4.87567544e-01
1.27660286e+00 -2.04122853e+00 8.14305171e-02 5.47961056e-01
3.73472691e-01 7.35732680e-03 4.86532860e-02 2.85276890e-01
-3.61846894e-01 2.11836442e-01 -4.78877842e-01 -2.92399555e-01
-5.53983487e-02 2.54612505e-01 2.24894628e-01 3.47855181e-01
2.31110543e-01 5.66155553e-01 -9.76864636e-01 -4.67617214e-01
-3.31627056e-02 6.27024889e-01 -8.47017407e-01 -6.89287484e-02
1.86484188e-01 3.39840651e-01 -3.14301699e-01 -1.56385466e-01
8.26949060e-01 -4.40654784e-01 2.61777967e-01 2.31242031e-02
-1.71520978e-01 4.23262298e-01 -1.17977083e+00 1.27616704e+00
-6.21540904e-01 9.66983855e-01 1.99071765e-02 -9.68339920e-01
7.44105875e-01 1.54149190e-01 4.34220940e-01 -3.38531643e-01
2.63633490e-01 3.79022181e-01 2.80738056e-01 2.03491375e-02
4.76871550e-01 -3.90411973e-01 1.82176679e-01 6.82533026e-01
1.35074824e-01 -3.69884297e-02 1.46667019e-01 1.19508475e-01
8.56278479e-01 -8.49289540e-03 4.01690215e-01 -5.77044964e-01
2.52819866e-01 -5.81937134e-01 4.48404074e-01 7.60435402e-01
-1.90519378e-01 5.72302520e-01 8.33919168e-01 -2.62502879e-01
-1.29510593e+00 -1.10646355e+00 -3.53806227e-01 6.08744919e-01
3.18541490e-02 -3.84421438e-01 -1.05877686e+00 -8.48466873e-01
-1.40796095e-01 7.40998209e-01 -7.11707294e-01 -2.55016714e-01
-3.02054822e-01 -1.04172802e+00 3.05740327e-01 3.73569548e-01
6.29757583e-01 -9.50882971e-01 -7.05405831e-01 5.80726601e-02
-3.40632737e-01 -3.78197223e-01 -3.87166977e-01 7.53200233e-01
-1.01454437e+00 -7.05063999e-01 -9.81584907e-01 -3.52285713e-01
7.65758634e-01 -3.42267931e-01 1.03808367e+00 -1.22129545e-01
1.01093911e-01 -2.95223743e-02 2.26352606e-02 -1.47970423e-01
-4.93035465e-01 2.50112772e-01 2.47760974e-02 -3.97454113e-01
4.50101674e-01 -7.19365418e-01 -7.54492760e-01 1.54475451e-01
-7.98333108e-01 3.30297984e-02 5.20207226e-01 1.02627075e+00
1.69512570e-01 8.98194388e-02 5.60540140e-01 -6.04180038e-01
7.23974407e-01 -5.57638109e-01 -6.05467319e-01 3.18813557e-03
-7.21721947e-01 8.60735238e-01 5.58903694e-01 -4.22566414e-01
-9.72380519e-01 -1.25074670e-01 -2.59567022e-01 -1.77898243e-01
1.17186330e-01 4.98008847e-01 1.90804631e-01 -7.53013939e-02
6.54612303e-01 -4.18238677e-02 2.54376203e-01 -4.41253424e-01
9.78642777e-02 6.18474305e-01 1.26861125e-01 -3.36206466e-01
3.13709974e-01 1.04640489e-02 -3.84038389e-02 -5.91630816e-01
-6.71918333e-01 -1.05902858e-01 -5.38858473e-01 -1.76601395e-01
7.57988751e-01 -4.81618255e-01 -7.79797018e-01 2.02775031e-01
-1.10919666e+00 -4.10457373e-01 -5.64795136e-01 7.95672238e-01
-6.78790629e-01 3.42994809e-01 -7.89952397e-01 -1.09490585e+00
-2.43253306e-01 -1.13259709e+00 8.01253736e-01 2.68690109e-01
-4.50741738e-01 -1.45057797e+00 2.64157474e-01 -3.66074622e-01
5.06158233e-01 -3.47200334e-02 8.63499880e-01 -8.26804519e-01
-1.85701102e-01 -1.70865208e-01 -3.03641766e-01 5.78241765e-01
2.40474761e-01 -6.67604357e-02 -7.60330796e-01 -2.90013224e-01
2.03463688e-01 -1.44294336e-01 1.14370620e+00 5.52220464e-01
9.22976077e-01 -4.85438854e-01 -2.71961749e-01 5.66756964e-01
1.56341040e+00 1.05927229e-01 7.36734092e-01 2.32103989e-01
3.98877770e-01 5.57537138e-01 1.92137718e-01 3.70946765e-01
1.63451642e-01 3.12072426e-01 -4.31411117e-02 -1.14906415e-01
2.20653027e-01 -2.88661689e-01 2.85926938e-01 9.53692079e-01
-1.39032915e-01 -2.54398659e-02 -7.86494553e-01 4.01930690e-01
-1.69921505e+00 -1.02525365e+00 2.84003109e-01 2.56478286e+00
1.21972775e+00 1.36445761e-01 1.51804686e-01 8.03639814e-02
6.67277753e-01 2.24690251e-02 -4.45312500e-01 -4.91025865e-01
9.09626335e-02 -3.54143269e-02 7.65363395e-01 8.39421749e-01
-9.59346354e-01 5.12949347e-01 7.93339014e+00 9.41017509e-01
-1.00258195e+00 -5.88976219e-02 8.20827246e-01 -1.95858907e-02
-3.51694673e-01 8.78312811e-02 -6.50869727e-01 7.66000271e-01
1.18897295e+00 4.32914756e-02 4.44191098e-01 8.44545484e-01
-1.26463786e-01 -5.98817945e-01 -9.60802495e-01 8.32830310e-01
-3.00996631e-01 -1.00068212e+00 -2.42335036e-01 2.50892460e-01
7.48383939e-01 8.53460208e-02 -1.09237634e-01 1.06476799e-01
4.22312558e-01 -7.39031613e-01 3.40736270e-01 5.72432637e-01
6.97035432e-01 -1.03100288e+00 9.89213526e-01 5.83791256e-01
-9.26368833e-01 4.65766378e-02 -6.09643698e-01 -8.32375959e-02
-1.28323242e-01 8.77352715e-01 -7.48753309e-01 2.13324621e-01
4.29526538e-01 3.87119889e-01 -2.25605369e-01 8.82585287e-01
1.64041400e-01 3.67948234e-01 -8.32639098e-01 -2.48308614e-01
1.99920535e-01 -5.24357557e-01 5.97697377e-01 1.12648439e+00
3.90346199e-01 -3.96866262e-01 -2.57206827e-01 1.17861032e+00
-4.44266088e-02 -4.41965945e-02 -7.12128878e-01 2.75772382e-02
2.55117029e-01 9.67292666e-01 -7.10181117e-01 -3.68444145e-01
-1.66669369e-01 1.02965677e+00 5.98622501e-01 3.99911910e-01
-6.70436621e-01 -8.47292006e-01 3.16425204e-01 -1.55750722e-01
2.85480201e-01 -9.20799375e-02 -1.87648296e-01 -1.16904521e+00
-7.26919472e-02 -4.54521924e-01 1.71006933e-01 -3.58118683e-01
-1.16746926e+00 4.50184524e-01 3.48333478e-01 -9.32005644e-01
-7.21554756e-01 -5.11001825e-01 -6.07648313e-01 1.04903996e+00
-1.29229558e+00 -2.71932334e-01 2.31735915e-01 1.83422655e-01
2.81215440e-02 7.09142387e-02 7.11507678e-01 2.67965287e-01
-3.84537250e-01 8.29697788e-01 5.04331589e-01 1.38343066e-01
5.10102630e-01 -1.56484091e+00 6.89246953e-02 4.83464926e-01
-1.24625057e-01 7.14999676e-01 9.85567153e-01 -5.19794226e-01
-5.68095803e-01 -6.37102962e-01 9.19518292e-01 -2.10954159e-01
5.39341271e-01 -1.74814075e-01 -9.29586828e-01 5.79765439e-01
9.29911658e-02 -2.86126256e-01 5.98001719e-01 1.87765241e-01
-2.84572244e-01 1.68635443e-01 -1.29588807e+00 6.58671260e-01
5.69009602e-01 -5.97823441e-01 -3.44750702e-01 6.15813546e-02
3.55745584e-01 2.62564979e-02 -8.79272938e-01 3.08414578e-01
6.95460439e-01 -1.23401010e+00 4.93496865e-01 -1.04898147e-01
3.52051288e-01 -1.37624741e-01 -4.55000326e-02 -1.37837481e+00
1.78716853e-02 -5.46753526e-01 -2.66778581e-02 9.14083123e-01
7.51004100e-01 -1.00713849e+00 7.85606146e-01 7.81731009e-01
3.38850021e-01 -1.01016772e+00 -1.05630648e+00 -6.95104063e-01
4.21558261e-01 -1.46095574e-01 2.59125859e-01 7.78575778e-01
4.03692037e-01 1.58585072e-01 -3.01676512e-01 -4.45579439e-01
5.15257895e-01 -1.93994552e-01 2.02786341e-01 -1.30669045e+00
-5.90660155e-01 -5.50933063e-01 -3.29015106e-01 -1.30646574e+00
5.25685586e-02 -7.40736604e-01 3.93859714e-01 -1.08687973e+00
4.67902124e-01 -5.35981655e-01 -3.60968739e-01 2.92035371e-01
-1.84522063e-01 1.06444478e-01 1.61201745e-01 1.57325253e-01
-4.07596916e-01 4.93429065e-01 1.11908185e+00 2.77786225e-01
-4.37498778e-01 3.67488787e-02 -5.98114908e-01 7.82309771e-01
9.24825668e-01 -5.94309926e-01 -2.81875372e-01 4.23117355e-03
4.31980580e-01 -3.07939071e-02 9.72820297e-02 -1.05487943e+00
1.27119154e-01 1.60509333e-01 6.27097070e-01 -3.98896247e-01
3.19206476e-01 -5.63244104e-01 9.51369777e-02 4.06836957e-01
-5.65022349e-01 -1.16704039e-01 3.48403938e-02 2.68886268e-01
-2.04613075e-01 -7.14089930e-01 1.03839350e+00 -2.62420774e-01
1.02173932e-01 -1.51704073e-01 -4.41413313e-01 -1.05195023e-01
5.73936343e-01 -3.37590069e-01 1.74498558e-01 -6.63418770e-01
-7.19856203e-01 -1.07140154e-01 6.31144524e-01 -2.61522919e-01
2.62098521e-01 -1.24136186e+00 -3.70322376e-01 2.99415678e-01
-3.99115801e-01 -1.10275418e-01 5.41677186e-03 1.09577596e+00
-3.94478261e-01 2.65385151e-01 -1.17073268e-01 -3.54469538e-01
-6.99626088e-01 2.22225025e-01 7.96456873e-01 -5.67913055e-01
-3.71571898e-01 8.29707384e-01 2.45161697e-01 -4.49020684e-01
1.32172033e-01 -1.62175730e-01 -1.01985717e-02 2.30874214e-02
4.82105166e-01 3.26431513e-01 -1.82684794e-01 -4.53761667e-01
-1.07155263e-01 4.44613606e-01 -2.03436062e-01 -4.90733653e-01
1.07082534e+00 -1.19958773e-01 -1.73746973e-01 6.30098283e-01
1.76084435e+00 -1.09719917e-01 -1.36475015e+00 8.95128772e-02
-5.26981335e-03 -2.75394261e-01 1.64539427e-01 -5.28054237e-01
-1.02005017e+00 8.17218065e-01 6.95157468e-01 2.61086226e-01
9.68660712e-01 -8.84725302e-02 5.69526136e-01 6.07756913e-01
2.01904580e-01 -1.26927125e+00 -3.02346081e-01 6.98775709e-01
4.58229870e-01 -1.11657929e+00 6.66887760e-02 9.90616903e-02
-5.45938671e-01 1.12718046e+00 3.46716881e-01 -1.91737920e-01
7.82353580e-01 2.93044358e-01 -2.09832758e-01 1.39241278e-01
-5.84864497e-01 -6.24465272e-02 2.79090494e-01 2.61561424e-01
7.03875422e-01 -5.87480329e-02 -3.67633075e-01 1.95488170e-01
-2.42609397e-01 -2.04918027e-01 2.65321493e-01 6.28070176e-01
-3.70513618e-01 -1.01477933e+00 -4.40868028e-02 6.77741647e-01
-6.44918442e-01 -3.33541989e-01 -2.79320925e-01 8.79067659e-01
-1.23436749e-01 5.82828283e-01 4.84586716e-01 -2.46689484e-01
-3.41178328e-01 3.04675967e-01 4.60958898e-01 -1.26645565e-01
-3.19454670e-01 -5.31527400e-03 -1.58488795e-01 -4.70802277e-01
-3.72580647e-01 -5.46790481e-01 -1.14144444e+00 -2.97380805e-01
-7.79720008e-01 5.95309079e-01 7.76083887e-01 9.53891277e-01
1.48672625e-01 1.81810185e-01 5.11790514e-01 -1.02044392e+00
-5.19419491e-01 -1.03787851e+00 -7.54300535e-01 2.48532116e-01
5.40708542e-01 -6.36367440e-01 -9.72149372e-01 -5.12888610e-01] | [7.797804832458496, 3.6735384464263916] |
541f5843-9dfa-474b-9699-fb1f227d61ab | intelligent-detect-for-substation-insulator | 2208.14598 | null | https://arxiv.org/abs/2208.14598v1 | https://arxiv.org/pdf/2208.14598v1.pdf | Intelligent detect for substation insulator defects based on CenterMask | With the development of intelligent operation and maintenance of substations, the daily inspection of substations needs to process massive video and image data. This puts forward higher requirements on the processing speed and accuracy of defect detection. Based on the end-to-end learning paradigm, this paper proposes an intelligent detection method for substation insulator defects based on CenterMask. First, the backbone network VoVNet is improved according to the residual connection and eSE module, which effectively solves the problems of deep network saturation and gradient information loss. On this basis, an insulator mask generation method based on a spatial attentiondirected mechanism is proposed. Insulators with complex image backgrounds are accurately segmented. Then, three strategies of pixel-wise regression prediction, multi-scale features and centerness are introduced. The anchor-free single-stage target detector accurately locates the defect points of insulators. Finally, an example analysis is carried out with the substation inspection image of a power supply company in a certain area to verify the effectiveness and robustness of the proposed method. | ['Lihua Wang', 'Huiting Yang', 'Peipei Yan', 'Mingxuan Li', 'Feng Li', 'Bo Ye'] | 2022-08-31 | null | null | null | null | ['defect-detection'] | ['computer-vision'] | [ 1.93507239e-01 -2.88368583e-01 4.52123582e-01 -2.37553075e-01
-3.44099522e-01 -8.58727768e-02 -1.82818770e-01 -3.44656259e-02
-8.10711011e-02 4.98870552e-01 -1.54567972e-01 -2.42422700e-01
-4.04632777e-01 -8.47656965e-01 -9.77926776e-02 -1.11123061e+00
-6.85602650e-02 -9.74805504e-02 1.75382808e-01 -4.37235236e-02
5.59996665e-01 6.12746954e-01 -1.17198467e+00 2.16745958e-01
1.43696022e+00 1.03326607e+00 5.40974796e-01 3.64050627e-01
8.41300786e-02 5.39450109e-01 -9.91704166e-01 2.04170555e-01
2.51619875e-01 -4.48106557e-01 -5.64938784e-01 6.45160079e-01
-3.48174945e-02 -3.96062732e-01 -5.90919256e-01 1.50323486e+00
9.41021860e-01 1.47486702e-01 5.92781067e-01 -9.70114350e-01
-4.52379942e-01 2.97659338e-01 -8.77839446e-01 8.31695020e-01
-7.49301985e-02 3.23830634e-01 8.77528191e-01 -6.80969596e-01
1.82186604e-01 7.24809349e-01 6.99259758e-01 -1.18585743e-01
-7.64723539e-01 -1.85860112e-01 3.52190942e-01 8.47552538e-01
-1.46960425e+00 -1.71633661e-02 1.09802485e+00 -2.67603159e-01
5.47239125e-01 1.41740024e-01 8.56145859e-01 1.23353429e-01
1.83882058e-01 1.02437246e+00 7.09713817e-01 -5.18414915e-01
4.28197198e-02 7.63347819e-02 1.84027821e-01 6.25910282e-01
2.27361441e-01 -1.90268829e-02 1.12093650e-01 5.30732095e-01
7.87600994e-01 1.25370026e-01 -6.38845801e-01 -1.34915277e-01
-6.41345918e-01 5.50332367e-01 7.87247598e-01 7.47298777e-01
-3.46472263e-01 -5.82704425e-01 4.88448143e-01 2.90936619e-01
6.17626369e-01 1.93529576e-01 -4.24727499e-01 2.29743809e-01
-1.10167813e+00 -4.37012672e-01 3.00233364e-01 6.16281569e-01
3.78744662e-01 4.79712546e-01 -2.96666175e-01 8.66680205e-01
2.16621563e-01 2.55791813e-01 1.56202316e-01 -3.34545821e-01
5.26970685e-01 7.47567832e-01 1.21480506e-03 -1.24822450e+00
-8.28805983e-01 -9.98670042e-01 -1.00084066e+00 4.72809494e-01
1.09515721e-02 -4.41457182e-01 -7.89927542e-01 9.35739934e-01
3.50805730e-01 1.93172634e-01 -1.87859997e-01 1.22793067e+00
5.65019310e-01 9.84766006e-01 -2.50057489e-01 -3.68396074e-01
1.12310445e+00 -1.08874154e+00 -9.61069465e-01 -4.02213223e-02
7.50265062e-01 -4.39291209e-01 9.08402443e-01 6.22712731e-01
-1.00420785e+00 -7.27774560e-01 -1.26450658e+00 4.00054634e-01
-3.13914418e-01 5.56242704e-01 4.21978325e-01 4.80077118e-01
-8.00346613e-01 5.84588706e-01 -4.56784278e-01 -1.71564683e-01
6.81650698e-01 1.16520464e-01 1.40023502e-02 -2.81899422e-01
-1.10708869e+00 9.35603857e-01 6.03667796e-01 8.56313348e-01
-8.03895772e-01 -5.16151071e-01 -5.81208169e-01 2.96430558e-01
6.17923252e-02 -2.07837209e-01 7.28031695e-01 -1.12152016e+00
-1.06178176e+00 5.82246900e-01 3.39723200e-01 -2.26287991e-01
5.71797192e-01 7.86926656e-04 -6.93183959e-01 4.36292559e-01
1.56068981e-01 2.32236050e-02 8.02133858e-01 -9.44947481e-01
-1.26205313e+00 -6.08779132e-01 -8.67832154e-02 4.71940607e-01
-3.23740751e-01 4.06037197e-02 -1.91444889e-01 -6.03328228e-01
4.70577568e-01 -1.43041834e-01 -2.70084828e-01 -6.88364580e-02
-4.82208699e-01 -6.86441064e-02 1.31371427e+00 -1.28487098e+00
1.28517306e+00 -2.42256713e+00 6.26234338e-02 4.92521375e-01
3.75219807e-02 3.70995998e-01 -2.86710281e-02 8.51388928e-03
-3.40733767e-01 -4.19830114e-01 -3.71693522e-01 2.40527675e-01
-2.81582922e-01 -2.34310001e-01 3.64874601e-01 8.66735339e-01
1.52504994e-02 4.49891031e-01 -6.80442572e-01 -5.98901212e-01
7.02384055e-01 9.25939456e-02 -2.39387333e-01 1.59519523e-01
2.70287842e-01 6.73733652e-01 -4.36284810e-01 8.59950185e-01
1.08326781e+00 -6.24811091e-02 -2.15761706e-01 -6.15078688e-01
-3.69898707e-01 -2.29362711e-01 -1.21249878e+00 1.55910468e+00
-2.78915256e-01 5.18960476e-01 4.22036260e-01 -1.53782833e+00
1.16384637e+00 3.22006851e-01 6.77412391e-01 -9.37582850e-01
1.81279376e-01 2.64901817e-01 -3.70973870e-02 -8.53562295e-01
8.52475092e-02 2.91444749e-01 2.12682590e-01 -8.95633399e-02
-1.87804565e-01 -9.43336338e-02 2.99563199e-01 -5.51496446e-02
1.00418282e+00 7.23227952e-03 -3.52907538e-01 -3.46687526e-01
8.02470863e-01 -3.53933056e-03 7.37333477e-01 5.09726405e-01
-2.64035165e-01 4.87010270e-01 2.19416887e-01 -5.82618952e-01
-9.26756859e-01 -7.90672541e-01 -3.28994393e-01 3.54612231e-01
5.65370917e-01 3.41658711e-01 -8.16081583e-01 -7.65358329e-01
-2.21598223e-01 5.62800586e-01 -1.86215192e-01 -3.49827647e-01
-4.17132646e-01 -1.13228798e+00 6.51769787e-02 4.55211103e-01
9.16258752e-01 -1.30131066e+00 -2.24260315e-01 4.77708548e-01
-1.05951406e-01 -7.26202130e-01 -2.99241364e-01 1.51033327e-01
-9.31364357e-01 -1.08304417e+00 -8.45472634e-01 -1.46256304e+00
1.18205440e+00 4.57895219e-01 6.16321146e-01 4.62476194e-01
-6.53586686e-01 -2.22995684e-01 -3.33689332e-01 -1.49881229e-01
-4.99386936e-02 -6.44819885e-02 -1.65593281e-01 8.81190076e-02
2.68538415e-01 -6.41599834e-01 -8.01483154e-01 1.78339109e-01
-5.98991156e-01 3.44637148e-02 7.36526191e-01 1.00072706e+00
1.75677344e-01 9.85524476e-01 7.48469889e-01 -3.80481839e-01
5.11270285e-01 -1.91949487e-01 -1.09760880e+00 1.08380578e-01
-5.85473537e-01 -7.51072049e-01 7.52213538e-01 9.16726049e-03
-1.12170398e+00 -1.92655146e-01 -3.89999837e-01 -2.94664472e-01
-2.69991577e-01 6.60295308e-01 -5.08625090e-01 -1.69701338e-01
2.54085571e-01 4.66618240e-01 -1.01655461e-01 -5.20101786e-01
-3.74056362e-02 7.60602832e-01 5.51813662e-01 2.45281190e-01
8.15859139e-01 3.77915114e-01 -2.72256315e-01 -8.85822117e-01
-6.12846315e-01 -4.16700125e-01 -7.68801033e-01 -6.90603554e-01
8.10133874e-01 -7.47682095e-01 -7.62137651e-01 9.40949082e-01
-1.14909232e+00 9.13953334e-02 -3.80094409e-01 6.41732931e-01
-1.82682112e-01 6.17134035e-01 -9.25002158e-01 -7.52023041e-01
-5.37099183e-01 -9.29290712e-01 5.97261250e-01 5.64647615e-01
5.53299248e-01 -8.76500249e-01 -4.92253363e-01 2.82885045e-01
2.31405422e-01 -6.26683757e-02 1.03330767e+00 -3.35717350e-01
-7.48450160e-01 -4.27229077e-01 -5.41879833e-01 7.11391330e-01
3.05856764e-01 -1.40489250e-01 -6.77660704e-01 -5.10890067e-01
3.59247923e-01 2.33078256e-01 5.91754615e-01 1.02881444e+00
1.21362877e+00 -1.12673856e-01 -3.09992850e-01 6.22815669e-01
1.61038125e+00 7.04753697e-01 8.44489753e-01 5.25755942e-01
7.41384983e-01 6.06253326e-01 8.39998245e-01 5.87931395e-01
8.62625316e-02 2.55021304e-01 6.65357292e-01 -8.45744789e-01
1.05935432e-01 7.70415589e-02 6.64797798e-02 9.40485060e-01
2.81967193e-01 -2.28841588e-01 -4.16345239e-01 8.14418316e-01
-1.63552105e+00 -1.02760231e+00 -4.13618684e-01 1.86380446e+00
2.55025148e-01 3.91990989e-01 -2.52357513e-01 6.31981850e-01
1.16607594e+00 1.09570220e-01 -6.60910010e-01 2.36128271e-02
-3.90474439e-01 -4.25260842e-01 3.77823859e-01 3.74734432e-01
-1.23735511e+00 4.98355031e-01 5.56147623e+00 1.06271744e+00
-9.26405132e-01 5.94827272e-02 8.97796333e-01 2.58540064e-01
9.04368609e-02 -1.15392409e-01 -5.43525457e-01 8.26939583e-01
1.10534206e-01 3.94590616e-01 2.45356798e-01 6.04164243e-01
7.35443115e-01 -3.09987336e-01 -4.68371511e-01 9.35170829e-01
1.91364065e-01 -1.26333213e+00 -3.24605674e-01 -2.37379387e-01
8.32147777e-01 -1.35232538e-01 -2.16548830e-01 4.81990725e-02
-3.00475299e-01 -5.05895197e-01 6.26226783e-01 4.59174335e-01
3.44935775e-01 -8.69313836e-01 1.06301343e+00 2.63349473e-01
-1.04878414e+00 -6.13024294e-01 -3.39733392e-01 1.06607109e-01
5.58879375e-01 9.80053127e-01 -4.46627408e-01 7.87020385e-01
8.11730087e-01 9.34910536e-01 -3.61588657e-01 1.65853536e+00
-4.93565887e-01 5.57907879e-01 -2.67210364e-01 1.61880791e-01
4.06429768e-01 -7.07331240e-01 5.63765764e-01 1.00420439e+00
3.97074878e-01 -6.84930086e-02 2.51058519e-01 8.41021299e-01
3.11244786e-01 1.38388485e-01 -5.20918846e-01 5.73484540e-01
2.83792138e-01 1.51915133e+00 -8.51814628e-01 -2.13486701e-01
-5.23697197e-01 1.00023174e+00 -2.32333660e-01 4.35018986e-01
-8.78475964e-01 -8.97890866e-01 4.07136641e-02 1.11793811e-02
4.90533382e-01 5.83914705e-02 -5.76342463e-01 -8.39770377e-01
2.11530462e-01 -6.61324501e-01 3.85673523e-01 -1.09707487e+00
-1.24363589e+00 3.53226572e-01 -4.34767187e-01 -1.48586798e+00
5.77957571e-01 -4.58402932e-01 -1.25709748e+00 9.27541375e-01
-1.72484398e+00 -8.68316710e-01 -4.56543982e-01 5.62989354e-01
8.78229439e-01 -2.76452571e-01 2.28004321e-01 8.58471215e-01
-1.05625677e+00 2.56502181e-01 3.65232050e-01 1.72243059e-01
-5.83344288e-02 -9.15369987e-01 1.04461960e-01 1.24105489e+00
-3.58860642e-01 -3.25812489e-01 5.87867320e-01 -5.95356286e-01
-7.82077193e-01 -9.86812174e-01 5.01801491e-01 4.82006431e-01
4.74386156e-01 -3.95327248e-02 -1.01432908e+00 1.43900484e-01
3.35627139e-01 -9.82282534e-02 4.34224531e-02 -2.78050333e-01
6.55915320e-01 -4.64903474e-01 -1.40091848e+00 5.42621195e-01
6.75286114e-01 -2.02475160e-01 -3.88467491e-01 9.36896026e-01
2.94122040e-01 -2.77413517e-01 -6.63278759e-01 5.00248671e-01
-1.45284340e-01 -7.12443173e-01 6.94551349e-01 -1.57159463e-01
-8.14743862e-02 -7.68418074e-01 3.99718702e-01 -1.37246621e+00
-7.34795988e-01 -2.72560865e-01 2.29705557e-01 1.25331450e+00
3.50286037e-01 -5.34783721e-01 8.58440816e-01 -5.81257865e-02
-6.52483821e-01 -6.36734307e-01 -9.51781511e-01 -4.37331557e-01
-5.34704924e-01 3.91339250e-02 4.86763388e-01 8.82377923e-01
-8.99216253e-03 3.56341988e-01 -2.28574559e-01 7.35935032e-01
5.68293929e-01 5.10148741e-02 1.04911551e-01 -1.16550434e+00
1.27999499e-01 -2.60807812e-01 -6.85130358e-01 -1.19867158e+00
-1.02293357e-01 -6.49083316e-01 7.91992545e-02 -1.79962003e+00
-1.25734210e-01 -3.87835234e-01 -4.94061351e-01 1.10540733e-01
-1.31069362e-01 1.27493560e-01 -1.80073872e-01 5.80839962e-02
-5.50699294e-01 6.08104169e-01 1.52946079e+00 -6.25149012e-01
-7.27029741e-02 4.23554778e-01 -2.43627384e-01 6.55534565e-01
1.07453501e+00 -1.08678646e-01 -3.70138258e-01 -5.62886178e-01
-2.59741783e-01 1.60276741e-01 1.62773237e-01 -1.21684682e+00
3.45472306e-01 4.07763094e-01 7.60120928e-01 -1.06454992e+00
9.83856097e-02 -1.17617214e+00 -4.97470379e-01 6.66161597e-01
1.43845022e-01 2.11297765e-01 8.19059163e-02 2.12941304e-01
-4.32528973e-01 -6.53634071e-01 9.63894963e-01 -1.29043415e-01
-9.91563141e-01 2.13024408e-01 -7.66714334e-01 -3.20871711e-01
1.18796051e+00 -7.10792601e-01 -1.79161474e-01 -2.11523980e-01
-7.26504683e-01 5.45067191e-01 -2.06220318e-02 2.60009944e-01
8.72310817e-01 -1.12332785e+00 -8.47077727e-01 7.92635381e-01
-2.09417641e-01 -9.28279310e-02 9.30237114e-01 1.15738523e+00
-9.28617775e-01 1.04580015e-01 -1.78657204e-01 -7.07885206e-01
-1.15233731e+00 5.03408492e-01 7.24734545e-01 -2.41495520e-01
-9.78365719e-01 7.52572358e-01 -2.15645060e-02 -1.48276284e-01
3.53096604e-01 -2.06641257e-01 -6.18829489e-01 4.36336063e-02
5.30712306e-01 5.40761411e-01 3.76764268e-01 -4.20653462e-01
-6.48442209e-02 5.49168825e-01 -1.53957695e-01 3.36826473e-01
1.31108844e+00 -3.77927870e-01 -3.08473349e-01 -6.90468922e-02
8.73942494e-01 -3.57257396e-01 -1.38325632e+00 1.32503480e-01
-1.18695304e-01 -6.46959722e-01 4.79848415e-01 -9.25133467e-01
-1.69099867e+00 9.45000947e-01 1.16519487e+00 4.94345307e-01
1.56800520e+00 -4.79236424e-01 7.10773349e-01 -3.18262586e-03
3.09505463e-01 -1.43546116e+00 -1.77597985e-01 2.23903343e-01
4.72264022e-01 -1.15055430e+00 -4.55391221e-02 -2.89653897e-01
-4.40371484e-01 9.17578578e-01 9.87003863e-01 -4.24067527e-02
6.26198053e-01 2.49059573e-01 5.90352751e-02 -2.49544278e-01
-2.40558572e-02 1.27261087e-01 -1.40207455e-01 7.55245388e-01
-3.39918621e-02 -3.07024360e-01 -3.96595836e-01 2.26365656e-01
5.99756420e-01 -2.54429549e-01 2.95312077e-01 9.27616358e-01
-7.62487531e-01 -4.91686314e-01 -5.86176991e-01 5.78422368e-01
-5.50846398e-01 -1.32674828e-01 4.39056993e-01 3.79165381e-01
2.68901467e-01 1.15898013e+00 1.40290380e-01 -2.20428765e-01
5.12984276e-01 -5.91558099e-01 2.49756709e-01 -3.04558903e-01
-4.88761276e-01 1.02658138e-01 -2.72645921e-01 4.14571390e-02
-1.40794173e-01 -4.80886489e-01 -1.35564923e+00 6.49003908e-02
-1.00431025e+00 5.50548971e-01 5.29160261e-01 8.01531255e-01
1.09897412e-01 1.07042050e+00 1.18649471e+00 -8.53823662e-01
-3.43470067e-01 -1.07229817e+00 -1.34147286e+00 1.65984750e-01
2.20073834e-01 -2.81672090e-01 -4.81366962e-01 -2.34792262e-01] | [7.468461513519287, 1.6704504489898682] |
afa37973-a521-480f-895f-4246a332b35f | when-bots-take-over-the-stock-market-evasion | 2010.09246 | null | https://arxiv.org/abs/2010.09246v2 | https://arxiv.org/pdf/2010.09246v2.pdf | Taking Over the Stock Market: Adversarial Perturbations Against Algorithmic Traders | In recent years, machine learning has become prevalent in numerous tasks, including algorithmic trading. Stock market traders utilize machine learning models to predict the market's behavior and execute an investment strategy accordingly. However, machine learning models have been shown to be susceptible to input manipulations called adversarial examples. Despite this risk, the trading domain remains largely unexplored in the context of adversarial learning. In this study, we present a realistic scenario in which an attacker influences algorithmic trading systems by using adversarial learning techniques to manipulate the input data stream in real time. The attacker creates a universal perturbation that is agnostic to the target model and time of use, which, when added to the input stream, remains imperceptible. We evaluate our attack on a real-world market data stream and target three different trading algorithms. We show that when added to the input stream, our perturbation can fool the trading algorithms at future unseen data points, in both white-box and black-box settings. Finally, we present various mitigation methods and discuss their limitations, which stem from the algorithmic trading domain. We believe that these findings should serve as an alert to the finance community about the threats in this area and promote further research on the risks associated with using automated learning models in the trading domain. | ['Yuval Elovici', 'Asaf Shabtai', 'Yael Mathov', 'Elior Nehemya'] | 2020-10-19 | null | null | null | null | ['real-world-adversarial-attack', 'algorithmic-trading'] | ['adversarial', 'time-series'] | [ 2.99881995e-01 -5.78767434e-03 4.05125283e-02 1.00075997e-01
-4.88811731e-01 -1.53839648e+00 9.27126288e-01 9.89644378e-02
-3.66146088e-01 3.93723249e-01 -3.99515718e-01 -8.14881504e-01
1.64297491e-01 -1.10314655e+00 -9.68104005e-01 -4.77499634e-01
-5.16146362e-01 2.88679898e-01 9.19123292e-02 -1.84088498e-01
4.50389922e-01 5.94957352e-01 -9.68564928e-01 2.38732114e-01
1.42899990e-01 9.73421216e-01 -9.43322659e-01 8.92035007e-01
2.88529038e-01 1.02588940e+00 -1.30141687e+00 -8.13981771e-01
1.20300198e+00 -3.46131772e-01 -2.18024343e-01 -2.70134419e-01
1.62109852e-01 -5.66819787e-01 -3.03479642e-01 1.28767288e+00
2.58878410e-01 -3.33696604e-01 3.92639101e-01 -1.67429638e+00
-2.86487818e-01 9.09403861e-01 -3.77303362e-01 4.10218894e-01
-1.59998447e-01 8.35569859e-01 7.20695853e-01 -1.25014231e-01
2.42437720e-01 1.02636206e+00 4.37180340e-01 5.42681336e-01
-1.32519889e+00 -1.37180245e+00 -1.33036487e-02 -4.24577802e-01
-7.35480070e-01 -1.97520107e-01 9.68563199e-01 -4.65371907e-01
5.51716566e-01 4.58125472e-01 6.33083642e-01 1.35455203e+00
6.55326724e-01 5.67062616e-01 1.32251191e+00 -2.43597239e-01
5.37645221e-01 3.18737507e-01 -1.77019775e-01 1.45655140e-01
5.58066964e-01 1.11037242e+00 -4.28848445e-01 -9.63907242e-01
6.82874084e-01 -8.28783661e-02 -3.14284787e-02 -2.72208095e-01
-9.68247473e-01 1.19033241e+00 1.09605387e-01 1.37648471e-02
-3.49838674e-01 5.41504741e-01 5.20230711e-01 9.11417544e-01
3.43855679e-01 1.21043277e+00 -7.30782509e-01 -2.34554932e-01
-4.24402833e-01 4.48988378e-01 1.39314461e+00 5.66276968e-01
2.34078139e-01 6.28035665e-01 4.77830410e-01 -1.42191678e-01
8.61376822e-02 8.08072507e-01 5.85484385e-01 -8.63626361e-01
4.41943735e-01 1.00449860e-01 1.72648862e-01 -1.12552357e+00
-4.04265057e-03 -4.11356091e-01 -3.12845170e-01 8.18922758e-01
8.24336171e-01 -6.75507069e-01 -5.63440144e-01 1.46991086e+00
2.36006528e-01 4.85193133e-01 2.34631121e-01 5.08199692e-01
-2.64052749e-01 4.28555578e-01 -1.05529934e-01 -1.81491509e-01
9.73893166e-01 -4.80767965e-01 -3.96374762e-01 -1.98355615e-01
4.45834041e-01 -7.76003599e-01 8.25716317e-01 6.58820510e-01
-8.55921030e-01 -1.96469743e-02 -1.32315874e+00 1.01025891e+00
-6.54569387e-01 -1.03827477e+00 6.00809515e-01 1.24924409e+00
-3.59812200e-01 7.19769895e-01 -1.02397597e+00 4.31893766e-01
4.10529017e-01 4.57776427e-01 9.66632664e-02 8.55053902e-01
-1.56955349e+00 8.01470995e-01 2.06466377e-01 -1.09421298e-01
-1.12301874e+00 -1.02934480e+00 -4.80252326e-01 -1.72000960e-01
4.48034704e-01 -4.00866598e-01 1.61847949e+00 -1.23514533e+00
-1.34269404e+00 5.42239785e-01 6.20301843e-01 -1.25248003e+00
1.16272855e+00 -2.10030660e-01 -6.68400526e-01 6.51549101e-02
-3.98399681e-01 -2.67427355e-01 1.45163929e+00 -1.06796360e+00
-4.51721877e-01 -1.58146173e-01 1.71688706e-01 -2.32747167e-01
-2.39471063e-01 1.70149744e-01 6.27483070e-01 -1.29723072e+00
-4.84011590e-01 -9.78297591e-01 -3.50821644e-01 -1.47321254e-01
-4.06567723e-01 5.86601198e-01 1.14230728e+00 -1.54918373e-01
1.05558991e+00 -2.18191385e+00 -4.87990201e-01 6.93565249e-01
1.34113058e-01 3.76141191e-01 2.93056548e-01 5.91635048e-01
-1.60149172e-01 6.48076534e-01 -2.23738670e-01 1.59429342e-01
2.00340122e-01 -7.80682638e-02 -1.44715905e+00 6.46364748e-01
1.50490314e-01 1.04005384e+00 -7.49422193e-01 2.71103024e-01
1.36854231e-01 -1.57879107e-02 -4.52356696e-01 3.15707684e-01
-5.78227162e-01 3.40241790e-01 -7.14213014e-01 4.78840262e-01
2.92671919e-01 9.14924145e-02 -1.29723787e-01 3.88604343e-01
2.48516619e-01 1.23005815e-01 -1.10749626e+00 4.20477331e-01
-1.94665283e-01 6.73253775e-01 -1.67228691e-02 -5.89740872e-01
7.75632977e-01 1.72204673e-01 9.24968049e-02 -3.88178229e-01
2.69312531e-01 2.64621675e-01 4.77397501e-01 -1.58718172e-02
1.37978390e-01 -4.12383348e-01 -3.22443843e-01 1.33933890e+00
-6.68229401e-01 -3.73797715e-01 -4.32356149e-01 -1.14924386e-02
1.39866018e+00 -3.52186143e-01 4.05260623e-01 3.49446312e-02
9.03985500e-02 1.57829463e-01 2.60396421e-01 1.15504241e+00
-3.17140311e-01 5.60361184e-02 6.86722159e-01 -7.04750538e-01
-9.71107364e-01 -1.22278857e+00 1.11416519e-01 6.15953803e-01
-7.44838268e-02 -1.41473353e-01 -5.88179290e-01 -1.03218853e+00
6.85805559e-01 9.97674048e-01 -8.35057855e-01 -6.45776391e-01
-8.47957253e-01 -6.86847031e-01 1.13646758e+00 3.64940643e-01
4.24553216e-01 -1.18704975e+00 -9.26261902e-01 2.54844368e-01
7.70654559e-01 -8.12957466e-01 -6.48552418e-01 1.78331822e-01
-7.29260206e-01 -1.41076744e+00 -8.49884152e-02 -8.97202939e-02
4.00319874e-01 -1.78684220e-01 9.60602522e-01 8.86417031e-02
-1.98340565e-01 6.48066401e-01 -1.43972293e-01 -1.21123648e+00
-1.18381906e+00 -1.53054059e-01 6.62319958e-02 2.47591749e-01
5.29075861e-01 -5.13379753e-01 -4.17513281e-01 2.52928138e-01
-1.28425515e+00 -6.32964492e-01 1.42455608e-01 5.98832726e-01
6.65187016e-02 4.32785958e-01 6.87791169e-01 -1.50619245e+00
8.89564157e-01 -6.42484248e-01 -1.23377514e+00 2.37144493e-02
-7.48449564e-01 8.69930722e-03 1.15805387e+00 -1.09972072e+00
-5.20408988e-01 -3.82860065e-01 4.69375849e-01 -6.68691576e-01
1.07343823e-01 2.08863780e-01 -1.60735697e-02 -3.95897329e-01
8.13293099e-01 3.24560523e-01 2.97411203e-01 -1.50201544e-01
1.30300388e-01 3.20304394e-01 2.52235055e-01 -5.33553600e-01
1.81319845e+00 4.86438453e-01 6.88117743e-02 -5.44341445e-01
-3.90246242e-01 3.95442784e-01 -7.61610419e-02 -1.22430742e-01
1.34489328e-01 -4.38922524e-01 -7.90622830e-01 1.01241672e+00
-7.44604945e-01 -5.49114823e-01 -3.90781760e-01 2.13658914e-01
-6.86654091e-01 2.86186002e-02 -6.26219511e-01 -9.49410260e-01
-2.25404322e-01 -1.03671515e+00 3.66741002e-01 -5.68666570e-02
-5.07734120e-01 -1.32603836e+00 8.09062272e-02 1.40410781e-01
7.30935335e-01 6.28820837e-01 1.01718688e+00 -1.68827224e+00
-8.95762205e-01 -6.16528869e-01 5.90459108e-01 4.63674843e-01
2.93187231e-01 1.58915088e-01 -1.02687788e+00 -4.59384620e-01
7.08778620e-01 -4.00385037e-02 3.00173640e-01 -2.24003255e-01
9.15174007e-01 -9.74577188e-01 1.32908642e-01 6.95153236e-01
1.06082964e+00 7.26507306e-01 1.14430830e-01 8.22325408e-01
2.87793338e-01 4.80762959e-01 5.99206269e-01 5.15240550e-01
-5.63094854e-01 1.70356631e-01 6.99649751e-01 1.03328213e-01
6.87111437e-01 -5.32763064e-01 6.75593615e-01 -1.05332658e-01
5.62079608e-01 -9.51600820e-02 -8.44519436e-01 -5.96274957e-02
-1.26698887e+00 -1.15315795e+00 4.24520075e-01 2.35773277e+00
9.14641321e-01 8.67259622e-01 4.20220137e-01 2.40305260e-01
5.59700489e-01 1.92679897e-01 -1.11539555e+00 -5.97722769e-01
-6.25047535e-02 5.45560718e-01 1.08312368e+00 5.09757102e-01
-1.36330092e+00 8.06889415e-01 6.36481953e+00 4.14023638e-01
-1.40171218e+00 -3.08192968e-01 6.25006914e-01 -2.66885340e-01
-2.87609726e-01 -6.10353984e-03 -4.62815225e-01 7.89708555e-01
1.29025841e+00 -9.65887606e-01 6.51956797e-01 7.96713412e-01
2.74155326e-02 4.99215692e-01 -1.32267272e+00 3.82297218e-01
-2.27724284e-01 -1.31052375e+00 1.25494540e-01 3.61146957e-01
3.20360273e-01 -2.59088278e-01 6.89252615e-01 3.64620984e-01
8.67460907e-01 -1.20232320e+00 7.00905919e-01 6.30809665e-02
1.60668761e-01 -1.25825310e+00 6.30992770e-01 5.35633147e-01
-6.99967861e-01 -1.40267104e-01 1.88431323e-01 -2.77487934e-01
-9.44144577e-02 7.50279278e-02 -1.10790515e+00 -1.40083767e-02
3.42737436e-01 3.55375618e-01 -5.07120192e-01 5.98001420e-01
-2.35990509e-01 1.20348120e+00 -5.12057126e-01 -4.20015678e-02
3.36273730e-01 -1.25841558e-01 8.53275776e-01 7.94356227e-01
-1.95451733e-03 -3.52369249e-02 -9.84367207e-02 9.75847006e-01
-1.51993364e-01 -3.94221604e-01 -1.15672994e+00 -6.06072843e-01
6.90443993e-01 6.44196987e-01 -5.98122120e-01 4.36912924e-02
-2.02681512e-01 6.66723788e-01 -2.68621564e-01 5.13288140e-01
-8.50982308e-01 -3.77836168e-01 9.91248786e-01 1.24048918e-01
6.33373410e-02 -9.22031105e-02 -4.97074246e-01 -1.01820540e+00
7.62331039e-02 -1.57071543e+00 5.04372954e-01 -9.82990265e-02
-1.52592468e+00 2.45630637e-01 -1.34298593e-01 -1.45873606e+00
-7.02120483e-01 -5.80045938e-01 -7.50427842e-01 6.98720217e-01
-1.02482843e+00 -5.25518477e-01 7.15542436e-01 4.70923007e-01
1.91338092e-01 -7.84487844e-01 6.93664610e-01 -3.71202290e-01
-3.48198116e-01 8.61934185e-01 1.78643659e-01 7.26404846e-01
5.17247677e-01 -1.27378654e+00 1.01205039e+00 1.00280273e+00
4.44548011e-01 8.28626037e-01 9.56494451e-01 -8.16860318e-01
-1.58151662e+00 -1.25894964e+00 9.98242795e-02 -8.81114781e-01
1.55502057e+00 -4.75604504e-01 -9.89638984e-01 1.07564628e+00
8.88075829e-02 -7.92806000e-02 7.95003295e-01 -4.69509423e-01
-7.49767542e-01 2.22180039e-02 -1.58081675e+00 9.15594757e-01
1.72006845e-01 -6.57209516e-01 -8.14788878e-01 1.60101518e-01
9.68013823e-01 -3.28640461e-01 -7.68657029e-01 1.58587366e-01
5.53570509e-01 -8.35520208e-01 8.98011923e-01 -1.06825078e+00
2.10708659e-02 -9.19020921e-02 1.22551993e-01 -1.29230332e+00
5.09973645e-01 -1.40849972e+00 -5.54333091e-01 9.99832630e-01
4.75538135e-01 -1.50270092e+00 8.14525723e-01 8.97488534e-01
7.07462907e-01 -4.14620280e-01 -9.22151923e-01 -1.09281993e+00
6.59474492e-01 -6.00547373e-01 7.89278448e-01 1.09635961e+00
-1.63332164e-01 -2.13046059e-01 -4.18393731e-01 4.77065623e-01
9.04708803e-01 1.35461390e-01 1.08363163e+00 -8.05160284e-01
-7.79901743e-01 -6.24844849e-01 -4.13852453e-01 -3.59178901e-01
2.03091234e-01 -5.59249401e-01 -4.06082451e-01 1.60441935e-01
-4.15150344e-01 -2.43830979e-01 -3.69997978e-01 1.88718557e-01
-5.91930151e-02 9.46923643e-02 7.90872335e-01 3.81151468e-01
1.63783282e-01 8.48271549e-02 7.95599461e-01 -3.22449565e-01
-1.12846211e-01 7.63196826e-01 -7.27752447e-01 8.01077485e-01
1.19670606e+00 -9.09116149e-01 -3.82720500e-01 1.69758976e-01
1.94861948e-01 -1.97019398e-01 4.86805737e-01 -6.23187542e-01
2.44579956e-01 -4.64359671e-01 1.83836699e-01 -9.28632692e-02
7.95571655e-02 -1.19939256e+00 2.16447979e-01 1.11657834e+00
-5.87955356e-01 3.18634450e-01 3.75696331e-01 7.59341300e-01
-5.53331617e-03 -1.69669375e-01 7.52164483e-01 -6.28931597e-02
-3.55652645e-02 2.98282444e-01 -5.61309814e-01 4.55924034e-01
1.60017920e+00 -2.56769545e-02 -5.09224296e-01 -6.95178151e-01
-4.54161137e-01 2.92014837e-01 6.36042535e-01 4.74398851e-01
4.19667840e-01 -8.56516421e-01 -6.37071371e-01 7.01252580e-01
-2.75687605e-01 -4.50247288e-01 -3.97803336e-01 8.55548084e-02
-6.32281721e-01 5.33482619e-02 1.05055451e-01 -5.91676235e-02
-1.03105617e+00 1.07135177e+00 7.40341783e-01 -2.01208711e-01
-4.10906106e-01 4.05536175e-01 1.00242786e-01 -2.04601496e-01
3.60028505e-01 -8.91728848e-02 3.38213265e-01 -6.49527907e-02
6.01363182e-01 2.73986936e-01 -2.40529791e-01 -3.18228826e-02
-1.41852528e-01 7.95460045e-02 -2.62511045e-01 -2.35298514e-01
9.14517522e-01 5.42845368e-01 1.24730229e-01 6.75789416e-01
9.35530841e-01 3.65760148e-01 -1.19160259e+00 -2.54171908e-01
3.47622395e-01 -6.12026393e-01 -4.46326256e-01 -9.02490020e-01
-1.21717763e+00 6.96143389e-01 2.65041262e-01 8.82674575e-01
9.64124799e-01 -4.47637558e-01 8.06587994e-01 5.77447057e-01
5.28793991e-01 -6.37080550e-01 -5.73270991e-02 2.88377672e-01
6.88459516e-01 -8.76402736e-01 -1.97986141e-01 -1.05272988e-02
-6.54221654e-01 1.14823294e+00 3.30417812e-01 -5.38384914e-01
9.90309000e-01 8.83902669e-01 6.90145791e-01 9.62892026e-02
-9.65746522e-01 6.43305838e-01 -2.21200630e-01 4.91691023e-01
-2.69971341e-01 4.30422947e-02 2.91595101e-01 6.10259056e-01
-7.05109835e-01 -1.98190525e-01 9.71524298e-01 1.32057977e+00
-1.28828973e-01 -1.23269880e+00 -7.85833478e-01 5.59982896e-01
-9.58488941e-01 -4.29853797e-02 -9.37983155e-01 1.31457567e+00
-4.24789667e-01 9.26991522e-01 8.37781578e-02 -4.58403170e-01
4.18854892e-01 1.05321467e-01 -2.16352314e-01 -5.09948969e-01
-1.43457174e+00 -1.59689322e-01 -5.23342751e-02 -4.64607567e-01
1.02568835e-01 -7.13557065e-01 -9.82369006e-01 -6.33821011e-01
1.25338575e-02 2.35377982e-01 1.34575516e-01 6.42580688e-01
1.73581645e-01 1.77183956e-01 1.15383840e+00 -5.46599030e-01
-1.68211222e+00 -5.31989694e-01 -7.70538092e-01 4.04048115e-01
7.51868129e-01 -3.63674223e-01 -1.17916942e+00 -9.08884853e-02] | [5.712240695953369, 7.668399810791016] |
5279a5ba-d5ad-4416-ad87-eca5d397e132 | training-like-a-medical-resident-universal | 2306.02416 | null | https://arxiv.org/abs/2306.02416v2 | https://arxiv.org/pdf/2306.02416v2.pdf | Training Like a Medical Resident: Universal Medical Image Segmentation via Context Prior Learning | A major enduring focus of clinical workflows is disease analytics and diagnosis, leading to medical imaging datasets where the modalities and annotations are strongly tied to specific clinical objectives. To date, building task-specific segmentation models is intuitive yet a restrictive approach, lacking insights gained from widespread imaging cohorts. Inspired by the training of medical residents, we explore universal medical image segmentation, whose goal is to learn from diverse medical imaging sources covering a range of clinical targets, body regions, and image modalities. Following this paradigm, we propose Hermes, a context prior learning approach that addresses the challenges related to the heterogeneity on data, modality, and annotations in the proposed universal paradigm. In a collection of seven diverse datasets, we demonstrate the appealing merits of the universal paradigm over the traditional task-specific training paradigm. By leveraging the synergy among various tasks, Hermes shows superior performance and model scalability. Our in-depth investigation on two additional datasets reveals Hermes' strong capabilities for transfer learning, incremental learning, and generalization to different downstream tasks. The code is available: https://github.com/yhygao/universal-medical-image-segmentation. | ['Dimitris N. Metaxas', 'Shaoting Zhang', 'Mu Zhou', 'Di Liu', 'Zhuowei Li', 'Yunhe Gao'] | 2023-06-04 | null | null | null | null | ['incremental-learning'] | ['methodology'] | [ 3.94687444e-01 1.37203366e-01 -5.31948626e-01 -4.29367483e-01
-1.26125860e+00 -5.39774776e-01 3.17478150e-01 1.65481135e-01
-2.60758251e-01 5.45953929e-01 3.46024036e-01 -5.35488069e-01
-1.62439570e-01 -1.96689785e-01 -5.70530713e-01 -6.68553889e-01
-5.10599464e-02 4.82176542e-01 1.97211444e-01 6.43986091e-02
-1.87504217e-01 2.69163013e-01 -8.22356761e-01 5.29461384e-01
8.09559703e-01 1.07926750e+00 3.78173143e-01 6.06190562e-01
1.37556866e-01 6.82874024e-01 -1.08049087e-01 -2.57545501e-01
2.97023863e-01 -4.26957071e-01 -8.92155051e-01 1.84674591e-01
4.70949620e-01 -2.76301533e-01 -2.99971104e-01 7.13438094e-01
7.75523722e-01 -2.59213597e-01 6.01021171e-01 -9.40943897e-01
-7.32389510e-01 4.09664363e-01 -4.54836339e-01 4.32049751e-01
1.95751097e-02 4.47485536e-01 8.27847779e-01 -5.57497084e-01
6.85487330e-01 5.00743508e-01 9.86510992e-01 8.31503689e-01
-1.25490177e+00 -5.21997035e-01 9.67246294e-02 -9.39335003e-02
-1.08092475e+00 -2.67404586e-01 3.88221115e-01 -8.34692836e-01
3.49200010e-01 2.65400410e-01 4.56131965e-01 1.34483051e+00
1.45602047e-01 9.64068592e-01 1.17232621e+00 -1.94617301e-01
8.38082656e-02 -9.25741508e-04 1.85856208e-01 7.61750817e-01
5.55643998e-02 -3.08788810e-02 -3.29553038e-01 -7.71342218e-02
6.58589244e-01 5.05278349e-01 -4.50895965e-01 -4.38802749e-01
-1.44661915e+00 4.11025673e-01 5.49701333e-01 2.70989686e-01
-3.93914193e-01 1.01980902e-01 5.42633176e-01 -7.50253024e-03
2.55059928e-01 3.93945694e-01 -6.02960408e-01 6.10101707e-02
-9.20132279e-01 4.46562283e-03 6.28804088e-01 8.58344078e-01
3.96031260e-01 -4.37354505e-01 -5.08572578e-01 6.27292752e-01
1.32908642e-01 2.77647555e-01 6.56147599e-01 -8.37158501e-01
2.88714796e-01 5.17720520e-01 -1.83758244e-01 -4.35012519e-01
-5.87454557e-01 -7.34473228e-01 -9.84293103e-01 -1.10083140e-01
5.97475231e-01 -3.73756438e-01 -1.34859514e+00 1.70762563e+00
4.64177072e-01 2.90948272e-01 -2.00579643e-01 8.49720061e-01
9.83952105e-01 -3.19977067e-02 6.12307847e-01 1.52514771e-01
1.58210611e+00 -1.01188612e+00 -4.29007530e-01 -1.88088909e-01
8.00995946e-01 -5.14942586e-01 1.24403703e+00 2.96501577e-01
-9.12179053e-01 -3.53267789e-01 -7.17258990e-01 -1.06685378e-01
-2.60378867e-01 1.31826699e-01 8.58006775e-01 4.66214478e-01
-9.95999694e-01 4.06387597e-01 -1.05426455e+00 -4.61881101e-01
8.46632183e-01 1.97166905e-01 -2.71413326e-01 -3.44489813e-01
-8.08020949e-01 7.47244298e-01 3.23354304e-01 6.00948334e-02
-1.09744656e+00 -1.34496748e+00 -4.96893227e-01 -1.54717550e-01
5.47167540e-01 -1.08305633e+00 1.43477738e+00 -7.31230259e-01
-1.22457445e+00 1.14382982e+00 7.14922100e-02 -4.37858284e-01
8.62206459e-01 -1.84569210e-01 -2.33181566e-01 3.52336377e-01
2.05927417e-01 7.74652839e-01 5.52098095e-01 -1.23806608e+00
-5.68607688e-01 -4.24093962e-01 -2.28429183e-01 1.19127288e-01
-1.27488270e-01 -2.87723660e-01 -6.65275097e-01 -7.01116741e-01
-8.73214304e-02 -1.01933455e+00 -6.46034360e-01 1.32158533e-01
-4.38803911e-01 9.78418067e-02 4.96124864e-01 -7.95976043e-01
1.08480608e+00 -2.21791434e+00 4.70399261e-02 9.92462859e-02
6.87232435e-01 8.04762840e-02 1.15354903e-01 9.64174792e-02
-1.25188366e-01 2.62636125e-01 -3.83279681e-01 -2.99751937e-01
-1.46612763e-01 1.51608363e-01 1.29011005e-01 2.91455775e-01
4.75285016e-02 1.27526069e+00 -9.94805396e-01 -7.30852008e-01
1.61347777e-01 3.37236553e-01 -5.79529464e-01 2.38360345e-01
-2.21261784e-01 1.21539962e+00 -7.08033860e-01 1.03303266e+00
2.38946810e-01 -1.00506186e+00 3.00883025e-01 -3.75820518e-01
3.04905236e-01 -3.36268023e-02 -5.04756033e-01 2.16408992e+00
-3.45278263e-01 1.36550590e-01 2.25745097e-01 -1.11329293e+00
3.75196457e-01 6.13237321e-01 1.04675925e+00 -6.31162167e-01
1.94418907e-01 3.70938033e-01 5.07079251e-02 -7.79149771e-01
-1.17091104e-01 -2.83979923e-01 -1.19375242e-02 2.14866996e-01
2.45067060e-01 4.51553315e-02 5.94985262e-02 1.79396704e-01
1.27980947e+00 9.94445607e-02 1.96420595e-01 -2.62083918e-01
1.35115147e-01 2.29884326e-01 5.99549234e-01 8.57583046e-01
-7.43892968e-01 8.79912913e-01 4.23339069e-01 -4.85157490e-01
-7.91326582e-01 -1.20494711e+00 -5.07334828e-01 1.06624079e+00
-3.68198082e-02 -2.28890345e-01 -5.81177533e-01 -9.38221812e-01
5.08710183e-02 9.05873179e-02 -1.02570784e+00 4.54434417e-02
-4.52930331e-01 -9.21349943e-01 5.22055447e-01 5.90166748e-01
2.72492319e-01 -8.60552192e-01 -7.92310059e-01 4.96160388e-02
-1.92471489e-01 -1.29040134e+00 -4.92057830e-01 1.69202760e-01
-1.13529325e+00 -1.21540749e+00 -1.09503722e+00 -7.14779317e-01
6.15223527e-01 -1.32288749e-03 1.36155784e+00 4.30176221e-02
-7.15187311e-01 6.79811299e-01 -2.80899256e-01 -4.34247643e-01
-3.68860275e-01 3.92352104e-01 -4.70609784e-01 -8.42646658e-02
5.32043651e-02 -5.31128049e-01 -1.11063373e+00 8.37602317e-02
-8.48989725e-01 3.48039895e-01 9.55972970e-01 8.90366018e-01
7.96613991e-01 -6.92396164e-01 7.05456138e-01 -1.43720794e+00
2.98948348e-01 -9.05989885e-01 -8.21687132e-02 5.04586875e-01
-6.02269888e-01 -2.68563360e-01 2.18209326e-01 -3.31333071e-01
-1.03734791e+00 1.82744533e-01 -1.16377532e-01 -4.20721680e-01
-4.33240891e-01 6.88650429e-01 2.89020360e-01 3.67007777e-02
8.15565884e-01 1.27772242e-01 1.99842796e-01 -5.13116837e-01
3.23806435e-01 3.35024208e-01 7.76174903e-01 -8.53730559e-01
3.12283069e-01 5.83067596e-01 2.81326473e-02 -4.05651182e-01
-1.08744550e+00 -6.17121875e-01 -7.02939630e-01 -1.55170739e-01
1.11485136e+00 -9.03588772e-01 -4.42992300e-01 2.31360883e-01
-6.74242318e-01 -7.25938499e-01 -3.64564747e-01 4.00603801e-01
-4.43794161e-01 3.18990737e-01 -8.19895148e-01 -1.92231625e-01
-4.85325813e-01 -1.49305165e+00 1.14935827e+00 1.48391023e-01
-2.03453720e-01 -1.29784405e+00 -1.43214734e-02 6.52025223e-01
5.38897038e-01 6.48353517e-01 1.00211322e+00 -7.84242690e-01
-5.66335917e-01 -1.42835751e-01 -2.95964450e-01 2.36892208e-01
3.58189464e-01 -2.96752989e-01 -8.04221332e-01 -2.37493798e-01
-2.03000903e-01 -4.76730257e-01 8.32025528e-01 6.08363330e-01
1.62832284e+00 2.35402837e-01 -5.12721837e-01 9.93859529e-01
1.33759642e+00 6.34603426e-02 2.27973580e-01 2.05756366e-01
8.33044887e-01 3.80466759e-01 2.13367388e-01 3.55990052e-01
4.99683648e-01 3.75170082e-01 3.39993000e-01 -6.34155810e-01
-3.02372098e-01 -5.40392213e-02 -2.30419040e-01 6.20064318e-01
-1.15577452e-01 9.33928192e-02 -1.47891247e+00 6.49217963e-01
-1.79591286e+00 -5.26448190e-01 1.08992562e-01 1.88654757e+00
1.09414566e+00 -1.63154546e-02 9.32915360e-02 -5.63459516e-01
4.47669119e-01 -1.32664025e-01 -7.45132565e-01 1.72166213e-01
2.22070187e-01 1.20511048e-01 5.16324759e-01 6.54428527e-02
-1.24301279e+00 6.48371100e-01 7.02917528e+00 6.71325743e-01
-1.34002078e+00 4.42322701e-01 1.04884791e+00 -1.06425323e-01
-1.73463225e-01 -2.83852100e-01 -4.62912947e-01 4.58156943e-01
7.55552888e-01 1.37591809e-01 4.75591123e-02 7.02835202e-01
4.46286388e-02 1.12753421e-01 -1.31884873e+00 7.85727262e-01
-7.78386146e-02 -1.42776513e+00 -6.37736991e-02 -5.16694505e-03
7.86513507e-01 3.51490557e-01 3.49721462e-01 3.59725922e-01
3.74687344e-01 -1.12195480e+00 2.93730438e-01 4.28630769e-01
1.05411148e+00 2.39881314e-02 5.84092319e-01 1.18403524e-01
-7.67222166e-01 -6.22728467e-02 2.55955964e-01 4.03267682e-01
1.32412001e-01 4.77828681e-01 -9.72030222e-01 6.81136489e-01
8.20147276e-01 7.97724068e-01 -6.17669880e-01 1.02763116e+00
9.22274441e-02 7.53410757e-01 -5.39044626e-02 7.31122077e-01
2.70374954e-01 -9.86126363e-02 1.35384247e-01 1.58783746e+00
2.02430919e-01 2.54407823e-01 5.12031734e-01 6.46092355e-01
-1.49156526e-01 9.71152335e-02 -3.88709426e-01 3.75631601e-02
1.56076580e-01 1.49700749e+00 -7.59602308e-01 -4.82085198e-01
-6.55485094e-01 6.69566453e-01 1.19614795e-01 5.13256788e-01
-9.05189455e-01 2.81594247e-01 3.02408993e-01 2.62887001e-01
8.46686438e-02 3.84403802e-02 -6.17252588e-01 -1.12066543e+00
-4.78097163e-02 -9.72799957e-01 8.53019476e-01 -4.60337549e-01
-1.46096861e+00 4.45557356e-01 6.83604851e-02 -1.19442809e+00
7.51989633e-02 -6.80353880e-01 -5.14333308e-01 7.98217297e-01
-1.69026172e+00 -1.53593516e+00 -6.13356411e-01 7.41593361e-01
4.68790323e-01 -7.98218474e-02 6.56870365e-01 6.89964116e-01
-6.55582368e-01 6.39470935e-01 5.95679618e-02 2.12876081e-01
9.32838738e-01 -1.29918349e+00 -5.94759881e-02 5.01860499e-01
-2.38142088e-01 6.17471576e-01 2.49904454e-01 -5.46201706e-01
-1.24310458e+00 -1.08867192e+00 2.26633117e-01 -6.51946783e-01
7.96246946e-01 -9.29498151e-02 -8.95796955e-01 1.04918444e+00
1.02473751e-01 5.15798450e-01 1.17791450e+00 2.78622091e-01
-2.73525298e-01 -1.01720341e-01 -1.08175135e+00 5.02829969e-01
1.11914015e+00 -4.22033489e-01 -4.25934136e-01 5.51213861e-01
6.18405700e-01 -7.74561644e-01 -1.25891006e+00 6.69767976e-01
5.56383491e-01 -6.79218471e-01 1.00117278e+00 -8.61054003e-01
7.02422559e-01 3.85379642e-02 -4.91909310e-03 -9.33845639e-01
-2.32203230e-01 -4.39000070e-01 -1.49893180e-01 8.64507437e-01
4.96535897e-01 -5.75482726e-01 9.20573771e-01 6.83466852e-01
-4.85886335e-01 -1.17178071e+00 -7.50721812e-01 -3.37906539e-01
2.92617202e-01 -3.07641417e-01 2.96076477e-01 1.12594593e+00
-1.21965043e-01 8.68099853e-02 -1.73109487e-01 1.21257521e-01
6.15481675e-01 1.62850305e-01 5.50073564e-01 -1.07821512e+00
-7.00819433e-01 -5.55584013e-01 3.38123902e-03 -8.93597305e-01
-2.26661414e-01 -1.26471996e+00 -5.03352890e-03 -1.59826517e+00
6.04263127e-01 -8.00710917e-01 -7.82585502e-01 6.54515803e-01
-4.98328924e-01 3.83941919e-01 9.15732458e-02 4.87593889e-01
-8.91379178e-01 8.98659974e-02 1.62449419e+00 -1.06817126e-01
7.00698001e-03 4.64127176e-02 -9.01906133e-01 6.51758790e-01
7.98154116e-01 -3.18942934e-01 -4.61632133e-01 -6.20021105e-01
-1.07334711e-01 2.67995954e-01 5.14654636e-01 -9.49349999e-01
2.33082518e-01 -1.25043839e-01 3.80656183e-01 -2.53508598e-01
-1.90842338e-02 -7.50060499e-01 2.24584222e-01 6.43691719e-01
-4.05073404e-01 -4.16095145e-02 2.48216718e-01 5.52963555e-01
-1.95617214e-01 1.57392934e-01 7.16415465e-01 -5.26697516e-01
-7.38006651e-01 7.35597014e-01 2.33094469e-02 4.77359742e-01
1.07133389e+00 -1.52691171e-01 -2.40911856e-01 3.08574531e-02
-1.19748306e+00 5.03140509e-01 1.72338143e-01 2.08553001e-01
3.29743296e-01 -9.40653443e-01 -8.28848541e-01 -1.04910247e-01
2.55953372e-01 4.07557011e-01 6.49460137e-01 1.43641210e+00
-4.81096655e-01 3.45443517e-01 -2.36298323e-01 -1.03586328e+00
-8.86255205e-01 5.19845068e-01 5.76743186e-01 -5.97250819e-01
-7.91993797e-01 6.26453161e-01 6.03208721e-01 -5.32556832e-01
1.75937980e-01 -3.86756510e-01 1.61013916e-01 -1.94375128e-01
1.42906323e-01 4.92517911e-02 2.13849500e-01 -2.03762278e-01
-2.15710491e-01 3.57600868e-01 -3.24894518e-01 1.80894956e-01
1.30375695e+00 -6.76895455e-02 2.42483437e-01 3.49128067e-01
9.56544161e-01 -2.76823938e-01 -1.41746950e+00 -4.28022325e-01
1.98731974e-01 -2.18825907e-01 -8.38204548e-02 -1.20265162e+00
-1.20772910e+00 9.09795046e-01 7.68641710e-01 -7.53186345e-02
1.18405497e+00 2.10101813e-01 8.30215156e-01 -4.48616259e-02
1.53751105e-01 -7.25793779e-01 1.36116013e-01 1.57633677e-01
6.06858730e-01 -1.64138186e+00 -9.12142098e-02 -4.99106616e-01
-8.29793751e-01 6.79745138e-01 5.98690629e-01 2.52970010e-01
6.67245388e-01 2.72048414e-01 4.36612248e-01 -3.37445617e-01
-5.97057700e-01 -1.59933597e-01 3.41374546e-01 5.89867711e-01
6.36240542e-01 2.16824755e-01 -3.81416976e-02 6.94723010e-01
1.84676766e-01 3.34588826e-01 7.75690600e-02 1.02089882e+00
2.66264305e-02 -1.13533962e+00 -2.60482840e-02 6.21885180e-01
-8.69480848e-01 -1.50017843e-01 9.04119238e-02 8.21949422e-01
2.88294971e-01 3.24015081e-01 -2.74308354e-01 -5.30744717e-02
2.11746171e-01 1.18310496e-01 4.26826388e-01 -7.98628867e-01
-6.99324846e-01 4.38747667e-02 -3.52977902e-01 -5.67583978e-01
-2.89411694e-01 -6.63746536e-01 -1.25482750e+00 1.38119608e-01
2.80558258e-01 -1.69810072e-01 4.84701037e-01 8.74634981e-01
6.82378113e-01 8.63779604e-01 2.30665773e-01 -6.79018795e-01
-6.52022421e-01 -7.13955700e-01 -2.81985044e-01 7.80655146e-01
5.06802738e-01 -4.61097091e-01 1.89948846e-02 3.77444804e-01] | [14.798856735229492, -2.2038090229034424] |
48c0166b-03b3-4886-a875-4e209fcba607 | v3det-vast-vocabulary-visual-detection | 2304.03752 | null | https://arxiv.org/abs/2304.03752v1 | https://arxiv.org/pdf/2304.03752v1.pdf | V3Det: Vast Vocabulary Visual Detection Dataset | Recent advances in detecting arbitrary objects in the real world are trained and evaluated on object detection datasets with a relatively restricted vocabulary. To facilitate the development of more general visual object detection, we propose V3Det, a vast vocabulary visual detection dataset with precisely annotated bounding boxes on massive images. V3Det has several appealing properties: 1) Vast Vocabulary: It contains bounding boxes of objects from 13,029 categories on real-world images, which is 10 times larger than the existing large vocabulary object detection dataset, e.g., LVIS. 2) Hierarchical Category Organization: The vast vocabulary of V3Det is organized by a hierarchical category tree which annotates the inclusion relationship among categories, encouraging the exploration of category relationships in vast and open vocabulary object detection. 3) Rich Annotations: V3Det comprises precisely annotated objects in 245k images and professional descriptions of each category written by human experts and a powerful chatbot. By offering a vast exploration space, V3Det enables extensive benchmarks on both vast and open vocabulary object detection, leading to new observations, practices, and insights for future research. It has the potential to serve as a cornerstone dataset for developing more general visual perception systems. | ['Dahua Lin', 'Conghui He', 'Bin Wang', 'Tong Wu', 'Yujie Zhou', 'Yuhang Cao', 'Tao Chu', 'Pan Zhang', 'Jiaqi Wang'] | 2023-04-07 | null | null | null | null | ['open-vocabulary-object-detection'] | ['computer-vision'] | [-3.24141562e-01 -7.82103986e-02 -2.46858731e-01 -1.69881880e-01
-2.74569064e-01 -8.95966828e-01 7.55634725e-01 5.91332838e-02
-4.43663955e-01 2.34677359e-01 1.50179133e-01 -1.32204518e-01
2.80175239e-01 -4.86212581e-01 -6.99154317e-01 -4.62672889e-01
-2.25048568e-02 4.15695935e-01 8.27460110e-01 -3.91145319e-01
2.02628329e-01 1.95564836e-01 -1.87722969e+00 3.15628678e-01
2.29149207e-01 1.34158516e+00 5.49216390e-01 5.06613135e-01
-1.58805475e-01 8.89949441e-01 -6.25715017e-01 -4.94937181e-01
3.78279060e-01 2.79316664e-01 -8.55813801e-01 3.32444310e-01
9.52829301e-01 -4.26720887e-01 -3.08171004e-01 1.19183075e+00
1.29100144e-01 -5.66287600e-02 7.19605863e-01 -1.43102455e+00
-1.05812085e+00 5.85035384e-01 -5.84727764e-01 4.69478548e-01
1.90129519e-01 5.59324145e-01 1.45147371e+00 -1.32448113e+00
8.58215630e-01 1.36975324e+00 4.63227153e-01 4.69449401e-01
-1.15306103e+00 -7.19523013e-01 3.63137871e-01 3.71139348e-01
-1.93751597e+00 -2.87965596e-01 3.58532697e-01 -1.01624763e+00
9.93298531e-01 3.90869319e-01 7.61748910e-01 1.01117134e+00
-3.16325784e-01 8.44232857e-01 7.40354538e-01 -2.07172230e-01
2.12706849e-02 4.83255208e-01 4.98274446e-01 8.08852553e-01
5.20170867e-01 5.51683195e-02 -3.87029290e-01 3.34858410e-02
7.61334479e-01 4.05267812e-02 2.84472741e-02 -8.38974237e-01
-1.45304620e+00 1.01576781e+00 9.45386410e-01 -1.92392245e-02
8.98643136e-02 1.87791944e-01 8.41618299e-01 2.12382227e-01
8.46108422e-02 5.10292947e-01 -1.00975290e-01 2.77239501e-01
-3.57534409e-01 2.63374001e-01 5.80319583e-01 1.66844249e+00
7.32807696e-01 5.00193387e-02 -2.90315539e-01 9.37397361e-01
3.94449204e-01 5.76243937e-01 2.75398612e-01 -8.71853650e-01
4.83952731e-01 9.68988061e-01 1.58315703e-01 -9.59628046e-01
-3.25414181e-01 -2.40229249e-01 -6.33992374e-01 3.91640402e-02
3.73024821e-01 3.96110684e-01 -8.45073402e-01 1.34732425e+00
3.91427130e-01 -3.84586751e-01 -2.08588123e-01 1.21331429e+00
1.57938123e+00 3.27630013e-01 1.72595873e-01 1.37005553e-01
1.93661380e+00 -1.01949656e+00 -4.33467984e-01 -3.56811047e-01
5.52201033e-01 -6.05874836e-01 1.32193613e+00 2.67972618e-01
-6.05659187e-01 -6.24772489e-01 -8.18106771e-01 -2.99725235e-01
-6.27258360e-01 1.18996784e-01 8.26279640e-01 4.51731920e-01
-9.27588642e-01 -4.15669620e-01 -1.70184523e-01 -6.63850963e-01
8.48898709e-01 7.53514543e-02 -3.49279493e-01 -1.81477785e-01
-8.14293325e-01 7.87059247e-01 7.47977376e-01 -3.24100673e-01
-1.45253229e+00 -4.73073483e-01 -9.83746350e-01 1.30325533e-03
7.94417560e-01 -3.69199574e-01 1.20171762e+00 -5.75605452e-01
-5.83137870e-01 1.44450438e+00 2.28804350e-01 -4.40243423e-01
4.86176491e-01 1.05493451e-02 -3.46509069e-01 8.48744810e-02
4.42182630e-01 1.12325573e+00 9.79089141e-01 -1.18063593e+00
-9.19306219e-01 -2.25856766e-01 2.48641148e-01 1.51807249e-01
-5.66282690e-01 3.01493019e-01 -8.17429543e-01 -6.88812613e-01
-4.91823368e-02 -6.63804293e-01 2.79614367e-02 4.33485985e-01
-5.38602889e-01 -8.96100581e-01 8.16675544e-01 -2.40271658e-01
1.02794540e+00 -2.34235811e+00 -9.71183479e-02 -2.45391220e-01
1.03440416e+00 2.01555938e-01 -4.90029752e-02 1.71400219e-01
4.52687919e-01 1.36636958e-01 4.25847918e-01 -1.77456662e-02
1.92638546e-01 5.45441844e-02 -6.54567301e-01 5.11977315e-01
-2.76219342e-02 1.23457479e+00 -9.00829017e-01 -9.67752695e-01
3.98685098e-01 -1.31339421e-02 -5.70264757e-01 1.20458215e-01
-2.94072956e-01 -9.41468105e-02 -4.21827197e-01 9.14399564e-01
3.94220293e-01 -6.11230910e-01 -1.75375760e-01 -1.78927153e-01
-7.32865930e-02 2.18320582e-02 -1.00200462e+00 1.18791854e+00
1.42058119e-01 1.10665882e+00 1.03672720e-01 -8.75710487e-01
9.93059456e-01 -1.33448103e-02 5.33929802e-02 -5.33803105e-01
3.38909626e-01 -8.28468725e-02 -6.63414504e-03 -4.67345297e-01
6.63396955e-01 3.44657600e-01 -3.93776655e-01 1.76618934e-01
2.93874204e-01 -2.78928518e-01 5.29270232e-01 7.57917583e-01
8.34304631e-01 -5.94055295e-01 6.64951622e-01 -3.13734859e-01
2.06537649e-01 4.06263560e-01 2.35544696e-01 1.28595352e+00
-7.70287335e-01 2.99693316e-01 3.21925312e-01 -7.32740462e-01
-1.22585845e+00 -1.23020971e+00 -4.45364416e-01 1.66626763e+00
7.16391802e-01 -5.36928415e-01 -2.31897250e-01 -5.59079349e-01
3.14251840e-01 9.58998725e-02 -7.41283596e-01 1.31565034e-01
-7.24059790e-02 -2.46258974e-01 5.87681711e-01 7.49214888e-01
4.82823879e-01 -1.35634530e+00 -5.19521236e-01 -6.55264631e-02
-1.86985612e-01 -1.51001167e+00 -5.60974121e-01 1.26939774e-01
-3.57026100e-01 -1.33474946e+00 -4.03210461e-01 -1.15062630e+00
3.85631144e-01 9.01872218e-01 1.41349268e+00 1.93726823e-01
-8.15680683e-01 4.38336611e-01 -4.64576632e-01 -8.64901483e-01
-2.81152397e-01 -2.80152947e-01 3.68679792e-01 -2.54651695e-01
6.78840339e-01 9.59220827e-02 -5.05492866e-01 8.01554084e-01
-4.02548134e-01 -5.27299605e-02 4.26339954e-01 7.35853970e-01
4.25879508e-01 -3.42338204e-01 2.14146227e-01 -4.47886288e-01
1.40020505e-01 -3.94366771e-01 -1.04635262e+00 1.87711164e-01
-2.56235272e-01 -4.61478025e-01 2.78490335e-01 -7.84301996e-01
-6.82975531e-01 -1.56686027e-02 2.72666782e-01 -5.03595591e-01
-1.77512527e-01 -1.17124647e-01 6.39085844e-02 -2.13183761e-01
1.00994754e+00 4.16263878e-01 -2.54503459e-01 -3.56663615e-01
7.46116161e-01 7.77615666e-01 7.05305815e-01 -9.93022695e-02
8.35785806e-01 6.04858577e-01 -5.48974752e-01 -1.07583117e+00
-9.96288478e-01 -1.19052649e+00 -7.08595455e-01 -3.58040392e-01
9.27331507e-01 -1.30523777e+00 -1.01920962e+00 2.22871229e-01
-1.06647623e+00 -3.23134929e-01 -2.16192603e-01 2.02325240e-01
-3.19995344e-01 3.58596504e-01 -5.18857956e-01 -4.81786191e-01
-2.74980009e-01 -8.87460649e-01 1.20096087e+00 -6.08579116e-03
-2.15165019e-01 -5.77047706e-01 -3.58512461e-01 5.98195314e-01
1.11314908e-01 -2.65057743e-01 5.59692144e-01 -7.76306450e-01
-8.62429380e-01 -2.68858939e-01 -8.68075609e-01 3.01156640e-01
-2.58695930e-01 -5.42807989e-02 -7.94601142e-01 -4.26971316e-01
-5.56998491e-01 -8.36139202e-01 1.05185902e+00 7.48616606e-02
1.14924860e+00 -2.21246779e-01 -6.61570668e-01 4.89441544e-01
1.21634912e+00 -6.42294511e-02 1.26682907e-01 3.32672030e-01
1.08030713e+00 4.25298929e-01 7.74246931e-01 4.94662642e-01
4.03052747e-01 8.75122130e-01 8.42414260e-01 -8.45091715e-02
-2.06327394e-01 -7.81654119e-02 1.27632748e-02 6.07960165e-01
9.21731368e-02 -1.76224634e-01 -1.14356065e+00 9.62801576e-01
-1.58213055e+00 -9.58099186e-01 -2.68000007e-01 1.72149265e+00
6.92073822e-01 3.04093122e-01 5.05692959e-01 -2.37615123e-01
8.80735636e-01 1.78103164e-01 -5.66366673e-01 7.40951719e-03
-1.89991325e-01 -5.68675280e-01 4.78436917e-01 -1.72912821e-01
-1.57133842e+00 1.26233947e+00 7.00983667e+00 1.03107071e+00
-7.56186247e-01 3.57904702e-01 -8.77394993e-03 -2.63370741e-02
3.25652272e-01 -2.05023944e-01 -1.37703788e+00 3.19518089e-01
6.58071637e-02 -1.68610543e-01 1.32270247e-01 1.51275969e+00
-2.86626458e-01 1.47766814e-01 -1.06363308e+00 1.22573292e+00
1.55038774e-01 -1.66410911e+00 2.47406930e-01 1.81287453e-01
6.42988920e-01 3.82037520e-01 -3.72218229e-02 6.79156899e-01
8.83279741e-01 -8.91882539e-01 1.11347020e+00 -1.26809284e-01
1.13653624e+00 -1.46682680e-01 5.27393401e-01 4.39450473e-01
-1.46050394e+00 -5.07004321e-01 -8.54345918e-01 -1.50348350e-01
-1.52620345e-01 2.65740510e-02 -1.06248522e+00 -2.45602280e-01
1.22418714e+00 8.15607965e-01 -9.10858989e-01 1.12145209e+00
-6.20845445e-02 7.02707767e-01 -3.01305771e-01 -4.07959402e-01
3.78616661e-01 2.37460792e-01 6.02430642e-01 1.40271211e+00
-3.97416741e-01 8.04515928e-02 6.71121538e-01 9.28449094e-01
-2.62659580e-01 1.84476361e-01 -7.79817283e-01 1.02367578e-02
8.43054116e-01 1.34975016e+00 -1.01691616e+00 -6.22478068e-01
-6.72820628e-01 4.83539701e-01 6.22926414e-01 1.53952375e-01
-8.63592863e-01 -3.16793680e-01 8.21570694e-01 5.99007271e-02
7.48104453e-01 -1.84404299e-01 -1.84543028e-01 -1.06760776e+00
-7.00146481e-02 -7.67177999e-01 5.85422754e-01 -9.54053044e-01
-1.39779496e+00 5.30780077e-01 -4.05312106e-02 -1.20117843e+00
3.75843465e-01 -9.13894355e-01 -2.69683361e-01 2.42634952e-01
-1.09607196e+00 -1.29782915e+00 -8.69645238e-01 7.04589963e-01
9.00002420e-01 -5.18180966e-01 4.44468886e-01 6.01116978e-02
-5.89309990e-01 5.66224158e-01 2.95592360e-02 6.55021071e-01
6.90823615e-01 -1.13193810e+00 5.88360786e-01 6.33551002e-01
6.12340629e-01 5.18209100e-01 4.83172774e-01 -5.33003807e-01
-1.19822252e+00 -1.43838131e+00 3.07180583e-01 -1.12824917e+00
9.85758543e-01 -9.41651285e-01 -8.56400788e-01 8.22754800e-01
-1.53957143e-01 3.92856449e-01 2.60351211e-01 1.08185314e-01
-1.03915262e+00 3.08479704e-02 -6.58313036e-01 5.27782083e-01
1.43292558e+00 -6.75409853e-01 -8.02891016e-01 7.16465890e-01
1.06984949e+00 -3.37711006e-01 -6.61007345e-01 2.14551076e-01
5.91112375e-01 -6.68147326e-01 1.30673361e+00 -7.19066203e-01
2.27073714e-01 -2.70577788e-01 -4.15837467e-01 -6.10866725e-01
-5.72631657e-01 -2.07355201e-01 -2.70294309e-01 1.13637543e+00
3.22106965e-02 -3.85507882e-01 5.66354990e-01 1.44135326e-01
-7.25985765e-02 -3.37840199e-01 -7.16487765e-01 -1.10270870e+00
-2.06417352e-01 -5.73794186e-01 3.69159609e-01 7.23971665e-01
-7.19472021e-02 5.18916845e-01 -3.20647717e-01 4.32569943e-02
7.49555409e-01 2.38907889e-01 1.17669904e+00 -1.54889607e+00
9.13275555e-02 -7.05429077e-01 -9.51864243e-01 -1.42478120e+00
-1.21770896e-01 -9.64404583e-01 2.16146529e-01 -1.25184703e+00
7.14918733e-01 -6.20769441e-01 -9.19980407e-02 6.52351081e-01
-1.35927990e-01 1.06218624e+00 3.54145974e-01 6.98986113e-01
-1.44559479e+00 4.09213454e-01 1.20442951e+00 -5.06292939e-01
6.47394806e-02 -3.16415250e-01 -8.27823281e-01 8.37066412e-01
2.43520558e-01 -3.46060634e-01 -1.04855783e-01 -2.47901127e-01
1.05406344e-01 -5.75450242e-01 8.67715895e-01 -7.39329576e-01
1.34571299e-01 -2.51011640e-01 1.32762507e-01 -8.74259055e-01
3.98658782e-01 -7.30114996e-01 -4.17061657e-01 2.66676098e-01
-2.80515492e-01 -4.42541420e-01 5.97041883e-02 8.65754843e-01
-1.34839416e-01 1.02601647e-01 8.16224217e-01 -2.29533494e-01
-1.81978798e+00 2.89018184e-01 -3.24275166e-01 4.39158976e-01
1.34731805e+00 -4.30717677e-01 -6.39774323e-01 -1.52686194e-01
-6.84314430e-01 5.37092090e-01 4.39790428e-01 8.32313120e-01
6.01755917e-01 -1.26506066e+00 -6.54069483e-01 1.94755960e-02
1.01501691e+00 1.26058552e-02 2.45562121e-02 5.41877091e-01
-4.36236650e-01 6.31225348e-01 -2.38812774e-01 -1.26524198e+00
-1.65409184e+00 1.24553597e+00 5.28202094e-02 2.91324586e-01
-9.44045186e-01 1.22611344e+00 9.89511907e-01 -3.92028600e-01
5.55950880e-01 -1.83287248e-01 -4.55091864e-01 1.42578632e-01
7.57048965e-01 2.31900245e-01 -2.85545826e-01 -9.17437971e-01
-5.30150712e-01 4.26071554e-01 -2.16077566e-01 5.32970369e-01
1.02545214e+00 -3.50423992e-01 -7.43846521e-02 4.02750224e-01
9.44848597e-01 -3.65140468e-01 -1.26492751e+00 -7.27458596e-01
1.13239149e-02 -4.64987367e-01 -1.94830939e-01 -5.47681630e-01
-7.19659328e-01 7.64513433e-01 5.07394075e-01 3.11662972e-01
6.20533049e-01 9.11012650e-01 2.57270098e-01 7.42079616e-01
6.40848994e-01 -9.50326860e-01 6.25458360e-01 6.05036020e-01
1.06163383e+00 -1.71335936e+00 1.06113270e-01 -7.76026189e-01
-8.21923733e-01 6.10162914e-01 9.24317062e-01 -7.11133378e-03
5.11952996e-01 6.79304674e-02 2.13323012e-01 -7.14069426e-01
-7.45602548e-01 -6.49112284e-01 6.31694376e-01 7.67170787e-01
-1.43246174e-01 2.81945378e-01 1.68212548e-01 4.97970462e-01
-1.12257645e-01 -5.43254137e-01 4.20355529e-01 3.85730147e-01
-7.75249958e-01 -1.75226361e-01 -4.62565243e-01 6.04748487e-01
4.31413241e-02 -1.61702484e-01 -5.62227607e-01 1.15411580e+00
2.08411798e-01 1.02933800e+00 2.28287742e-01 -2.14000538e-01
2.02791452e-01 -2.15426430e-01 1.82803914e-01 -1.02723980e+00
-3.02887321e-01 -2.67046064e-01 -7.04422146e-02 -6.02987409e-01
-1.92957502e-02 -4.20111954e-01 -7.67710805e-01 -1.19477831e-01
-6.90272093e-01 -2.09297776e-01 4.55497593e-01 7.12197483e-01
2.39768773e-01 2.40984038e-01 2.86469758e-01 -8.66336346e-01
-5.00436068e-01 -1.10781229e+00 -7.50384867e-01 7.85977542e-01
3.92407089e-01 -1.14920306e+00 -2.66404152e-01 1.50626019e-01] | [9.697671890258789, 1.5315479040145874] |
3c29a36a-def2-4314-8e5d-cb2f834116e2 | cola-weakly-supervised-temporal-action | 2103.16392 | null | https://arxiv.org/abs/2103.16392v2 | https://arxiv.org/pdf/2103.16392v2.pdf | CoLA: Weakly-Supervised Temporal Action Localization with Snippet Contrastive Learning | Weakly-supervised temporal action localization (WS-TAL) aims to localize actions in untrimmed videos with only video-level labels. Most existing models follow the "localization by classification" procedure: locate temporal regions contributing most to the video-level classification. Generally, they process each snippet (or frame) individually and thus overlook the fruitful temporal context relation. Here arises the single snippet cheating issue: "hard" snippets are too vague to be classified. In this paper, we argue that learning by comparing helps identify these hard snippets and we propose to utilize snippet Contrastive learning to Localize Actions, CoLA for short. Specifically, we propose a Snippet Contrast (SniCo) Loss to refine the hard snippet representation in feature space, which guides the network to perceive precise temporal boundaries and avoid the temporal interval interruption. Besides, since it is infeasible to access frame-level annotations, we introduce a Hard Snippet Mining algorithm to locate the potential hard snippets. Substantial analyses verify that this mining strategy efficaciously captures the hard snippets and SniCo Loss leads to more informative feature representation. Extensive experiments show that CoLA achieves state-of-the-art results on THUMOS'14 and ActivityNet v1.2 datasets. CoLA code is publicly available at https://github.com/zhang-can/CoLA. | ['Yuexian Zou', 'Jie Chen', 'Dongming Yang', 'Meng Cao', 'Can Zhang'] | 2021-03-30 | null | http://openaccess.thecvf.com//content/CVPR2021/html/Zhang_CoLA_Weakly-Supervised_Temporal_Action_Localization_With_Snippet_Contrastive_Learning_CVPR_2021_paper.html | http://openaccess.thecvf.com//content/CVPR2021/papers/Zhang_CoLA_Weakly-Supervised_Temporal_Action_Localization_With_Snippet_Contrastive_Learning_CVPR_2021_paper.pdf | cvpr-2021-1 | ['weakly-supervised-action-localization', 'weakly-supervised-temporal-action'] | ['computer-vision', 'computer-vision'] | [ 2.50750035e-01 -6.99633285e-02 -9.90575135e-01 -1.51287794e-01
-7.76805401e-01 -4.05989081e-01 3.76477510e-01 -7.45147094e-02
-1.72231793e-01 6.20204151e-01 2.37871185e-01 -2.03596354e-02
-4.35763299e-01 -1.37501478e-01 -5.89002252e-01 -8.00993383e-01
-5.35826504e-01 -1.68298021e-01 4.38660920e-01 2.24967748e-01
1.51900262e-01 4.20443974e-02 -1.43342185e+00 6.17163301e-01
5.79933405e-01 1.36356807e+00 2.65453398e-01 2.80540049e-01
2.08265513e-01 1.40819931e+00 -4.87289637e-01 7.53899962e-02
2.01966405e-01 -5.56608498e-01 -8.21125150e-01 1.94985613e-01
3.39765787e-01 -2.61296809e-01 -4.51515526e-01 8.46505642e-01
7.29360878e-02 8.35205168e-02 2.97204077e-01 -1.68198788e+00
-1.62397742e-01 8.22648764e-01 -8.90927851e-01 7.19400942e-01
5.07199228e-01 2.18141422e-01 1.29384708e+00 -9.01595235e-01
5.43476403e-01 9.51997340e-01 7.62454569e-01 2.81419963e-01
-8.75230014e-01 -6.44362271e-01 4.41459090e-01 7.16279566e-01
-1.73069108e+00 -4.27536666e-01 9.33354139e-01 -3.43315870e-01
6.99148774e-01 2.83010662e-01 5.86568475e-01 1.38421047e+00
-1.73900388e-02 1.30844903e+00 8.58136833e-01 -1.35036200e-01
2.07110435e-01 -3.63879830e-01 9.30511802e-02 8.99462759e-01
-1.39613524e-01 -2.78162044e-02 -1.05190253e+00 1.84409142e-01
7.12620676e-01 2.93895900e-01 -3.04247826e-01 -2.05612436e-01
-1.43672168e+00 4.72277939e-01 1.52400330e-01 4.94820416e-01
-3.15528750e-01 2.99875945e-01 5.87574005e-01 1.21188432e-01
3.49537462e-01 1.47114724e-01 -4.56408739e-01 -6.58841729e-01
-1.01187980e+00 -5.76912798e-02 2.37964898e-01 9.61589217e-01
7.20153928e-01 -4.10838500e-02 -4.64111954e-01 6.62603438e-01
1.32766170e-02 1.14182541e-02 3.62619609e-01 -1.34869528e+00
6.19692445e-01 6.79918706e-01 1.27277020e-02 -1.09171879e+00
-3.63023967e-01 -2.66905129e-01 -6.95142150e-01 -1.55051574e-01
5.29875755e-01 5.87663315e-02 -5.56158900e-01 1.56395411e+00
3.83219384e-02 7.40273297e-01 -4.21447754e-01 8.97825658e-01
4.65136558e-01 5.03816545e-01 9.58705917e-02 -5.05041480e-01
1.29251492e+00 -1.12649679e+00 -8.45223904e-01 -1.87098891e-01
7.04369485e-01 -3.02789062e-01 1.28703761e+00 4.75920171e-01
-9.67357993e-01 -6.48906708e-01 -9.05678809e-01 2.34063253e-01
-1.16312727e-01 3.25620890e-01 5.68262219e-01 2.40435839e-01
-7.05515027e-01 6.11815333e-01 -1.00446188e+00 -2.15184823e-01
9.46180224e-01 8.14335272e-02 -2.73623139e-01 8.22377279e-02
-1.06477892e+00 4.72250730e-01 5.07871091e-01 5.93526615e-03
-1.03423059e+00 -6.53232634e-01 -9.41236854e-01 5.96891483e-03
1.13524759e+00 -1.83023021e-01 1.20005679e+00 -1.11090553e+00
-1.04189539e+00 5.99813402e-01 -5.01487494e-01 -7.50918746e-01
4.97985721e-01 -2.05013931e-01 -5.02912760e-01 6.70240521e-01
4.64390308e-01 5.24344742e-01 9.64719772e-01 -1.02333903e+00
-1.03613150e+00 -7.10243061e-02 8.71331990e-02 1.06742822e-01
-4.52414423e-01 4.31571947e-03 -6.23367608e-01 -9.08743620e-01
2.56551832e-01 -6.08552873e-01 1.24611914e-01 1.26134336e-01
-4.30368304e-01 -6.53080881e-01 1.14015102e+00 -3.56345356e-01
1.82419729e+00 -2.30353379e+00 -1.59268662e-01 2.18581483e-01
5.44629633e-01 5.78029230e-02 -8.43391754e-04 9.97692272e-02
-1.27573341e-01 2.16978118e-01 3.57056782e-02 -3.00131291e-01
-4.30387259e-02 3.38679343e-01 -9.36778188e-02 6.26751244e-01
1.30189851e-01 8.89594972e-01 -1.12800753e+00 -8.70526254e-01
3.50624323e-01 1.97008058e-01 -3.80017459e-01 -7.85049200e-02
-1.71733439e-01 3.87121946e-01 -5.68582833e-01 1.18257546e+00
1.48034438e-01 -4.90214795e-01 1.08470909e-01 -4.81389552e-01
-1.69584811e-01 2.04362512e-01 -9.89209414e-01 1.75669730e+00
2.43590642e-02 7.54926383e-01 -2.21834227e-01 -1.23651469e+00
5.32677889e-01 2.32451633e-01 1.08963823e+00 -7.20706403e-01
-6.66198228e-03 1.54774599e-02 -2.11312994e-01 -7.71360099e-01
1.27482951e-01 3.25596631e-01 -1.59873217e-01 1.51307225e-01
1.15985580e-01 8.28992307e-01 4.31360334e-01 2.59181619e-01
1.53619909e+00 5.02840221e-01 5.20411193e-01 -7.03334510e-02
4.36412185e-01 -1.28899440e-01 1.01254165e+00 7.42722034e-01
-7.27674961e-01 5.74894369e-01 7.14867890e-01 -4.36939955e-01
-5.18249631e-01 -1.12876332e+00 9.32115465e-02 1.24956453e+00
3.64347875e-01 -8.89930904e-01 -6.84899986e-01 -1.16164553e+00
-3.54365289e-01 3.12173188e-01 -7.99647331e-01 -1.37567922e-01
-6.72720730e-01 -2.48931229e-01 6.02759182e-01 7.17367887e-01
6.72461033e-01 -1.12578440e+00 -7.89101005e-01 9.21065360e-02
-7.12120354e-01 -1.26736438e+00 -8.09390247e-01 2.20076114e-01
-5.40228546e-01 -1.32485855e+00 -3.34236115e-01 -6.30433083e-01
5.44394195e-01 5.33728361e-01 1.02072227e+00 3.95961776e-02
-6.92485422e-02 3.05207402e-01 -8.65282357e-01 4.58818721e-03
2.17391506e-01 -1.43038735e-01 1.34977490e-01 2.22525388e-01
6.88582957e-01 -6.44917667e-01 -7.34593153e-01 8.08689237e-01
-5.06305039e-01 5.00250645e-02 4.51594979e-01 5.80850363e-01
9.00023460e-01 4.55932558e-01 6.78622186e-01 -4.48580563e-01
8.97587538e-02 -6.31352842e-01 -2.89295137e-01 1.25538662e-01
-4.77719963e-01 -3.17205429e-01 5.71412385e-01 -5.66984594e-01
-7.58067727e-01 8.60352069e-02 3.31740797e-01 -9.54446733e-01
-2.91361362e-01 4.65415835e-01 -1.80493414e-01 1.72782838e-01
4.31173623e-01 3.76082599e-01 -2.34274611e-01 -2.76782215e-01
-7.83395469e-02 2.20461696e-01 6.28124833e-01 -5.39802730e-01
5.62557459e-01 7.05325603e-01 -1.21376403e-01 -7.66100526e-01
-1.18903983e+00 -9.01348472e-01 -4.96664524e-01 -7.38831222e-01
8.55275571e-01 -9.39118683e-01 -9.86183405e-01 4.89991792e-02
-7.61823237e-01 -2.67282695e-01 -4.81805414e-01 4.35604632e-01
-8.34362686e-01 5.22481084e-01 -4.21110690e-01 -8.08509707e-01
1.02261342e-01 -9.32296872e-01 1.07894313e+00 1.12438962e-01
-5.75113177e-01 -7.17323482e-01 -1.82001770e-01 6.34776592e-01
-2.56175607e-01 3.03401649e-01 3.45679671e-01 -5.25444686e-01
-6.10715866e-01 -7.59326443e-02 -1.19390868e-01 2.40436167e-01
1.80349410e-01 -1.85325831e-01 -8.81060660e-01 -4.78495359e-02
-8.41382816e-02 -2.75367945e-01 9.70131099e-01 6.68673754e-01
1.90565455e+00 -5.85209846e-01 -3.79302174e-01 5.60068309e-01
9.30444956e-01 3.28460306e-01 7.10754633e-01 2.72346944e-01
7.94750810e-01 4.35607016e-01 1.05744910e+00 8.14707339e-01
3.99806589e-01 8.14579606e-01 5.12918115e-01 7.98145831e-02
-5.41159883e-02 -4.63331342e-01 7.86816299e-01 4.35992956e-01
-3.44782680e-01 -2.96705097e-01 -7.73767114e-01 6.56606138e-01
-2.27727342e+00 -1.45241785e+00 -4.64750174e-03 1.85618687e+00
7.74462461e-01 4.22520518e-01 4.19419080e-01 3.85690123e-01
6.37033522e-01 4.99479949e-01 -4.29902583e-01 3.06795061e-01
-2.79090106e-01 -1.07193246e-01 4.61476177e-01 1.57710001e-01
-1.46518064e+00 9.56104040e-01 4.54503727e+00 1.39646840e+00
-5.84090769e-01 2.89367467e-01 6.68888807e-01 -4.62427080e-01
3.26073587e-01 6.42205123e-03 -6.77508354e-01 9.17393208e-01
6.09237075e-01 8.63830745e-02 2.62201250e-01 7.81516790e-01
1.04043448e+00 -3.88821989e-01 -1.39285791e+00 1.22793269e+00
2.53850669e-02 -1.50729144e+00 -2.54161566e-01 -7.12363645e-02
5.86770117e-01 -2.49242842e-01 -8.61613154e-02 3.49349201e-01
-1.69586599e-01 -7.86300719e-01 8.95944715e-01 5.52579880e-01
5.58995724e-01 -5.49088120e-01 5.38588226e-01 2.07231924e-01
-1.69874334e+00 -4.29027587e-01 1.03153966e-01 -2.88650785e-02
2.37751395e-01 4.25126255e-01 -4.97126728e-01 4.41870868e-01
1.04058421e+00 1.37833607e+00 -5.78240871e-01 1.02617490e+00
-2.33230859e-01 9.40788031e-01 -1.69534177e-01 2.31600791e-01
5.01671076e-01 9.62964445e-02 5.96301675e-01 1.07641077e+00
2.03432515e-01 8.17611068e-02 5.83961189e-01 6.60862982e-01
-6.93094060e-02 -3.34923297e-01 -4.78279173e-01 -1.22598767e-01
5.83800316e-01 1.02657986e+00 -8.11692834e-01 -2.60328054e-01
-4.20666575e-01 8.06706727e-01 3.94288264e-03 4.90189970e-01
-1.22523117e+00 -4.55441140e-02 5.64387619e-01 3.22227478e-01
3.65182310e-01 -1.10500254e-01 -2.90421546e-01 -1.14971578e+00
2.81023383e-01 -7.53057182e-01 7.58802652e-01 -6.67355359e-01
-1.13985574e+00 1.16466329e-01 2.13861614e-01 -1.89727724e+00
-1.54660657e-01 -1.38198823e-01 -6.63251758e-01 2.97134966e-02
-1.28170645e+00 -1.11351502e+00 -3.55506718e-01 8.84877145e-01
1.16641617e+00 -8.32869858e-02 2.22092539e-01 5.05203366e-01
-7.13022828e-01 7.07050383e-01 -4.32108074e-01 3.20360601e-01
4.89093006e-01 -1.00180817e+00 -3.88788730e-01 9.21375751e-01
3.54414582e-01 3.22522432e-01 5.02334177e-01 -5.90836406e-01
-1.10197222e+00 -1.26247787e+00 8.49797308e-01 -5.88301480e-01
9.01386738e-01 -1.79851472e-01 -7.28546023e-01 7.55855322e-01
-4.96504866e-02 2.81582892e-01 6.21844888e-01 -3.17737646e-03
-2.93723762e-01 -2.59256423e-01 -8.81091535e-01 5.43505907e-01
1.45918834e+00 -5.52362204e-01 -4.94241267e-01 4.34540659e-01
5.22697687e-01 3.74648012e-02 -7.05042064e-01 4.37675625e-01
5.18900812e-01 -1.04042149e+00 9.42655563e-01 -2.78812081e-01
5.17293155e-01 -6.12852454e-01 -1.08123593e-01 -6.38820231e-01
-3.23873997e-01 -9.37803268e-01 -8.33937645e-01 1.23240745e+00
2.98098415e-01 -6.89446554e-02 9.46372807e-01 5.93929999e-02
-3.18007410e-01 -8.68054569e-01 -1.12134337e+00 -1.10306203e+00
-6.95961297e-01 -7.33805299e-01 1.34341806e-01 1.12038958e+00
4.92372394e-01 8.62538666e-02 -7.24188626e-01 1.06759407e-01
6.29459381e-01 -1.22982949e-01 3.82436335e-01 -8.40877056e-01
-1.79248288e-01 -6.69461489e-01 -3.75155598e-01 -1.13145602e+00
2.01196343e-01 -6.10168874e-01 1.67166591e-01 -1.11967194e+00
2.34841183e-01 -3.09850395e-01 -6.42050028e-01 9.52188134e-01
9.36131030e-02 3.24745387e-01 1.03857972e-01 3.43211472e-01
-1.51328754e+00 4.79031473e-01 1.06624985e+00 -1.86033979e-01
-2.66679227e-01 1.06585585e-01 -4.62406605e-01 1.02215624e+00
8.23178113e-01 -3.68712425e-01 -5.90054691e-01 -7.82684311e-02
5.95084354e-02 2.17730422e-02 7.15285122e-01 -1.20359349e+00
2.96414882e-01 -3.56530249e-01 3.26655269e-01 -7.98892081e-01
3.17939788e-01 -8.41401696e-01 -2.05769688e-01 2.27179751e-01
-4.82135057e-01 -2.45615095e-01 -1.17310978e-01 8.04552913e-01
-3.09371144e-01 1.27391908e-02 5.61135530e-01 1.98331010e-02
-1.23576510e+00 5.20278394e-01 -4.54870343e-01 1.79255933e-01
1.30018258e+00 -6.00232899e-01 -2.99787462e-01 -3.01992118e-01
-8.80993724e-01 5.33805251e-01 8.27053785e-02 4.31657881e-01
7.59598851e-01 -1.45990014e+00 -3.55402827e-01 -2.83337831e-02
3.27822238e-01 -2.54350275e-01 4.72493201e-01 1.34707963e+00
-1.73432119e-02 2.24623054e-01 2.39390344e-03 -7.60428727e-01
-1.31070328e+00 4.31110680e-01 1.51743770e-01 -1.73649833e-01
-7.82177389e-01 7.84580588e-01 2.72559226e-01 5.29082298e-01
6.85762703e-01 -5.31560004e-01 -2.87173003e-01 2.96806604e-01
6.26956165e-01 6.62181973e-01 -3.95646125e-01 -6.46511078e-01
-7.01239765e-01 3.78125101e-01 8.88926312e-02 3.04643512e-01
1.04375982e+00 -3.25189590e-01 2.60727704e-01 4.33741510e-01
1.09931135e+00 -3.63632560e-01 -1.78342032e+00 -3.46816063e-01
2.16042295e-01 -5.90933383e-01 -4.80923951e-02 -7.04140067e-01
-1.01920295e+00 6.73892438e-01 5.44159532e-01 2.30160892e-01
1.35691321e+00 4.31284040e-01 8.21854591e-01 3.34374368e-01
3.54413778e-01 -1.43925130e+00 5.45882106e-01 3.98993522e-01
7.39426613e-01 -1.31256163e+00 -1.49098728e-02 -3.30733210e-01
-7.70911455e-01 7.38704324e-01 8.78115296e-01 1.72207877e-01
4.84477609e-01 1.90286741e-01 -3.62173647e-01 -2.75768161e-01
-8.34350288e-01 -3.64229202e-01 2.19220370e-01 4.94345039e-01
1.17891757e-02 9.31639597e-02 -2.93710887e-01 8.94439399e-01
2.68471509e-01 1.59567222e-01 4.03017730e-01 1.00240779e+00
-5.18497705e-01 -6.09422684e-01 -1.36882439e-01 5.02498448e-01
-4.67905492e-01 1.38117507e-01 -1.64806560e-01 6.90117359e-01
5.09553492e-01 1.05760539e+00 -1.78968851e-02 -7.80812502e-01
1.23051889e-01 -1.56855434e-01 2.44960442e-01 -3.97871017e-01
-2.92909384e-01 4.10140127e-01 1.84597164e-01 -1.19668853e+00
-8.77031386e-01 -8.28291774e-01 -1.45028913e+00 -1.41183555e-01
5.69693781e-02 2.37929691e-02 -1.48437977e-01 1.03417528e+00
2.91669071e-01 6.10906959e-01 6.78576291e-01 -7.55558848e-01
-1.14193000e-01 -6.98943675e-01 -6.31854355e-01 2.81132013e-01
4.39884573e-01 -9.92730379e-01 -3.26449364e-01 4.76888090e-01] | [8.466031074523926, 0.6451312303543091] |
a54dacd1-00ed-4589-b558-fa8911ab50ad | quantifying-and-learning-static-vs-dynamic | 2211.01783 | null | https://arxiv.org/abs/2211.01783v1 | https://arxiv.org/pdf/2211.01783v1.pdf | Quantifying and Learning Static vs. Dynamic Information in Deep Spatiotemporal Networks | There is limited understanding of the information captured by deep spatiotemporal models in their intermediate representations. For example, while evidence suggests that action recognition algorithms are heavily influenced by visual appearance in single frames, no quantitative methodology exists for evaluating such static bias in the latent representation compared to bias toward dynamics. We tackle this challenge by proposing an approach for quantifying the static and dynamic biases of any spatiotemporal model, and apply our approach to three tasks, action recognition, automatic video object segmentation (AVOS) and video instance segmentation (VIS). Our key findings are: (i) Most examined models are biased toward static information. (ii) Some datasets that are assumed to be biased toward dynamics are actually biased toward static information. (iii) Individual channels in an architecture can be biased toward static, dynamic or a combination of the two. (iv) Most models converge to their culminating biases in the first half of training. We then explore how these biases affect performance on dynamically biased datasets. For action recognition, we propose StaticDropout, a semantically guided dropout that debiases a model from static information toward dynamics. For AVOS, we design a better combination of fusion and cross connection layers compared with previous architectures. | ['Konstantinos G. Derpanis', 'Richard P. Wildes', 'Neil D. B. Bruce', 'Md Amirul Islam', 'Mennatullah Siam', 'Matthew Kowal'] | 2022-11-03 | null | null | null | null | ['video-instance-segmentation', 'video-object-segmentation'] | ['computer-vision', 'computer-vision'] | [ 3.31810832e-01 -1.14056356e-01 -4.24464196e-01 -4.50265557e-01
-2.30788901e-01 -7.11232483e-01 9.43118989e-01 -2.69064277e-01
-4.77193147e-01 5.13997138e-01 4.59546924e-01 -2.83862744e-02
1.39509186e-01 -3.97847086e-01 -9.80861127e-01 -8.99924040e-01
5.93152456e-02 3.97931695e-01 5.88130832e-01 -8.14606547e-02
1.03090808e-01 4.94240373e-01 -1.48036802e+00 7.43790030e-01
4.92820144e-01 1.02989769e+00 -6.57941774e-02 6.36138558e-01
-8.46648589e-02 1.21114552e+00 -7.90386498e-01 -1.03952013e-01
2.52527624e-01 -6.32908404e-01 -8.74836206e-01 8.20665732e-02
8.86741817e-01 -5.92146456e-01 -4.74139392e-01 1.02465701e+00
1.47363424e-01 7.40923434e-02 6.76137388e-01 -1.35530710e+00
-7.07786739e-01 8.47151041e-01 -4.66287225e-01 7.47062862e-01
2.00905979e-01 7.78554499e-01 6.79680705e-01 -5.50940037e-01
8.58897924e-01 1.64869881e+00 5.15298009e-01 7.40303695e-01
-1.43974972e+00 -4.26166564e-01 9.78231609e-01 1.60434529e-01
-1.09363651e+00 -6.07212365e-01 7.78780103e-01 -7.40075350e-01
7.62692094e-01 2.06631482e-01 8.34093153e-01 1.88569844e+00
1.79280996e-01 1.11587560e+00 1.16533387e+00 1.43669203e-01
3.37248325e-01 7.14902133e-02 3.48794699e-01 3.22931528e-01
1.81249678e-01 1.89082310e-01 -7.93982804e-01 1.18694931e-01
6.85431123e-01 2.65240986e-02 -3.19005042e-01 -4.42897320e-01
-1.16652429e+00 4.49149847e-01 4.46927398e-01 4.48565125e-01
-3.61270189e-01 6.11105323e-01 3.19613487e-01 3.00944000e-01
4.06882942e-01 3.14763278e-01 -3.16084385e-01 -3.09929490e-01
-1.11446881e+00 3.76658529e-01 2.39726648e-01 5.49998522e-01
6.43721342e-01 3.78955364e-01 -6.37448132e-01 4.94755715e-01
1.73616990e-01 3.74940813e-01 6.73786104e-01 -1.02403510e+00
3.32103878e-01 6.24447227e-01 3.56097311e-01 -9.74282384e-01
-4.22727704e-01 -9.83587056e-02 -3.34994614e-01 2.66789526e-01
8.45205247e-01 -4.85116281e-02 -1.19110465e+00 2.09465837e+00
7.60318828e-04 1.41921520e-01 -3.68040353e-02 1.11384749e+00
6.58570349e-01 5.25937974e-01 4.10307437e-01 3.69515941e-02
9.46924627e-01 -8.63507450e-01 -7.09436595e-01 -3.22047144e-01
3.77606720e-01 -2.80637592e-01 1.11458075e+00 2.36991093e-01
-1.26425958e+00 -7.43148208e-01 -8.73900890e-01 -7.54215121e-02
-3.70332569e-01 2.51646675e-02 6.79686844e-01 5.77752888e-01
-1.25690556e+00 7.43050218e-01 -1.18494391e+00 -4.96574253e-01
7.59885550e-01 2.28100300e-01 -9.96581465e-02 2.34824196e-01
-9.24685895e-01 7.50954747e-01 2.39427894e-01 1.67814180e-01
-1.43967283e+00 -6.85998857e-01 -4.67124283e-01 -2.11597100e-01
1.38327301e-01 -5.49047828e-01 1.15334558e+00 -2.06523871e+00
-1.24378943e+00 7.60795891e-01 -3.59224379e-01 -7.87063122e-01
9.06368554e-01 -3.30879241e-01 -2.05169335e-01 3.13926131e-01
-1.70225337e-01 1.13009346e+00 1.04729700e+00 -1.40016437e+00
-5.58829188e-01 -4.31053221e-01 3.97919714e-01 1.40314460e-01
-2.87282377e-01 -1.21372931e-01 -3.03128928e-01 -7.57911980e-01
7.00859055e-02 -9.44798887e-01 -8.35820846e-03 1.15868017e-01
-2.62578875e-01 -1.20329551e-01 1.10228181e+00 -3.58507127e-01
1.22663999e+00 -2.04440069e+00 3.73760760e-01 -2.64972299e-01
1.94130987e-01 3.11753213e-01 -1.69998720e-01 1.21339425e-01
-2.22384050e-01 2.26524726e-01 -3.22629809e-02 -1.85043827e-01
-1.51437402e-01 2.49954283e-01 -5.83169699e-01 5.15158713e-01
3.91112328e-01 1.06707191e+00 -9.15806770e-01 -2.54744262e-01
2.26423040e-01 3.60917151e-01 -4.50353950e-01 -1.03181668e-01
-4.12833691e-01 6.92584157e-01 -3.07856172e-01 7.44177699e-01
5.62668681e-01 -2.19285622e-01 7.52647966e-03 -3.94971460e-01
-1.04607448e-01 2.47235015e-01 -9.57006931e-01 1.48024142e+00
-6.20635338e-02 1.02567041e+00 -2.38710150e-01 -9.47085023e-01
4.94672030e-01 2.27867201e-01 6.21924281e-01 -7.12464035e-01
1.35335535e-01 -7.67960399e-02 1.74905598e-01 -5.09566009e-01
4.44960952e-01 7.92312175e-02 2.80929625e-01 4.98559088e-01
1.22552894e-01 4.06747788e-01 2.92257637e-01 3.10694605e-01
1.10386896e+00 4.97074157e-01 -3.67409289e-01 -2.03819647e-01
1.59510747e-01 6.72573075e-02 5.63238263e-01 9.65555608e-01
-6.50990903e-01 7.29679704e-01 8.72860551e-01 -7.06201792e-01
-7.42926240e-01 -1.01026857e+00 1.53062511e-02 1.12755883e+00
3.06109101e-01 8.94423425e-02 -8.65402997e-01 -9.45925772e-01
6.33128881e-02 6.71773970e-01 -9.95911598e-01 -5.40107906e-01
-5.78253329e-01 -7.37785935e-01 6.23995066e-01 6.81695879e-01
5.78611195e-01 -1.04945934e+00 -9.76842821e-01 1.98586993e-02
-1.47383884e-01 -1.10033250e+00 -4.06209737e-01 5.64536490e-02
-1.00574100e+00 -1.10238278e+00 -6.84664965e-01 -1.48571342e-01
5.73968530e-01 2.33956262e-01 1.18851876e+00 -9.28847715e-02
4.51475345e-02 6.53811932e-01 -3.25463057e-01 -2.96496868e-01
-2.82525390e-01 -1.73079774e-01 1.50354300e-02 4.33160454e-01
4.84193861e-01 -3.57522279e-01 -9.38122571e-01 4.83397543e-01
-1.09336281e+00 1.36180944e-03 4.28881288e-01 3.01706672e-01
2.87747949e-01 -3.43928009e-01 2.88562238e-01 -8.80816877e-01
3.74712318e-01 -4.75335062e-01 -3.67930353e-01 7.43243396e-02
-3.83174360e-01 4.59936224e-02 2.82109231e-01 -9.66424942e-01
-1.03749514e+00 1.42720148e-01 2.60905683e-01 -8.70159805e-01
-3.18175733e-01 1.45197526e-01 5.92886768e-02 1.28752321e-01
7.64422834e-01 1.44024104e-01 -7.92834535e-02 -3.28796893e-01
2.53641784e-01 2.39908800e-01 2.95373142e-01 -5.34904063e-01
4.48289067e-01 1.14036822e+00 -4.63091344e-01 -5.97170472e-01
-8.25727224e-01 -1.06816962e-01 -6.98794484e-01 -6.14927530e-01
1.11851668e+00 -9.02789235e-01 -5.01711965e-01 7.91552067e-01
-9.92776394e-01 -7.08437443e-01 -5.54120779e-01 3.20208579e-01
-4.75697547e-01 5.62937520e-02 -4.36099589e-01 -7.50795245e-01
2.53830403e-01 -1.33191502e+00 1.10534000e+00 2.17912927e-01
-6.16399586e-01 -1.00441313e+00 4.31545675e-02 2.56239325e-01
4.36476082e-01 2.75243193e-01 6.33741617e-01 -5.30652285e-01
-5.51463723e-01 -5.76758385e-03 -1.02083206e-01 3.33154440e-01
6.23568073e-02 4.27522033e-01 -1.13820505e+00 -2.68312961e-01
-1.45874143e-01 -5.08595288e-01 1.28404808e+00 7.50024199e-01
1.01403069e+00 -2.72529811e-01 -3.38190913e-01 5.83765149e-01
1.06413579e+00 3.23602706e-01 6.66951895e-01 2.62863755e-01
7.11680949e-01 5.83987892e-01 4.12780315e-01 1.70303509e-01
1.41271070e-01 7.60240674e-01 4.99997050e-01 -1.25456274e-01
-4.03713435e-01 -2.91711122e-01 8.80695403e-01 3.97285670e-02
-1.76944107e-01 -3.09590608e-01 -8.89139652e-01 7.40932703e-01
-1.95613289e+00 -1.26118279e+00 -6.31837323e-02 2.04366088e+00
5.71911514e-01 3.30571055e-01 4.02482957e-01 -1.97271451e-01
5.42095065e-01 6.37192130e-01 -9.01779950e-01 -3.00382972e-01
-3.98390591e-01 -2.99731761e-01 4.94246036e-01 2.54135489e-01
-1.21433568e+00 1.11309671e+00 7.15750265e+00 4.08473939e-01
-1.59314346e+00 1.08878531e-01 9.14363027e-01 -5.80807924e-01
-3.80128592e-01 9.45345685e-02 -7.20206857e-01 7.74970949e-01
8.85508597e-01 2.86935478e-01 1.52428687e-01 7.35342860e-01
5.59546411e-01 -2.88600892e-01 -1.40501666e+00 7.18318343e-01
-6.72691613e-02 -1.39018369e+00 6.02833152e-01 -7.21513182e-02
8.87164176e-01 5.94184808e-02 4.67194915e-01 2.46669039e-01
4.61061269e-01 -1.07502663e+00 1.37623739e+00 8.53222072e-01
3.08701456e-01 -1.88892201e-01 2.71285653e-01 8.30837861e-02
-7.69692898e-01 -3.08759958e-01 -2.09177539e-01 5.99399582e-02
5.33355363e-02 1.77290559e-01 -2.21871123e-01 -7.47226104e-02
8.06498885e-01 8.66020083e-01 -7.67118275e-01 6.82357609e-01
3.15598547e-02 8.38244617e-01 -1.62272766e-01 1.87256366e-01
5.83457053e-01 -1.32086098e-01 6.07640922e-01 1.09752011e+00
-8.59719887e-02 -1.62156537e-01 9.83849838e-02 1.02164233e+00
1.70468569e-01 -4.70732361e-01 -5.71484029e-01 -3.31586212e-01
9.10046101e-02 6.86041057e-01 -9.03692603e-01 -4.76128548e-01
-4.15799379e-01 1.02945352e+00 1.52629837e-01 8.77008796e-01
-9.99707520e-01 4.53942001e-01 8.43821287e-01 2.50701785e-01
3.81991506e-01 -4.63362448e-02 -2.84329295e-01 -1.17298400e+00
-4.30804119e-02 -9.43429291e-01 4.11657214e-01 -9.71200943e-01
-8.42801571e-01 4.65251684e-01 2.19085082e-01 -1.08185709e+00
-4.60794196e-02 -5.19428551e-01 -6.01964295e-01 5.46917796e-01
-1.21513307e+00 -1.00109732e+00 -2.87523001e-01 7.15580761e-01
7.88809776e-01 -3.70475464e-02 1.83807865e-01 2.21374705e-01
-7.42954433e-01 2.81013012e-01 -2.57148832e-01 1.42042309e-01
5.07156610e-01 -1.11079764e+00 1.78531647e-01 9.83874857e-01
4.05990720e-01 5.47411859e-01 8.42636347e-01 -5.92079878e-01
-1.16674078e+00 -1.08202720e+00 3.12850952e-01 -8.59658122e-01
5.54329932e-01 -1.53433174e-01 -8.76736045e-01 8.60981405e-01
2.52999943e-02 1.44322574e-01 2.41236091e-01 -1.61097959e-01
-4.12226975e-01 -5.36410846e-02 -8.82756531e-01 7.35352278e-01
1.14103615e+00 -3.89247268e-01 -3.73584598e-01 1.59299925e-01
4.06649560e-01 -3.40831459e-01 -4.02500123e-01 3.49271148e-01
7.38621593e-01 -1.31666684e+00 7.78304815e-01 -1.02603710e+00
4.59290087e-01 -3.55025142e-01 -4.05107811e-02 -1.01211274e+00
-1.43391371e-01 -4.19555902e-01 -2.99980342e-01 9.37873125e-01
3.50251198e-01 -3.40768486e-01 9.78543282e-01 5.50429046e-01
5.44493683e-02 -6.14030182e-01 -9.04078841e-01 -7.59872794e-01
2.09812269e-01 -4.68242347e-01 3.20772797e-01 8.61599684e-01
-5.51190734e-01 4.53444794e-02 -2.92797804e-01 2.57665813e-02
1.32606819e-01 5.53914867e-02 6.90703094e-01 -8.65817726e-01
-1.14490047e-01 -6.98968053e-01 -5.50093651e-01 -1.16106832e+00
-1.13483686e-02 -5.13795316e-01 -1.84072107e-01 -1.41867018e+00
2.18995437e-01 -1.89455971e-01 -4.95488703e-01 4.11252737e-01
-6.81912750e-02 3.50055784e-01 2.53010660e-01 4.55844194e-01
-7.07152963e-01 3.54521245e-01 1.10427654e+00 -3.14623177e-01
-1.59230858e-01 -8.50925818e-02 -5.18989980e-01 9.79702592e-01
5.55761814e-01 -3.93372536e-01 -6.57608688e-01 -6.94152355e-01
1.36556521e-01 -2.56163985e-01 7.37929821e-01 -9.82952654e-01
-5.88328466e-02 -3.35887581e-01 4.79060769e-01 -3.76542687e-01
3.46227795e-01 -6.68181360e-01 7.27828294e-02 5.95003843e-01
-5.29724061e-01 7.94522557e-03 1.65167913e-01 6.62566364e-01
-2.73106366e-01 2.80792527e-02 9.24232960e-01 -2.73290783e-01
-9.22265053e-01 3.08382273e-01 -6.93853378e-01 2.59333491e-01
9.91091132e-01 -6.56452775e-01 -2.72091836e-01 -4.52261239e-01
-9.86721456e-01 6.91049919e-02 5.44542968e-01 7.48017788e-01
3.50595832e-01 -1.15135455e+00 -3.99253011e-01 1.48692802e-02
-3.27706374e-02 -4.68001127e-01 2.33948767e-01 1.16345501e+00
-4.13401663e-01 3.41152638e-01 -2.48680532e-01 -9.77022707e-01
-1.14504838e+00 3.97706479e-01 6.85438335e-01 1.67346187e-02
-3.94884050e-01 9.49187636e-01 4.29507971e-01 1.39228657e-01
4.67269421e-01 -7.08361626e-01 -2.12103695e-01 4.79027867e-01
3.71693850e-01 2.29400858e-01 -1.35847971e-01 -9.60453629e-01
-2.91228652e-01 4.63422060e-01 -2.38616630e-01 -1.31252781e-01
1.06177855e+00 -6.70426637e-02 2.18949184e-01 9.52206433e-01
1.05803764e+00 -4.07324255e-01 -1.90728664e+00 3.64520214e-02
-9.68834609e-02 -4.71312284e-01 6.00957684e-03 -7.60241449e-01
-1.28278768e+00 9.73883271e-01 1.00085723e+00 2.17170596e-01
9.37471807e-01 8.24789777e-02 3.18917960e-01 1.78609416e-01
1.84287086e-01 -1.27386498e+00 3.59489441e-01 3.62467378e-01
8.37437034e-01 -1.23243093e+00 -1.38230175e-01 9.50862914e-02
-1.05576348e+00 9.00014162e-01 8.02588642e-01 -2.92926759e-01
4.01498973e-01 6.88589886e-02 2.77574390e-01 -3.75990182e-01
-1.00373411e+00 -2.76638329e-01 2.73814231e-01 4.97289062e-01
3.61253321e-01 -2.88229316e-01 -2.39883875e-03 2.00044021e-01
2.36804396e-01 -3.81674021e-02 4.09003258e-01 8.20769072e-01
-2.59280175e-01 -5.43815434e-01 -1.90821469e-01 2.78217256e-01
-5.09049058e-01 1.67601407e-01 -8.09056163e-01 8.20561290e-01
2.06577942e-01 6.79061770e-01 4.19632524e-01 -1.56735286e-01
2.87694484e-01 2.23079681e-01 4.70241845e-01 -3.29721123e-01
-7.16480017e-01 -9.62270983e-03 -7.38378242e-02 -1.04185820e+00
-6.50542736e-01 -9.81918335e-01 -9.64200079e-01 2.65844603e-04
-3.25793438e-02 -3.08732897e-01 3.89740378e-01 9.10307109e-01
2.68392086e-01 8.08238804e-01 1.69941142e-01 -9.97704744e-01
-3.43304455e-01 -8.58316064e-01 -2.19929948e-01 8.68195593e-01
5.65991998e-01 -8.95208955e-01 -5.22784710e-01 3.05437446e-01] | [8.804965019226074, 0.4334627091884613] |
98b6bc83-608f-4536-8f36-5ca3d94f3bfc | bnn-dp-robustness-certification-of-bayesian | 2306.10742 | null | https://arxiv.org/abs/2306.10742v1 | https://arxiv.org/pdf/2306.10742v1.pdf | BNN-DP: Robustness Certification of Bayesian Neural Networks via Dynamic Programming | In this paper, we introduce BNN-DP, an efficient algorithmic framework for analysis of adversarial robustness of Bayesian Neural Networks (BNNs). Given a compact set of input points $T\subset \mathbb{R}^n$, BNN-DP computes lower and upper bounds on the BNN's predictions for all the points in $T$. The framework is based on an interpretation of BNNs as stochastic dynamical systems, which enables the use of Dynamic Programming (DP) algorithms to bound the prediction range along the layers of the network. Specifically, the method uses bound propagation techniques and convex relaxations to derive a backward recursion procedure to over-approximate the prediction range of the BNN with piecewise affine functions. The algorithm is general and can handle both regression and classification tasks. On a set of experiments on various regression and classification tasks and BNN architectures, we show that BNN-DP outperforms state-of-the-art methods by up to four orders of magnitude in both tightness of the bounds and computational efficiency. | ['Luca Laurenti', 'Morteza Lahijanian', 'Andrea Patane', 'Steven Adams'] | 2023-06-19 | null | null | null | null | ['adversarial-robustness'] | ['adversarial'] | [ 1.71370924e-01 3.62115562e-01 4.91971821e-02 -3.36708754e-01
-7.48682737e-01 -7.02602267e-01 2.97491550e-01 -2.47169822e-01
-4.93452132e-01 7.97518015e-01 -4.32413965e-01 -5.93032479e-01
-5.76308072e-01 -6.93734169e-01 -1.08102548e+00 -1.00315833e+00
-3.99331242e-01 3.90582651e-01 4.92524654e-01 -2.05107898e-01
6.41863942e-02 8.35810423e-01 -8.01730037e-01 -2.86984518e-02
2.50309348e-01 1.39050424e+00 -4.66748387e-01 7.50214756e-01
4.46365923e-01 7.60079563e-01 -5.14590859e-01 -9.11205113e-01
5.81688702e-01 2.15964690e-02 -6.25316322e-01 -7.96528578e-01
2.93545872e-01 -2.52145320e-01 -8.49995136e-01 1.54014194e+00
3.40889096e-01 4.33538020e-01 7.74046063e-01 -1.33763850e+00
-2.65469939e-01 1.09347534e+00 -4.80793148e-01 3.73032033e-01
-3.54617476e-01 -1.47187024e-01 8.14645350e-01 -4.54966307e-01
3.47766280e-01 1.55639732e+00 1.19585991e+00 9.30218577e-01
-1.60965228e+00 -8.88861299e-01 4.71521050e-01 -2.04606876e-02
-1.35324335e+00 -4.12856817e-01 6.37914538e-01 -3.88361394e-01
8.63256872e-01 3.61787826e-01 3.13513607e-01 1.25200009e+00
3.88941884e-01 4.68646973e-01 7.04321563e-01 -9.32451263e-02
5.03122330e-01 -1.15707666e-01 4.11251694e-01 7.51990497e-01
2.56681055e-01 5.08416891e-01 -3.38849813e-01 -3.47236603e-01
6.83102012e-01 -9.98876840e-02 -1.09217130e-01 -2.82809913e-01
-5.90206861e-01 9.77776051e-01 4.46478158e-01 -3.14606279e-01
6.44590938e-03 7.70919204e-01 6.06463432e-01 1.25202820e-01
6.72457635e-01 1.81573436e-01 -6.53379321e-01 -4.21093814e-02
-7.21504986e-01 5.52029729e-01 1.03976202e+00 8.06349397e-01
2.43688926e-01 3.09434794e-02 1.75008640e-01 7.01839328e-01
4.62400109e-01 6.38951838e-01 -1.95622936e-01 -1.21005487e+00
6.45211935e-01 -5.70659563e-02 1.05699347e-02 -1.17357981e+00
-4.03068602e-01 -2.35227212e-01 -1.14878857e+00 2.78798848e-01
7.87846804e-01 -3.80040973e-01 -9.73086357e-01 2.14925027e+00
1.97754115e-01 2.19707862e-01 -3.19843702e-02 5.51760256e-01
1.70179203e-01 7.87380397e-01 -7.41552189e-02 -2.52934694e-01
7.31486261e-01 -4.42294747e-01 -2.27172196e-01 -2.80279368e-02
3.65563005e-01 2.31964178e-02 5.00386894e-01 5.49205005e-01
-1.22066367e+00 -2.30071694e-01 -1.02488458e+00 2.89941281e-01
-4.28729266e-01 -5.04960477e-01 3.19324821e-01 9.26890910e-01
-1.04480982e+00 1.10062921e+00 -1.37812936e+00 3.26093465e-01
6.89494669e-01 8.05403709e-01 -4.81307833e-03 8.05733427e-02
-1.23698342e+00 8.16062033e-01 4.24957693e-01 6.21540427e-01
-1.34859598e+00 -9.61440623e-01 -7.92285919e-01 -1.09279826e-01
2.60267794e-01 -3.51247221e-01 1.21069217e+00 -4.67968047e-01
-1.51162767e+00 4.36595351e-01 9.64589566e-02 -1.05013537e+00
7.45037377e-01 -2.89284408e-01 -3.17398249e-03 1.32188410e-01
-5.13048291e-01 4.27025735e-01 8.10834169e-01 -1.09002960e+00
-3.77675384e-01 -5.16142309e-01 2.29964763e-01 -3.52751553e-01
-1.54524758e-01 9.52215791e-02 -1.93853632e-01 -5.76216877e-01
1.57499105e-01 -1.11840975e+00 -6.86328292e-01 -2.78065470e-03
-7.27198780e-01 -1.50677279e-01 7.44196236e-01 -6.04077697e-01
1.03220022e+00 -1.93829083e+00 4.32696134e-01 6.47322953e-01
3.18748653e-01 1.57032833e-01 2.19904408e-01 -1.25253558e-01
-2.00488254e-01 2.21579164e-01 -3.77963960e-01 -3.61672163e-01
4.12748098e-01 4.51288581e-01 -9.61480379e-01 8.47345471e-01
3.77875119e-02 5.18918157e-01 -5.88612139e-01 -8.67295116e-02
-7.52421245e-02 4.52778250e-01 -6.39751315e-01 -3.18831913e-02
-4.96759623e-01 1.78591773e-01 -2.88397461e-01 3.50673974e-01
7.03837454e-01 2.50995457e-01 2.01158319e-02 7.68158361e-02
3.57449859e-01 9.62168574e-02 -1.12101758e+00 1.18002307e+00
-2.49950930e-01 7.07080364e-01 2.37500116e-01 -1.37017727e+00
7.42472351e-01 1.69706121e-01 3.87378812e-01 -8.44362192e-04
3.69630039e-01 3.82908508e-02 -5.23381121e-02 -9.97279771e-04
-1.21786837e-02 -2.55401909e-01 -2.95416206e-01 3.38891625e-01
-7.39375409e-03 1.77843440e-02 8.45932961e-02 -2.88704671e-02
1.29639018e+00 6.41556680e-02 -2.60732055e-01 -3.65377039e-01
5.03477931e-01 -4.09086555e-01 6.64593458e-01 1.17362440e+00
-2.95618743e-01 8.33419897e-03 1.12999964e+00 -4.85797733e-01
-1.02769172e+00 -1.42236102e+00 -3.78237486e-01 1.31428659e+00
-9.60376933e-02 2.22554412e-02 -1.16318488e+00 -7.24656701e-01
1.37453064e-01 7.97523677e-01 -9.49518681e-01 -3.93426359e-01
-5.95967531e-01 -9.98287141e-01 8.88944328e-01 6.63155913e-01
2.77304560e-01 -8.91891122e-01 -5.08353293e-01 1.68124475e-02
2.14553371e-01 -9.72993076e-01 -1.92455903e-01 5.05936980e-01
-9.63297665e-01 -8.81771743e-01 -4.34426755e-01 -4.03245926e-01
6.91668570e-01 -5.19971728e-01 8.30221832e-01 -4.68662858e-01
-7.22497627e-02 1.95655122e-01 3.34640682e-01 -6.36690676e-01
-5.43583035e-01 4.82078977e-02 4.10797864e-01 -2.97312826e-01
-2.65668407e-02 -8.28675449e-01 -2.73264617e-01 4.43716139e-01
-7.90569782e-01 -3.62844467e-01 1.05389297e-01 7.04255998e-01
7.55102873e-01 3.58080328e-01 3.21958542e-01 -8.38268280e-01
4.63333338e-01 -4.14297253e-01 -1.41399527e+00 2.96210915e-01
-3.64716649e-01 3.00405174e-01 8.78670216e-01 -7.39539921e-01
-6.30312324e-01 -3.70034762e-02 -1.97077364e-01 -7.89769590e-01
2.72149384e-01 3.12761992e-01 -1.70354068e-01 -2.14761749e-01
7.48558342e-01 -5.35107628e-02 -1.42061964e-01 -3.14129978e-01
2.79753208e-01 1.25728443e-01 6.82165086e-01 -8.48643959e-01
9.38321114e-01 4.39978182e-01 5.55124640e-01 -4.85390455e-01
-1.21435976e+00 1.43564552e-01 -5.87406099e-01 -3.48372996e-01
5.71472764e-01 -3.58442575e-01 -1.28305280e+00 4.48352218e-01
-1.06875634e+00 -6.64642394e-01 -2.97725260e-01 1.42291129e-01
-8.01926613e-01 1.18574826e-02 -8.05102587e-01 -1.29457724e+00
-3.09260488e-01 -1.12321603e+00 4.87556815e-01 2.24269051e-02
2.24780053e-01 -1.05774927e+00 2.66692787e-02 -3.12526003e-02
1.79760858e-01 5.57438910e-01 9.97318447e-01 -1.02100611e+00
-2.48944327e-01 -5.12017965e-01 -1.32428542e-01 6.39477789e-01
-7.18623459e-01 1.80899709e-01 -1.09834266e+00 -3.14164281e-01
3.56718123e-01 -1.74084261e-01 1.06563628e+00 7.85378158e-01
1.74322629e+00 -7.74322391e-01 -4.29790288e-01 1.00827014e+00
1.29584217e+00 2.92281270e-01 5.62296033e-01 1.04381233e-01
5.50687313e-01 4.72088367e-01 1.79859713e-01 2.93937653e-01
-1.03380613e-01 2.57941067e-01 8.43520105e-01 5.39223969e-01
5.87980032e-01 -1.05470426e-01 5.64622641e-01 4.22095716e-01
-4.07534689e-02 -1.33720234e-01 -9.89060640e-01 1.37611076e-01
-1.95187759e+00 -1.08547914e+00 3.21560919e-01 2.07316518e+00
8.80497873e-01 6.78767800e-01 1.78298771e-01 2.37276509e-01
7.99341917e-01 7.90208876e-02 -1.05547667e+00 -6.27267659e-01
2.44416863e-01 2.62583464e-01 9.74350333e-01 6.93913221e-01
-1.18830895e+00 7.37850487e-01 7.35902023e+00 9.21761692e-01
-7.25722611e-01 5.84274791e-02 1.09966135e+00 -3.69282901e-01
1.22497857e-01 -3.69349390e-01 -1.20854712e+00 3.32217038e-01
1.46090043e+00 -4.41346318e-02 8.89217198e-01 1.11773849e+00
-2.15609327e-01 3.90293807e-01 -1.23861039e+00 5.61880648e-01
-3.17354530e-01 -1.35428345e+00 -3.51303518e-01 1.33782834e-01
5.76820552e-01 2.15446711e-01 5.31207502e-01 3.59297037e-01
9.46139634e-01 -1.18830681e+00 9.16277945e-01 5.76141238e-01
7.10819781e-01 -1.36425793e+00 7.00298667e-01 4.46600318e-01
-7.88237810e-01 -5.24040878e-01 -7.27926970e-01 2.21251324e-01
-1.77966878e-01 5.02772331e-01 -5.80450654e-01 6.55459762e-02
1.00164592e+00 4.40425247e-01 -1.18514352e-01 4.86462057e-01
-1.88034669e-01 7.38578379e-01 -7.83877850e-01 -1.43234134e-01
1.79049566e-01 -1.57831967e-01 6.94887400e-01 1.28596187e+00
6.20314553e-02 1.89812437e-01 -1.75326064e-01 9.97859180e-01
-2.90827483e-01 -5.31499565e-01 -5.09474039e-01 9.93680209e-02
4.63280618e-01 9.09837127e-01 -5.79548657e-01 9.40177813e-02
1.79559842e-01 2.85333931e-01 5.27972162e-01 4.36294526e-01
-1.33183122e+00 -5.58927000e-01 8.72200131e-01 -2.15537593e-01
6.00805819e-01 -3.50117594e-01 -5.16235232e-01 -6.40317380e-01
-1.76132713e-02 -6.32448554e-01 5.33318579e-01 -4.11445796e-01
-1.35683501e+00 6.32177949e-01 2.86152303e-01 -6.11094713e-01
-3.23308051e-01 -1.20197821e+00 -4.03290153e-01 7.85910010e-01
-8.47588718e-01 -6.29382789e-01 4.05426800e-01 6.39239073e-01
-8.82163923e-03 -1.97754607e-01 7.83485651e-01 -2.25615487e-01
-1.11028111e+00 9.74327743e-01 4.71532881e-01 4.20960903e-01
4.36500050e-02 -1.17808843e+00 4.52735722e-01 1.22369611e+00
-8.94226432e-02 5.54261386e-01 9.74349141e-01 -4.19106930e-01
-1.21033502e+00 -1.14454412e+00 9.56201404e-02 -5.93151033e-01
1.13433290e+00 -4.97105330e-01 -6.87762201e-01 8.47807884e-01
-4.53856558e-01 2.25270882e-01 6.09246910e-01 2.25812584e-01
-7.31979609e-01 -4.70366150e-01 -1.45255363e+00 8.61748040e-01
1.04749715e+00 -3.83833259e-01 -2.53384918e-01 3.65647823e-01
8.97803247e-01 -7.84571826e-01 -9.98529971e-01 3.98279905e-01
6.32471383e-01 -8.47912133e-01 1.27082276e+00 -9.59908187e-01
4.34786201e-01 1.21333897e-01 -5.42041898e-01 -9.48525608e-01
-2.95071490e-02 -1.21263754e+00 -6.96543276e-01 8.76734078e-01
6.14766777e-01 -7.82318175e-01 7.73144841e-01 1.06174076e+00
-8.31177309e-02 -1.11067772e+00 -1.52846205e+00 -8.08430016e-01
5.39675891e-01 -1.03017592e+00 3.17560732e-01 2.39230454e-01
-6.75845966e-02 -3.57914299e-01 -1.16104804e-01 5.40362239e-01
9.57794309e-01 -4.08750951e-01 2.96584040e-01 -1.13975477e+00
-3.57179850e-01 -7.03531802e-01 -4.91905004e-01 -9.26055491e-01
5.77065706e-01 -7.48184443e-01 3.69397163e-01 -7.99450636e-01
7.38909170e-02 -5.49029231e-01 -6.55170143e-01 4.59229231e-01
2.73577720e-01 1.51791781e-01 1.17141239e-01 -2.58674063e-02
-5.96755981e-01 1.58973902e-01 6.91118956e-01 -1.17252976e-01
-1.24615394e-02 5.11770129e-01 -6.30296469e-01 1.10557389e+00
7.60543346e-01 -8.58464420e-01 -2.66607195e-01 -2.93914080e-01
3.38812113e-01 1.42403170e-01 5.46398282e-01 -1.02328682e+00
3.12941074e-01 -2.01506674e-01 4.90548790e-01 -6.30612373e-01
5.16857505e-01 -8.20338666e-01 -2.82170298e-03 7.42680490e-01
-8.36158276e-01 -3.55983265e-02 4.14729655e-01 9.39869702e-01
5.21426141e-01 -3.98516864e-01 1.18823695e+00 2.22177252e-01
1.81674406e-01 6.58078313e-01 -3.09636205e-01 2.64671654e-01
8.75342071e-01 2.99529254e-01 -3.73374820e-01 -1.96500540e-01
-8.67396414e-01 1.40345171e-01 -3.99516113e-02 -1.53554797e-01
6.23322725e-01 -1.16645479e+00 -3.88811737e-01 9.89911631e-02
-6.50977433e-01 4.56176072e-01 1.34146899e-01 6.77939832e-01
-5.77442467e-01 3.72297138e-01 6.03826791e-02 -4.70695555e-01
-9.50140893e-01 6.61784053e-01 8.54682803e-01 -5.83975732e-01
-4.23444211e-01 1.38149619e+00 -8.21032096e-03 -2.32147828e-01
8.76295209e-01 -5.26925802e-01 2.06372485e-01 -3.36359292e-01
4.82899398e-01 6.23669147e-01 -3.52811277e-01 -2.97316402e-01
-2.96045631e-01 1.53765500e-01 -2.95332819e-01 -4.59523082e-01
1.53802335e+00 2.28232816e-01 -2.14974776e-01 4.31310624e-01
1.18414760e+00 -4.67945963e-01 -1.85356581e+00 -1.56936944e-01
7.22125694e-02 -4.77168299e-02 -1.11782268e-01 -7.27918684e-01
-1.22532403e+00 9.16306317e-01 3.87542933e-01 5.04964411e-01
1.11842036e+00 -1.62658729e-02 5.88031769e-01 8.64495337e-01
1.99430212e-01 -1.03471422e+00 -2.52037436e-01 8.33784878e-01
9.07393336e-01 -5.75244606e-01 -8.65390524e-02 -1.05752379e-01
-1.32511899e-01 1.25258207e+00 3.72112185e-01 -5.62023342e-01
1.23386550e+00 4.33932781e-01 -3.63210529e-01 1.94744557e-01
-8.11174273e-01 6.80907547e-01 1.37614638e-01 6.30229175e-01
-1.91403508e-01 -5.94894551e-02 3.46600980e-01 1.01369572e+00
-5.10131121e-01 -5.90627730e-01 1.98666573e-01 5.90871036e-01
-4.05343652e-01 -5.88595271e-01 -3.72243643e-01 3.30552936e-01
-8.37572813e-01 2.52815280e-02 -3.54273207e-02 7.67223954e-01
-1.78791597e-01 7.38080680e-01 -7.27527440e-02 -5.71269572e-01
2.05302417e-01 1.11745084e-02 6.25206590e-01 -3.77845988e-02
-3.94678533e-01 -3.28028172e-01 -1.91515788e-01 -6.84043586e-01
1.27380297e-01 -8.99475098e-01 -1.03840339e+00 -6.11935198e-01
-2.34049037e-01 -7.31259808e-02 7.82932281e-01 1.06906664e+00
-1.19524918e-01 4.21761245e-01 7.26850986e-01 -1.00817823e+00
-1.31266224e+00 -9.03234363e-01 -5.66788316e-01 -1.10382356e-01
3.97329450e-01 -4.95624304e-01 -7.79943407e-01 -7.79769793e-02] | [5.706859588623047, 7.688052177429199] |
6173496e-c9ee-40ef-965c-fb0f9a7eb4a5 | personalized-federated-learning-for-multi | 2211.09406 | null | https://arxiv.org/abs/2211.09406v1 | https://arxiv.org/pdf/2211.09406v1.pdf | Personalized Federated Learning for Multi-task Fault Diagnosis of Rotating Machinery | Intelligent fault diagnosis is essential to safe operation of machinery. However, due to scarce fault samples and data heterogeneity in field machinery, deep learning based diagnosis methods are prone to over-fitting with poor generalization ability. To solve the problem, this paper proposes a personalized federated learning framework, enabling multi-task fault diagnosis method across multiple factories in a privacypreserving manner. Firstly, rotating machines from different factories with similar vibration feature data are categorized into machine groups using a federated clustering method. Then, a multi-task deep learning model based on convolutional neural network is constructed to diagnose the multiple faults of machinery with heterogeneous information fusion. Finally, a personalized federated learning framework is proposed to solve data heterogeneity across different machines using adaptive hierarchical aggregation strategy. The case study on collected data from real machines verifies the effectiveness of the proposed framework. The result shows that the diagnosis accuracy could be improved significantly using the proposed personalized federated learning, especially for those machines with scarce fault samples. | ['Cheng Hao Jin', 'Shubao Zhao', 'Hui Liu', 'Zengxiang Li', 'Sheng Guo'] | 2022-11-17 | null | null | null | null | ['personalized-federated-learning'] | ['methodology'] | [-1.90959856e-01 -1.57179430e-01 1.36565462e-01 -1.54696926e-01
-5.76200128e-01 -3.51898342e-01 -2.79259324e-01 -2.12515146e-01
4.57683712e-01 3.91935259e-01 -1.51328042e-01 -1.83408577e-02
-7.43121922e-01 -7.91472673e-01 -6.58426940e-01 -1.11680794e+00
8.66033584e-02 3.70359302e-01 -4.76320058e-01 2.36883044e-01
9.89430472e-02 5.05312562e-01 -1.95387328e+00 5.67533672e-01
1.20779765e+00 1.45768070e+00 1.63593873e-01 2.17327967e-01
1.43440112e-01 8.33007216e-01 -8.04348528e-01 -2.64821768e-01
4.10062551e-01 4.98685725e-02 -9.65272605e-01 3.82081747e-01
-4.24048528e-02 -4.88841474e-01 -5.22189498e-01 1.07073891e+00
6.45971477e-01 3.14220041e-01 4.70300704e-01 -1.64664543e+00
-8.04774821e-01 3.68450642e-01 -9.20061320e-02 -4.14037108e-01
-9.57351401e-02 -1.23377934e-01 5.22916436e-01 -6.04170799e-01
7.45722577e-02 9.16933179e-01 5.28672218e-01 2.69428223e-01
-5.68363786e-01 -6.21387064e-01 -1.08560741e-01 5.71758270e-01
-1.27915955e+00 6.81131557e-02 7.75101662e-01 -4.37680215e-01
5.80204070e-01 2.91420251e-01 3.25912774e-01 7.86343157e-01
8.07395756e-01 8.08583140e-01 7.77181685e-01 1.61077335e-01
2.92678624e-01 -1.29815340e-01 1.08810604e-01 7.14384258e-01
5.58140099e-01 -2.79666662e-01 -4.26649690e-01 -2.35565051e-01
4.64176089e-01 9.90412652e-01 -1.63055301e-01 -3.55166018e-01
-1.05435956e+00 6.35013282e-01 4.35038894e-01 2.34180272e-01
-4.79543149e-01 -3.70921880e-01 1.04226720e+00 6.51062429e-01
6.55782580e-01 2.52634555e-01 -1.02239954e+00 4.14668769e-01
-4.50202405e-01 1.75075650e-01 7.47419775e-01 8.50534379e-01
8.86405170e-01 1.85346216e-01 -9.66438949e-02 6.56208038e-01
-3.64133269e-02 3.63207459e-01 7.51166463e-01 -1.07580304e+00
3.13751638e-01 9.01077330e-01 7.26516023e-02 -1.34381461e+00
-4.86744821e-01 -5.18095493e-01 -1.21974528e+00 1.52287930e-01
-3.33729684e-02 -6.26512170e-01 -4.20073062e-01 1.04488838e+00
5.16247094e-01 9.95697901e-02 2.55118132e-01 1.01510727e+00
4.33954507e-01 3.81195962e-01 -3.20781499e-01 -1.38332723e-02
1.43440044e+00 -8.42444062e-01 -1.01417911e+00 4.81613904e-01
7.68148839e-01 -4.77063924e-01 5.04459798e-01 7.64735937e-01
-3.92360687e-01 -6.74223900e-01 -9.69663203e-01 2.27071881e-01
-4.60562289e-01 5.44729888e-01 7.04325318e-01 4.11730319e-01
-5.48929751e-01 6.97335005e-01 -6.27674103e-01 -6.12542257e-02
6.70678258e-01 4.65818912e-01 -7.97156990e-01 -6.42818511e-01
-1.13706458e+00 5.42515516e-01 4.61956620e-01 2.49933675e-01
-1.09540951e+00 -4.96502727e-01 -7.95991719e-01 2.14631915e-01
4.71191913e-01 -9.62989092e-01 9.56658185e-01 -7.28519857e-01
-1.11372173e+00 1.87345237e-01 2.84998387e-01 -3.93897235e-01
1.92224652e-01 -5.81028700e-01 -5.88424921e-01 1.46354318e-01
6.86534271e-02 -3.60292315e-01 1.09478295e+00 -1.27062011e+00
-8.04814875e-01 -8.88739944e-01 -2.69922018e-01 9.26910192e-02
-5.30664265e-01 -1.64336160e-01 2.23861247e-01 -4.93686199e-01
1.25026494e-01 -6.02890491e-01 -5.91827258e-02 -4.51328665e-01
-5.66871345e-01 -4.17530298e-01 1.59875643e+00 -9.88541186e-01
8.09258223e-01 -2.22338533e+00 1.25035048e-01 8.40095133e-02
4.27597970e-01 -6.59170523e-02 1.32342398e-01 1.42982870e-01
-8.53013396e-02 -2.99213231e-01 -1.01602182e-01 -4.35766205e-02
2.04659641e-01 2.94255286e-01 9.09494609e-02 7.31378078e-01
5.27379848e-02 5.22225738e-01 -6.12248957e-01 -2.51691341e-01
5.06804883e-01 1.69102013e-01 -1.14827290e-01 5.10656536e-01
9.64991748e-03 3.80596131e-01 -9.18953538e-01 1.09673321e+00
8.37107778e-01 -2.11868018e-01 -2.44476134e-03 -6.71661437e-01
2.32025966e-01 -6.62152946e-01 -1.12464440e+00 1.89865983e+00
-4.91613179e-01 1.93798188e-02 6.84821010e-01 -1.57633162e+00
1.13012278e+00 8.21359754e-01 1.29402208e+00 -1.67260423e-01
6.51705325e-01 9.40015092e-02 -3.02942812e-01 -1.08223736e+00
3.84203047e-01 1.38505757e-01 -4.02648270e-01 3.98982525e-01
2.94518709e-01 2.74283975e-01 -5.82774282e-01 -1.21502779e-01
1.38671017e+00 -1.73601285e-01 -3.87168169e-01 -1.47285134e-01
4.28126156e-01 -7.57791176e-02 7.63667047e-01 3.08521748e-01
-2.38639683e-01 3.80649716e-01 -1.33764252e-01 -6.59358203e-01
-5.83922029e-01 -5.77320576e-01 1.67717542e-02 7.63584197e-01
2.32520610e-01 -5.26382476e-02 -7.20969975e-01 -1.03399563e+00
5.18968225e-01 2.79315710e-01 -4.51211810e-01 -6.34949803e-01
4.86920513e-02 -6.48539007e-01 4.69903350e-01 4.33701724e-01
6.87083960e-01 -7.96091020e-01 -5.33340871e-01 6.50207996e-02
-2.38124460e-01 -6.45053029e-01 -1.87827311e-02 4.19644505e-01
-7.22035885e-01 -1.56291211e+00 -6.30672157e-01 -9.63563621e-01
5.57166338e-01 4.06935334e-01 5.79975605e-01 1.21882088e-01
-5.90383947e-01 4.01881307e-01 -6.34765089e-01 -3.48487735e-01
-8.30663368e-02 -4.99679297e-02 4.56357747e-01 4.16082978e-01
3.60635042e-01 -1.49821773e-01 -5.91330647e-01 3.91894758e-01
-9.39293504e-01 -5.12716472e-01 5.07775366e-01 1.27009952e+00
2.10723728e-01 1.38889921e+00 1.05905998e+00 -6.55176461e-01
6.02011144e-01 -1.03754687e+00 -2.28642479e-01 3.13643247e-01
-7.75101125e-01 -3.44711214e-01 8.82844448e-01 -5.77907041e-02
-1.28810334e+00 1.46436334e-01 2.95374453e-01 -1.04335117e+00
-6.45147204e-01 4.94162589e-01 -7.32719421e-01 -7.40865916e-02
1.89225644e-01 -1.03501104e-01 4.18608248e-01 -8.02390635e-01
2.55956382e-01 1.17852795e+00 7.35679924e-01 -6.87678695e-01
4.16263878e-01 1.14983559e-01 -5.09250024e-03 -2.33912140e-01
-3.60769391e-01 -2.81890482e-01 -3.69682580e-01 -3.15192103e-01
9.12331939e-01 -1.14703214e+00 -1.12573886e+00 9.69770908e-01
-1.02215004e+00 2.97761619e-01 -1.52380869e-01 8.27210069e-01
-3.58066648e-01 3.78507197e-01 -1.01688969e+00 -7.18975782e-01
-7.01471508e-01 -1.18649864e+00 1.07186604e+00 1.71412982e-03
3.99048507e-01 -7.63673663e-01 -5.62706411e-01 8.06769252e-01
3.40311170e-01 5.55657744e-01 9.72820282e-01 -9.57011759e-01
-2.48097748e-01 -6.83112264e-01 2.14696620e-02 8.15937817e-01
8.66002679e-01 -2.68120676e-01 -9.11794364e-01 -5.29173851e-01
4.67947364e-01 -4.59127545e-01 3.65498841e-01 3.02196741e-01
1.55394983e+00 -4.94892359e-01 -3.27300161e-01 4.73482788e-01
1.26068664e+00 8.54375437e-02 1.69749781e-01 2.95136869e-01
1.04499424e+00 9.13711548e-01 9.75835383e-01 8.38469207e-01
2.69134402e-01 -9.88847315e-02 8.45454156e-01 1.48015141e-01
4.77344751e-01 2.89622456e-01 1.35446981e-01 1.11492538e+00
3.27027082e-01 -2.24704370e-01 -5.78475952e-01 5.67751229e-01
-1.99003804e+00 -7.36326396e-01 -3.92447487e-02 1.80172634e+00
1.80462569e-01 -6.60025597e-01 -2.93680549e-01 4.47927684e-01
1.01419127e+00 -3.51930648e-01 -9.70769405e-01 -2.19691455e-01
-1.66871458e-01 -1.91509500e-01 5.25877118e-01 -3.45188439e-01
-1.32519841e+00 2.84842581e-01 5.25358057e+00 8.73786688e-01
-1.01814413e+00 2.70416170e-01 4.77169633e-01 3.66131291e-02
5.45341931e-02 -3.22006702e-01 5.73406648e-03 5.06539404e-01
6.85150743e-01 -1.69894323e-01 4.08019602e-01 1.17966664e+00
-9.32577103e-02 3.63138825e-01 -7.35019267e-01 1.03635085e+00
6.48038536e-02 -1.17816436e+00 -2.13994592e-01 4.39367741e-02
8.50873530e-01 2.42446903e-02 2.27062702e-02 1.65082719e-02
4.63731855e-01 -7.53583610e-01 3.43183279e-01 7.22128391e-01
6.75354779e-01 -1.20793319e+00 1.25182867e+00 3.93440366e-01
-1.11381757e+00 -8.43921721e-01 -4.53707010e-01 -6.75224811e-02
-2.92220920e-01 9.44593370e-01 -4.16077524e-01 1.73761797e+00
1.21490645e+00 8.71357799e-01 -3.10297042e-01 7.57280529e-01
6.05502367e-01 3.35599393e-01 9.35927033e-02 5.19832790e-01
-3.72839868e-01 -2.47139573e-01 1.58253178e-01 4.10774082e-01
8.21151018e-01 -2.52768725e-01 4.80237931e-01 3.47767174e-01
-6.00790307e-02 2.55031437e-02 -1.04465652e+00 1.05954498e-01
6.28476083e-01 1.46154690e+00 -2.71058053e-01 -1.86937489e-02
-4.40874249e-01 1.20963132e+00 -6.31040102e-03 1.49558708e-01
-4.56161916e-01 -6.50650501e-01 8.06012690e-01 -4.00354028e-01
2.05544651e-01 2.35878490e-02 -2.74434060e-01 -1.11190867e+00
-1.45766690e-01 -1.09359169e+00 6.88120186e-01 -7.46522427e-01
-1.92905879e+00 5.51687002e-01 -3.09534162e-01 -1.40884459e+00
-7.61593282e-02 -7.34179497e-01 -5.18032014e-01 7.53716111e-01
-7.72641540e-01 -1.26824486e+00 -3.12798858e-01 1.24336767e+00
3.88106585e-01 -7.57525384e-01 9.70892012e-01 5.79954624e-01
-9.81005430e-01 5.01215458e-01 5.34711003e-01 4.92593944e-02
6.41217291e-01 -9.46515620e-01 -3.24003875e-01 8.65858495e-01
-5.69718301e-01 4.51857597e-01 2.01561362e-01 -7.53844500e-01
-2.09509754e+00 -1.69242239e+00 1.29477948e-01 -5.63793853e-02
4.26415354e-01 1.29921958e-01 -1.03746855e+00 5.51262736e-01
3.49366128e-01 3.41177523e-01 8.24213743e-01 6.33519143e-02
-6.73103854e-02 -4.01895255e-01 -1.82368791e+00 -1.88864395e-01
4.03409600e-01 -6.48843527e-01 -3.60753238e-01 6.80877924e-01
1.02653670e+00 -2.05486417e-01 -1.71756399e+00 6.67317748e-01
2.32741803e-01 -5.11101067e-01 5.12685239e-01 -5.55293322e-01
1.71864882e-01 -5.37561774e-01 -3.96778584e-01 -1.59533191e+00
-6.38008773e-01 -3.37550372e-01 -4.54786211e-01 1.41617537e+00
-3.96418810e-01 -8.90737593e-01 6.41541481e-01 6.88161016e-01
-6.19188726e-01 -6.12526476e-01 -1.06652462e+00 -5.83691418e-01
-4.05085310e-02 -2.55664557e-01 1.28832197e+00 1.27686226e+00
-1.28560513e-01 -1.85154587e-01 -5.27947903e-01 5.72559536e-01
6.47576571e-01 2.08859146e-01 6.54971540e-01 -1.39531660e+00
1.27478279e-02 2.21531615e-01 -6.10217512e-01 -1.83391467e-01
5.46662450e-01 -9.41371381e-01 1.12733372e-01 -1.35260773e+00
-1.57869592e-01 -5.43733239e-01 -7.38470435e-01 7.30798066e-01
-1.47074116e-02 -2.34317988e-01 -2.03419834e-01 9.09102038e-02
-5.51854670e-01 7.08195925e-01 1.23810959e+00 -4.61371511e-01
5.30331850e-01 -5.42162638e-03 -6.55831456e-01 2.54823536e-01
1.04120040e+00 -1.90311700e-01 -4.03282195e-01 -7.12782502e-01
-4.67269778e-01 2.92185098e-01 5.11817217e-01 -1.16780031e+00
3.33353013e-01 1.15020499e-01 5.93776703e-01 -6.16439223e-01
-8.92301500e-02 -1.47785270e+00 5.36099374e-01 5.64579189e-01
2.09803388e-01 9.31733549e-02 3.22889276e-02 6.74479723e-01
-3.82123530e-01 1.83887601e-01 1.63092539e-02 -2.96304859e-02
-5.58028877e-01 5.08697450e-01 -3.41822654e-01 -7.83613145e-01
1.20092988e+00 1.28963783e-01 -4.29387242e-01 2.17208549e-01
-6.09278440e-01 3.99858207e-01 4.38943118e-01 5.82476318e-01
7.62286067e-01 -1.49408889e+00 -6.03106022e-01 4.53151852e-01
2.18107089e-01 4.20236290e-01 1.03042698e+00 6.78929448e-01
-2.29894266e-01 3.43919843e-01 -2.63552934e-01 -4.85644042e-01
-1.03930223e+00 1.05933499e+00 3.12695742e-01 1.50622949e-01
-6.82946980e-01 5.91836870e-01 -2.27320999e-01 -8.24222505e-01
2.52050191e-01 -1.93314090e-01 6.77189603e-02 -6.09233566e-02
1.75369412e-01 9.05110359e-01 6.95219159e-01 -6.24545753e-01
-2.60222644e-01 2.08263040e-01 2.00354889e-01 8.41580510e-01
1.12778890e+00 -2.73522496e-01 -4.60612535e-01 2.26356894e-01
1.42189980e+00 -5.49338281e-01 -1.18610311e+00 -1.57209501e-01
-3.27539474e-01 -5.69140911e-01 2.18202665e-01 -7.06062734e-01
-1.81106043e+00 5.55023968e-01 8.52235496e-01 2.89908379e-01
1.45471287e+00 -2.82887936e-01 1.10565436e+00 4.12935853e-01
8.15637112e-01 -1.02075672e+00 -2.94330239e-01 1.46302089e-01
5.87662816e-01 -1.28913927e+00 -3.39806139e-01 4.22438905e-02
-5.96106708e-01 1.09360087e+00 7.62279212e-01 -1.84016228e-01
8.47680211e-01 4.02497530e-01 2.95106191e-02 -4.33888286e-01
-6.45477057e-01 3.81654859e-01 5.67225181e-03 7.25610614e-01
-7.47439340e-02 2.97726870e-01 1.44255430e-01 1.31733108e+00
1.05303772e-01 9.00259912e-02 -4.32712920e-02 1.20882928e+00
-2.60324299e-01 -7.98267961e-01 -7.55431533e-01 9.43449676e-01
-5.81243217e-01 5.69908738e-01 -1.87023710e-02 1.04575180e-01
6.56358242e-01 1.56024015e+00 4.07607518e-02 -1.19511211e+00
1.73909485e-01 1.89787224e-01 2.14758307e-01 -2.31370419e-01
-7.60860682e-01 -3.87031913e-01 -3.69749874e-01 -5.89013636e-01
-2.66116589e-01 -4.71520633e-01 -1.24285364e+00 -4.05080676e-01
-6.30476475e-01 5.10036409e-01 6.01360559e-01 1.01937401e+00
8.82815599e-01 9.83148098e-01 1.35104275e+00 -8.26795459e-01
-8.17682207e-01 -8.56065452e-01 -1.32488251e+00 5.86569488e-01
3.49601835e-01 -8.41539502e-01 -3.39199454e-01 -5.45074977e-02] | [7.028772354125977, 2.400217056274414] |
cf72c7bf-fb1e-4de0-b0e5-cb3c12411b3d | ju_cse-a-crf-based-approach-to-annotation-of | null | null | https://aclanthology.org/S13-2011 | https://aclanthology.org/S13-2011.pdf | JU\_CSE: A CRF Based Approach to Annotation of Temporal Expression, Event and Temporal Relations | null | ['B', 'Asif Ekbal', 'Rajdeep Gupta', 'Anup Kumar Kolya', 'Sivaji yopadhyay', 'Amitava Kundu'] | 2013-06-01 | null | null | null | semeval-2013-6 | ['temporal-information-extraction'] | ['natural-language-processing'] | [-8.63703638e-02 1.71006292e-01 -6.22772932e-01 -4.08054382e-01
-8.41685571e-03 -9.08429027e-01 6.55310392e-01 -6.53472245e-01
-2.85945535e-01 1.06888819e+00 -4.63127941e-02 -1.01159286e+00
-3.91567826e-01 -9.63214397e-01 -4.95059669e-01 -6.31337762e-01
-9.79754329e-01 7.25764990e-01 3.30370307e-01 -6.93831444e-01
7.03166842e-01 7.88774848e-01 -1.68942046e+00 7.18545914e-01
7.04417467e-01 8.52217197e-01 2.49141872e-01 1.14950800e+00
-1.95044339e-01 1.55633950e+00 -7.48382092e-01 -5.46825826e-01
3.13719302e-01 -1.23176083e-01 -7.22945035e-01 -1.01074085e-01
9.28529128e-02 -8.59008506e-02 -2.09758401e-01 9.22211111e-01
5.37373662e-01 4.49454933e-02 1.08379531e+00 -1.42548037e+00
-5.91619551e-01 6.10313773e-01 -4.01565880e-02 1.21627934e-01
1.03678203e+00 -5.39447069e-01 1.19919395e+00 -1.13026452e+00
7.20913768e-01 1.26888943e+00 8.66221786e-01 5.44149756e-01
-1.22286928e+00 -1.94712028e-01 -3.26822817e-01 -9.51717794e-02
-1.46558487e+00 -3.25250506e-01 4.25783843e-02 -2.08119690e-01
1.66093647e+00 1.26596653e+00 1.20609856e+00 1.01401424e+00
1.26658809e+00 8.34431887e-01 1.04267764e+00 -5.13792276e-01
3.35295945e-01 3.66983831e-01 1.54683650e-01 6.33519173e-01
8.40953708e-01 5.26628852e-01 -7.06372619e-01 -9.13127720e-01
9.33553874e-01 -2.94925272e-01 1.71355158e-01 -5.05680561e-01
-9.05919552e-01 6.91228509e-01 1.78732842e-01 3.83959889e-01
-1.39880210e-01 9.89067405e-02 1.26390755e-01 5.30987144e-01
-2.58292928e-02 6.47037446e-01 -9.11868811e-01 -1.33165747e-01
-8.71728659e-01 5.10332465e-01 1.25398111e+00 1.52653182e+00
1.24482810e-01 2.94908643e-01 -9.34252143e-02 3.17179203e-01
8.92314315e-01 1.01808000e+00 4.28362608e-01 -1.36146402e+00
-6.87414408e-02 1.72361732e-01 5.01781464e-01 -8.52631688e-01
-6.33224547e-01 -9.64177120e-03 -8.93263519e-01 4.49267089e-01
3.49161088e-01 4.57367361e-01 -8.02827001e-01 5.07305264e-01
4.33481112e-02 -2.34125629e-01 4.53833073e-01 5.55570945e-02
4.99930978e-01 3.76208365e-01 -1.34477139e-01 -5.73289394e-01
1.06082785e+00 -1.36716676e+00 -1.35299087e+00 2.33215362e-01
9.05734658e-01 -1.07320261e+00 4.35900748e-01 5.33875942e-01
-1.55548143e+00 -1.37560293e-01 -1.08699942e+00 1.78573877e-01
-7.27255583e-01 -3.14239264e-01 8.57801437e-01 1.43120694e+00
-1.60129595e+00 9.73287821e-01 -4.91727620e-01 6.59165755e-02
1.20568443e-02 8.24621081e-01 -2.64718989e-03 4.62812334e-01
-1.33193445e+00 1.08501506e+00 2.22979754e-01 -1.21242590e-01
-1.65216476e-01 -2.13068098e-01 -8.23704481e-01 -5.63443303e-01
-4.78693932e-01 -5.29636025e-01 1.44139910e+00 -2.59346128e-01
-1.65295815e+00 9.71794367e-01 -1.42069459e-01 -1.97814897e-01
6.14786744e-01 -1.28011424e-02 -8.31891418e-01 2.42498964e-01
-1.89849049e-01 5.76383233e-01 9.28263724e-01 -1.35132408e+00
-7.59897232e-01 -1.67359829e-01 -1.23336017e-01 2.66287565e-01
-1.25510961e-01 1.89734384e-01 2.11616129e-01 -1.12999000e-01
3.27147305e-01 -7.20919967e-01 -2.53068686e-01 -5.32041907e-01
-1.46512717e-01 -7.10518599e-01 7.70373225e-01 -4.51523662e-01
1.83705616e+00 -1.67618537e+00 -2.06720144e-01 3.98590982e-01
3.57815564e-01 -1.24705513e-03 2.40583986e-01 1.08380008e+00
-2.76906848e-01 7.32199550e-01 4.11965609e-01 -1.10722095e-01
2.24991128e-01 4.85861301e-01 -4.16602850e-01 3.05609167e-01
-1.29282743e-01 1.15307164e+00 -1.16605783e+00 -5.23096442e-01
4.11106765e-01 9.42391157e-02 -4.58732933e-01 4.79237735e-01
2.99364805e-01 1.47170946e-01 -3.56553018e-01 1.39399457e+00
1.15709066e+00 -1.31984919e-01 1.45911396e-01 5.30878425e-01
-3.79135728e-01 3.55090618e-01 -6.96863770e-01 1.05554795e+00
7.08333924e-02 5.00173986e-01 1.02364108e-01 -7.94621468e-01
3.33247900e-01 8.61540735e-01 4.37155962e-01 -9.67555881e-01
-2.26398129e-02 6.92409754e-01 1.12803578e-01 -6.07703328e-01
7.58228302e-01 3.94563079e-02 -3.64872098e-01 6.40070081e-01
-2.37588286e-01 -6.59476995e-01 9.64643434e-02 3.08100313e-01
6.22585893e-01 -5.18246442e-02 5.62923312e-01 -1.05613089e+00
7.86340594e-01 -1.82965681e-01 -1.81299388e-01 1.03415680e+00
-3.09923887e-01 3.19085121e-01 2.99841821e-01 -6.65102363e-01
-6.45341039e-01 -1.12307119e+00 -4.89381433e-01 1.30636716e+00
3.24267983e-01 -4.39044595e-01 -9.54439282e-01 -2.49762803e-01
1.77620783e-01 6.89606130e-01 -5.90509653e-01 3.84124845e-01
-5.03739953e-01 -8.32535863e-01 7.39044368e-01 3.45434904e-01
-5.07752821e-02 -1.33414865e+00 -6.58416986e-01 1.25490099e-01
-2.20292807e-01 -6.63697243e-01 -6.23428151e-02 4.48765576e-01
-1.35989368e+00 -5.18594682e-01 -6.66252747e-02 -8.20914626e-01
5.87345481e-01 2.46782884e-01 1.27047324e+00 5.39230824e-01
-2.31483161e-01 4.26904231e-01 -1.21292919e-01 -4.95818377e-01
-4.59671497e-01 -8.00336525e-02 5.28869390e-01 -5.87835789e-01
5.19427478e-01 -2.50617653e-01 -7.29350567e-01 5.37953973e-01
-6.88540697e-01 1.62748516e-01 1.79803044e-01 1.04410267e+00
1.35816500e-01 -9.34035778e-02 1.22507080e-01 -6.38007045e-01
8.72274399e-01 -1.69219792e-01 -3.78732830e-01 5.77745810e-02
-6.77108407e-01 -3.74140263e-01 3.21430594e-01 -3.25342178e-01
-1.01981449e+00 -4.87835288e-01 -9.82677937e-02 2.45538145e-01
1.11353043e-02 -1.46784872e-01 6.47139177e-02 -5.24923325e-01
8.02199244e-01 9.25758183e-02 1.99174434e-02 -6.80815242e-03
3.01039815e-01 7.09525108e-01 -6.82967342e-03 -6.68678164e-01
8.44880998e-01 4.91470337e-01 7.98524171e-02 -9.57177758e-01
-1.52186140e-01 -2.60129690e-01 -9.51962709e-01 -6.54426932e-01
6.56643391e-01 -6.78531289e-01 -9.10833478e-01 3.91110867e-01
-9.38691139e-01 -3.38627815e-01 -3.91645581e-01 4.25431967e-01
-1.01278400e+00 2.75717527e-02 -3.90154392e-01 -1.27895141e+00
-5.10977268e-01 -1.02017939e+00 9.43384409e-01 5.30070923e-02
-5.10597289e-01 -1.26927447e+00 5.87685481e-02 2.71537274e-01
1.81734428e-01 -1.73075795e-01 6.90226793e-01 -2.38256708e-01
-4.24233019e-01 -1.53791070e-01 2.34436691e-02 -1.39755070e-01
1.70832314e-02 4.95917559e-01 -9.81751978e-01 -5.31145096e-01
6.65065646e-02 -1.92070693e-01 -1.08835101e-01 6.52520418e-01
5.91872573e-01 -2.29931593e-01 -8.56000841e-01 5.40386558e-01
1.38545322e+00 3.85070026e-01 5.32770038e-01 7.28214979e-01
1.41836226e-01 5.53460240e-01 9.17806149e-01 4.63203549e-01
1.30579369e-02 3.28798652e-01 2.40537539e-01 1.49327129e-01
1.11720070e-01 -1.54819340e-01 3.77893507e-01 1.16112018e+00
-8.18235934e-01 -2.69281328e-01 -5.07867396e-01 4.42987174e-01
-1.72482407e+00 -1.40330648e+00 -4.32368398e-01 6.90478683e-01
6.25676990e-01 1.56016424e-01 -1.48347050e-01 3.35214496e-01
4.99015123e-01 -2.03574806e-01 -1.19133167e-01 -1.06291151e+00
-1.43546045e-01 3.15233678e-01 7.37729073e-01 1.00061214e+00
-7.20721722e-01 1.03317809e+00 1.29781246e+01 1.02230716e+00
2.21112028e-01 1.03134915e-01 5.16071796e-01 3.48020852e-01
-4.36954498e-01 -4.56139445e-02 -1.04416132e+00 2.72933897e-02
1.38140702e+00 -4.30666685e-01 6.85999811e-01 5.44219851e-01
3.44648361e-01 -4.23268199e-01 -1.26188684e+00 5.26221812e-01
9.73738134e-02 -1.40886843e+00 -2.83300440e-04 6.85225725e-01
7.73699820e-01 -5.08050561e-01 6.22419357e-01 3.24184299e-01
6.09259963e-01 -1.14389277e+00 8.60300779e-01 2.53660440e-01
1.03040910e+00 -6.05088234e-01 5.67372203e-01 1.68872893e-01
-1.14389896e+00 -2.20873043e-01 -8.77727985e-01 -1.00755692e+00
3.93533185e-02 -1.81779593e-01 -4.29956943e-01 3.48861217e-01
9.58353162e-01 2.99398601e-01 -3.93658698e-01 9.95779395e-01
-4.78694476e-02 1.04875881e-02 -2.71853864e-01 -4.48467314e-01
4.83122796e-01 -3.54241252e-01 4.66730654e-01 1.00164843e+00
2.48499006e-01 3.51035744e-01 -9.84472036e-02 4.01770771e-01
5.45058846e-01 3.29446048e-02 -1.19659424e+00 -1.78908288e-01
2.83276141e-01 9.16795909e-01 -4.83487815e-01 -4.22520459e-01
-2.00212970e-01 8.62069130e-01 -3.55488248e-02 5.01107454e-01
-6.11489356e-01 -4.35615242e-01 9.72222984e-01 -1.27327025e-01
-1.14700586e-01 -3.48497719e-01 -6.23769283e-01 -7.30352640e-01
-5.89872956e-01 -4.54965204e-01 5.93606755e-02 -5.53365827e-01
-1.39813089e+00 5.79277515e-01 -2.27688253e-02 -1.40553558e+00
-6.99901402e-01 -1.27676582e+00 -4.76714373e-01 4.92853165e-01
-1.11898029e+00 -1.10984349e+00 2.50124663e-01 4.52870727e-01
1.64141744e-01 -5.34416080e-01 1.39563632e+00 3.57715860e-02
1.00637585e-01 9.24474537e-01 6.69434488e-01 -7.37814724e-01
5.56605101e-01 -1.27867436e+00 5.68737745e-01 -1.37897313e-01
-4.31265175e-01 9.05828118e-01 6.28349900e-01 -5.39804697e-01
-1.41196322e+00 -3.66917729e-01 1.08350635e+00 -9.83769417e-01
6.55218959e-01 -3.86345625e-01 4.23767231e-02 7.88592756e-01
7.15902448e-01 -6.18741751e-01 8.21781039e-01 -1.83753878e-01
1.80774391e-01 5.75296998e-01 -1.39248300e+00 6.12354755e-01
1.66275799e+00 -4.63594139e-01 -6.25784039e-01 7.60327101e-01
8.13696027e-01 -6.94087505e-01 -1.30082703e+00 3.34633321e-01
8.65424156e-01 -8.75409484e-01 1.61978090e+00 -1.32660246e+00
-4.43697497e-02 2.81152606e-01 -2.61993498e-01 -9.32519078e-01
-5.96193194e-01 -1.23518765e+00 -5.33532679e-01 -5.83747849e-02
5.96577883e-01 -1.13057327e+00 3.42365682e-01 8.74560475e-01
-2.82833427e-01 -6.42737269e-01 -1.06996536e+00 -1.32016802e+00
-3.35779637e-02 -1.45572275e-01 4.90409225e-01 7.63798356e-01
6.83744550e-01 1.09839931e-01 -6.36873543e-02 -1.00294888e-01
5.33176839e-01 9.55312885e-03 4.41501856e-01 -1.34294486e+00
3.86843324e-01 -5.75816095e-01 -3.07655483e-01 -9.45992947e-01
-8.85957032e-02 -8.12076271e-01 -6.53862000e-01 -1.28511906e+00
-8.32044985e-03 -1.91056758e-01 -1.11109078e-01 -1.65725678e-01
3.67937148e-01 2.13746816e-01 1.20859891e-02 1.03788137e-01
-3.71160030e-01 6.18435517e-02 1.29639816e+00 6.91750320e-05
-1.62315920e-01 4.85058486e-01 -4.81304944e-01 7.84440815e-01
8.58408585e-02 -2.99253196e-01 -6.78878546e-01 6.11881316e-02
6.69384480e-01 4.61409837e-02 3.24159935e-02 -7.42885649e-01
5.37211418e-01 -3.75702560e-01 4.78586555e-01 -1.32223868e+00
1.30741090e-01 -9.61415648e-01 6.95283338e-02 9.40189242e-01
2.74610907e-01 1.20424610e-02 8.66204947e-02 5.39500564e-02
-1.44871444e-01 -5.70943117e-01 9.21121240e-01 -4.07591403e-01
-4.92852688e-01 -5.20386267e-03 -1.03226590e+00 8.97834301e-02
9.92593169e-01 -7.84614205e-01 -3.59281451e-01 -4.20183957e-01
-8.29068601e-01 -1.95836127e-02 6.50830388e-01 3.03609259e-02
7.30431557e-01 -1.51530886e+00 -2.30721906e-01 7.18729138e-01
-3.22939813e-01 -3.74400020e-01 -1.70157343e-01 6.58265352e-01
-1.32361674e+00 1.02442718e+00 -5.41665435e-01 -4.55340147e-01
-1.14228773e+00 4.78126436e-01 4.28307921e-01 -2.41845414e-01
-2.15481281e-01 1.11954463e+00 2.71224789e-02 -8.23025763e-01
1.85185194e-01 -9.21545625e-02 -7.54407048e-01 3.55081353e-03
6.88606799e-01 1.05194807e+00 -2.91290224e-01 -6.04341030e-01
-4.56784427e-01 6.65885091e-01 2.32151806e-01 -2.87484169e-01
9.17833567e-01 -2.19243199e-01 -9.89108324e-01 4.28274393e-01
8.38715494e-01 -1.36269778e-01 -3.69319022e-02 4.16855574e-01
1.25943512e-01 -8.02164078e-01 -4.32406247e-01 -3.55811834e-01
-1.85641110e-01 5.54822803e-01 5.34874737e-01 8.99602413e-01
8.57008278e-01 -3.02566767e-01 8.18335712e-01 9.66778398e-01
5.72402716e-01 -1.68019545e+00 -2.13140488e-01 6.89524531e-01
9.12339568e-01 -9.28350806e-01 5.44190466e-01 -7.27165341e-01
-4.14997995e-01 1.32979155e+00 4.68304873e-01 -1.55325383e-01
1.27306652e+00 5.74917436e-01 1.14069022e-02 -3.70670199e-01
-9.44949508e-01 1.12705544e-01 3.75366658e-01 1.11147714e+00
5.06513238e-01 5.07374525e-01 -9.66500878e-01 3.21953118e-01
-7.28706717e-01 -2.34555230e-01 4.90474731e-01 1.41972518e+00
-6.43810987e-01 -1.20391917e+00 -7.23931909e-01 4.61561680e-01
-5.49773455e-01 -1.16372630e-01 -5.28106689e-01 8.46754074e-01
-5.76629937e-02 1.49448860e+00 -1.97535474e-03 -5.03491640e-01
4.11356747e-01 1.41089618e-01 7.21762300e-01 -1.23501487e-01
-9.04846430e-01 4.15413082e-01 3.84890139e-01 -1.23056793e+00
-8.58632207e-01 -1.05834293e+00 -1.40667629e+00 -1.19437599e+00
-5.12782812e-01 1.89310342e-01 3.83317530e-01 3.90289724e-01
-2.06836104e-01 2.85260603e-02 9.80917513e-01 -1.07949340e+00
-3.90341938e-01 -9.39418614e-01 -1.00026262e+00 -7.84516707e-02
2.89751232e-01 -8.00943017e-01 -7.83523321e-01 2.83909619e-01] | [-7.2375712394714355, 3.7010931968688965] |
2111c1d3-6202-4839-91e6-39e0e8023c0a | unicon-combating-label-noise-through-uniform | 2203.14542 | null | https://arxiv.org/abs/2203.14542v4 | https://arxiv.org/pdf/2203.14542v4.pdf | UNICON: Combating Label Noise Through Uniform Selection and Contrastive Learning | Supervised deep learning methods require a large repository of annotated data; hence, label noise is inevitable. Training with such noisy data negatively impacts the generalization performance of deep neural networks. To combat label noise, recent state-of-the-art methods employ some sort of sample selection mechanism to select a possibly clean subset of data. Next, an off-the-shelf semi-supervised learning method is used for training where rejected samples are treated as unlabeled data. Our comprehensive analysis shows that current selection methods disproportionately select samples from easy (fast learnable) classes while rejecting those from relatively harder ones. This creates class imbalance in the selected clean set and in turn, deteriorates performance under high label noise. In this work, we propose UNICON, a simple yet effective sample selection method which is robust to high label noise. To address the disproportionate selection of easy and hard samples, we introduce a Jensen-Shannon divergence based uniform selection mechanism which does not require any probabilistic modeling and hyperparameter tuning. We complement our selection method with contrastive learning to further combat the memorization of noisy labels. Extensive experimentation on multiple benchmark datasets demonstrates the effectiveness of UNICON; we obtain an 11.4% improvement over the current state-of-the-art on CIFAR100 dataset with a 90% noise rate. Our code is publicly available | ['Mubarak Shah', 'Ajmal Mian', 'Nazanin Rahnavard', 'Mamshad Nayeem Rizve', 'Nazmul Karim'] | 2022-03-28 | null | http://openaccess.thecvf.com//content/CVPR2022/html/Karim_UniCon_Combating_Label_Noise_Through_Uniform_Selection_and_Contrastive_Learning_CVPR_2022_paper.html | http://openaccess.thecvf.com//content/CVPR2022/papers/Karim_UniCon_Combating_Label_Noise_Through_Uniform_Selection_and_Contrastive_Learning_CVPR_2022_paper.pdf | cvpr-2022-1 | ['learning-with-noisy-labels', 'learning-with-noisy-labels'] | ['computer-vision', 'natural-language-processing'] | [ 2.36779734e-01 -1.11243874e-01 -2.50239402e-01 -7.67523229e-01
-1.03132522e+00 -4.07979190e-01 2.95594513e-01 3.75398189e-01
-9.07047927e-01 9.87231791e-01 -2.84409732e-01 -6.04461506e-03
-1.12501411e-02 -8.05997074e-01 -5.88479757e-01 -1.11567032e+00
2.97015727e-01 4.90647376e-01 1.96265221e-01 1.68281779e-01
9.16144401e-02 1.64473325e-01 -1.71994770e+00 2.87079036e-01
1.07388401e+00 1.33610189e+00 1.23398624e-01 1.16423517e-01
-1.60259381e-01 5.78828156e-01 -9.45744634e-01 -3.15829545e-01
2.39457428e-01 -3.46199244e-01 -6.07065916e-01 -1.88321590e-01
6.37385666e-01 -1.32462785e-01 -3.87964211e-02 1.23797739e+00
9.66227293e-01 2.32993692e-01 7.45135605e-01 -1.07352662e+00
-5.32960117e-01 9.69099820e-01 -5.00098109e-01 2.58394092e-01
-4.06060129e-01 -9.94954538e-03 9.30006564e-01 -1.12195027e+00
2.72875965e-01 9.21768486e-01 7.49832511e-01 9.33449507e-01
-1.37336493e+00 -1.14798522e+00 2.79431999e-01 9.11154673e-02
-1.53151309e+00 -4.89100486e-01 8.27350378e-01 -1.80934682e-01
4.13078189e-01 2.45627016e-01 2.90400326e-01 1.21996748e+00
5.66488653e-02 8.77237141e-01 1.25871170e+00 -4.65948611e-01
7.65720546e-01 3.11221749e-01 5.39078534e-01 3.91091079e-01
5.29817402e-01 3.63308303e-02 -6.60302997e-01 -2.54895926e-01
1.79428503e-01 -1.67170465e-02 -2.77805775e-01 -1.72157407e-01
-6.70953274e-01 8.48090112e-01 3.58481616e-01 9.07315232e-04
-1.48350254e-01 -1.90574738e-05 5.74675083e-01 2.52803624e-01
6.77678347e-01 2.83373684e-01 -6.05527878e-01 1.10937200e-01
-9.72810447e-01 6.75215796e-02 6.30586565e-01 8.08835328e-01
7.13745773e-01 7.19375238e-02 -3.79690051e-01 1.23688149e+00
2.79899210e-01 4.97338980e-01 7.18815923e-01 -6.46220803e-01
4.80927825e-01 4.79408711e-01 -1.16834700e-01 -6.80374622e-01
-3.71599168e-01 -1.12761021e+00 -1.02230048e+00 1.03251062e-01
2.52423435e-01 -2.95954674e-01 -1.20882976e+00 1.88352406e+00
3.34292293e-01 1.33218959e-01 1.64745580e-02 8.22157443e-01
9.55853701e-01 3.67732882e-01 3.22049320e-01 -3.29011649e-01
9.35537457e-01 -7.24533319e-01 -7.48800337e-01 -3.12796414e-01
6.95089996e-01 -3.87115419e-01 1.09820938e+00 7.51631260e-01
-6.29678726e-01 -4.24857736e-01 -1.24541783e+00 2.36168951e-01
-2.61046201e-01 2.98308074e-01 5.73905706e-01 9.64845896e-01
-6.90161526e-01 6.64243758e-01 -6.32088363e-01 2.24357873e-01
9.36913133e-01 5.45184016e-01 -2.50423819e-01 -2.20700443e-01
-1.22161818e+00 5.01372159e-01 5.22032857e-01 1.15856424e-01
-8.67221057e-01 -5.43962121e-01 -7.53772318e-01 1.32679775e-01
4.95440125e-01 -2.04957590e-01 1.26587069e+00 -9.66327727e-01
-1.41668856e+00 6.95581555e-01 6.48102462e-02 -4.55317080e-01
5.41039884e-01 -4.39883232e-01 -1.98287919e-01 -1.17536716e-01
-6.95533603e-02 7.05344796e-01 7.02139318e-01 -1.51522875e+00
-5.90859175e-01 -3.73132288e-01 -5.72330773e-01 1.45247355e-01
-6.59181237e-01 -2.46113598e-01 -2.21942589e-01 -8.75287056e-01
3.99084985e-01 -8.87186050e-01 -1.22533031e-01 -1.68704599e-01
-5.08280694e-01 -4.78303075e-01 5.45324862e-01 1.33740809e-02
1.25866055e+00 -2.11815643e+00 -3.70447487e-01 3.25692832e-01
3.59238803e-01 5.42537332e-01 -1.55367881e-01 -8.33462700e-02
-1.89873520e-02 3.64992023e-01 -2.91453451e-01 -6.22371972e-01
2.43373867e-02 -1.13505227e-02 -2.38440543e-01 6.14633918e-01
8.69093090e-02 3.69656801e-01 -1.02799094e+00 -3.04209471e-01
-6.93022311e-02 4.06055808e-01 -4.96765822e-01 6.65351376e-02
1.29789871e-03 1.38933361e-01 -3.67778718e-01 6.36978626e-01
8.80082369e-01 -8.77587646e-02 4.09624493e-03 -1.54583722e-01
3.88322055e-01 3.67245018e-01 -1.32990527e+00 1.34917462e+00
-3.23545247e-01 2.66631842e-01 -2.92282343e-01 -9.44876015e-01
1.21987295e+00 1.34597406e-01 2.36281246e-01 -6.05691612e-01
4.90484893e-01 4.01601642e-01 4.18876968e-02 -2.01927558e-01
2.47365072e-01 -1.97855502e-01 1.38370460e-02 4.17455196e-01
2.35167593e-01 8.82501230e-02 1.29207671e-01 2.29520835e-02
1.00150692e+00 -1.20794840e-01 -8.17961544e-02 -4.18561995e-01
1.70400873e-01 -3.66979331e-01 1.05776978e+00 1.18963754e+00
-6.08228147e-01 7.46326387e-01 2.42298529e-01 -3.34048927e-01
-5.82528591e-01 -8.61894727e-01 -4.34418350e-01 1.24772906e+00
1.13535836e-01 -1.89041913e-01 -8.53382766e-01 -1.01315141e+00
-6.86723068e-02 8.03114295e-01 -6.82230115e-01 -6.07399464e-01
-2.88509846e-01 -1.34515977e+00 6.08528674e-01 5.20679116e-01
4.74065572e-01 -1.08951378e+00 -3.71740729e-01 9.88697931e-02
-1.20302467e-02 -7.85374582e-01 -2.88378537e-01 9.99905825e-01
-7.77928472e-01 -8.71350646e-01 -4.52550411e-01 -8.81929994e-01
8.87121737e-01 2.10724384e-01 1.08809769e+00 8.53827447e-02
-4.19361033e-02 -2.55810350e-01 -5.80062747e-01 -7.19714940e-01
-2.73859739e-01 1.47071466e-01 3.89618218e-01 5.10051362e-02
7.05050170e-01 -3.31214041e-01 -6.06605113e-01 4.10339445e-01
-1.03416288e+00 -2.96677768e-01 4.43394154e-01 1.17763162e+00
9.19125319e-01 4.46780026e-01 1.18884289e+00 -1.18364477e+00
5.29175580e-01 -5.65687358e-01 -4.83345032e-01 1.30282730e-01
-7.05734313e-01 1.87085807e-01 8.66332710e-01 -8.13820064e-01
-1.01462460e+00 1.42492145e-01 -1.07501082e-01 -1.38962299e-01
-1.15142934e-01 4.66523349e-01 -2.94768363e-01 2.56622825e-02
1.03316510e+00 1.80075262e-02 -3.08467418e-01 -6.85088575e-01
-2.23239674e-03 9.86022294e-01 2.69192755e-01 -5.45555055e-01
5.23465693e-01 2.87761033e-01 -2.67954320e-01 -3.14090669e-01
-1.25620997e+00 -3.94041330e-01 -4.28416073e-01 -1.73448399e-02
2.84303695e-01 -9.97125447e-01 -3.76731277e-01 8.11221004e-01
-6.77290380e-01 -5.33106327e-01 -2.25343555e-01 5.36464393e-01
-6.90650493e-02 -2.73170811e-03 -5.53010285e-01 -8.93437386e-01
-6.25227153e-01 -1.21633291e+00 8.68043184e-01 4.07359898e-01
-1.03211291e-01 -5.57248652e-01 -2.69963622e-01 3.95557046e-01
3.91764671e-01 -7.42052915e-03 6.28824234e-01 -1.27016962e+00
-1.14029236e-01 -2.95580953e-01 -1.06357642e-01 7.79390752e-01
1.05276771e-01 -2.51468837e-01 -1.25056756e+00 -3.76783282e-01
1.54973134e-01 -7.63594210e-01 1.35866034e+00 3.09510440e-01
1.51390302e+00 -5.61871864e-02 -2.06792772e-01 5.21695435e-01
1.40333915e+00 1.63853943e-01 2.85380095e-01 3.29196990e-01
6.63586855e-01 3.43411744e-01 7.94284165e-01 4.80657995e-01
9.78322923e-02 2.61078626e-01 3.29164356e-01 2.19918694e-02
-1.31314173e-01 1.85719989e-02 7.54460022e-02 6.86940551e-01
5.60418665e-01 -4.49220479e-01 -9.16098177e-01 4.03442055e-01
-1.68382740e+00 -6.31102860e-01 3.02253086e-02 2.42336774e+00
1.62621176e+00 5.27148664e-01 -2.02463403e-01 5.70521057e-01
8.01501036e-01 -3.13301235e-01 -8.76654923e-01 9.11171064e-02
-6.75865039e-02 2.80757159e-01 6.99090779e-01 2.06710249e-01
-1.47757244e+00 8.73283744e-01 5.70278597e+00 1.33945119e+00
-1.32775915e+00 1.04514353e-01 1.27528071e+00 -3.61340404e-01
-1.69001967e-01 -5.23820877e-01 -1.22167706e+00 6.97560191e-01
9.16844964e-01 3.46007168e-01 7.05171674e-02 1.05554509e+00
5.70236444e-02 -2.89150685e-01 -1.01599669e+00 1.01497018e+00
4.12811302e-02 -1.02403665e+00 -2.00741038e-01 -2.99752533e-01
8.65203321e-01 2.85591453e-01 9.17614475e-02 5.31262815e-01
3.38746101e-01 -1.00021541e+00 8.91968668e-01 2.86061287e-01
7.16138542e-01 -1.01743698e+00 1.13587296e+00 4.39176351e-01
-5.65884471e-01 -2.73141772e-01 -6.44464731e-01 2.19175965e-01
-2.50967503e-01 1.42171216e+00 -5.90814054e-01 9.88523215e-02
7.78958261e-01 2.82396942e-01 -7.55771279e-01 1.36947095e+00
-2.37119645e-01 1.04843223e+00 -4.20173287e-01 -9.02392417e-02
-1.17802061e-01 1.08283274e-01 2.56876588e-01 1.15304291e+00
6.18122220e-02 -1.06681868e-01 3.30625743e-01 5.78526855e-01
-5.36265433e-01 1.04372837e-01 -2.07692444e-01 3.69539440e-01
9.61206794e-01 1.04101837e+00 -1.01452518e+00 -4.86200601e-01
2.85293460e-02 5.70651293e-01 5.80849349e-01 3.27309221e-01
-7.58436739e-01 -5.19115984e-01 2.15867355e-01 -2.20757946e-01
1.78291976e-01 1.09436564e-01 -7.98907578e-01 -1.08743441e+00
1.38946414e-01 -9.75761533e-01 2.77707964e-01 -8.60208422e-02
-1.60409343e+00 8.07578564e-01 -1.63196549e-01 -1.19282234e+00
3.64843667e-01 -4.30700511e-01 -2.66355425e-01 6.68556869e-01
-1.47954500e+00 -4.46841776e-01 -2.82424599e-01 9.61408988e-02
6.18695378e-01 -2.82023996e-01 7.07875371e-01 5.29883146e-01
-7.67366767e-01 1.08842742e+00 3.96760523e-01 2.05637917e-01
9.52467263e-01 -1.31855452e+00 1.25992954e-01 6.49397612e-01
4.40673679e-02 6.86549902e-01 5.85774302e-01 -6.54444218e-01
-9.78674948e-01 -1.47457910e+00 6.42771602e-01 -1.10155858e-01
1.94567814e-01 -6.15434945e-01 -1.11319065e+00 1.49734750e-01
-2.46529087e-01 4.62438166e-01 1.04622388e+00 9.58367586e-02
-5.09186745e-01 -5.60646355e-01 -1.39307261e+00 3.46293479e-01
8.13927174e-01 -1.94927439e-01 -3.60769838e-01 3.24451268e-01
6.02372825e-01 -3.37458014e-01 -5.26965261e-01 7.96776414e-01
3.22520465e-01 -8.68605614e-01 6.36180758e-01 -1.87424153e-01
-8.73158779e-03 -2.26085380e-01 -1.12016365e-01 -1.52727950e+00
-1.59297571e-01 -3.64245385e-01 -1.19031921e-01 1.34190071e+00
5.11409163e-01 -5.90778589e-01 9.07450140e-01 6.27122879e-01
-8.82839784e-02 -1.10120058e+00 -8.94206524e-01 -8.66479993e-01
4.07024771e-02 -3.55233461e-01 3.27027172e-01 1.04405010e+00
-2.93792397e-01 3.11415523e-01 -2.13087305e-01 -1.24867171e-01
7.88071454e-01 -3.04574013e-01 2.87890941e-01 -1.54259396e+00
-5.40189520e-02 -3.24606836e-01 -1.42313823e-01 -7.23096848e-01
1.66975796e-01 -8.68074298e-01 6.04472697e-01 -1.14819288e+00
2.65588224e-01 -9.72131670e-01 -8.08081985e-01 7.93031812e-01
-5.12319446e-01 4.96917158e-01 -6.22907234e-03 1.24140523e-01
-8.97192299e-01 8.68929386e-01 9.25284684e-01 -1.35924712e-01
-2.77824551e-01 1.09884046e-01 -9.29354072e-01 6.53107703e-01
1.06068099e+00 -9.91141021e-01 -7.17083633e-01 -4.06777114e-01
9.10958499e-02 -6.94242299e-01 -2.95790080e-02 -1.27907777e+00
-2.12583691e-03 -5.42449253e-03 6.00976288e-01 -3.65374923e-01
1.35459125e-01 -6.00063622e-01 -3.69673014e-01 2.94685304e-01
-7.66716421e-01 -4.23388749e-01 1.73055962e-01 7.01798797e-01
-4.89636660e-02 -5.77101946e-01 1.10972857e+00 1.13604099e-01
-2.55351484e-01 2.82614648e-01 -3.28532279e-01 2.63260782e-01
8.56352031e-01 1.39452130e-01 -4.22635555e-01 4.28119339e-02
-5.65012217e-01 2.46542320e-01 1.71671093e-01 3.06989312e-01
5.19734144e-01 -1.27004731e+00 -6.57142937e-01 2.44861096e-01
4.36357595e-02 3.95077467e-01 7.92123899e-02 4.94627416e-01
-2.42993400e-01 -1.49835702e-02 1.82715252e-01 -5.41434526e-01
-1.16770005e+00 2.10461140e-01 2.42508560e-01 -4.13213931e-02
-3.72097939e-01 1.32959116e+00 -1.56501561e-01 -5.33521950e-01
8.63500535e-01 -2.43560791e-01 -3.08687925e-01 2.08564922e-01
7.13602066e-01 1.57679126e-01 4.82076168e-01 -3.12209487e-01
-3.82915646e-01 -3.94083522e-02 -4.00804937e-01 1.09693244e-01
1.37609041e+00 2.40440574e-02 5.66652603e-02 5.37018597e-01
1.24816239e+00 -1.78541720e-01 -1.29661047e+00 -5.10836065e-01
6.54288977e-02 -2.89861739e-01 4.78118032e-01 -1.03790331e+00
-1.24096251e+00 7.28706837e-01 8.78636062e-01 1.85510069e-01
1.21406662e+00 -3.05227250e-01 8.07817400e-01 5.76849282e-01
3.82285684e-01 -1.43438768e+00 -3.94865638e-04 5.70222437e-01
4.51890677e-01 -1.60246205e+00 9.17846570e-04 -3.38034153e-01
-6.25951171e-01 8.45084369e-01 9.36714053e-01 -6.93136733e-03
8.29011381e-01 3.85594070e-01 3.01372349e-01 -1.75967831e-02
-7.85873413e-01 5.84643595e-02 1.93585575e-01 4.23729122e-01
4.82661813e-01 4.44183014e-02 -5.00236034e-01 1.04962087e+00
-3.55869159e-02 -7.39496574e-02 4.52281058e-01 1.05113387e+00
-6.47091985e-01 -1.07319462e+00 -3.17966461e-01 7.72601843e-01
-6.41766489e-01 -3.46657306e-01 -1.45032033e-01 2.52222091e-01
4.47142124e-01 1.11686802e+00 -7.89993107e-02 -4.19904679e-01
-1.24821309e-02 1.84354782e-01 8.77225921e-02 -8.60466480e-01
-6.00054264e-01 6.09609187e-02 -3.55442762e-02 -1.91087246e-01
-2.35689744e-01 -3.71540725e-01 -1.35186017e+00 1.24199256e-01
-7.44240522e-01 2.15108588e-01 6.80801868e-01 9.72883284e-01
3.34193915e-01 5.69166005e-01 6.34313881e-01 -7.85497069e-01
-1.04911315e+00 -1.04866374e+00 -5.64751446e-01 4.29971606e-01
4.56901848e-01 -9.44957197e-01 -5.35074770e-01 -1.94763079e-01] | [9.340409278869629, 3.857887029647827] |
2d9a2062-3d35-4a17-b9ac-a741e45128e6 | global-wheat-head-dataset-2021-an-update-to | 2105.07660 | null | https://arxiv.org/abs/2105.07660v2 | https://arxiv.org/pdf/2105.07660v2.pdf | Global Wheat Head Dataset 2021: more diversity to improve the benchmarking of wheat head localization methods | The Global Wheat Head Detection (GWHD) dataset was created in 2020 and has assembled 193,634 labelled wheat heads from 4,700 RGB images acquired from various acquisition platforms and 7 countries/institutions. With an associated competition hosted in Kaggle, GWHD has successfully attracted attention from both the computer vision and agricultural science communities. From this first experience in 2020, a few avenues for improvements have been identified, especially from the perspective of data size, head diversity and label reliability. To address these issues, the 2020 dataset has been reexamined, relabeled, and augmented by adding 1,722 images from 5 additional countries, allowing for 81,553 additional wheat heads to be added. We now release a new version of the Global Wheat Head Detection (GWHD) dataset in 2021, which is bigger, more diverse, and less noisy than the 2020 version. The GWHD 2021 is now publicly available at http://www.global-wheat.com/ and a new data challenge has been organized on AIcrowd to make use of this updated dataset. | ['Wei Guo', 'Ian Stavness', 'Frédéric Baret', 'Benoit de Solan', 'Scott Chapman', 'Jesse Poland', 'Morten Lilimo', 'David Shaner LeBauer', 'Curtis Pozniak', 'Minhajul A. Badhon', 'Masanori Ishii', 'Haozhou Wang', 'Ken Kuroki', 'Benoit Mercatoris', 'Alexis Carlier', 'Sébastien Dandrifosse', 'Goro Ishikawa', 'Koichi Nagasawa', 'Pouria Sadhegi-Tehran', 'Andreas Hund', 'Helge Aasen', 'Norbert Kichgessner', 'Bangyou Zheng', 'Shuhei Nasuda', 'Hisashi Tsujimoto', 'Izzat S. A. Tahir', 'Shahameh Shafiee', 'Francisco Pinto Espinosa', 'Xu Wang', 'Shouyang Liu', 'Kaaviya Velumani', 'Simon Madec', 'Daniel Smith', 'Mario Serouart', 'Etienne David'] | 2021-05-17 | null | null | null | null | ['head-detection'] | ['computer-vision'] | [-7.63516268e-03 6.82018921e-02 2.46663485e-02 -3.92170161e-01
-6.30839646e-01 -1.04621053e+00 3.46028030e-01 3.44135404e-01
-1.50035322e-01 7.30451345e-01 7.23331198e-02 -2.62483150e-01
6.54770136e-02 -9.81934786e-01 -4.91607666e-01 -7.55842626e-01
-1.62424281e-01 2.08722159e-01 1.32925838e-01 -3.06510597e-01
-2.65341967e-01 4.92577255e-01 -1.76593578e+00 -1.40534461e-01
7.05847204e-01 1.09854090e+00 7.36062825e-01 6.58273637e-01
2.97810942e-01 3.25472206e-01 -3.69084746e-01 -2.56873488e-01
5.37350059e-01 -3.11976731e-01 -7.09339499e-01 1.96771383e-01
5.15016496e-01 -3.12134027e-01 2.28207231e-01 1.07709885e+00
7.78756618e-01 -3.27039540e-01 1.62571892e-02 -1.29796863e+00
-9.24556315e-01 6.70062900e-01 -9.94606793e-01 -1.96451426e-01
1.87534899e-01 6.45476207e-02 1.07783520e+00 -8.59646380e-01
7.08838761e-01 9.39437628e-01 1.10997105e+00 -4.74184006e-02
-1.05512083e+00 -8.04925740e-01 6.96582571e-02 1.87447369e-01
-1.37813020e+00 -3.57016742e-01 2.14164227e-01 -2.74494946e-01
5.50287962e-01 6.29706264e-01 9.35929835e-01 5.67212880e-01
-2.99730012e-03 8.81478667e-01 1.50195742e+00 -7.29244173e-01
1.65871710e-01 -3.09570581e-01 2.80442595e-01 6.45003319e-01
8.09147239e-01 2.99650013e-01 -2.80345887e-01 1.36512160e-01
4.89222229e-01 -4.19695348e-01 -3.44794422e-01 -6.14555240e-01
-1.10467005e+00 9.55603480e-01 8.38026643e-01 2.07034349e-02
-5.93689203e-01 -5.56225359e-01 4.02540445e-01 1.70159727e-01
5.34013987e-01 6.21039152e-01 -8.86634886e-01 2.94245541e-01
-8.10650587e-01 7.52989426e-02 7.18346059e-01 1.13169599e+00
9.54317570e-01 -1.49544016e-01 1.84066936e-01 8.50927532e-01
4.06996816e-01 1.14353514e+00 -7.10684881e-02 -1.13844752e+00
-3.36812697e-02 6.93869233e-01 3.57146472e-01 -1.07051682e+00
-9.75458026e-01 -2.78065443e-01 -7.50522435e-01 9.72263515e-02
2.72033215e-01 -5.01254082e-01 -1.34239781e+00 1.53539526e+00
3.51698309e-01 -3.28797817e-01 -7.31832236e-02 9.60571945e-01
1.11992502e+00 5.22472978e-01 -7.76109770e-02 -6.86542764e-02
1.51155651e+00 -7.90414989e-01 -7.10372567e-01 -3.97979796e-01
7.14808822e-01 -1.00352788e+00 5.15294611e-01 5.39392292e-01
-5.99911511e-01 -5.80232680e-01 -1.07934284e+00 2.61409491e-01
-9.49953437e-01 3.61663520e-01 8.50723445e-01 6.82909966e-01
-1.25322711e+00 -1.40911341e-01 -7.04480767e-01 -1.23104990e+00
2.40512133e-01 1.03584193e-01 -6.46003723e-01 -4.42499757e-01
-1.06419945e+00 1.00942802e+00 8.18071842e-01 5.41540861e-01
-7.93253958e-01 -4.27135795e-01 -9.68403220e-01 -3.90689611e-01
2.33208090e-01 -2.07058311e-01 9.94196594e-01 -3.87026817e-01
-1.18314433e+00 1.21314275e+00 3.11034292e-01 -2.95712233e-01
1.46644011e-01 -3.49408537e-01 -6.63434505e-01 -3.61150712e-01
6.76564217e-01 9.88457382e-01 5.60262688e-02 -1.30927896e+00
-1.34164464e+00 -7.18609214e-01 -6.45967945e-02 1.09607965e-01
1.03464060e-01 1.64565891e-01 -4.61315453e-01 -6.31038725e-01
2.63826281e-01 -1.31644094e+00 -4.80777919e-02 -1.08103901e-01
-3.99936885e-01 3.55982363e-01 8.26162755e-01 -1.01340222e+00
6.76538169e-01 -2.01572323e+00 -1.90359712e-01 -1.69217631e-01
-1.03711545e-01 4.06279236e-01 -4.23964113e-01 2.72248030e-01
-7.51161426e-02 -4.08896387e-01 -5.04059553e-01 1.93520654e-02
-7.66535103e-02 2.92554796e-01 2.42801681e-02 6.39480412e-01
1.19130135e-01 8.68937433e-01 -1.12639773e+00 -4.38691443e-03
7.37722456e-01 6.83803856e-02 3.13968621e-02 1.56310499e-01
1.98983744e-01 2.65989989e-01 -1.29939660e-01 1.39023936e+00
1.61258459e+00 1.48265168e-01 2.62471080e-01 -3.90102178e-01
-7.19915450e-01 -6.54370129e-01 -1.12759292e+00 1.56061649e+00
-4.99241129e-02 6.66150451e-01 4.94154543e-01 -7.04602182e-01
1.13830888e+00 8.35850090e-02 4.28853780e-01 -6.22175038e-01
-4.94252741e-02 4.02069628e-01 -2.14749113e-01 -1.19815193e-01
7.11880088e-01 2.93799251e-01 -4.47481215e-01 -1.52544126e-01
3.30523580e-01 -3.53710055e-01 4.86953557e-01 -1.77686185e-01
9.84484434e-01 4.94979978e-01 4.25151080e-01 -5.67763090e-01
6.17993250e-02 8.05537462e-01 7.32950032e-01 7.55132854e-01
-7.20803380e-01 7.74274051e-01 1.26493752e-01 -5.54533601e-01
-9.24210787e-01 -8.98098052e-01 -3.53591263e-01 1.37981737e+00
2.89864875e-02 1.32541982e-02 -6.36261582e-01 -2.60400295e-01
1.59404084e-01 4.41540390e-01 -8.39524627e-01 1.22168824e-01
-2.57920831e-01 -1.52974248e+00 7.37325966e-01 4.63714421e-01
9.42066252e-01 -1.30356181e+00 -8.38774562e-01 3.21812958e-01
-2.13373452e-01 -1.04385936e+00 1.84866354e-01 9.15310085e-01
-3.15434307e-01 -1.18747401e+00 -1.08961689e+00 -9.02335465e-01
3.60000223e-01 5.20464957e-01 1.27786827e+00 -5.13301611e-01
-5.18254817e-01 -1.35604098e-01 -9.69092548e-01 -9.52429950e-01
-7.18895197e-02 3.15327525e-01 -1.25871971e-01 -5.43738544e-01
5.30209363e-01 2.86286056e-01 -3.67436230e-01 3.86611104e-01
-6.26531124e-01 -8.74810368e-02 6.04882956e-01 7.47933626e-01
5.41659892e-01 -2.06733510e-01 5.99295914e-01 -8.11590254e-01
-9.10244957e-02 -5.10176480e-01 -1.13080740e+00 5.04827023e-01
-3.69812876e-01 -6.54365420e-01 6.70491830e-02 3.97291750e-01
-1.15880072e+00 5.92373788e-01 -9.08360407e-02 2.23817259e-01
-5.16205430e-01 7.80914426e-01 -3.68786931e-01 -2.19980359e-01
4.52257544e-01 -4.29218918e-01 -1.45952240e-01 -5.98080039e-01
8.75802577e-01 7.23816037e-01 1.04888725e+00 -1.34257004e-01
5.76552391e-01 3.23233813e-01 -2.56748378e-01 -8.51775408e-01
-1.01486540e+00 -7.52683520e-01 -7.82609224e-01 -1.90922856e-01
7.34654963e-01 -1.22637546e+00 -2.08639979e-01 1.36501455e+00
-7.13841259e-01 -4.04056072e-01 -3.34078133e-01 3.10928017e-01
-2.80716121e-02 3.42337228e-02 -5.10634363e-01 -5.52984655e-01
-3.42729807e-01 -9.14240539e-01 1.20267522e+00 3.66955161e-01
1.89638838e-01 -6.22234285e-01 1.34374991e-01 -6.91604316e-02
4.22277272e-01 8.64713013e-01 3.38858157e-01 -8.14232007e-02
1.28599957e-01 -5.64361811e-01 -6.82842970e-01 3.03259641e-01
6.97300076e-01 2.84697205e-01 -1.06604433e+00 -6.19224668e-01
-4.64487940e-01 -3.57290357e-01 8.87376606e-01 8.26337039e-01
3.17165703e-01 5.42866766e-01 -2.69199580e-01 6.53329492e-01
1.50760245e+00 3.40963840e-01 3.27858031e-01 8.34020495e-01
5.99085569e-01 8.47448826e-01 1.27171433e+00 5.79983652e-01
5.70577443e-01 3.96480292e-01 1.01376045e+00 -7.44534969e-01
-2.20209986e-01 1.82767510e-01 1.11564733e-01 8.56543720e-01
-7.62053058e-02 -2.97526032e-01 -1.04290414e+00 6.94349706e-01
-1.73207974e+00 -4.29589689e-01 -4.05893803e-01 2.05862212e+00
3.57520312e-01 -2.78579861e-01 1.61753610e-01 2.73070461e-03
9.70212996e-01 1.62359968e-01 -6.14876509e-01 2.10110676e-02
-1.02340734e+00 2.28601217e-01 1.31208348e+00 2.91624278e-01
-1.55189860e+00 1.21886861e+00 7.08264351e+00 2.44530857e-01
-8.72814894e-01 -1.47430196e-01 5.20955563e-01 5.19223750e-01
3.94903123e-01 1.96762625e-02 -7.38817453e-01 9.12961140e-02
6.42234683e-01 1.68552026e-02 8.33084881e-02 6.05065882e-01
-5.83548136e-02 -4.43378001e-01 -1.68694392e-01 3.37025821e-01
-1.68533385e-01 -1.03409922e+00 -4.66258287e-01 1.05144963e-01
8.03722143e-01 6.76219881e-01 -1.47525981e-01 3.07441890e-01
9.46224391e-01 -5.57485163e-01 8.02711725e-01 3.09827447e-01
9.82165277e-01 -6.81490839e-01 1.38812661e+00 -9.72051471e-02
-1.63941848e+00 -2.21823096e-01 -5.76956213e-01 -4.39904444e-02
1.69049934e-01 7.72201598e-01 -3.70577753e-01 1.18985415e+00
1.40378106e+00 6.73462689e-01 -1.21139443e+00 1.20974970e+00
-2.08667591e-01 3.76829058e-01 -4.99878824e-01 2.99610585e-01
3.39705408e-01 -2.51547635e-01 1.21847853e-01 9.13205326e-01
5.81828892e-01 -7.31887147e-02 2.69236803e-01 1.07431397e-01
1.57930255e-01 -2.33406633e-01 -5.49905360e-01 2.55178139e-02
7.00556278e-01 1.68119276e+00 -1.09763348e+00 5.23856021e-02
-4.51189011e-01 8.54450762e-01 -1.65546283e-01 7.43280351e-02
-6.63599491e-01 -4.49621677e-01 4.42670286e-01 -5.28255641e-01
3.04681033e-01 7.50634447e-02 9.93601754e-02 -6.65021479e-01
-5.10139704e-01 -8.42532218e-01 5.51408291e-01 -1.17793751e+00
-1.24092185e+00 5.17625868e-01 -1.64173394e-01 -9.04351950e-01
8.89844261e-03 -1.08326411e+00 1.71320513e-01 9.98843014e-01
-1.42238176e+00 -1.80414832e+00 -9.19444084e-01 2.79706735e-02
2.99569905e-01 -1.06577188e-01 1.41056800e+00 2.57157505e-01
-7.06318259e-01 2.22951904e-01 3.84856641e-01 -4.72457632e-02
8.65531027e-01 -1.43122995e+00 8.39629173e-01 9.64363456e-01
-2.01099768e-01 -1.46490902e-01 7.61422873e-01 -5.40736258e-01
-1.36303341e+00 -1.39487159e+00 7.43887663e-01 -4.90311950e-01
6.94014847e-01 -1.81659997e-01 -5.01361430e-01 9.48928833e-01
3.27313930e-01 6.74437508e-02 5.26144087e-01 -1.32682472e-01
2.40749121e-02 -1.37200922e-01 -1.19468319e+00 -2.23444756e-02
7.03498900e-01 -9.12657231e-02 -1.47704184e-01 2.94679612e-01
5.58201492e-01 -5.67135453e-01 -9.22116518e-01 8.34006608e-01
6.36004090e-01 -8.26736212e-01 6.01969242e-01 1.93653882e-01
-1.26218304e-01 -5.79270899e-01 -8.16192210e-01 -1.58362150e+00
-7.39048600e-01 -3.05099934e-01 2.85572648e-01 1.44534218e+00
2.56710887e-01 -6.47937953e-01 6.76296473e-01 -1.75639004e-01
-4.10164356e-01 2.19717110e-03 -5.72213650e-01 -7.16085196e-01
9.19034854e-02 -8.11656117e-02 9.50756609e-01 1.04849541e+00
-5.44143260e-01 -2.69429553e-02 -4.48474646e-01 5.46034157e-01
5.50631762e-01 3.77784520e-01 7.79100239e-01 -1.54881418e+00
4.94135201e-01 -3.62877734e-02 -3.71562093e-01 -5.90978384e-01
-5.14045537e-01 -6.73383296e-01 4.97761756e-01 -1.71032524e+00
1.94663048e-01 -4.24738497e-01 -4.36865121e-01 9.71557617e-01
-2.95632243e-01 7.09346712e-01 2.68959850e-01 1.22232355e-01
-2.78854281e-01 1.68514792e-02 9.03040826e-01 -2.20453776e-02
1.16215006e-01 -1.96567968e-01 -8.43511343e-01 5.46942294e-01
8.25580180e-01 4.72058449e-03 3.28094453e-01 -5.12465239e-01
-1.54942134e-02 -5.65062523e-01 1.99159041e-01 -9.61051166e-01
-3.28132153e-01 -1.71770398e-02 5.90918541e-01 -1.10176861e+00
1.05834484e-01 -8.11932862e-01 4.72593874e-01 5.68966627e-01
3.75139832e-01 4.10124391e-01 5.65172672e-01 -1.90626606e-02
-8.84760469e-02 1.35815311e-02 6.45899713e-01 -2.33532920e-01
-1.57094812e+00 1.19098812e-01 -3.53775322e-01 -2.67476171e-01
1.12004554e+00 -1.46753564e-01 -5.15459538e-01 1.61460489e-01
-8.11365783e-01 5.10716796e-01 7.85836816e-01 6.40259802e-01
7.28996545e-02 -1.16813612e+00 -1.05363989e+00 3.56064469e-01
6.79728925e-01 -6.52189031e-02 -3.94055769e-02 5.28734148e-01
-1.07958484e+00 4.82321292e-01 -7.84619868e-01 -6.30797029e-01
-9.90106404e-01 4.33590591e-01 -5.08483080e-03 -1.30011752e-01
-4.88231957e-01 9.34960604e-01 3.73922661e-02 -1.08872640e+00
2.28495225e-02 -1.60077780e-01 -4.50523198e-01 7.04338610e-01
3.74883473e-01 4.22211021e-01 3.53948623e-01 -9.89495635e-01
-5.44532418e-01 4.52538192e-01 2.46454835e-01 1.70489684e-01
1.57583296e+00 -3.13142657e-01 -2.24884525e-01 3.11918467e-01
6.81011617e-01 -4.67532098e-01 -1.21292448e+00 6.62960187e-02
3.73632729e-01 -2.40697727e-01 1.58441037e-01 -1.21202552e+00
-1.32889569e+00 3.82044464e-01 1.39795721e+00 6.41858935e-01
1.61161029e+00 -4.66927402e-02 6.06849730e-01 1.53669259e-02
7.59718239e-01 -9.16493714e-01 -6.68619096e-01 6.73477352e-01
8.45510364e-01 -1.52675271e+00 -3.45517509e-02 -4.86481488e-01
-3.24189514e-01 7.52168059e-01 4.29156601e-01 1.12245493e-01
4.86294538e-01 6.28286421e-01 5.76995969e-01 -2.75899798e-01
-4.37988900e-02 -6.78960502e-01 -2.24218741e-01 1.24152815e+00
5.59038579e-01 7.61911273e-01 -3.27318087e-02 3.22872043e-01
-2.47390643e-01 1.23758480e-01 4.50758427e-01 1.38906062e+00
-5.53324103e-01 -8.45191836e-01 -7.09722042e-01 4.31956112e-01
3.17793489e-02 -2.12772071e-01 -5.25284588e-01 9.00983155e-01
4.96813387e-01 1.17297494e+00 -1.28183648e-01 -1.96770579e-01
7.28135347e-01 -3.80974889e-01 3.80260468e-01 -5.30551255e-01
-2.50625849e-01 1.40914796e-02 2.07287073e-01 -3.85990471e-01
-7.50109732e-01 -7.70401001e-01 -6.15106404e-01 -5.14910936e-01
-9.82116163e-01 -7.28006940e-03 6.64937079e-01 3.80850673e-01
2.43110403e-01 4.19579536e-01 6.26697481e-01 -1.24155414e+00
-9.12255719e-02 -1.04056108e+00 -1.13632667e+00 -1.91029429e-01
1.56675056e-01 -6.86782360e-01 -2.09170192e-01 8.97727087e-02] | [9.229683876037598, -1.5168362855911255] |
2d173faa-98c6-4d08-a10c-e1daef2efd98 | video-object-tracking-based-on-yolov7-and | 2207.12202 | null | https://arxiv.org/abs/2207.12202v1 | https://arxiv.org/pdf/2207.12202v1.pdf | Video object tracking based on YOLOv7 and DeepSORT | Multiple object tracking (MOT) is an important technology in the field of computer vision, which is widely used in automatic driving, intelligent monitoring, behavior recognition and other directions. Among the current popular MOT methods based on deep learning, Detection Based Tracking (DBT) is the most widely used in industry, and the performance of them depend on their object detection network. At present, the DBT algorithm with good performance and the most widely used is YOLOv5-DeepSORT. Inspired by YOLOv5-DeepSORT, with the proposal of YOLOv7 network, which performs better in object detection, we apply YOLOv7 as the object detection part to the DeepSORT, and propose YOLOv7-DeepSORT. After experimental evaluation, compared with the previous YOLOv5-DeepSORT, YOLOv7-DeepSORT performances better in tracking accuracy. | ['Bo Liu', 'Xingle Zhang', 'Feng Yang'] | 2022-07-25 | null | null | null | null | ['video-object-tracking'] | ['computer-vision'] | [-4.07508969e-01 -6.51102245e-01 -2.31010333e-01 4.17108804e-01
9.73664597e-02 1.74485609e-01 3.64031225e-01 1.11509059e-02
-6.83660388e-01 4.91449356e-01 -4.55544740e-01 -1.35990858e-01
-1.12441465e-01 -8.15891862e-01 -6.43195570e-01 -8.55909288e-01
1.36955917e-01 3.43917936e-01 1.23696661e+00 -4.66285080e-01
9.65916216e-02 3.54479998e-01 -1.59037673e+00 -2.76361853e-01
6.62530243e-01 1.09121478e+00 4.44491863e-01 4.10088480e-01
-1.69863284e-01 7.77076244e-01 -5.71566045e-01 -1.63935959e-01
4.22031768e-02 -3.72935027e-01 1.05944738e-01 -4.43028212e-01
2.39664689e-01 -2.17056781e-01 -5.04347324e-01 1.07963896e+00
5.84133804e-01 4.93175350e-03 5.16914427e-01 -1.54287124e+00
-4.36635315e-01 4.15969968e-01 -8.34043562e-01 6.20181441e-01
-2.28664681e-01 3.15173984e-01 3.74744177e-01 -8.01952541e-01
2.89058506e-01 1.62408161e+00 9.96886432e-01 6.61560893e-01
-6.08311057e-01 -1.19497013e+00 2.24689230e-01 9.44809616e-01
-1.44672060e+00 -1.06184408e-01 5.82986951e-01 -4.82326746e-01
5.25882840e-01 2.42683925e-02 9.20051336e-01 8.18417668e-01
7.43000925e-01 1.28579974e+00 9.60083187e-01 -1.00726239e-01
-1.03136562e-01 8.10955372e-03 3.90691519e-01 7.66905367e-01
4.89218265e-01 4.13090706e-01 -2.15717062e-01 4.05852795e-01
5.57691336e-01 2.13947237e-01 1.00708857e-01 -2.78934121e-01
-1.18147886e+00 6.74142957e-01 7.50611782e-01 5.82202017e-01
-2.02673078e-01 5.88576019e-01 5.00348508e-01 1.06987342e-01
1.59351364e-01 -2.89265007e-01 -4.69973758e-02 5.60240299e-02
-7.27823377e-01 4.38716710e-01 2.25964505e-02 1.01891482e+00
5.85799336e-01 5.61416507e-01 -4.34568256e-01 4.02385443e-01
5.83179593e-01 9.74414229e-01 7.45970130e-01 -3.94313365e-01
6.21287823e-02 8.97951245e-01 1.63676783e-01 -1.03546906e+00
-4.79241818e-01 -6.19062185e-01 -6.20111287e-01 5.49533188e-01
2.31755108e-01 -9.42101404e-02 -7.17930555e-01 1.25868559e+00
3.57574731e-01 6.87509358e-01 2.06917360e-01 8.71508658e-01
1.18249691e+00 6.69268072e-01 1.17698245e-01 -1.21642850e-01
1.40356958e+00 -1.09139824e+00 -7.94053137e-01 -1.46722257e-01
5.59860408e-01 -7.08913326e-01 3.57314587e-01 3.36447984e-01
-3.22281063e-01 -1.07720959e+00 -1.08488572e+00 2.14121178e-01
-4.44565326e-01 2.04040140e-01 4.20501620e-01 6.78688467e-01
-7.84045875e-01 2.40209311e-01 -8.86780441e-01 -6.15220249e-01
5.26918828e-01 4.03660178e-01 1.89250186e-02 -1.17517747e-01
-1.10127306e+00 1.24489605e+00 6.82136536e-01 2.37582520e-01
-1.32979095e+00 -1.63491875e-01 -4.73174453e-01 -1.45277143e-01
5.36268950e-01 -3.06461096e-01 1.20149744e+00 -5.13739645e-01
-1.15331781e+00 3.14353049e-01 -6.92267418e-02 -9.33655977e-01
8.12737584e-01 -2.82739252e-01 -6.57355130e-01 -2.31687456e-01
1.63524523e-01 7.20161796e-01 8.48347485e-01 -9.39682603e-01
-9.00743365e-01 -2.75938332e-01 -3.20858926e-01 -9.93832573e-03
-3.37765813e-01 3.21789384e-02 -3.26452792e-01 -9.17996988e-02
-2.31665030e-01 -9.25220191e-01 -7.32672140e-02 3.74403037e-02
-2.55973727e-01 -9.86837387e-01 1.86206162e+00 -1.92077041e-01
1.00020838e+00 -2.23370075e+00 -1.20980598e-01 -4.24392998e-01
3.63557279e-01 9.27893281e-01 7.50116408e-02 1.50780782e-01
5.02167583e-01 -5.31690776e-01 2.77248323e-01 -1.31913170e-01
4.52366769e-02 2.00327232e-01 6.46176711e-02 5.29587328e-01
-8.42024237e-02 1.00957549e+00 -9.28026259e-01 -7.87498355e-01
8.50389302e-01 3.39871645e-01 -2.30146527e-01 -1.27277121e-01
-4.01303053e-01 5.36052465e-01 -8.70185554e-01 6.28944099e-01
8.20652723e-01 6.64357319e-02 -6.17723584e-01 -2.01856941e-01
-7.28872061e-01 -4.36951160e-01 -1.07476020e+00 1.17223871e+00
-5.64198829e-02 1.09011912e+00 -8.36276338e-02 -1.06428635e+00
1.25384927e+00 1.70884386e-01 5.28988659e-01 -7.54170179e-01
5.97478449e-01 4.61366415e-01 3.96889359e-01 -6.72822952e-01
5.09344876e-01 1.54158417e-02 1.66917369e-01 -4.10726279e-01
-1.60578847e-01 3.37801397e-01 3.28225762e-01 -1.93825569e-02
9.73967493e-01 2.00931400e-01 -4.05856334e-02 -3.10594112e-01
7.08063066e-01 2.80186862e-01 8.90770912e-01 6.48434341e-01
-6.35059953e-01 1.70969844e-01 -5.95460385e-02 -5.93793631e-01
-6.66100204e-01 -5.67503035e-01 -2.89377868e-01 6.92165494e-01
1.03674841e+00 -8.25890992e-03 -4.78510469e-01 -6.33733213e-01
3.02797854e-01 5.78831136e-01 -5.02879620e-01 -3.79441202e-01
-7.53512442e-01 -5.91255844e-01 7.22208440e-01 6.29341185e-01
1.25842655e+00 -1.42536700e+00 -8.17667544e-01 4.54148054e-01
2.32221127e-01 -1.04465556e+00 -7.17428932e-03 -5.40892109e-02
-6.22969449e-01 -1.11400938e+00 -1.04022920e+00 -8.95710349e-01
1.04424015e-01 6.35934412e-01 4.55015033e-01 2.77956456e-01
-2.46010274e-01 2.37204269e-01 -3.67833048e-01 -9.57727194e-01
-3.13501269e-01 -6.93633258e-02 1.74429104e-01 1.23539567e-01
7.10366726e-01 -2.15401039e-01 -5.92734754e-01 5.07347465e-01
-6.08461797e-01 -1.78231716e-01 8.05978775e-01 4.39484239e-01
2.81733871e-01 1.20705947e-01 5.63250363e-01 -4.55327988e-01
2.14838505e-01 -3.13151449e-01 -9.33648288e-01 5.94330542e-02
-5.55651188e-01 -3.45144540e-01 6.33525312e-01 -3.99158418e-01
-6.41109765e-01 -1.42053872e-01 -3.86345625e-01 -8.76987994e-01
-1.51456878e-01 -2.31097662e-03 -3.68613154e-02 -4.51054990e-01
2.76073009e-01 5.20673633e-01 -1.03286318e-01 -5.81163704e-01
2.83423942e-02 5.73224962e-01 4.59694684e-01 7.97532871e-02
7.87319005e-01 3.63869637e-01 1.52491003e-01 -9.03809428e-01
-5.01669705e-01 -4.84281629e-01 -4.43388283e-01 -6.66919529e-01
1.36653972e+00 -9.40830171e-01 -1.18220115e+00 8.54881585e-01
-1.15656042e+00 7.06801936e-03 1.57476187e-01 9.40468371e-01
-6.83891103e-02 7.96095431e-02 -2.24480614e-01 -9.97609735e-01
-2.70692825e-01 -1.38960445e+00 1.20116603e+00 8.36515307e-01
4.21197295e-01 -9.66911077e-01 -6.68625310e-02 2.28396043e-01
3.87464732e-01 2.25739211e-01 4.01941538e-01 -4.93655086e-01
-1.02833748e+00 -2.43586197e-01 -3.43474865e-01 2.14755669e-01
-1.50623813e-01 -8.48317295e-02 -7.09973156e-01 -4.58703071e-01
-2.62177020e-01 3.09907664e-02 1.25156331e+00 4.89472032e-01
8.58311057e-01 1.86866477e-01 -1.02652860e+00 5.30342221e-01
1.54835415e+00 7.78508961e-01 4.99189466e-01 8.61192942e-01
9.59023237e-01 7.34283254e-02 9.25734818e-01 -3.14225778e-02
4.67008531e-01 8.25758100e-01 8.52917910e-01 -3.36268637e-03
-4.37518626e-01 -1.80005208e-01 6.90135062e-01 8.65484655e-01
-2.20422819e-02 -2.79005140e-01 -6.63121581e-01 5.40275276e-01
-2.02159357e+00 -1.23311234e+00 -6.22550130e-01 1.80677462e+00
5.16000378e-04 6.84329808e-01 4.26456213e-01 1.76121846e-01
9.16071057e-01 2.02388659e-01 -9.50510383e-01 -3.74490954e-02
-1.19534835e-01 -1.59289733e-01 7.92182803e-01 8.19318816e-02
-1.05286860e+00 1.03516710e+00 5.24029732e+00 1.30969036e+00
-1.24245059e+00 3.00008535e-01 -1.90992519e-01 3.30050945e-01
2.90963233e-01 -2.36789897e-01 -1.56392837e+00 7.59998679e-01
5.85211277e-01 -2.29243144e-01 -5.23254126e-02 1.01999068e+00
3.30029428e-01 -1.50345013e-01 -6.08618319e-01 9.98579979e-01
-7.39859743e-03 -1.22345328e+00 -1.52906641e-01 -1.44646600e-01
4.36839223e-01 1.38591841e-01 1.07136883e-01 7.60274529e-01
2.08640769e-01 -5.26692510e-01 8.03660274e-01 2.52356768e-01
2.49716133e-01 -6.50547802e-01 1.03788090e+00 6.46481395e-01
-1.72042477e+00 -2.80198544e-01 -6.79882288e-01 5.12694046e-02
3.14797573e-02 3.61412168e-01 -6.26910865e-01 6.76423848e-01
9.75710928e-01 1.41617370e+00 -6.34651661e-01 1.72347558e+00
-1.20440498e-02 6.39878929e-01 -2.24312916e-01 -6.16625071e-01
3.42675656e-01 -1.99757159e-01 9.19884562e-01 1.06375766e+00
4.14739549e-01 -4.03432846e-01 6.56069100e-01 7.49126136e-01
1.00936882e-01 -1.91105470e-01 -8.17863107e-01 3.08670074e-01
5.11624098e-01 1.39000010e+00 -8.66374791e-01 -6.41643107e-01
-3.32389653e-01 3.24472606e-01 -2.66998470e-01 -9.39795822e-02
-1.16511917e+00 -4.34091806e-01 5.32723725e-01 4.96063801e-03
7.47945428e-01 -2.62543499e-01 2.26659268e-01 -7.28676379e-01
-2.07380176e-01 -5.30461729e-01 1.97745427e-01 -9.92928803e-01
-8.98752928e-01 5.42203546e-01 8.13066512e-02 -1.84257400e+00
3.07782829e-01 -9.46127415e-01 -7.51860201e-01 5.94776750e-01
-1.75780857e+00 -1.12118018e+00 -6.41374052e-01 4.92374957e-01
6.54860139e-01 -4.47391093e-01 1.81012541e-01 7.23412037e-01
-1.02499032e+00 4.41645026e-01 4.81518030e-01 2.54209906e-01
3.38731349e-01 -8.39578152e-01 1.64719805e-01 8.73098016e-01
-9.43355486e-02 2.36566469e-01 6.76271498e-01 -6.00360572e-01
-1.48781013e+00 -1.39497328e+00 1.04715638e-01 -1.90617830e-01
6.02533817e-01 -8.07601437e-02 -7.84733593e-01 7.45235860e-01
4.68106449e-01 1.65868044e-01 -1.26448080e-01 -6.42763853e-01
1.87760040e-01 -6.03554070e-01 -9.75314200e-01 2.91495681e-01
8.45059276e-01 1.85955182e-01 -5.52260399e-01 3.73527318e-01
8.08433652e-01 -6.43470705e-01 -5.21468759e-01 4.75153744e-01
5.49403250e-01 -1.08704078e+00 7.02733815e-01 1.53726369e-01
-3.51964414e-01 -9.47507739e-01 1.44998714e-01 -1.09058428e+00
-3.62020016e-01 -1.95150614e-01 2.85192169e-02 1.12366700e+00
-1.83030307e-01 -8.56295109e-01 9.71820652e-01 -5.93613207e-01
-5.94154060e-01 -3.84281516e-01 -1.15428984e+00 -1.37223673e+00
-9.24642533e-02 -1.64620697e-01 5.77800274e-01 5.26696384e-01
-8.25782418e-01 3.44726533e-01 -5.18235564e-01 1.64334893e-01
8.79469216e-01 1.40647382e-01 9.21563029e-01 -1.54726529e+00
6.34390339e-02 -5.37348688e-01 -1.00772023e+00 -1.45394063e+00
-3.33552748e-01 -7.05471218e-01 2.90945172e-01 -1.84333301e+00
-4.09095585e-02 -4.97718006e-01 -6.67539060e-01 2.66904533e-01
-1.97593033e-01 3.07675928e-01 2.23628715e-01 3.75955850e-01
-1.15052450e+00 8.00740480e-01 1.57050991e+00 -5.45087278e-01
3.56711214e-03 2.26179808e-01 -1.13278307e-01 6.33286655e-01
4.72897619e-01 -6.93027437e-01 -8.26563463e-02 -4.59079802e-01
-5.60550451e-01 -1.02511197e-01 5.03628075e-01 -1.70424581e+00
4.97061282e-01 7.93140847e-03 4.74120915e-01 -1.21569097e+00
5.30613244e-01 -9.42610741e-01 2.51326114e-02 1.33405519e+00
2.53866643e-01 3.00635189e-01 4.40992087e-01 6.89441562e-01
-2.23886073e-01 -1.21967360e-01 8.18506658e-01 -9.56846029e-02
-1.46061122e+00 6.54075325e-01 -5.89242935e-01 -1.85060218e-01
1.40658700e+00 -4.81392622e-01 -6.02656782e-01 2.58317560e-01
-3.72407883e-01 5.69185853e-01 -1.10999011e-02 6.22168243e-01
7.18413115e-01 -1.59838080e+00 -5.89294970e-01 -4.94891480e-02
1.44979164e-01 -7.72778615e-02 8.43361616e-02 1.20516562e+00
-5.49937427e-01 5.99845886e-01 -4.03603077e-01 -1.19420350e+00
-1.40263355e+00 6.33780181e-01 4.66237724e-01 8.92060697e-02
-8.44806492e-01 6.78148150e-01 2.50068814e-01 2.57668942e-02
1.58762217e-01 -2.46785432e-01 -7.68521249e-01 -2.28985488e-01
3.51471841e-01 6.91842020e-01 -1.07333146e-01 -8.67232859e-01
-5.62117279e-01 9.24403846e-01 -2.22797655e-02 3.08986545e-01
9.45671976e-01 8.61195400e-02 1.05925865e-01 6.59257829e-01
7.30577946e-01 -1.46792293e-01 -1.03641808e+00 -5.30780219e-02
6.99146613e-02 -3.42624396e-01 6.41934350e-02 -2.57578909e-01
-1.03816712e+00 1.15547013e+00 1.25627315e+00 4.16565269e-01
6.29572034e-01 -3.22999507e-01 1.24715090e+00 5.11396348e-01
7.40453720e-01 -6.21284544e-01 2.31602162e-01 5.14081836e-01
3.84352148e-01 -1.30346680e+00 -1.46891205e-02 7.31167849e-03
-2.77523398e-01 9.86049950e-01 1.37090480e+00 -4.20683652e-01
5.90756238e-01 4.66596186e-02 2.05862254e-01 -1.68233603e-01
-4.75749224e-01 -4.95767355e-01 1.25904530e-01 5.29236317e-01
-5.34289069e-02 -8.97277966e-02 -4.11921501e-01 4.25339341e-02
1.43884048e-01 1.73274264e-01 1.21193446e-01 8.42806399e-01
-9.72686708e-01 -1.05394387e+00 -6.15201354e-01 4.50163037e-01
-5.86566068e-02 3.11865270e-01 5.50279021e-02 1.09348381e+00
8.61825109e-01 9.04650569e-01 2.44151726e-02 -7.97832310e-01
5.34743726e-01 -4.83038336e-01 3.43402505e-01 -3.32979918e-01
-4.89456177e-01 -3.34466174e-02 -3.93926531e-01 -7.90567324e-02
-6.04564428e-01 -7.08792269e-01 -1.44395423e+00 -3.17552716e-01
-8.30508947e-01 3.09502363e-01 5.20126104e-01 1.23365748e+00
1.84365466e-01 8.97054493e-01 4.68169749e-01 -6.92767262e-01
-3.38080749e-02 -1.11095965e+00 -4.72215027e-01 4.62684780e-02
4.45646912e-01 -1.30800676e+00 -4.41629775e-02 -5.69524825e-01] | [6.46063756942749, -2.103574514389038] |
9f782cd5-dfc6-4f35-a520-0e48649c51ff | learning-spatial-knowledge-for-text-to-3d | null | null | https://aclanthology.org/D14-1217 | https://aclanthology.org/D14-1217.pdf | Learning Spatial Knowledge for Text to 3D Scene Generation | null | ['Angel Chang', 'Manolis Savva', 'Christopher D. Manning'] | 2014-10-01 | null | null | null | emnlp-2014-10 | ['scene-generation', 'text-to-3d'] | ['computer-vision', 'computer-vision'] | [-8.63703638e-02 1.71006292e-01 -6.22772932e-01 -4.08054382e-01
-8.41685571e-03 -9.08429027e-01 6.55310392e-01 -6.53472245e-01
-2.85945535e-01 1.06888819e+00 -4.63127941e-02 -1.01159286e+00
-3.91567826e-01 -9.63214397e-01 -4.95059669e-01 -6.31337762e-01
-9.79754329e-01 7.25764990e-01 3.30370307e-01 -6.93831444e-01
7.03166842e-01 7.88774848e-01 -1.68942046e+00 7.18545914e-01
7.04417467e-01 8.52217197e-01 2.49141872e-01 1.14950800e+00
-1.95044339e-01 1.55633950e+00 -7.48382092e-01 -5.46825826e-01
3.13719302e-01 -1.23176083e-01 -7.22945035e-01 -1.01074085e-01
9.28529128e-02 -8.59008506e-02 -2.09758401e-01 9.22211111e-01
5.37373662e-01 4.49454933e-02 1.08379531e+00 -1.42548037e+00
-5.91619551e-01 6.10313773e-01 -4.01565880e-02 1.21627934e-01
1.03678203e+00 -5.39447069e-01 1.19919395e+00 -1.13026452e+00
7.20913768e-01 1.26888943e+00 8.66221786e-01 5.44149756e-01
-1.22286928e+00 -1.94712028e-01 -3.26822817e-01 -9.51717794e-02
-1.46558487e+00 -3.25250506e-01 4.25783843e-02 -2.08119690e-01
1.66093647e+00 1.26596653e+00 1.20609856e+00 1.01401424e+00
1.26658809e+00 8.34431887e-01 1.04267764e+00 -5.13792276e-01
3.35295945e-01 3.66983831e-01 1.54683650e-01 6.33519173e-01
8.40953708e-01 5.26628852e-01 -7.06372619e-01 -9.13127720e-01
9.33553874e-01 -2.94925272e-01 1.71355158e-01 -5.05680561e-01
-9.05919552e-01 6.91228509e-01 1.78732842e-01 3.83959889e-01
-1.39880210e-01 9.89067405e-02 1.26390755e-01 5.30987144e-01
-2.58292928e-02 6.47037446e-01 -9.11868811e-01 -1.33165747e-01
-8.71728659e-01 5.10332465e-01 1.25398111e+00 1.52653182e+00
1.24482810e-01 2.94908643e-01 -9.34252143e-02 3.17179203e-01
8.92314315e-01 1.01808000e+00 4.28362608e-01 -1.36146402e+00
-6.87414408e-02 1.72361732e-01 5.01781464e-01 -8.52631688e-01
-6.33224547e-01 -9.64177120e-03 -8.93263519e-01 4.49267089e-01
3.49161088e-01 4.57367361e-01 -8.02827001e-01 5.07305264e-01
4.33481112e-02 -2.34125629e-01 4.53833073e-01 5.55570945e-02
4.99930978e-01 3.76208365e-01 -1.34477139e-01 -5.73289394e-01
1.06082785e+00 -1.36716676e+00 -1.35299087e+00 2.33215362e-01
9.05734658e-01 -1.07320261e+00 4.35900748e-01 5.33875942e-01
-1.55548143e+00 -1.37560293e-01 -1.08699942e+00 1.78573877e-01
-7.27255583e-01 -3.14239264e-01 8.57801437e-01 1.43120694e+00
-1.60129595e+00 9.73287821e-01 -4.91727620e-01 6.59165755e-02
1.20568443e-02 8.24621081e-01 -2.64718989e-03 4.62812334e-01
-1.33193445e+00 1.08501506e+00 2.22979754e-01 -1.21242590e-01
-1.65216476e-01 -2.13068098e-01 -8.23704481e-01 -5.63443303e-01
-4.78693932e-01 -5.29636025e-01 1.44139910e+00 -2.59346128e-01
-1.65295815e+00 9.71794367e-01 -1.42069459e-01 -1.97814897e-01
6.14786744e-01 -1.28011424e-02 -8.31891418e-01 2.42498964e-01
-1.89849049e-01 5.76383233e-01 9.28263724e-01 -1.35132408e+00
-7.59897232e-01 -1.67359829e-01 -1.23336017e-01 2.66287565e-01
-1.25510961e-01 1.89734384e-01 2.11616129e-01 -1.12999000e-01
3.27147305e-01 -7.20919967e-01 -2.53068686e-01 -5.32041907e-01
-1.46512717e-01 -7.10518599e-01 7.70373225e-01 -4.51523662e-01
1.83705616e+00 -1.67618537e+00 -2.06720144e-01 3.98590982e-01
3.57815564e-01 -1.24705513e-03 2.40583986e-01 1.08380008e+00
-2.76906848e-01 7.32199550e-01 4.11965609e-01 -1.10722095e-01
2.24991128e-01 4.85861301e-01 -4.16602850e-01 3.05609167e-01
-1.29282743e-01 1.15307164e+00 -1.16605783e+00 -5.23096442e-01
4.11106765e-01 9.42391157e-02 -4.58732933e-01 4.79237735e-01
2.99364805e-01 1.47170946e-01 -3.56553018e-01 1.39399457e+00
1.15709066e+00 -1.31984919e-01 1.45911396e-01 5.30878425e-01
-3.79135728e-01 3.55090618e-01 -6.96863770e-01 1.05554795e+00
7.08333924e-02 5.00173986e-01 1.02364108e-01 -7.94621468e-01
3.33247900e-01 8.61540735e-01 4.37155962e-01 -9.67555881e-01
-2.26398129e-02 6.92409754e-01 1.12803578e-01 -6.07703328e-01
7.58228302e-01 3.94563079e-02 -3.64872098e-01 6.40070081e-01
-2.37588286e-01 -6.59476995e-01 9.64643434e-02 3.08100313e-01
6.22585893e-01 -5.18246442e-02 5.62923312e-01 -1.05613089e+00
7.86340594e-01 -1.82965681e-01 -1.81299388e-01 1.03415680e+00
-3.09923887e-01 3.19085121e-01 2.99841821e-01 -6.65102363e-01
-6.45341039e-01 -1.12307119e+00 -4.89381433e-01 1.30636716e+00
3.24267983e-01 -4.39044595e-01 -9.54439282e-01 -2.49762803e-01
1.77620783e-01 6.89606130e-01 -5.90509653e-01 3.84124845e-01
-5.03739953e-01 -8.32535863e-01 7.39044368e-01 3.45434904e-01
-5.07752821e-02 -1.33414865e+00 -6.58416986e-01 1.25490099e-01
-2.20292807e-01 -6.63697243e-01 -6.23428151e-02 4.48765576e-01
-1.35989368e+00 -5.18594682e-01 -6.66252747e-02 -8.20914626e-01
5.87345481e-01 2.46782884e-01 1.27047324e+00 5.39230824e-01
-2.31483161e-01 4.26904231e-01 -1.21292919e-01 -4.95818377e-01
-4.59671497e-01 -8.00336525e-02 5.28869390e-01 -5.87835789e-01
5.19427478e-01 -2.50617653e-01 -7.29350567e-01 5.37953973e-01
-6.88540697e-01 1.62748516e-01 1.79803044e-01 1.04410267e+00
1.35816500e-01 -9.34035778e-02 1.22507080e-01 -6.38007045e-01
8.72274399e-01 -1.69219792e-01 -3.78732830e-01 5.77745810e-02
-6.77108407e-01 -3.74140263e-01 3.21430594e-01 -3.25342178e-01
-1.01981449e+00 -4.87835288e-01 -9.82677937e-02 2.45538145e-01
1.11353043e-02 -1.46784872e-01 6.47139177e-02 -5.24923325e-01
8.02199244e-01 9.25758183e-02 1.99174434e-02 -6.80815242e-03
3.01039815e-01 7.09525108e-01 -6.82967342e-03 -6.68678164e-01
8.44880998e-01 4.91470337e-01 7.98524171e-02 -9.57177758e-01
-1.52186140e-01 -2.60129690e-01 -9.51962709e-01 -6.54426932e-01
6.56643391e-01 -6.78531289e-01 -9.10833478e-01 3.91110867e-01
-9.38691139e-01 -3.38627815e-01 -3.91645581e-01 4.25431967e-01
-1.01278400e+00 2.75717527e-02 -3.90154392e-01 -1.27895141e+00
-5.10977268e-01 -1.02017939e+00 9.43384409e-01 5.30070923e-02
-5.10597289e-01 -1.26927447e+00 5.87685481e-02 2.71537274e-01
1.81734428e-01 -1.73075795e-01 6.90226793e-01 -2.38256708e-01
-4.24233019e-01 -1.53791070e-01 2.34436691e-02 -1.39755070e-01
1.70832314e-02 4.95917559e-01 -9.81751978e-01 -5.31145096e-01
6.65065646e-02 -1.92070693e-01 -1.08835101e-01 6.52520418e-01
5.91872573e-01 -2.29931593e-01 -8.56000841e-01 5.40386558e-01
1.38545322e+00 3.85070026e-01 5.32770038e-01 7.28214979e-01
1.41836226e-01 5.53460240e-01 9.17806149e-01 4.63203549e-01
1.30579369e-02 3.28798652e-01 2.40537539e-01 1.49327129e-01
1.11720070e-01 -1.54819340e-01 3.77893507e-01 1.16112018e+00
-8.18235934e-01 -2.69281328e-01 -5.07867396e-01 4.42987174e-01
-1.72482407e+00 -1.40330648e+00 -4.32368398e-01 6.90478683e-01
6.25676990e-01 1.56016424e-01 -1.48347050e-01 3.35214496e-01
4.99015123e-01 -2.03574806e-01 -1.19133167e-01 -1.06291151e+00
-1.43546045e-01 3.15233678e-01 7.37729073e-01 1.00061214e+00
-7.20721722e-01 1.03317809e+00 1.29781246e+01 1.02230716e+00
2.21112028e-01 1.03134915e-01 5.16071796e-01 3.48020852e-01
-4.36954498e-01 -4.56139445e-02 -1.04416132e+00 2.72933897e-02
1.38140702e+00 -4.30666685e-01 6.85999811e-01 5.44219851e-01
3.44648361e-01 -4.23268199e-01 -1.26188684e+00 5.26221812e-01
9.73738134e-02 -1.40886843e+00 -2.83300440e-04 6.85225725e-01
7.73699820e-01 -5.08050561e-01 6.22419357e-01 3.24184299e-01
6.09259963e-01 -1.14389277e+00 8.60300779e-01 2.53660440e-01
1.03040910e+00 -6.05088234e-01 5.67372203e-01 1.68872893e-01
-1.14389896e+00 -2.20873043e-01 -8.77727985e-01 -1.00755692e+00
3.93533185e-02 -1.81779593e-01 -4.29956943e-01 3.48861217e-01
9.58353162e-01 2.99398601e-01 -3.93658698e-01 9.95779395e-01
-4.78694476e-02 1.04875881e-02 -2.71853864e-01 -4.48467314e-01
4.83122796e-01 -3.54241252e-01 4.66730654e-01 1.00164843e+00
2.48499006e-01 3.51035744e-01 -9.84472036e-02 4.01770771e-01
5.45058846e-01 3.29446048e-02 -1.19659424e+00 -1.78908288e-01
2.83276141e-01 9.16795909e-01 -4.83487815e-01 -4.22520459e-01
-2.00212970e-01 8.62069130e-01 -3.55488248e-02 5.01107454e-01
-6.11489356e-01 -4.35615242e-01 9.72222984e-01 -1.27327025e-01
-1.14700586e-01 -3.48497719e-01 -6.23769283e-01 -7.30352640e-01
-5.89872956e-01 -4.54965204e-01 5.93606755e-02 -5.53365827e-01
-1.39813089e+00 5.79277515e-01 -2.27688253e-02 -1.40553558e+00
-6.99901402e-01 -1.27676582e+00 -4.76714373e-01 4.92853165e-01
-1.11898029e+00 -1.10984349e+00 2.50124663e-01 4.52870727e-01
1.64141744e-01 -5.34416080e-01 1.39563632e+00 3.57715860e-02
1.00637585e-01 9.24474537e-01 6.69434488e-01 -7.37814724e-01
5.56605101e-01 -1.27867436e+00 5.68737745e-01 -1.37897313e-01
-4.31265175e-01 9.05828118e-01 6.28349900e-01 -5.39804697e-01
-1.41196322e+00 -3.66917729e-01 1.08350635e+00 -9.83769417e-01
6.55218959e-01 -3.86345625e-01 4.23767231e-02 7.88592756e-01
7.15902448e-01 -6.18741751e-01 8.21781039e-01 -1.83753878e-01
1.80774391e-01 5.75296998e-01 -1.39248300e+00 6.12354755e-01
1.66275799e+00 -4.63594139e-01 -6.25784039e-01 7.60327101e-01
8.13696027e-01 -6.94087505e-01 -1.30082703e+00 3.34633321e-01
8.65424156e-01 -8.75409484e-01 1.61978090e+00 -1.32660246e+00
-4.43697497e-02 2.81152606e-01 -2.61993498e-01 -9.32519078e-01
-5.96193194e-01 -1.23518765e+00 -5.33532679e-01 -5.83747849e-02
5.96577883e-01 -1.13057327e+00 3.42365682e-01 8.74560475e-01
-2.82833427e-01 -6.42737269e-01 -1.06996536e+00 -1.32016802e+00
-3.35779637e-02 -1.45572275e-01 4.90409225e-01 7.63798356e-01
6.83744550e-01 1.09839931e-01 -6.36873543e-02 -1.00294888e-01
5.33176839e-01 9.55312885e-03 4.41501856e-01 -1.34294486e+00
3.86843324e-01 -5.75816095e-01 -3.07655483e-01 -9.45992947e-01
-8.85957032e-02 -8.12076271e-01 -6.53862000e-01 -1.28511906e+00
-8.32044985e-03 -1.91056758e-01 -1.11109078e-01 -1.65725678e-01
3.67937148e-01 2.13746816e-01 1.20859891e-02 1.03788137e-01
-3.71160030e-01 6.18435517e-02 1.29639816e+00 6.91750320e-05
-1.62315920e-01 4.85058486e-01 -4.81304944e-01 7.84440815e-01
8.58408585e-02 -2.99253196e-01 -6.78878546e-01 6.11881316e-02
6.69384480e-01 4.61409837e-02 3.24159935e-02 -7.42885649e-01
5.37211418e-01 -3.75702560e-01 4.78586555e-01 -1.32223868e+00
1.30741090e-01 -9.61415648e-01 6.95283338e-02 9.40189242e-01
2.74610907e-01 1.20424610e-02 8.66204947e-02 5.39500564e-02
-1.44871444e-01 -5.70943117e-01 9.21121240e-01 -4.07591403e-01
-4.92852688e-01 -5.20386267e-03 -1.03226590e+00 8.97834301e-02
9.92593169e-01 -7.84614205e-01 -3.59281451e-01 -4.20183957e-01
-8.29068601e-01 -1.95836127e-02 6.50830388e-01 3.03609259e-02
7.30431557e-01 -1.51530886e+00 -2.30721906e-01 7.18729138e-01
-3.22939813e-01 -3.74400020e-01 -1.70157343e-01 6.58265352e-01
-1.32361674e+00 1.02442718e+00 -5.41665435e-01 -4.55340147e-01
-1.14228773e+00 4.78126436e-01 4.28307921e-01 -2.41845414e-01
-2.15481281e-01 1.11954463e+00 2.71224789e-02 -8.23025763e-01
1.85185194e-01 -9.21545625e-02 -7.54407048e-01 3.55081353e-03
6.88606799e-01 1.05194807e+00 -2.91290224e-01 -6.04341030e-01
-4.56784427e-01 6.65885091e-01 2.32151806e-01 -2.87484169e-01
9.17833567e-01 -2.19243199e-01 -9.89108324e-01 4.28274393e-01
8.38715494e-01 -1.36269778e-01 -3.69319022e-02 4.16855574e-01
1.25943512e-01 -8.02164078e-01 -4.32406247e-01 -3.55811834e-01
-1.85641110e-01 5.54822803e-01 5.34874737e-01 8.99602413e-01
8.57008278e-01 -3.02566767e-01 8.18335712e-01 9.66778398e-01
5.72402716e-01 -1.68019545e+00 -2.13140488e-01 6.89524531e-01
9.12339568e-01 -9.28350806e-01 5.44190466e-01 -7.27165341e-01
-4.14997995e-01 1.32979155e+00 4.68304873e-01 -1.55325383e-01
1.27306652e+00 5.74917436e-01 1.14069022e-02 -3.70670199e-01
-9.44949508e-01 1.12705544e-01 3.75366658e-01 1.11147714e+00
5.06513238e-01 5.07374525e-01 -9.66500878e-01 3.21953118e-01
-7.28706717e-01 -2.34555230e-01 4.90474731e-01 1.41972518e+00
-6.43810987e-01 -1.20391917e+00 -7.23931909e-01 4.61561680e-01
-5.49773455e-01 -1.16372630e-01 -5.28106689e-01 8.46754074e-01
-5.76629937e-02 1.49448860e+00 -1.97535474e-03 -5.03491640e-01
4.11356747e-01 1.41089618e-01 7.21762300e-01 -1.23501487e-01
-9.04846430e-01 4.15413082e-01 3.84890139e-01 -1.23056793e+00
-8.58632207e-01 -1.05834293e+00 -1.40667629e+00 -1.19437599e+00
-5.12782812e-01 1.89310342e-01 3.83317530e-01 3.90289724e-01
-2.06836104e-01 2.85260603e-02 9.80917513e-01 -1.07949340e+00
-3.90341938e-01 -9.39418614e-01 -1.00026262e+00 -7.84516707e-02
2.89751232e-01 -8.00943017e-01 -7.83523321e-01 2.83909619e-01] | [-7.271348476409912, 3.734673261642456] |
407e8331-ef3e-4787-ad80-5b50bbabe8cd | learning-spatial-semantic-context-with-fully | 1610.02616 | null | http://arxiv.org/abs/1610.02616v2 | http://arxiv.org/pdf/1610.02616v2.pdf | Learning Spatial-Semantic Context with Fully Convolutional Recurrent Network for Online Handwritten Chinese Text Recognition | Online handwritten Chinese text recognition (OHCTR) is a challenging problem
as it involves a large-scale character set, ambiguous segmentation, and
variable-length input sequences. In this paper, we exploit the outstanding
capability of path signature to translate online pen-tip trajectories into
informative signature feature maps using a sliding window-based method,
successfully capturing the analytic and geometric properties of pen strokes
with strong local invariance and robustness. A multi-spatial-context fully
convolutional recurrent network (MCFCRN) is proposed to exploit the multiple
spatial contexts from the signature feature maps and generate a prediction
sequence while completely avoiding the difficult segmentation problem.
Furthermore, an implicit language model is developed to make predictions based
on semantic context within a predicting feature sequence, providing a new
perspective for incorporating lexicon constraints and prior knowledge about a
certain language in the recognition procedure. Experiments on two standard
benchmarks, Dataset-CASIA and Dataset-ICDAR, yielded outstanding results, with
correct rates of 97.10% and 97.15%, respectively, which are significantly
better than the best result reported thus far in the literature. | ['Zenghui Sun', 'Lianwen Jin', 'Hao Ni', 'Terry Lyons', 'Zecheng Xie'] | 2016-10-09 | null | null | null | null | ['handwritten-chinese-text-recognition', 'handwritten-chinese-text-recognition'] | ['computer-vision', 'natural-language-processing'] | [ 6.04497075e-01 -7.14704931e-01 -1.51114687e-01 -4.09155846e-01
-6.72808766e-01 -9.04010653e-01 3.48859489e-01 -3.35718453e-01
-3.89001161e-01 5.08351207e-01 -4.41617556e-02 -4.11930472e-01
-1.40734985e-01 -5.87351322e-01 -5.69127381e-01 -7.63982058e-01
3.26601475e-01 2.88208961e-01 2.75434107e-01 -1.32870883e-01
8.76612425e-01 9.08622026e-01 -1.09157658e+00 4.90376055e-01
1.07142377e+00 9.74427223e-01 5.14524937e-01 8.23106229e-01
-4.26963627e-01 7.89563119e-01 -6.43525779e-01 -4.87625539e-01
1.04039676e-01 -5.10823071e-01 -4.43656772e-01 2.29871467e-01
2.61107594e-01 -2.74243236e-01 -5.66561520e-01 8.17345858e-01
2.99179703e-01 1.33937716e-01 6.71270072e-01 -5.85033774e-01
-1.01652980e+00 6.38705611e-01 -4.40699399e-01 -1.61568031e-01
2.63280302e-01 3.57919224e-02 9.16032970e-01 -1.07785428e+00
8.82044375e-01 8.32240164e-01 5.93704224e-01 4.03826892e-01
-5.75471520e-01 -5.92205048e-01 3.33753169e-01 3.32079351e-01
-1.38749397e+00 -9.79424715e-02 1.04085684e+00 -2.68555135e-01
8.51114631e-01 3.90622854e-01 3.24841559e-01 1.27184141e+00
-5.72308637e-02 1.29349256e+00 9.57961977e-01 -5.59930861e-01
-1.79906636e-02 -7.94811174e-02 3.63646209e-01 7.31042981e-01
-1.28706887e-01 -2.25919396e-01 -4.62349981e-01 2.93424904e-01
1.03648436e+00 2.83227324e-01 -1.90324008e-01 -7.59527786e-03
-1.17434263e+00 4.62546289e-01 3.92643541e-01 4.16711956e-01
-1.95864975e-01 -1.78466588e-01 4.54856455e-01 -5.97026898e-03
-8.55325088e-02 3.54030311e-01 -2.81327099e-01 -4.28858995e-01
-9.84240115e-01 -5.17996699e-02 7.15557694e-01 1.38818133e+00
3.19710404e-01 3.80161792e-01 -3.85076016e-01 9.89349484e-01
4.03506085e-02 8.90823603e-01 6.58099949e-01 -4.71860319e-01
8.78017068e-01 6.72149420e-01 8.24775696e-02 -1.18623722e+00
-1.27033323e-01 -3.17138731e-01 -8.68769526e-01 -2.63819933e-01
5.83197415e-01 -1.77870020e-02 -1.15023983e+00 1.06436920e+00
-2.74804801e-01 -3.23847756e-02 9.46612433e-02 9.48455930e-01
3.66600633e-01 7.88169503e-01 -3.92759830e-01 3.29961479e-02
9.99043167e-01 -1.16109598e+00 -7.90517449e-01 -1.81244716e-01
4.49550718e-01 -9.25334036e-01 1.17002022e+00 6.71359837e-01
-7.05544233e-01 -5.24092674e-01 -1.21634793e+00 -1.10270679e-01
-5.41420281e-01 8.99615049e-01 5.64714968e-01 6.03261650e-01
-5.04218161e-01 6.64097965e-01 -6.75978422e-01 1.45546813e-02
5.43053329e-01 9.01089162e-02 -3.83894183e-02 -2.07699254e-01
-8.21183026e-01 4.04826581e-01 3.74982029e-01 6.37030363e-01
-4.17275846e-01 -2.66611993e-01 -5.29529810e-01 -1.46437988e-01
3.88118267e-01 2.55769551e-01 8.25320184e-01 -1.05223608e+00
-2.07528830e+00 2.30659857e-01 -2.68896580e-01 -2.11158305e-01
9.33201313e-01 -4.00952280e-01 -5.73209703e-01 2.99549490e-01
-3.14925164e-01 1.70765191e-01 9.73720491e-01 -8.09895158e-01
-4.66063589e-01 -3.94155294e-01 -6.27678990e-01 1.02984287e-01
-3.24715316e-01 6.28281981e-02 -1.04077613e+00 -1.10341692e+00
2.88023949e-01 -7.79896498e-01 -1.86055064e-01 -1.57145143e-01
-6.72428310e-01 -4.69770841e-02 9.54897225e-01 -1.11483121e+00
1.18396556e+00 -2.05454135e+00 1.28533021e-01 4.66907591e-01
-4.31813240e-01 5.99916518e-01 -2.93354690e-01 3.14328402e-01
4.24150884e-01 -2.93557942e-02 -3.42481583e-01 -1.19090423e-01
2.37881299e-02 2.32351273e-02 -9.02435005e-01 3.79930884e-01
4.55233574e-01 1.11289454e+00 -6.97097540e-01 -2.06150949e-01
1.48413643e-01 3.60959888e-01 -2.34978274e-01 7.13998228e-02
-4.40426737e-01 1.48916990e-01 -6.90547466e-01 1.00532341e+00
7.11926877e-01 -8.75577480e-02 3.87779623e-01 8.04768726e-02
-3.27691346e-01 -1.04062259e-01 -9.99932468e-01 1.54547107e+00
-3.32097501e-01 8.72645259e-01 -3.97747993e-01 -8.38859677e-01
1.48774743e+00 -2.39146873e-01 -2.14751679e-02 -1.09552968e+00
1.34461550e-02 4.58082646e-01 -2.48958319e-01 -3.25280011e-01
9.36719179e-01 4.91429657e-01 -1.05095766e-01 3.45037758e-01
-3.93442363e-01 1.57174870e-01 1.56679563e-02 -1.33585379e-01
7.00926602e-01 5.01934528e-01 -2.66644686e-01 4.71045423e-05
6.29385114e-01 -1.40304402e-01 5.20625770e-01 1.03507960e+00
1.41867742e-01 9.61763859e-01 3.82677078e-01 -2.86783278e-01
-1.11704135e+00 -7.05521286e-01 -1.20187975e-01 8.09626162e-01
3.03770661e-01 -1.03813313e-01 -6.20262384e-01 -4.94519413e-01
-1.45789132e-01 5.47461748e-01 -4.27187979e-01 1.58297360e-01
-1.02748466e+00 -4.81590241e-01 8.96086693e-01 1.09830272e+00
7.01636255e-01 -1.04082990e+00 -5.10021627e-01 4.49834585e-01
1.56177610e-01 -1.37261295e+00 -7.66039908e-01 3.57427523e-02
-7.87795365e-01 -9.22195673e-01 -1.17786932e+00 -9.81811941e-01
7.06165612e-01 6.10476993e-02 3.87206137e-01 -1.74418949e-02
-5.29098868e-01 -6.36002421e-02 -6.68409526e-01 3.72581393e-03
-6.55503646e-02 1.43354356e-01 -3.47668409e-01 2.44449034e-01
1.13186106e-01 -1.76938493e-02 -2.77312219e-01 5.59045672e-01
-7.19472945e-01 1.80213839e-01 9.25698757e-01 1.24741340e+00
7.79485345e-01 -1.60319999e-01 5.49193561e-01 -8.81337225e-01
4.35098827e-01 1.61390975e-02 -8.39610994e-01 7.56997705e-01
-3.74690711e-01 -4.63238284e-02 1.15183890e+00 -6.92623556e-01
-1.32488680e+00 2.66731352e-01 9.43366345e-03 -3.35093230e-01
-3.35096009e-02 4.07562673e-01 -3.54701698e-01 -1.26435116e-01
2.37222597e-01 9.92101371e-01 -1.95781052e-01 -5.81853747e-01
2.80634552e-01 8.94788504e-01 8.18345606e-01 -6.99501753e-01
5.50193608e-01 3.33301008e-01 -2.46982634e-01 -1.08164644e+00
-4.52984869e-01 -3.22422266e-01 -1.01450670e+00 6.81489781e-02
6.51364982e-01 -5.56557655e-01 -6.04114890e-01 1.26694775e+00
-1.24222529e+00 -5.03091395e-01 2.64801055e-01 1.61254391e-01
-2.93908596e-01 7.21296072e-01 -7.39749730e-01 -8.66949141e-01
-3.96373451e-01 -1.18907952e+00 8.82024348e-01 5.26757598e-01
3.66442055e-01 -8.38478029e-01 -4.90056425e-01 1.97778225e-01
4.37916666e-01 1.32462412e-01 7.60748446e-01 -7.68840730e-01
-1.09847772e+00 -6.52947545e-01 -5.67734599e-01 3.43151838e-01
1.01465598e-01 4.26421762e-01 -6.79526806e-01 -7.61897787e-02
-4.30341095e-01 -2.13313207e-01 8.36543381e-01 -9.95066017e-02
1.61744702e+00 -2.09958598e-01 -1.78093553e-01 7.62620091e-01
1.32500994e+00 6.09624088e-01 8.56383145e-01 1.29186749e-01
9.06427622e-01 1.07928693e-01 6.79246902e-01 6.67274773e-01
6.97718933e-02 7.07011580e-01 -2.40952194e-01 1.79898947e-01
-4.44374681e-02 -5.25859416e-01 3.92420590e-01 1.01793301e+00
7.54916817e-02 -3.21045280e-01 -9.85512495e-01 2.49069110e-01
-1.74313581e+00 -7.50708461e-01 -2.61437073e-02 2.02581024e+00
7.56591678e-01 2.23206565e-01 -1.59321547e-01 2.23588534e-02
8.47584605e-01 1.85907334e-01 -8.50613832e-01 -4.11568701e-01
-6.01632893e-01 1.93961546e-01 7.17704713e-01 3.91796082e-01
-1.12349594e+00 1.34329772e+00 5.47697830e+00 1.16529346e+00
-1.37107408e+00 -6.59843922e-01 6.11650050e-01 4.51488674e-01
-1.98696926e-01 -2.32347652e-01 -1.06141925e+00 5.61987758e-01
5.59665680e-01 6.31492808e-02 6.02361321e-01 6.96643293e-01
-7.77333044e-03 2.32460126e-01 -8.05993199e-01 9.36854959e-01
3.22306633e-01 -1.52566099e+00 2.98979729e-01 -1.23363018e-01
7.55092144e-01 -2.66050100e-01 2.48677269e-01 1.33289859e-01
8.61367062e-02 -1.16397548e+00 8.63533497e-01 1.06135702e+00
1.06257093e+00 -8.24017942e-01 4.76358563e-01 3.30041707e-01
-1.20961893e+00 -1.69864923e-01 -4.14330840e-01 2.07880691e-01
-6.74933791e-02 2.53688663e-01 -6.81478560e-01 7.29191542e-01
1.87106088e-01 1.10823750e+00 -7.17622638e-01 1.04892051e+00
-3.00113082e-01 6.10624135e-01 -1.67764172e-01 -6.46589875e-01
4.50471073e-01 -4.30411011e-01 3.00432444e-01 1.58237660e+00
3.24288905e-01 1.18595727e-01 -4.22229543e-02 9.76610005e-01
-1.05002038e-01 1.23737872e-01 -2.76631355e-01 -3.91224146e-01
5.83355069e-01 9.63777244e-01 -1.03430033e+00 -1.69980466e-01
-5.39530218e-02 1.50390613e+00 1.81669310e-01 4.93756413e-01
-8.27726483e-01 -1.07948077e+00 4.93473150e-02 -6.75081134e-01
7.96243072e-01 -5.06789625e-01 -6.87466979e-01 -1.23333097e+00
3.53419483e-01 -8.89452457e-01 2.40306761e-02 -5.78373790e-01
-1.13093793e+00 5.08953512e-01 -7.49459743e-01 -1.23284101e+00
-1.07143514e-01 -1.14244425e+00 -6.88822031e-01 9.66100216e-01
-1.45712936e+00 -1.32733977e+00 -1.91000119e-01 4.77028459e-01
8.68714273e-01 -4.21345323e-01 7.09764898e-01 1.17465459e-01
-9.89793181e-01 1.22318077e+00 7.48732865e-01 6.95982635e-01
4.50245887e-01 -1.09245396e+00 5.78870237e-01 1.03411305e+00
2.13656843e-01 6.42376184e-01 2.84651890e-02 -7.69716322e-01
-2.00875640e+00 -1.05994320e+00 5.51623166e-01 -3.40228319e-01
6.50931239e-01 -6.16939485e-01 -1.11676037e+00 4.50698227e-01
-3.27872902e-01 -8.78568515e-02 4.03399259e-01 -3.68548453e-01
-5.73591769e-01 3.32479440e-02 -6.78642809e-01 7.15584755e-01
1.00645661e+00 -7.28056490e-01 -2.93971241e-01 1.49565473e-01
4.29751426e-01 -8.41890395e-01 -5.11506021e-01 1.56153843e-01
8.19397390e-01 -5.33416390e-01 7.61599243e-01 -4.61030066e-01
5.77046692e-01 -1.68774307e-01 -2.29496688e-01 -6.49689436e-01
-6.88263550e-02 -7.05569327e-01 8.72696564e-03 1.34596944e+00
5.57618737e-01 -4.44117486e-01 8.45062554e-01 5.54143906e-01
-2.73590267e-01 -8.50241601e-01 -6.82819188e-01 -9.56631064e-01
2.22730383e-01 -6.12668395e-01 6.10400736e-01 7.87807941e-01
-1.48025960e-01 -2.91321337e-01 -4.91098732e-01 1.33396536e-01
4.46997315e-01 5.82188070e-01 4.79911238e-01 -6.61104620e-01
-1.96197197e-01 -7.04167306e-01 -3.09907645e-01 -1.64328015e+00
2.99004436e-01 -7.82734990e-01 1.25946388e-01 -9.72089291e-01
-4.95044626e-02 -5.33261299e-01 -1.40553311e-01 4.69413131e-01
-5.04627228e-02 6.29009679e-02 3.50066572e-01 2.33215094e-01
-6.05430424e-01 6.33099258e-01 1.25983977e+00 -2.34103754e-01
-2.50537813e-01 1.49550274e-01 -3.85448545e-01 4.89649713e-01
5.57763457e-01 1.11921020e-01 -1.23179607e-01 -6.09088898e-01
-1.32202551e-01 2.27946833e-01 1.81213513e-01 -7.11295307e-01
6.11340225e-01 -2.43031546e-01 7.43258059e-01 -8.98258567e-01
1.70895010e-01 -5.58456361e-01 -3.76565546e-01 2.06757009e-01
-5.64538598e-01 -8.21793526e-02 2.39251405e-01 6.94640577e-01
-1.46987706e-01 -3.38799119e-01 5.42184293e-01 1.81850359e-01
-1.17996991e+00 1.64380848e-01 -2.61032850e-01 -2.58296654e-02
7.59045899e-01 -5.97373188e-01 -2.71089822e-01 -2.08422933e-02
-2.80422032e-01 9.21314657e-02 3.45765948e-01 6.18682325e-01
1.13377190e+00 -1.18683636e+00 -5.29281676e-01 4.48814154e-01
6.13684207e-03 -1.92738429e-01 3.70609045e-01 3.79637659e-01
-8.19616735e-01 7.41067111e-01 -1.23983786e-01 -3.88543427e-01
-9.40999746e-01 1.81103706e-01 2.48316929e-01 -9.06222910e-02
-1.04993796e+00 7.46703982e-01 -4.54077065e-01 -4.52083707e-01
4.44254071e-01 -3.52175653e-01 -6.55098930e-02 -1.16440408e-01
6.65187776e-01 3.98499519e-01 3.51147552e-04 -4.44825113e-01
-2.74364352e-01 8.34903896e-01 -4.41768706e-01 2.54045124e-03
1.30882418e+00 1.02606580e-01 1.11177891e-01 3.29086274e-01
1.07675123e+00 2.76208576e-02 -1.62458825e+00 -4.14003372e-01
4.52703983e-01 -6.42192066e-01 -4.01384741e-01 -1.17092204e+00
-9.66643214e-01 1.03007734e+00 2.95084924e-01 -3.18506598e-01
8.69712472e-01 -5.59634566e-01 8.81709158e-01 8.10806215e-01
3.14905256e-01 -1.47002506e+00 3.06178123e-01 1.07210779e+00
9.75858390e-01 -9.08691108e-01 -2.52818078e-01 -2.14924440e-01
-1.00079405e+00 1.74484324e+00 4.70963508e-01 -1.82251081e-01
2.17155695e-01 1.33681521e-01 -4.58871834e-02 2.70887256e-01
-2.77358949e-01 2.74999708e-01 3.36920947e-01 3.35880935e-01
2.83644527e-01 1.56376749e-01 5.78496978e-02 1.12106812e+00
-3.96145992e-02 4.67146374e-02 5.81692815e-01 9.79404032e-01
-1.82996288e-01 -1.04126823e+00 -1.31905109e-01 3.62955183e-01
-2.23688096e-01 -9.36827660e-02 -6.30745113e-01 5.55344939e-01
-4.00892794e-01 5.27571857e-01 -3.77945080e-02 -3.98512423e-01
2.84622908e-01 4.71962318e-02 3.57358575e-01 -1.26495346e-01
-2.53938466e-01 9.94583145e-02 -2.94012934e-01 -3.54927599e-01
2.61259794e-01 -7.80774653e-01 -1.48801816e+00 9.22846347e-02
-4.75007206e-01 -1.77715525e-01 7.70236373e-01 1.01214957e+00
3.66283238e-01 4.11484748e-01 9.48881090e-01 -6.91035986e-01
-8.09350431e-01 -6.53875530e-01 -6.09099090e-01 2.45995745e-01
1.29835680e-01 -2.46155128e-01 -9.62517112e-02 3.59898470e-02] | [11.962557792663574, 2.331563711166382] |
4e79ad53-3094-4916-9475-35668f1e8d93 | transvos-video-object-segmentation-with | 2106.00588 | null | https://arxiv.org/abs/2106.00588v2 | https://arxiv.org/pdf/2106.00588v2.pdf | TransVOS: Video Object Segmentation with Transformers | Recently, Space-Time Memory Network (STM) based methods have achieved state-of-the-art performance in semi-supervised video object segmentation (VOS). A crucial problem in this task is how to model the dependency both among different frames and inside every frame. However, most of these methods neglect the spatial relationships (inside each frame) and do not make full use of the temporal relationships (among different frames). In this paper, we propose a new transformer-based framework, termed TransVOS, introducing a vision transformer to fully exploit and model both the temporal and spatial relationships. Moreover, most STM-based approaches employ two separate encoders to extract features of two significant inputs, i.e., reference sets (history frames with predicted masks) and query frame (current frame), respectively, increasing the models' parameters and complexity. To slim the popular two-encoder pipeline while keeping the effectiveness, we design a single two-path feature extractor to encode the above two inputs in a unified way. Extensive experiments demonstrate the superiority of our TransVOS over state-of-the-art methods on both DAVIS and YouTube-VOS datasets. | ['Yong liu', 'Yi Yuan', 'Yeneng Lin', 'Mengmeng Wang', 'Jianbiao Mei'] | 2021-06-01 | null | null | null | null | ['one-shot-visual-object-segmentation'] | ['computer-vision'] | [ 1.32890597e-01 -4.89942044e-01 -3.13537270e-01 -3.80424470e-01
-3.54311138e-01 -3.61412972e-01 4.80920643e-01 -1.09102100e-01
-6.62810504e-01 3.10257733e-01 -2.13307157e-01 -1.44152015e-01
2.35814437e-01 -6.75419271e-01 -7.44096816e-01 -5.02265811e-01
3.08160514e-01 1.31624127e-02 1.10742664e+00 8.83064643e-02
1.03017479e-01 3.38045478e-01 -1.57905149e+00 3.33823204e-01
7.62896061e-01 1.19646728e+00 6.96877956e-01 4.58214581e-01
-3.20067167e-01 8.98508310e-01 -3.76131356e-01 -2.06917524e-01
1.84302583e-01 -4.51432705e-01 -8.09344947e-01 2.67353326e-01
3.63066316e-01 -6.04223967e-01 -7.40953505e-01 9.09239531e-01
2.02234760e-01 1.48492724e-01 2.35956952e-01 -1.27460814e+00
-4.02742594e-01 2.65122294e-01 -6.56264067e-01 5.24389029e-01
-8.25772621e-03 3.43383998e-01 7.85310864e-01 -7.81922817e-01
7.29694009e-01 9.85496163e-01 3.79706860e-01 4.11697417e-01
-9.91540670e-01 -5.90598881e-01 4.48512197e-01 6.97967649e-01
-1.28701639e+00 -5.65679550e-01 8.52825701e-01 -4.69213903e-01
8.84943902e-01 4.30262126e-02 8.23998749e-01 7.24598110e-01
3.28371041e-02 1.23945034e+00 8.88931394e-01 -1.24742657e-01
6.98089525e-02 -3.21394205e-02 1.57192558e-01 7.88070560e-01
-1.57453239e-01 -7.29672611e-02 -6.56334519e-01 3.95415932e-01
1.03154731e+00 2.71088600e-01 -4.41300243e-01 -5.55460691e-01
-1.22214580e+00 5.85161150e-01 3.87332678e-01 4.66101348e-01
-3.03065091e-01 1.21366523e-01 4.71153110e-01 5.14601730e-02
3.00833791e-01 -1.94923550e-01 -4.85869557e-01 -1.70571923e-01
-1.27849674e+00 -8.47947970e-02 3.89912784e-01 1.09885561e+00
8.41216266e-01 -4.54583205e-02 -3.67854744e-01 6.22757554e-01
3.06660950e-01 1.94968760e-01 5.71706057e-01 -7.94754684e-01
5.60443878e-01 5.97286403e-01 -1.70309115e-02 -1.03587294e+00
-2.25228727e-01 -4.22877610e-01 -7.80518413e-01 -1.78536355e-01
3.58907968e-01 1.64886534e-01 -1.20314169e+00 1.62662280e+00
4.64139909e-01 6.67098463e-01 -3.58976908e-02 1.02075613e+00
8.73122871e-01 7.62775004e-01 -9.00264382e-02 -4.10003662e-01
1.25513005e+00 -1.41298735e+00 -7.01198757e-01 -3.38443130e-01
4.84382063e-01 -6.64538801e-01 8.12653840e-01 1.55580208e-01
-1.22057438e+00 -8.97397459e-01 -1.08583105e+00 -2.57626534e-01
-1.87665224e-01 1.84211180e-01 4.60236847e-01 2.23292142e-01
-9.13185537e-01 7.26944566e-01 -1.24344933e+00 -7.75944665e-02
4.20524567e-01 4.03362334e-01 -2.27332041e-01 -1.23636380e-01
-1.01917672e+00 5.26085019e-01 4.28016752e-01 2.95027941e-01
-1.04018748e+00 -4.57442403e-01 -8.84555459e-01 7.80410469e-02
7.09026873e-01 -5.26867926e-01 1.18483305e+00 -9.66637433e-01
-1.37733746e+00 4.45794523e-01 -6.06743336e-01 -4.55376476e-01
5.38838804e-01 -1.86502889e-01 -2.25198209e-01 4.29500043e-01
2.86990888e-02 7.79342055e-01 9.92637098e-01 -1.02535355e+00
-8.55485201e-01 -2.59774715e-01 2.24348664e-01 1.25026077e-01
-4.80625629e-01 1.10641934e-01 -1.15856516e+00 -4.57327306e-01
2.78394192e-01 -7.68003583e-01 -2.62676049e-02 2.09385157e-01
-3.42896909e-01 -2.68275738e-01 1.13353157e+00 -6.80636346e-01
1.30890071e+00 -2.23481297e+00 3.21438015e-01 -2.76767164e-01
2.78149426e-01 8.20693016e-01 -1.93559498e-01 1.01152353e-01
2.61328340e-01 -7.00567067e-02 -1.95071623e-01 -6.61692619e-01
-2.93467551e-01 4.40712988e-01 -1.41935438e-01 4.36828732e-01
2.98283935e-01 1.00316894e+00 -8.51586044e-01 -9.70328987e-01
5.67645311e-01 6.30632579e-01 -3.74776423e-01 2.78813183e-01
-3.18382800e-01 6.51199639e-01 -4.58771646e-01 4.23032850e-01
6.88033104e-01 -3.36569220e-01 6.65466338e-02 -3.76729935e-01
-3.63489389e-01 3.28888267e-01 -1.07301366e+00 1.98460889e+00
-2.55287677e-01 6.56378567e-01 -1.39747083e-01 -8.48370612e-01
6.58387244e-01 3.06722283e-01 4.66051579e-01 -9.03955162e-01
1.89968243e-01 1.40938565e-01 -1.50269777e-01 -5.44637322e-01
3.88756841e-01 1.48947164e-01 4.81971711e-01 1.77206576e-01
3.11486840e-01 3.62117171e-01 4.65092987e-01 1.54530853e-01
7.95928419e-01 4.87496167e-01 1.47061780e-01 9.34745446e-02
7.96444774e-01 -2.62199342e-01 1.01865041e+00 4.30006653e-01
-4.56745535e-01 7.90845692e-01 4.48776066e-01 -4.71453369e-01
-8.09122443e-01 -6.91894293e-01 9.45535377e-02 8.32399905e-01
6.01101816e-01 -6.93503320e-01 -6.76452875e-01 -7.48393595e-01
-2.97525227e-01 3.54480386e-01 -5.10238230e-01 9.79198143e-02
-8.29813719e-01 -3.04732442e-01 3.29951257e-01 6.85339630e-01
7.81493008e-01 -9.38060820e-01 -1.15003157e+00 3.07840168e-01
-5.01755536e-01 -1.63402426e+00 -7.58962989e-01 -5.94265983e-02
-8.82998049e-01 -1.01228750e+00 -7.55240083e-01 -7.19256282e-01
4.00898576e-01 7.47411728e-01 7.74240494e-01 1.41456306e-01
-9.98940393e-02 1.75636616e-02 -4.00852472e-01 -1.80651415e-02
2.46494263e-02 6.22468479e-02 -3.69391859e-01 2.94941664e-01
1.47454455e-01 -5.39648831e-01 -6.72541916e-01 6.34537697e-01
-9.55592155e-01 6.12194121e-01 5.66535234e-01 7.69976556e-01
8.40499282e-01 3.40371169e-02 1.32705376e-01 -5.59396982e-01
-1.74747795e-01 -2.62573719e-01 -6.32456481e-01 3.58900905e-01
-2.92095572e-01 2.87224203e-02 6.54182911e-01 -4.52407658e-01
-8.86371434e-01 1.83767587e-01 4.47907485e-02 -8.78034055e-01
-4.79878373e-02 3.98686886e-01 -2.88329452e-01 -6.45364001e-02
-1.15151510e-01 7.41908729e-01 -1.35751218e-01 -5.91249704e-01
1.62182316e-01 5.86802363e-01 6.44928277e-01 -1.80084541e-01
6.62172735e-01 6.70580685e-01 -1.97085038e-01 -7.07498312e-01
-8.27376604e-01 -6.38133645e-01 -1.00586224e+00 -2.18621328e-01
1.02116120e+00 -9.08213377e-01 -5.79442382e-01 7.28043258e-01
-1.40649009e+00 -2.42114514e-01 -1.12616144e-01 3.86511415e-01
-4.37895119e-01 5.69505811e-01 -7.23490834e-01 -6.64370179e-01
-1.25848666e-01 -1.50168979e+00 1.05763149e+00 5.09509385e-01
3.06130022e-01 -7.84371734e-01 -2.92670220e-01 3.35567594e-01
3.17657411e-01 1.05524831e-01 5.40839016e-01 -3.00785601e-01
-9.67204988e-01 2.23543227e-01 -5.40177584e-01 3.51013750e-01
1.54130518e-01 -4.99333888e-02 -8.60332310e-01 -2.35354304e-01
1.66171730e-01 2.28113905e-02 1.13617098e+00 2.74387836e-01
1.13513362e+00 -4.59813476e-02 -3.40588480e-01 8.57248902e-01
1.30907893e+00 3.51821065e-01 5.51492810e-01 2.93238133e-01
1.06844842e+00 4.74339724e-01 7.25900054e-01 3.11696142e-01
6.57462776e-01 8.84587348e-01 4.67191488e-01 -7.99920633e-02
-2.29235306e-01 -2.79003769e-01 4.12728548e-01 9.31560993e-01
-4.06102315e-02 -2.89896429e-01 -6.63492918e-01 6.36458278e-01
-2.16354179e+00 -7.76944876e-01 -1.40430808e-01 2.16117740e+00
6.68483198e-01 2.02827930e-01 1.84346624e-02 1.51098594e-01
7.64914513e-01 5.87284088e-01 -7.85756528e-01 8.69237334e-02
-9.66013968e-02 -1.08845174e-01 4.80895638e-01 2.01909378e-01
-1.23537207e+00 1.09975934e+00 5.48014021e+00 8.73116612e-01
-1.36130404e+00 2.53525764e-01 5.13493299e-01 -1.78579241e-01
-5.36350720e-02 1.81458771e-01 -7.71019816e-01 5.96272111e-01
7.17570603e-01 1.33219182e-01 3.35459054e-01 6.35448456e-01
1.99520513e-01 -2.27226108e-01 -1.12397516e+00 1.14518118e+00
4.41428386e-02 -1.30445933e+00 1.83550678e-02 -1.00039564e-01
5.70503414e-01 9.83109102e-02 -1.26815841e-01 7.91208372e-02
-3.36663604e-01 -7.83372998e-01 1.10498619e+00 4.32119846e-01
7.91159272e-01 -5.54860115e-01 6.73157275e-01 4.22578365e-01
-1.66281593e+00 6.82930201e-02 -2.87865549e-01 -1.35436316e-03
4.72011745e-01 4.09401059e-01 -3.28279585e-01 8.26584697e-01
7.15705633e-01 1.13313425e+00 -6.94187164e-01 9.81474161e-01
-1.94480270e-01 5.34485161e-01 -3.23926866e-01 3.48042518e-01
4.52230006e-01 -4.31792699e-02 4.19128358e-01 1.14517987e+00
1.17779143e-01 7.22293109e-02 2.86114484e-01 7.35971630e-01
1.29612669e-01 -1.85869336e-01 -1.64757043e-01 -4.07528393e-02
3.48318458e-01 1.08870196e+00 -8.98064375e-01 -5.35050035e-01
-7.37673581e-01 1.14491916e+00 1.65328711e-01 3.86859268e-01
-9.98732209e-01 -2.18566060e-01 5.75041711e-01 1.50174052e-01
9.14776921e-01 -4.83538836e-01 -6.48758654e-03 -1.38148332e+00
3.33027720e-01 -5.10011673e-01 1.72867194e-01 -5.28381884e-01
-7.85224259e-01 6.07344985e-01 -4.89813136e-03 -1.37001562e+00
-1.55861571e-01 -3.47017050e-01 -4.01806861e-01 6.95230424e-01
-1.98814607e+00 -1.29355109e+00 -3.86787295e-01 8.18513215e-01
7.97411561e-01 1.54278234e-01 2.65384138e-01 5.31526148e-01
-8.55225801e-01 3.69440615e-01 -1.56420276e-01 2.95345843e-01
4.94975895e-01 -8.98916066e-01 5.08318186e-01 1.14849460e+00
3.22926939e-01 4.62542027e-01 3.15911323e-01 -4.60583001e-01
-1.48236513e+00 -1.11823010e+00 8.12253773e-01 -7.79552460e-02
3.97806495e-01 -2.67732710e-01 -1.07893145e+00 6.29667163e-01
8.19666088e-02 5.47910750e-01 2.46815771e-01 -4.03765231e-01
-2.81763941e-01 -1.79909393e-01 -7.63541758e-01 4.56246912e-01
1.15079129e+00 -6.75781548e-01 -4.18400913e-01 -1.15981944e-01
9.25538957e-01 -5.37133396e-01 -5.00384927e-01 4.60083961e-01
5.92192769e-01 -1.37910414e+00 8.35395992e-01 -2.35306457e-01
4.31564540e-01 -7.45871484e-01 -5.14894836e-02 -8.05010140e-01
-1.41779467e-01 -6.40269399e-01 -4.64465380e-01 1.28550613e+00
-1.50070012e-01 -4.93442357e-01 6.56440794e-01 3.68813097e-01
-2.06553563e-01 -1.00579524e+00 -1.06416488e+00 -6.94738686e-01
-4.29864287e-01 -5.44650435e-01 5.59146106e-01 7.58670926e-01
-3.74919683e-01 2.99940169e-01 -5.86165309e-01 5.55860773e-02
4.01381493e-01 2.67560869e-01 6.83793426e-01 -1.05561936e+00
-4.15850610e-01 -2.26994663e-01 -6.63811982e-01 -1.52047741e+00
-6.42694309e-02 -5.46351731e-01 1.08925313e-01 -1.59460032e+00
2.96590835e-01 -2.86571175e-01 -5.90884268e-01 4.70720500e-01
-2.95781553e-01 1.52067363e-01 6.11994624e-01 5.05997717e-01
-9.49428141e-01 6.93743169e-01 1.35849214e+00 -2.10165456e-02
-2.64599413e-01 -2.62735426e-01 -2.76614904e-01 8.62537801e-01
4.94021416e-01 -4.27063167e-01 -5.64350724e-01 -8.99044156e-01
-4.06262308e-01 2.66107500e-01 5.09971082e-01 -1.18004477e+00
6.01833820e-01 -2.48485178e-01 3.02682042e-01 -8.57908726e-01
4.84108329e-01 -7.72402763e-01 7.75037855e-02 3.19775581e-01
-7.53796846e-02 8.78467932e-02 1.36272937e-01 6.36406720e-01
-5.14420152e-01 -1.03722848e-01 8.18975210e-01 -3.35099995e-02
-1.07912064e+00 6.03923321e-01 3.41056436e-02 -1.02063648e-01
1.19453812e+00 -3.06340694e-01 -1.79935247e-01 -9.26859751e-02
-4.21017319e-01 3.85159612e-01 5.30890584e-01 4.84781861e-01
8.20318043e-01 -1.04130661e+00 -3.61654639e-01 2.74096698e-01
-3.63425240e-02 4.62455601e-01 6.15299165e-01 1.06229568e+00
-3.76176655e-01 5.47870219e-01 -2.54664123e-01 -8.72660995e-01
-1.26189446e+00 6.55169368e-01 1.87774971e-01 -3.23541284e-01
-7.45734692e-01 8.10056865e-01 5.67673028e-01 1.62986621e-01
2.27347180e-01 -4.33266461e-01 -3.26311499e-01 5.13550192e-02
6.92405045e-01 1.72071770e-01 -1.56868786e-01 -8.74882758e-01
-3.16657037e-01 7.27501512e-01 -2.34965637e-01 -2.57438943e-02
1.21544313e+00 -3.37354392e-01 -6.93734065e-02 5.94225407e-01
1.20757926e+00 -3.99490565e-01 -1.79548967e+00 -6.14449203e-01
-1.65769309e-02 -7.30620801e-01 2.12504014e-01 -2.78185338e-01
-1.47657752e+00 1.20505643e+00 4.67338860e-01 -5.41095845e-02
1.34554100e+00 -2.49292254e-01 1.29786432e+00 -2.69519221e-02
4.48888928e-01 -1.01314557e+00 -1.28287245e-02 3.83138001e-01
3.49859834e-01 -1.08982229e+00 -1.24076188e-01 -6.80119336e-01
-6.66528046e-01 1.14971304e+00 6.47952259e-01 1.14368506e-01
4.22894180e-01 1.13813967e-01 -1.25550255e-01 1.20814480e-01
-8.10799599e-01 -3.87824893e-01 3.50047916e-01 4.02518630e-01
2.53057450e-01 -2.44449288e-01 -2.36647695e-01 4.54586297e-01
4.15640980e-01 3.21052939e-01 3.54612350e-01 1.02164507e+00
-2.90102839e-01 -1.14341545e+00 -4.46561463e-02 2.36956000e-01
-3.03366274e-01 -4.76265624e-02 5.01209497e-02 6.36889040e-01
3.60543281e-01 8.93018961e-01 1.61622912e-01 -5.64553678e-01
1.59367293e-01 -8.16768259e-02 3.74670446e-01 -4.53586072e-01
-4.20593143e-01 4.20889288e-01 -3.07466596e-01 -7.98571944e-01
-7.80125320e-01 -7.38953233e-01 -1.37433624e+00 -1.10003114e-01
-3.44915479e-01 -8.84317160e-02 5.66166043e-01 1.25305200e+00
4.97144699e-01 7.76507616e-01 5.10988355e-01 -1.06230915e+00
-1.58267483e-01 -7.23855913e-01 -4.67284113e-01 2.03006059e-01
4.72118437e-01 -7.59772241e-01 1.13407344e-01 1.74157590e-01] | [9.228320121765137, -0.04471762478351593] |
f50955a6-eb9c-4fe6-b213-68fd1f106c8d | proposal-based-multiple-instance-learning-for-1 | 2305.17861 | null | https://arxiv.org/abs/2305.17861v1 | https://arxiv.org/pdf/2305.17861v1.pdf | Proposal-Based Multiple Instance Learning for Weakly-Supervised Temporal Action Localization | Weakly-supervised temporal action localization aims to localize and recognize actions in untrimmed videos with only video-level category labels during training. Without instance-level annotations, most existing methods follow the Segment-based Multiple Instance Learning (S-MIL) framework, where the predictions of segments are supervised by the labels of videos. However, the objective for acquiring segment-level scores during training is not consistent with the target for acquiring proposal-level scores during testing, leading to suboptimal results. To deal with this problem, we propose a novel Proposal-based Multiple Instance Learning (P-MIL) framework that directly classifies the candidate proposals in both the training and testing stages, which includes three key designs: 1) a surrounding contrastive feature extraction module to suppress the discriminative short proposals by considering the surrounding contrastive information, 2) a proposal completeness evaluation module to inhibit the low-quality proposals with the guidance of the completeness pseudo labels, and 3) an instance-level rank consistency loss to achieve robust detection by leveraging the complementarity of RGB and FLOW modalities. Extensive experimental results on two challenging benchmarks including THUMOS14 and ActivityNet demonstrate the superior performance of our method. | ['Yongdong Zhang', 'Tianzhu Zhang', 'Wenfei Yang', 'Huan Ren'] | 2023-05-29 | proposal-based-multiple-instance-learning-for | http://openaccess.thecvf.com//content/CVPR2023/html/Ren_Proposal-Based_Multiple_Instance_Learning_for_Weakly-Supervised_Temporal_Action_Localization_CVPR_2023_paper.html | http://openaccess.thecvf.com//content/CVPR2023/papers/Ren_Proposal-Based_Multiple_Instance_Learning_for_Weakly-Supervised_Temporal_Action_Localization_CVPR_2023_paper.pdf | cvpr-2023-1 | ['weakly-supervised-action-localization', 'weakly-supervised-temporal-action', 'action-localization', 'action-recognition', 'multiple-instance-learning'] | ['computer-vision', 'computer-vision', 'computer-vision', 'computer-vision', 'methodology'] | [ 3.98021579e-01 -1.33379713e-01 -6.46460116e-01 -3.52434188e-01
-9.18660045e-01 -2.35207438e-01 4.08569366e-01 -2.29417786e-01
-3.92592937e-01 4.89346206e-01 2.96554416e-01 2.10874811e-01
3.90266329e-02 -3.82256478e-01 -7.66117811e-01 -7.97063291e-01
-5.13683539e-03 1.13935567e-01 8.92202079e-01 1.80859491e-01
2.64289528e-01 -1.27607621e-02 -1.47146440e+00 8.44276667e-01
9.09213543e-01 1.49949670e+00 9.94861424e-02 1.54949054e-01
-7.96446651e-02 1.42776000e+00 -2.70848572e-01 -5.15244482e-03
4.37026531e-01 -5.33568740e-01 -6.16820872e-01 5.21089733e-01
6.40021741e-01 -6.77987993e-01 -3.46198380e-01 9.99826312e-01
2.28386685e-01 3.11237216e-01 4.06170398e-01 -1.30638850e+00
4.47763912e-02 2.66411990e-01 -6.90957487e-01 4.49779421e-01
3.04186732e-01 6.36668086e-01 1.09851933e+00 -1.29838872e+00
4.79782879e-01 1.24816942e+00 2.71124631e-01 1.50667310e-01
-9.20674205e-01 -5.96418858e-01 7.04141140e-01 6.56228900e-01
-1.33799899e+00 -3.85517329e-01 1.07790411e+00 -5.92744529e-01
5.83323181e-01 -6.08940683e-02 7.41938233e-01 9.41889465e-01
-2.05865726e-01 1.33296669e+00 9.82636511e-01 9.63846296e-02
1.45108864e-01 -7.08743483e-02 2.37495881e-02 1.11310768e+00
-1.06287666e-01 1.05785564e-01 -6.60178483e-01 1.62452042e-01
9.26226258e-01 1.68594331e-01 -3.02979052e-01 -7.27629066e-01
-1.43407524e+00 3.73885453e-01 5.30893207e-01 1.84879169e-01
-4.02792424e-01 5.81802987e-02 5.46916783e-01 -5.96613847e-02
3.08532059e-01 5.27677871e-02 -5.31008959e-01 -5.14958389e-02
-1.04130232e+00 -1.10483631e-01 3.14632863e-01 9.75339234e-01
7.24740744e-01 -2.44368408e-02 -7.78135478e-01 6.74348176e-01
2.99958140e-01 1.46792874e-01 4.12059098e-01 -1.06370425e+00
9.14742410e-01 9.95111942e-01 9.21891853e-02 -8.25249791e-01
-1.46322116e-01 -6.26529932e-01 -4.94443536e-01 1.14782885e-01
5.56678474e-01 1.54635862e-01 -9.11746919e-01 1.59487188e+00
6.03756666e-01 7.39591420e-01 -3.40526700e-01 1.24552667e+00
9.07119334e-01 6.16545975e-01 1.83876619e-01 -4.00036842e-01
9.86230373e-01 -1.51529992e+00 -4.25587982e-01 -1.98552787e-01
6.75069630e-01 -4.82597470e-01 1.06052542e+00 3.39422762e-01
-9.70538616e-01 -9.10295129e-01 -9.98963833e-01 4.79787514e-02
2.10407674e-01 5.83522081e-01 4.14601922e-01 2.29915097e-01
-3.60687315e-01 5.10333896e-01 -9.17000711e-01 1.88894138e-01
8.91184211e-01 1.38936773e-01 -3.65526676e-01 -2.82119006e-01
-8.69347930e-01 3.87860805e-01 6.15615308e-01 4.12643224e-01
-1.28897333e+00 -5.20511508e-01 -8.36849272e-01 1.00865334e-01
9.77348864e-01 -1.59210652e-01 8.48153532e-01 -1.27510095e+00
-1.42218542e+00 3.99701834e-01 -3.56760472e-02 -5.69949560e-02
9.07533765e-01 -1.90387592e-01 -1.86364323e-01 4.62812006e-01
2.17279539e-01 8.18478584e-01 9.74664390e-01 -9.91650701e-01
-1.04463387e+00 -1.65396541e-01 1.90294042e-01 3.23579341e-01
-1.27753958e-01 -2.58148491e-01 -8.70839655e-01 -7.07762361e-01
4.13358450e-01 -7.92830944e-01 -1.32888198e-01 2.43265375e-01
-2.59256124e-01 -4.73122299e-01 7.34294176e-01 -7.89153159e-01
1.04290807e+00 -2.18341160e+00 1.93886861e-01 1.43974304e-01
2.38878354e-02 2.56739646e-01 -4.23669398e-01 -2.64414072e-01
-1.55766243e-02 -3.51227254e-01 -8.46688077e-02 -1.90375373e-01
-2.92989582e-01 1.13279633e-01 -1.20478064e-01 5.35800099e-01
6.39128208e-01 8.48484218e-01 -1.22559655e+00 -1.10437751e+00
5.51390707e-01 1.05116591e-01 -7.48388112e-01 3.18735033e-01
-1.79525003e-01 8.53019953e-01 -8.27304125e-01 1.11716449e+00
4.01435256e-01 -3.22016597e-01 8.86286725e-04 -6.18423343e-01
-2.27825548e-02 3.61863285e-01 -1.52020085e+00 1.77665699e+00
-8.61094799e-03 1.46604374e-01 -1.42390221e-01 -1.18723571e+00
4.15633857e-01 1.75852254e-01 8.98152351e-01 -8.03446412e-01
-9.24304873e-02 2.23425165e-01 6.59104884e-02 -7.07605064e-01
-6.90075010e-02 1.60256073e-01 1.07656166e-01 1.39267355e-01
3.02933186e-01 3.80865067e-01 4.30638611e-01 2.29727760e-01
1.19481146e+00 7.68576205e-01 7.45697543e-02 6.87759370e-02
8.90497386e-01 -4.03397046e-02 1.26747882e+00 6.69773698e-01
-8.92746389e-01 5.19462228e-01 5.31854272e-01 -5.23264885e-01
-5.96268594e-01 -8.84096622e-01 3.13511565e-02 1.06984973e+00
6.34844005e-01 -3.95238578e-01 -4.03893411e-01 -1.40660679e+00
-2.08455265e-01 2.29813680e-01 -5.54489553e-01 -1.95296302e-01
-7.91995108e-01 -3.71235877e-01 1.04013771e-01 7.89926112e-01
7.09370553e-01 -1.24767029e+00 -5.05561352e-01 1.48817778e-01
-5.51232100e-01 -1.24572575e+00 -8.48271132e-01 4.25355509e-02
-9.77946937e-01 -1.36522770e+00 -6.64860547e-01 -7.46248484e-01
7.87423074e-01 2.69749761e-01 7.15620875e-01 1.44874930e-01
-8.82489756e-02 1.65846094e-01 -5.41584134e-01 2.70553708e-01
1.20009892e-01 -2.96474427e-01 -1.07269369e-01 5.52838564e-01
1.51262647e-02 -2.94877559e-01 -9.08299565e-01 7.61391759e-01
-6.68845356e-01 1.86874792e-01 1.09573460e+00 8.42622101e-01
9.04655635e-01 -3.98795716e-02 3.25065672e-01 -5.07936597e-01
-2.88972199e-01 -5.86317927e-02 -3.03570509e-01 5.00943720e-01
-3.07820529e-01 -8.44593570e-02 4.51749861e-01 -6.77565753e-01
-1.02193213e+00 5.43094039e-01 2.23929018e-01 -8.11018348e-01
4.91852649e-02 2.51498461e-01 -3.21027309e-01 -2.24927381e-01
3.70252639e-01 4.55306590e-01 -1.65371895e-01 -1.96592495e-01
1.66606501e-01 1.78900376e-01 4.39901322e-01 -5.46191096e-01
6.66457474e-01 4.60921407e-01 -6.72902390e-02 -3.55683208e-01
-1.14794719e+00 -8.47284615e-01 -7.43100703e-01 -7.81252563e-01
8.84074092e-01 -1.16793227e+00 -3.70716929e-01 4.35888559e-01
-8.01964343e-01 -2.56053239e-01 -3.86722684e-01 8.15825343e-01
-6.05063379e-01 6.12155199e-01 -5.20167053e-01 -6.59118116e-01
-6.51149079e-02 -1.53999043e+00 1.15263522e+00 1.92010611e-01
2.46678114e-01 -4.12857443e-01 -2.81768262e-01 7.63527155e-01
-7.23657906e-02 7.23460913e-02 4.09013003e-01 -4.45475936e-01
-1.09882092e+00 -1.26226947e-01 -4.09402996e-01 7.18504667e-01
3.47141959e-02 -2.06111178e-01 -7.28385031e-01 -2.10250482e-01
-1.11911230e-01 -6.81623280e-01 1.16288781e+00 3.64206076e-01
1.36312294e+00 -2.35546887e-01 -2.56567508e-01 5.90959191e-01
1.08334506e+00 -5.50807565e-02 5.60982704e-01 3.48062776e-02
9.96771336e-01 6.66123450e-01 1.16305137e+00 4.34772611e-01
1.11103497e-01 8.34811866e-01 4.87070829e-01 7.93467537e-02
-2.94874102e-01 -4.27951217e-01 7.12698996e-01 4.52505082e-01
-2.90263414e-01 2.19266176e-01 -4.15393203e-01 4.12270367e-01
-2.20058179e+00 -1.10581875e+00 -7.06945583e-02 2.15275002e+00
5.97179472e-01 4.89579827e-01 2.90985256e-01 5.89183494e-02
7.80523837e-01 4.29321408e-01 -7.25807428e-01 6.50821924e-01
1.49875075e-01 -1.95375443e-01 3.48638594e-01 3.93541455e-02
-1.46886230e+00 7.80825198e-01 4.50913763e+00 1.15905249e+00
-8.62774014e-01 2.31169879e-01 6.92551672e-01 -2.07070589e-01
2.34887823e-01 2.63429791e-01 -6.92953467e-01 7.23609209e-01
9.76312533e-03 4.54195499e-01 5.24394168e-03 1.02030313e+00
4.33589667e-01 -1.77985594e-01 -1.11242282e+00 1.02950430e+00
4.15841267e-02 -1.15311837e+00 1.03440896e-01 -1.88509703e-01
7.36726165e-01 -8.73695239e-02 -1.96856603e-01 6.33801103e-01
-3.14612955e-01 -4.08640802e-01 1.00606859e+00 7.20118284e-01
4.65932995e-01 -5.16574442e-01 5.00803828e-01 3.34233761e-01
-1.57964480e+00 -4.39689428e-01 -2.42390320e-01 1.54568434e-01
2.45634034e-01 3.73337805e-01 -5.97402513e-01 5.72431207e-01
6.29985631e-01 1.26009369e+00 -4.91549641e-01 1.32438815e+00
-4.85455155e-01 8.03667009e-01 -1.52685225e-01 2.55857646e-01
4.37633127e-01 -1.84911087e-01 5.63487291e-01 9.98683631e-01
-4.82897460e-02 1.24579415e-01 8.64558458e-01 6.90535605e-01
1.89806446e-01 1.18866004e-01 1.85435444e-01 9.18146074e-02
1.42099157e-01 1.28500247e+00 -7.56667316e-01 -4.03115034e-01
-5.49427927e-01 9.41251338e-01 2.67576844e-01 3.20757002e-01
-1.04857385e+00 5.94473295e-02 1.36708513e-01 2.91476458e-01
4.85365003e-01 -1.43524334e-01 -4.04259097e-03 -1.41282892e+00
3.72872859e-01 -7.89819837e-01 6.74302220e-01 -6.42583013e-01
-1.04014444e+00 1.83759496e-01 -7.52769634e-02 -1.91267896e+00
1.21759333e-01 -3.65642220e-01 -7.28029490e-01 4.13752228e-01
-1.53732598e+00 -1.27209318e+00 -4.44533974e-01 6.83888733e-01
9.03417289e-01 -2.82913987e-02 5.99569594e-03 5.15154064e-01
-9.88975704e-01 4.36302841e-01 -4.03696895e-01 3.37098032e-01
7.50406563e-01 -9.51779783e-01 -4.22929078e-01 9.99364495e-01
2.07571670e-01 1.13414057e-01 3.47827405e-01 -6.90895200e-01
-9.75698709e-01 -1.32693911e+00 3.98789555e-01 -3.65643591e-01
4.51259881e-01 -2.18871492e-03 -8.94567966e-01 4.13130879e-01
-5.29194772e-01 6.38051033e-01 3.12671155e-01 -1.51789874e-01
-3.29794943e-01 -4.15440708e-01 -7.60435820e-01 2.10086375e-01
1.34711647e+00 -2.76416928e-01 -5.16902447e-01 5.23427248e-01
4.26956236e-01 -3.46946388e-01 -6.18752122e-01 8.64705741e-01
6.95134580e-01 -9.61837113e-01 8.73026609e-01 -4.68552500e-01
4.90450770e-01 -7.10190356e-01 -7.42245745e-03 -5.73935449e-01
-3.59357297e-01 -1.37766674e-01 -4.17297691e-01 1.17091763e+00
1.98041290e-01 2.37300359e-02 8.13943505e-01 3.57198000e-01
-3.67139548e-01 -1.06666362e+00 -1.21482229e+00 -7.13543117e-01
-6.30118132e-01 -4.43454653e-01 5.22267073e-02 6.93355441e-01
-1.18584506e-01 1.22764617e-01 -5.41938961e-01 1.59165710e-01
6.70925438e-01 1.97413966e-01 7.26152956e-01 -8.16648901e-01
-5.74235260e-01 -3.79805982e-01 -7.01547682e-01 -1.47907603e+00
1.08722247e-01 -6.69968605e-01 3.36610764e-01 -1.28099680e+00
4.55729306e-01 -4.50173259e-01 -7.36352623e-01 5.89950979e-01
-3.41228545e-01 1.91676900e-01 1.22129023e-01 6.37190223e-01
-1.49553156e+00 7.96340823e-01 1.52997959e+00 -3.32004756e-01
-2.00164661e-01 1.13325678e-01 -1.89121753e-01 8.21567953e-01
1.56670034e-01 -4.49236244e-01 -5.75658381e-01 -1.13362961e-01
-1.86156094e-01 1.40150413e-01 6.50395393e-01 -1.31492150e+00
2.16868788e-01 -3.88549209e-01 5.66618443e-01 -8.67982328e-01
1.59411296e-01 -5.99428594e-01 -3.97166550e-01 5.30778229e-01
-4.95203853e-01 -5.24737060e-01 -2.88958967e-01 1.00635850e+00
-3.56297821e-01 4.76302542e-02 9.66221154e-01 -1.22468956e-01
-1.33047378e+00 7.47433364e-01 1.66873470e-01 7.72078410e-02
1.14552402e+00 -3.03366482e-01 -1.53107002e-01 -3.82089689e-02
-7.06412077e-01 6.75414026e-01 2.13844523e-01 4.82665241e-01
7.58067012e-01 -1.29015243e+00 -6.44019186e-01 2.16455787e-01
4.15995270e-01 9.61049795e-02 6.01681590e-01 1.41085899e+00
-1.36874393e-01 2.16415852e-01 -1.00061044e-01 -9.24871624e-01
-1.03017604e+00 4.50249612e-01 4.57587421e-01 -4.13220227e-01
-6.14779472e-01 1.05487919e+00 7.12786376e-01 -3.88790555e-02
5.20137787e-01 -3.44937235e-01 -3.80769491e-01 -1.73560288e-02
5.96673846e-01 3.57847184e-01 -9.86794680e-02 -8.45824718e-01
-6.30439758e-01 4.39178318e-01 -1.34367198e-01 1.06538720e-01
1.05010164e+00 5.77889057e-03 2.62674809e-01 3.33945036e-01
1.05071962e+00 -3.70848507e-01 -1.92890644e+00 -3.24150920e-01
-1.92643359e-01 -7.93207586e-01 8.32416266e-02 -6.78528368e-01
-1.23757052e+00 7.98629642e-01 6.61932886e-01 -4.20343101e-01
1.07226610e+00 -9.12281424e-02 7.05615044e-01 2.58426309e-01
4.66807455e-01 -1.34105468e+00 8.28847468e-01 1.99623495e-01
7.17502117e-01 -1.48606563e+00 4.32850048e-02 -5.07933199e-01
-7.92206347e-01 9.25295532e-01 1.11839581e+00 -8.60629678e-02
5.19508064e-01 -3.65452170e-01 -3.26670825e-01 1.71900336e-02
-6.64339542e-01 -3.94091904e-01 6.68197751e-01 2.78886378e-01
8.11121017e-02 -3.17844301e-01 -3.53562832e-01 7.25959241e-01
8.66747856e-01 1.09787725e-01 6.38016388e-02 8.91445875e-01
-6.30797923e-01 -8.38995039e-01 -7.47582093e-02 6.98727310e-01
-2.30483189e-01 1.92574322e-01 -1.95548519e-01 4.72483724e-01
5.83824813e-01 8.34824026e-01 -1.54538214e-01 -5.57471275e-01
3.85452867e-01 -1.25659406e-01 4.27307487e-01 -6.58488035e-01
-3.04818094e-01 2.87231445e-01 6.26213700e-02 -1.00191748e+00
-9.13451254e-01 -8.73451471e-01 -1.23434877e+00 3.58590871e-01
-6.57270789e-01 4.38101254e-02 5.63438945e-02 1.27192867e+00
2.36424953e-01 4.90115345e-01 7.21466064e-01 -9.76737738e-01
-6.00877404e-01 -9.73003566e-01 -5.63848078e-01 6.20962083e-01
2.09707066e-01 -1.05547154e+00 -3.59832734e-01 2.84414049e-02] | [8.442315101623535, 0.6059369444847107] |
23098084-261a-4872-8ef3-76efbb62532b | 2305-14872 | 2305.14872 | null | https://arxiv.org/abs/2305.14872v2 | https://arxiv.org/pdf/2305.14872v2.pdf | Timeseries-aware Uncertainty Wrappers for Uncertainty Quantification of Information-Fusion-Enhanced AI Models based on Machine Learning | As the use of Artificial Intelligence (AI) components in cyber-physical systems is becoming more common, the need for reliable system architectures arises. While data-driven models excel at perception tasks, model outcomes are usually not dependable enough for safety-critical applications. In this work,we present a timeseries-aware uncertainty wrapper for dependable uncertainty estimates on timeseries data. The uncertainty wrapper is applied in combination with information fusion over successive model predictions in time. The application of the uncertainty wrapper is demonstrated with a traffic sign recognition use case. We show that it is possible to increase model accuracy through information fusion and additionally increase the quality of uncertainty estimates through timeseries-aware input quality features. | ['Pascal Gerber', 'Lisa Jöckel', 'Michael Kläs', 'Janek Groß'] | 2023-05-24 | null | null | null | null | ['traffic-sign-recognition'] | ['computer-vision'] | [ 1.53411493e-01 8.72246772e-02 1.33808389e-01 -7.41880655e-01
-7.67118037e-01 -3.57617110e-01 8.12399983e-01 4.45021749e-01
-2.33185261e-01 8.28002751e-01 -1.80576846e-01 -4.03304785e-01
-7.95802295e-01 -7.67020106e-01 -4.91865188e-01 -4.52809215e-01
-3.40109110e-01 5.31152606e-01 3.95779014e-01 -9.86005738e-02
1.53755680e-01 7.69858062e-01 -1.87070417e+00 2.22024053e-01
8.04475427e-01 1.59246075e+00 -2.01344699e-01 8.23355556e-01
6.16470352e-02 6.20612025e-01 -7.29720652e-01 -6.08158894e-02
4.08185661e-01 1.88495338e-01 -1.55069798e-01 -4.55785580e-02
-2.10090533e-01 -1.97059482e-01 -8.20226669e-02 6.60806656e-01
-1.20281830e-01 1.90469876e-01 5.54982543e-01 -1.82715535e+00
1.15867354e-01 4.22743052e-01 5.84122129e-02 -2.29214132e-01
2.60901242e-01 3.99049371e-01 4.76722360e-01 -5.73424160e-01
2.55606443e-01 1.23062766e+00 4.70537782e-01 1.08023420e-01
-1.07977200e+00 -3.52894157e-01 1.26864180e-01 3.75140518e-01
-1.30608785e+00 -3.21752310e-01 7.51824260e-01 -4.98175114e-01
1.04179168e+00 4.19544160e-01 6.13298237e-01 6.14914954e-01
7.86688626e-01 5.51848531e-01 1.30979085e+00 -4.49560612e-01
8.02335560e-01 3.43347564e-02 3.17569524e-01 7.09387511e-02
6.27485096e-01 9.47176576e-01 -4.88783538e-01 -2.00045556e-01
2.27364287e-01 1.04757033e-01 3.00137192e-01 -3.13243240e-01
-1.00379813e+00 3.11861366e-01 2.94196516e-01 1.77348539e-01
-6.63571954e-01 5.02086282e-01 1.60323545e-01 3.68815094e-01
2.98739910e-01 4.02339309e-01 -6.54892385e-01 -3.93698215e-01
-7.22176552e-01 3.85928154e-01 9.13892329e-01 1.02189898e+00
5.02779603e-01 2.46742040e-01 -1.35109825e-02 1.38322830e-01
6.92792356e-01 7.99080968e-01 3.34183164e-02 -9.63529408e-01
7.22969621e-02 6.88014388e-01 5.30018926e-01 -5.98226249e-01
-6.80480123e-01 -2.75750854e-03 -5.07308304e-01 8.69908214e-01
1.71698481e-01 -9.44909230e-02 -1.24301636e+00 1.32752180e+00
9.17837173e-02 -3.50585859e-03 2.43725866e-01 6.28593564e-01
-1.51831374e-01 5.93086481e-01 2.45483026e-01 -3.20899814e-01
1.11081660e+00 -8.47539958e-03 -1.03737736e+00 1.04898913e-02
2.00380564e-01 -4.22774047e-01 2.76847303e-01 9.41280723e-01
-6.49637461e-01 -5.91867745e-01 -1.45908391e+00 5.64704835e-01
-7.38746762e-01 -4.75201756e-01 5.77086329e-01 8.83606195e-01
-4.25188243e-01 6.96569920e-01 -1.07949948e+00 -1.14276066e-01
1.40542090e-02 3.80131572e-01 -1.47226259e-01 1.51480272e-01
-1.27449298e+00 1.57384622e+00 6.58295393e-01 7.97200948e-02
-6.23992205e-01 -5.96111059e-01 -7.96064615e-01 -2.16054991e-01
3.91943395e-01 -2.62664020e-01 1.34876454e+00 -4.08353925e-01
-1.63536048e+00 -2.45889276e-01 4.61847275e-01 -1.05151188e+00
4.67179924e-01 -2.23251343e-01 -1.12518275e+00 -1.89018190e-01
-5.56066871e-01 1.27780125e-01 8.32365394e-01 -1.35763919e+00
-6.84343696e-01 -3.43127578e-01 -2.83703059e-01 -4.58264709e-01
4.18081731e-01 -2.80236453e-01 2.86237031e-01 -1.39777273e-01
2.50034720e-01 -9.60677028e-01 -4.40399051e-01 2.51022428e-02
9.84343700e-03 -1.66446052e-03 8.37431729e-01 -6.06749535e-01
1.36107147e+00 -1.76244903e+00 -2.66624779e-01 7.83849835e-01
-1.36077106e-01 1.94131732e-01 2.42398351e-01 7.84648657e-01
9.61107239e-02 -7.92348906e-02 -3.82686853e-01 1.52330413e-01
3.05977881e-01 5.50146878e-01 -4.31722611e-01 2.54086137e-01
5.28441072e-01 5.86779952e-01 -5.86526871e-01 -2.32127905e-01
1.03848505e+00 2.30286747e-01 -1.15075879e-01 1.43574268e-01
-6.18005097e-01 1.43854737e-01 -5.08043647e-01 7.09912896e-01
4.84695643e-01 2.98818320e-01 -7.08295777e-02 -2.81093210e-01
-3.10103059e-01 -3.79560776e-02 -1.27422535e+00 1.38468766e+00
-6.85365438e-01 4.25550073e-01 -2.24795610e-01 -4.61539477e-01
1.09142923e+00 4.15471047e-01 5.65931082e-01 -6.27920270e-01
3.66676003e-01 3.03482801e-01 1.66528031e-01 -4.79527980e-01
5.99994898e-01 -2.54732549e-01 -4.25517768e-01 1.54097363e-01
-1.64251864e-01 -8.66302013e-01 1.25162870e-01 -8.38247463e-02
1.16063476e+00 5.01481116e-01 5.38381517e-01 -8.70229155e-02
2.97609121e-01 9.19452906e-02 5.39181590e-01 2.53482908e-01
-3.51525754e-01 3.76254231e-01 8.58413056e-02 -5.04591882e-01
-1.02303767e+00 -1.20886338e+00 -3.95215660e-01 2.80389130e-01
4.29640599e-02 -3.07239383e-01 -3.52627635e-01 -5.06847799e-01
5.49772680e-01 1.57660699e+00 -4.64384824e-01 -4.93136466e-01
-1.67797077e-02 -3.04182619e-01 1.53891474e-01 6.40045047e-01
1.54628590e-01 -5.42226374e-01 -1.06095374e+00 6.32361531e-01
3.89902025e-01 -8.86202276e-01 3.55770111e-01 3.37271303e-01
-8.64089251e-01 -9.08756196e-01 8.62131082e-03 7.57920086e-01
3.59392613e-01 -3.30414295e-01 9.21470404e-01 -3.04272413e-01
-3.07006627e-01 7.50748336e-01 -3.70232314e-01 -1.33353555e+00
-7.52018154e-01 -6.80710435e-01 3.37297499e-01 -1.22171052e-01
3.83619845e-01 -5.43606162e-01 -2.57533640e-01 3.31765324e-01
-9.91844714e-01 -1.98903441e-01 4.84766185e-01 5.64115942e-01
4.88064557e-01 2.55692303e-01 8.40086937e-01 -1.47909537e-01
6.90894663e-01 -3.51180971e-01 -1.17684746e+00 4.47651386e-01
-1.10409856e+00 4.24830943e-01 2.05795437e-01 -2.81672686e-01
-1.13504779e+00 2.78048992e-01 1.34435296e-01 -5.53365052e-01
-2.92155683e-01 6.89629853e-01 -1.02914676e-01 4.66886014e-02
8.11760902e-01 -4.02238339e-01 2.81617790e-01 -2.39690050e-01
2.57896632e-01 8.90205920e-01 4.57674056e-01 -6.74889266e-01
5.70664525e-01 2.09740326e-01 5.26003242e-01 -5.65534294e-01
-2.99750835e-01 -1.47113830e-01 -3.75759870e-01 -7.77959228e-01
5.11762619e-01 -4.71924573e-01 -1.02846944e+00 1.68899670e-01
-8.88272405e-01 9.71348956e-02 -8.61542702e-01 9.09343958e-01
-9.42814350e-01 -8.92842840e-03 2.12274939e-01 -1.69154251e+00
4.67979871e-02 -9.34069097e-01 8.88969958e-01 1.70139313e-01
-3.19625229e-01 -6.45910442e-01 6.58168942e-02 2.42282879e-02
5.06815135e-01 5.21476388e-01 4.22233075e-01 -8.20181251e-01
-7.09124863e-01 -1.12615132e+00 1.21896364e-01 5.21649063e-01
-6.29942417e-02 3.65571052e-01 -1.10223520e+00 1.67703867e-01
2.44322210e-03 -7.07067689e-03 5.50099909e-01 3.02051187e-01
9.04254138e-01 -2.18505785e-02 -2.65184045e-01 -3.08648109e-01
1.36815667e+00 7.03981459e-01 6.94740772e-01 1.43425390e-01
-7.54139125e-02 8.37291479e-01 1.16154540e+00 7.39670455e-01
1.88391805e-01 6.56209946e-01 6.83961987e-01 6.56112373e-01
1.63284019e-01 3.12281772e-02 1.13608152e-01 3.29159677e-01
-2.09196210e-01 -3.00397694e-01 -1.13324547e+00 4.10796702e-01
-2.11180305e+00 -8.46458793e-01 -7.86894280e-03 2.57668304e+00
3.70704770e-01 4.02741224e-01 -1.21861145e-01 4.58269775e-01
3.80814224e-01 -4.47182864e-01 -6.09888673e-01 -6.26798213e-01
3.18102390e-01 -9.90374461e-02 7.73253858e-01 4.82226849e-01
-7.62754738e-01 1.47398025e-01 6.47734070e+00 4.88037527e-01
-4.81843084e-01 -2.61158317e-01 2.96591878e-01 1.45797990e-02
-2.39082918e-01 2.57585675e-01 -4.22694534e-01 4.96588498e-01
1.60731864e+00 -4.07532841e-01 9.34202000e-02 8.26348841e-01
2.62775868e-01 -7.30500042e-01 -1.29605734e+00 7.57686615e-01
-4.57136810e-01 -1.08480585e+00 -3.77185732e-01 1.10402972e-01
4.36692059e-01 -2.77351171e-01 -6.62089288e-02 1.74661711e-01
5.45975387e-01 -8.29261899e-01 8.48855197e-01 1.37837088e+00
2.55126476e-01 -1.03117073e+00 8.78445983e-01 2.92456299e-01
-8.35468769e-01 -4.87083018e-01 1.51791453e-01 -1.47393763e-01
6.25155270e-01 1.10376143e+00 -9.20825362e-01 8.73319507e-01
3.34468126e-01 1.82978168e-01 -2.91547120e-01 1.37680626e+00
-5.64541714e-03 3.09942931e-01 -7.81844139e-01 -2.22689956e-01
-2.54215479e-01 -6.44307956e-02 6.98859632e-01 8.28691483e-01
7.21590936e-01 3.02366436e-01 -8.78970549e-02 8.10087442e-01
8.86355042e-01 -6.23287737e-01 -7.27737844e-01 -1.78120658e-01
4.85980719e-01 9.22347903e-01 -3.10315073e-01 -3.55982006e-01
-7.41128698e-02 2.87631214e-01 -5.37219405e-01 2.44909182e-01
-6.81839406e-01 -3.40000451e-01 7.32187331e-01 -4.76728976e-02
-2.81021632e-02 -5.80495536e-01 -5.55750191e-01 -5.96249402e-01
3.15463543e-02 -5.92208385e-01 2.93432057e-01 -1.15902019e+00
-1.41919148e+00 5.59650660e-01 6.83352411e-01 -1.81000590e+00
-8.78064930e-01 -7.14712203e-01 -2.73368537e-01 8.55266809e-01
-9.77216840e-01 -1.08305061e+00 -1.02857582e-01 1.36097863e-01
2.30314493e-01 -8.38731900e-02 8.14144373e-01 -2.35237703e-01
-2.65926093e-01 -2.85308778e-01 -4.84418161e-02 -6.90103829e-01
3.94584835e-01 -1.02955329e+00 2.63274729e-01 8.73594105e-01
-1.98944896e-01 3.09461236e-01 1.40195048e+00 -9.15053487e-01
-1.54093874e+00 -8.01791251e-01 4.94949341e-01 -6.46099091e-01
9.00038540e-01 1.95438415e-01 -7.40380466e-01 4.14025456e-01
3.82059477e-02 1.46386266e-01 3.64635974e-01 -1.52876407e-01
-4.07074809e-01 -5.26040852e-01 -1.66007113e+00 3.52474689e-01
3.55829746e-01 -3.47381741e-01 -8.97298098e-01 -1.33289516e-01
6.83561146e-01 -1.93168849e-01 -1.30284369e+00 9.21446681e-01
6.25560582e-01 -7.44631708e-01 7.55344868e-01 -3.85436475e-01
-3.61386538e-01 -7.79802620e-01 -5.06413400e-01 -1.37214684e+00
1.64358616e-01 -5.79297185e-01 -5.10961890e-01 9.42675531e-01
4.38382506e-01 -7.98675120e-01 2.63749003e-01 1.66324055e+00
-1.81907862e-01 -2.02928931e-01 -1.39715004e+00 -1.26414096e+00
-3.98974538e-01 -1.23194945e+00 9.24119532e-01 1.79038584e-01
3.84845614e-01 -2.89055347e-01 -1.33627504e-01 3.19398433e-01
8.97974074e-01 -2.31998920e-01 4.18706149e-01 -1.57430148e+00
8.47073738e-03 -1.60559937e-01 -9.36395645e-01 -7.58905616e-03
-3.76454294e-01 -4.91757803e-02 4.34867591e-01 -1.44748700e+00
-5.69379330e-01 -5.07036209e-01 -6.04015231e-01 2.61949211e-01
3.33029091e-01 -3.89576286e-01 4.04692411e-01 -2.65190244e-01
-2.91576654e-01 5.22516966e-01 5.17560959e-01 -1.42044187e-01
-2.18770429e-01 5.03861845e-01 1.41262516e-01 7.17350364e-01
1.02066779e+00 -3.43544841e-01 -6.75411820e-01 2.91166663e-01
4.36130583e-01 3.68006498e-01 5.14791071e-01 -1.64508259e+00
2.88333833e-01 -4.49612141e-01 3.38018030e-01 -1.02101851e+00
6.22393072e-01 -1.65510917e+00 8.04851055e-01 5.05931020e-01
-8.33540186e-02 -1.54934719e-01 6.50084674e-01 9.19859588e-01
-2.01221228e-01 -1.42585069e-01 4.86355811e-01 3.93890440e-01
-7.85297155e-01 3.39650325e-02 -7.52984881e-01 -9.73219454e-01
1.40598345e+00 -2.56429195e-01 -2.03687698e-01 -2.61834770e-01
-7.89777458e-01 1.90482080e-01 1.39098451e-01 6.02154374e-01
7.31816351e-01 -1.19130218e+00 -3.86029333e-01 2.38214314e-01
5.42253375e-01 -2.48160630e-01 3.11314613e-01 5.58911026e-01
-9.55550224e-02 5.93542159e-01 -3.41272891e-01 -7.10941970e-01
-8.93490493e-01 8.05890083e-01 3.21730256e-01 5.64026982e-02
-9.02944878e-02 2.51023769e-01 -6.49316132e-01 -2.07675248e-01
1.67361021e-01 -8.00994217e-01 2.20775709e-01 -1.15256757e-01
7.17589200e-01 6.67372584e-01 3.64570260e-01 -1.93393022e-01
-4.42641467e-01 2.69899964e-01 2.91839659e-01 -7.72850037e-01
1.05812907e+00 -2.72362798e-01 3.17579359e-01 1.04282093e+00
4.24492151e-01 -4.67392802e-01 -1.53414297e+00 1.14819510e-02
5.89461327e-01 -3.77896249e-01 2.14178756e-01 -1.39153075e+00
-3.80916864e-01 7.09577262e-01 9.92782176e-01 6.00753367e-01
1.29797649e+00 -3.23744416e-01 2.62491047e-01 6.29593968e-01
9.48293209e-01 -1.36879373e+00 -6.22380018e-01 1.76236838e-01
1.28854275e+00 -1.25148225e+00 3.96214277e-01 -6.48481399e-02
-7.34348059e-01 1.13410807e+00 2.96774238e-01 1.12114631e-01
1.10426438e+00 7.72342205e-01 -1.26904532e-01 2.76410934e-02
-1.26862037e+00 -3.35262269e-01 4.74708349e-01 8.97133291e-01
-4.97208349e-02 4.33070868e-01 -3.15078199e-01 7.47759342e-01
1.78401843e-01 3.69434953e-01 4.57919449e-01 1.21024024e+00
-6.30557418e-01 -1.18784857e+00 -7.33039618e-01 4.66648370e-01
9.47187319e-02 4.34215724e-01 -1.49442062e-01 8.76058638e-01
-7.38097215e-03 1.44763434e+00 1.27559319e-01 -8.32357705e-01
5.98234415e-01 2.76230186e-01 3.85769725e-01 -1.88155875e-01
-4.04882967e-01 -3.66655976e-01 6.17439926e-01 -8.25088620e-01
-3.67233396e-01 -7.29589522e-01 -1.43436027e+00 -2.06415281e-01
-4.74635810e-01 -1.08517513e-01 1.49331963e+00 1.01609802e+00
5.49453855e-01 8.35507035e-01 5.73680997e-01 -7.54169941e-01
-8.29211235e-01 -9.33879256e-01 -4.90539342e-01 -1.41401976e-01
3.13785553e-01 -8.73266935e-01 -3.33061367e-01 -1.03893317e-02] | [5.818093776702881, 2.0488767623901367] |
c5ae2f6a-a5ea-4943-9d1f-b555f3e56cd9 | label-correction-model-for-aspect-based | null | null | https://aclanthology.org/2020.coling-main.71 | https://aclanthology.org/2020.coling-main.71.pdf | Label Correction Model for Aspect-based Sentiment Analysis | Aspect-based sentiment analysis includes opinion aspect extraction and aspect sentiment classification. Researchers have attempted to discover the relationship between these two sub-tasks and have proposed the joint model for solving aspect-based sentiment analysis. However, they ignore a phenomenon: aspect boundary label and sentiment label of the same word can correct each other. To exploit this phenomenon, we propose a novel deep learning model named the label correction model. Specifically, given an input sentence, our model first predicts the aspect boundary label sequence and sentiment label sequence, then re-predicts the aspect boundary (sentiment) label sequence using the embeddings of the previously predicted sentiment (aspect boundary) label. The goal of the re-prediction operation (can be repeated multiple times) is to use the information of the sentiment (aspect boundary) label to correct the wrong aspect boundary (sentiment) label. Moreover, we explore two ways of using label embeddings: add and gate mechanism. We evaluate our model on three benchmark datasets. Experimental results verify that our model achieves state-of-the-art performance compared with several baselines. | ['Jiangtao Ren', 'Qianlong Wang'] | 2020-12-01 | null | null | null | coling-2020-8 | ['aspect-extraction'] | ['natural-language-processing'] | [ 3.60703796e-01 2.71562412e-02 -3.58742207e-01 -7.91782737e-01
-6.85527146e-01 -6.37388408e-01 3.66940886e-01 9.11855325e-02
-2.40199640e-01 4.40972000e-01 4.74264234e-01 -2.43113995e-01
6.06679380e-01 -8.64945829e-01 -4.32057470e-01 -6.69755638e-01
6.52413964e-01 3.96714240e-01 -5.11293150e-02 -1.81574121e-01
6.59692407e-01 -2.03095257e-01 -1.21761405e+00 6.27176642e-01
4.56654400e-01 1.00960016e+00 -2.52084881e-01 4.81462598e-01
-7.96903253e-01 8.48385215e-01 -6.32429838e-01 -7.24286437e-01
1.43738508e-01 -5.35189867e-01 -8.06448698e-01 4.40242350e-01
-8.91311914e-02 -9.87028778e-02 4.31439221e-01 1.29138756e+00
3.70730639e-01 -7.56734535e-02 7.67144084e-01 -1.21845019e+00
-1.06099284e+00 4.26209062e-01 -9.86176968e-01 -1.76332280e-01
2.11923167e-01 -7.92774484e-02 1.45810175e+00 -9.34805453e-01
4.17529523e-01 9.23407555e-01 7.40531147e-01 5.95318317e-01
-6.10341370e-01 -4.90161568e-01 7.40864158e-01 5.19658104e-02
-9.36756015e-01 1.31240815e-01 8.98807704e-01 -2.54425496e-01
1.07421148e+00 3.09909526e-02 8.87825906e-01 7.29386151e-01
5.25761724e-01 1.06963205e+00 1.07995093e+00 -3.19878340e-01
2.20530093e-01 3.88695449e-01 5.51871657e-01 7.63356149e-01
2.20195144e-01 -4.15709466e-01 -5.66950858e-01 1.33937849e-02
-1.26015589e-01 1.41166106e-01 3.88153046e-02 -8.12334493e-02
-9.79896128e-01 9.73236501e-01 1.13093354e-01 1.96123600e-01
-2.45088279e-01 4.48250957e-02 6.25458896e-01 3.95325869e-01
8.78559887e-01 6.13104224e-01 -1.16867805e+00 -1.14162952e-01
-6.20639086e-01 -1.19688421e-01 1.06101716e+00 9.07145500e-01
1.03714478e+00 -2.97400635e-02 -2.34214962e-01 6.14613116e-01
6.50081217e-01 3.54907185e-01 7.18288422e-01 -3.23457241e-01
2.74826199e-01 1.23325026e+00 -1.23930192e-02 -8.63848090e-01
-5.38405359e-01 -4.95295882e-01 -5.97610712e-01 -1.45800328e-02
-1.31975681e-01 -3.35846663e-01 -1.17911386e+00 1.34287095e+00
5.52625179e-01 8.74126703e-02 2.00428411e-01 8.17386210e-01
1.01529109e+00 7.94559240e-01 1.79575682e-01 -2.58072853e-01
1.77375293e+00 -1.65807962e+00 -8.22415113e-01 -5.97683191e-01
1.01387691e+00 -1.09499025e+00 1.08217549e+00 2.64195830e-01
-5.56610167e-01 -2.35328138e-01 -1.14725399e+00 -2.19724458e-02
-6.44538045e-01 2.56248772e-01 9.11467791e-01 5.72541773e-01
-8.68120551e-01 2.66649246e-01 -5.04943967e-01 -1.24863595e-01
2.10916638e-01 3.31662059e-01 -1.58485979e-01 1.39229253e-01
-1.16396761e+00 5.46315253e-01 -7.03429580e-02 2.12725565e-01
-5.77122748e-01 -6.87550128e-01 -9.18871880e-01 -2.85861501e-03
2.28261068e-01 -7.72265613e-01 1.43634915e+00 -1.57099104e+00
-1.34980440e+00 1.05549634e+00 -6.92647457e-01 2.25168630e-01
-1.25829577e-01 -3.01476061e-01 -5.58945537e-01 -3.61881971e-01
5.49741685e-01 3.89052480e-01 8.35286081e-01 -1.38052046e+00
-9.15614367e-01 -5.08258939e-01 2.35165447e-01 4.29424137e-01
-4.09361541e-01 -5.27846664e-02 -4.86059994e-01 -5.25415778e-01
2.42367625e-01 -8.32744718e-01 -4.12894517e-01 -3.49733800e-01
-5.00801027e-01 -4.40275788e-01 5.79330981e-01 -4.01928067e-01
1.32269168e+00 -2.02330375e+00 -2.38987133e-01 5.17729558e-02
5.56420386e-02 2.64234394e-01 -3.98217857e-01 2.53311813e-01
-2.50919372e-01 4.19780940e-01 -4.25788999e-01 -6.07097030e-01
-9.71677750e-02 2.63281055e-02 -5.04382253e-01 1.50434196e-01
2.74287730e-01 1.02236259e+00 -8.66313279e-01 -3.56340736e-01
-3.12495947e-01 2.90251344e-01 -5.93930066e-01 2.38771692e-01
-5.40421009e-01 2.59289563e-01 -7.58375168e-01 7.40476251e-01
7.99597859e-01 -2.77700543e-01 1.81418136e-01 -4.96140659e-01
7.46695399e-02 6.97126091e-01 -9.79575574e-01 1.46538174e+00
-5.16037822e-01 3.84967089e-01 -5.14413297e-01 -7.72465348e-01
1.06748116e+00 4.39681590e-01 3.25995713e-01 -5.12280107e-01
3.93416196e-01 1.14256099e-01 -3.23976815e-01 -6.50113285e-01
7.66618192e-01 -6.14564359e-01 -2.77203381e-01 1.18157756e+00
-2.81151123e-02 -1.74912736e-01 1.16278097e-01 -4.36247289e-02
7.97333121e-01 2.37264946e-01 4.61673528e-01 4.46009263e-02
8.63279402e-01 1.99129224e-01 8.29497397e-01 3.74195397e-01
-2.47947469e-01 5.55751264e-01 1.05012250e+00 -8.81049991e-01
-8.57296407e-01 -6.03771567e-01 2.29442716e-01 1.20697391e+00
3.28089416e-01 -5.80325365e-01 -5.53466797e-01 -1.35803938e+00
-4.57878858e-01 6.13644361e-01 -9.93004620e-01 -2.49995634e-01
-4.13479477e-01 -1.12098813e+00 8.68173614e-02 5.64032078e-01
4.84983295e-01 -1.29609835e+00 5.35596348e-02 1.19252920e-01
-1.06384836e-01 -8.55691433e-01 -6.43590868e-01 2.11994812e-01
-7.26074517e-01 -1.02162457e+00 -1.73906192e-01 -1.20446181e+00
9.35872078e-01 3.62184465e-01 1.25597501e+00 4.31273788e-01
3.40286255e-01 2.16642231e-01 -6.71720684e-01 -4.23623919e-01
-1.27842411e-01 2.60893732e-01 -2.35665590e-01 2.53211439e-01
1.18422878e+00 -3.61964047e-01 -6.73295140e-01 1.06019013e-01
-1.09066248e+00 -1.14777669e-01 5.86288214e-01 7.04471648e-01
9.70081270e-01 -1.85169093e-02 6.68274283e-01 -1.52112293e+00
7.52735496e-01 -4.44782972e-01 -2.87123054e-01 3.31448585e-01
-1.12495935e+00 2.55938228e-02 6.58984721e-01 -2.07756579e-01
-1.03074288e+00 -5.84725626e-02 -5.58724523e-01 2.48415262e-01
6.76681176e-02 7.63419628e-01 -3.26660037e-01 4.78378296e-01
-8.01607519e-02 4.38448280e-01 -3.35550666e-01 -2.77060896e-01
3.63253653e-01 8.45213830e-01 -1.73950508e-01 -1.26332968e-01
4.24012780e-01 6.23249471e-01 -1.87213600e-01 -2.18658239e-01
-1.87313485e+00 -5.34850717e-01 -5.14926672e-01 -3.73086445e-02
9.26954985e-01 -1.00299454e+00 -5.70288479e-01 5.90497255e-01
-1.49168074e+00 2.86621988e-01 -3.41156781e-01 2.12758407e-01
-2.31335983e-01 2.94310987e-01 -5.33745468e-01 -6.00327909e-01
-7.17801571e-01 -1.34042299e+00 1.39359045e+00 4.36784327e-01
-3.77686054e-01 -1.16562057e+00 4.51098710e-01 3.67415339e-01
2.54272223e-01 -2.35288128e-01 1.07578409e+00 -9.05566037e-01
-3.84581387e-01 -5.35190940e-01 -1.94194391e-01 4.78408307e-01
4.26424712e-01 -4.86949123e-02 -1.12120402e+00 1.27131462e-01
3.21019977e-01 -7.57797435e-02 9.65817809e-01 -9.59682651e-03
6.74001157e-01 -2.53232479e-01 -9.61623788e-02 5.28307796e-01
1.37266195e+00 2.60045320e-01 5.59682906e-01 6.52221799e-01
8.62449050e-01 4.64422524e-01 7.41671979e-01 1.94129661e-01
7.31286764e-01 3.45873386e-01 2.80724466e-01 8.05168077e-02
-3.99465561e-02 -3.54019374e-01 5.31225443e-01 1.36407983e+00
5.11788666e-01 -2.40665361e-01 -4.32023585e-01 7.52368569e-01
-1.70696127e+00 -4.52926964e-01 -2.96952873e-01 1.60704350e+00
8.83108616e-01 3.09142500e-01 -4.06441927e-01 1.33285942e-02
4.65841323e-01 5.74178457e-01 -6.09155655e-01 -1.00520849e+00
-1.35489792e-01 5.75398840e-02 7.13724867e-02 5.69276273e-01
-1.26791513e+00 1.30859149e+00 5.70682096e+00 6.44179821e-01
-1.04825640e+00 3.49004269e-01 7.59055138e-01 1.71354964e-01
-9.12014008e-01 3.78633201e-01 -1.05186784e+00 2.13224933e-01
4.52682137e-01 -1.25043347e-01 2.83820033e-02 9.70858812e-01
-2.53836792e-02 -1.57769397e-01 -8.72731924e-01 4.64641243e-01
6.26551926e-01 -1.06388795e+00 4.74488348e-01 -3.50253224e-01
1.10732067e+00 -2.99353927e-01 2.53164000e-03 6.14254177e-01
1.86057299e-01 -6.87823653e-01 3.68323028e-01 4.01035190e-01
3.66637290e-01 -7.55178571e-01 1.28572261e+00 -7.20794499e-02
-1.35579574e+00 2.45273471e-01 -3.70291412e-01 -2.41321683e-01
3.99408877e-01 9.66207981e-01 -5.71922421e-01 3.22215319e-01
5.21939576e-01 1.15129507e+00 -4.21708852e-01 5.26704073e-01
-9.54407454e-01 6.18778944e-01 3.02141339e-01 -3.62577438e-01
3.20099562e-01 -4.16850150e-01 4.88692284e-01 8.98746073e-01
1.11666001e-01 -1.25937939e-01 -1.00518659e-01 5.13175905e-01
-2.60716498e-01 3.97582233e-01 -4.94959891e-01 -3.52869898e-01
-4.78581637e-02 1.52132523e+00 -8.90508175e-01 -4.52361524e-01
-6.95560396e-01 1.16082895e+00 3.45159918e-01 2.95709044e-01
-5.78118503e-01 -2.53932685e-01 1.08153486e+00 -3.63763928e-01
6.19195580e-01 2.00965822e-01 -9.55238581e-01 -1.26281738e+00
2.02165544e-01 -7.79103756e-01 2.03165695e-01 -8.48136604e-01
-1.34763694e+00 8.12206864e-01 -8.78990293e-01 -1.40812254e+00
1.25245810e-01 -6.09751761e-01 -9.07085478e-01 6.71780765e-01
-1.93123853e+00 -1.29195714e+00 5.35676554e-02 8.61914679e-02
8.29748392e-01 -2.36661568e-01 1.03125048e+00 1.69879794e-01
-3.71277928e-01 5.47036767e-01 -2.35329986e-01 2.30386063e-01
6.26115978e-01 -1.22701383e+00 5.10626793e-01 7.44075477e-01
7.23338798e-02 5.90418279e-01 5.27894199e-01 -8.52948427e-01
-1.17144597e+00 -1.30186999e+00 1.60756981e+00 -6.60817862e-01
8.00020874e-01 -1.11664176e-01 -5.58252752e-01 8.56778145e-01
4.36280161e-01 -4.85278875e-01 1.29141331e+00 2.12366581e-01
-6.49131656e-01 -1.86240137e-01 -8.53044987e-01 6.47187829e-01
6.76228940e-01 -3.93855810e-01 -6.88534498e-01 3.74134570e-01
1.31762874e+00 -6.19235029e-03 -3.61426741e-01 4.76409525e-01
6.07315004e-01 -9.12554145e-01 4.40143913e-01 -8.80244792e-01
1.05310476e+00 -6.44446254e-01 -1.25341862e-01 -1.49644458e+00
-1.63226902e-01 3.13091800e-02 1.10654920e-01 1.46530080e+00
9.93975282e-01 -5.83225429e-01 8.99508715e-01 4.91539717e-01
-1.59167305e-01 -1.26688397e+00 -5.90739191e-01 -1.14186041e-01
-9.64340270e-02 -5.83749294e-01 9.88101423e-01 7.37136781e-01
-2.87955515e-02 9.94635344e-01 -4.15251225e-01 1.97749779e-01
1.38620257e-01 8.19865942e-01 5.90238690e-01 -9.00215924e-01
-1.42121300e-01 -3.83841306e-01 -1.79984137e-01 -1.18821073e+00
3.70820463e-01 -8.59947979e-01 9.64218825e-02 -1.82875621e+00
4.82662678e-01 -2.02233210e-01 -3.13750982e-01 5.53564310e-01
-6.45922303e-01 4.88169283e-01 2.01655701e-02 -1.55315618e-03
-9.70174551e-01 7.83851922e-01 1.47059429e+00 -4.51900959e-01
-6.19201921e-02 1.02034509e-01 -1.36457503e+00 8.78717899e-01
8.82899523e-01 -8.51010263e-01 -2.91642398e-01 -6.46533370e-01
1.11646903e+00 -5.93185663e-01 -5.50911486e-01 -4.09152329e-01
2.02114195e-01 -1.52475182e-02 7.98209459e-02 -7.31694281e-01
1.25906125e-01 -1.01823342e+00 -4.60962474e-01 1.73735335e-01
-3.63849223e-01 3.21529359e-01 -4.89698118e-03 5.74795663e-01
-5.44884145e-01 -5.72987080e-01 2.72391289e-01 -2.19307885e-01
-7.85035431e-01 3.03926766e-01 -5.21796644e-01 8.81728232e-02
9.85800922e-01 6.06843494e-02 -4.26242709e-01 -3.77151787e-01
-4.63234365e-01 3.30315024e-01 2.87532687e-01 6.30400717e-01
6.61802948e-01 -1.32536817e+00 -3.46807569e-01 4.44820106e-01
3.66039813e-01 -5.36167063e-02 9.75515321e-02 7.40595758e-01
-3.34901452e-01 2.87032872e-01 4.13920701e-01 -1.89839035e-01
-1.12772322e+00 5.76267838e-01 2.68661559e-01 -8.02985907e-01
-9.71715078e-02 9.86668527e-01 3.38384867e-01 -1.16276824e+00
-2.07340419e-01 1.02008715e-01 -6.96881473e-01 4.05002475e-01
5.77618957e-01 -2.71766454e-01 1.51416093e-01 -5.79635501e-01
-3.82359713e-01 9.66332674e-01 -4.58376288e-01 4.41301428e-02
1.39065850e+00 -3.37149143e-01 -6.78793192e-01 7.29659796e-01
1.46610081e+00 9.99565944e-02 -6.44317269e-01 -1.17694348e-01
-5.64016104e-02 -2.43381321e-01 -1.57714695e-01 -8.08323145e-01
-1.40595841e+00 7.40740001e-01 3.23342800e-01 9.19309333e-02
1.10258603e+00 -2.53623305e-03 1.11986244e+00 2.15414762e-01
2.84359846e-02 -1.04165041e+00 1.73593074e-01 9.09688711e-01
2.45738015e-01 -1.33980453e+00 6.29192963e-02 -4.71492112e-01
-1.02909839e+00 8.31131458e-01 8.30226481e-01 -2.55041689e-01
1.06963313e+00 2.27327287e-01 8.06930780e-01 -5.32726467e-01
-8.93919885e-01 -1.78726345e-01 9.28119197e-02 3.09367806e-01
7.56101906e-01 1.31336838e-01 -7.99088418e-01 1.03001380e+00
-1.91934273e-01 -1.35429613e-02 6.42287016e-01 9.34541702e-01
-5.14302194e-01 -1.35145175e+00 1.56066701e-01 6.29780591e-01
-6.11618519e-01 -6.31987095e-01 -5.54531813e-01 1.53256014e-01
1.24098793e-01 1.07886767e+00 4.11228165e-02 -6.53651476e-01
3.45561951e-01 2.79857248e-01 -2.63197273e-01 -8.07005405e-01
-7.83792019e-01 -2.17889939e-02 6.84190542e-02 -3.98084998e-01
-7.20143318e-01 -3.35207194e-01 -1.29945087e+00 -2.76661813e-02
-3.02255541e-01 4.45298880e-01 9.44963753e-01 1.36345422e+00
3.87550294e-01 8.00208509e-01 9.90985692e-01 -2.40358472e-01
-1.31920636e-01 -1.01343167e+00 -7.44748473e-01 4.17604804e-01
2.59178162e-01 -2.85310954e-01 -6.42793179e-01 1.28933936e-01] | [11.44726276397705, 6.645628929138184] |
16287424-550d-44f7-8391-98744c87b821 | learn-interpretable-word-embeddings | null | null | https://openreview.net/forum?id=Bke02gHYwB | https://openreview.net/pdf?id=Bke02gHYwB | Learn Interpretable Word Embeddings Efficiently with von Mises-Fisher Distribution | Word embedding plays a key role in various tasks of natural language processing. However, the dominant word embedding models don't explain what information is carried with the resulting embeddings. To generate interpretable word embeddings we intend to replace the word vector with a probability density distribution. The insight here is that if we regularize the mixture distribution of all words to be uniform, then we can prove that the inner product between word embeddings represent the point-wise mutual information between words. Moreover, our model can also handle polysemy. Each word's probability density distribution will generate different vectors for its various meanings. We have evaluated our model in several word similarity tasks. Results show that our model can outperform the dominant models consistently in these tasks. | ['Shafei Wang', 'Jian Yang', 'Houqiang Li', 'Liansheng Zhuang', 'Minghong Yao'] | 2019-09-25 | null | null | null | null | ['word-similarity'] | ['natural-language-processing'] | [-1.53844267e-01 1.49147660e-01 -2.29922831e-01 -2.01371908e-01
3.19304015e-03 -5.48977077e-01 8.63282144e-01 3.79392833e-01
-7.54097641e-01 3.59759390e-01 7.00346351e-01 -5.85551918e-01
-3.07312589e-02 -8.98406148e-01 -1.36394650e-01 -5.75439870e-01
3.07833627e-02 5.37184715e-01 6.57742098e-03 -5.30375540e-01
1.40080810e-01 2.96097070e-01 -1.25756776e+00 7.59837106e-02
6.99382544e-01 2.58964896e-01 3.20521057e-01 6.02293372e-01
-7.18266785e-01 2.22697303e-01 -8.36463034e-01 -7.26882935e-01
7.65070096e-02 -1.64094046e-01 -8.26245308e-01 -1.76602766e-01
-5.31526431e-02 6.73039332e-02 -3.75588506e-01 1.32193494e+00
1.57936335e-01 2.30998069e-01 1.18168020e+00 -1.14233208e+00
-1.20584214e+00 7.99109876e-01 -3.29965293e-01 3.21313947e-01
2.84619540e-01 -2.02891201e-01 1.70456159e+00 -9.47587490e-01
4.39373225e-01 1.50521433e+00 4.82603222e-01 5.01326799e-01
-1.51408160e+00 -2.00295463e-01 2.99817264e-01 1.91136852e-01
-1.39323413e+00 1.03476025e-01 6.30637169e-01 -3.56069267e-01
9.86669600e-01 2.58511245e-01 6.52316391e-01 1.17434943e+00
6.72238767e-01 6.24370217e-01 6.05644345e-01 -5.90978026e-01
6.74688742e-02 2.72356600e-01 5.02208889e-01 3.40359181e-01
8.50142479e-01 -2.93152899e-01 -1.31440639e-01 -2.73698568e-01
4.56909090e-01 2.66957194e-01 -1.73435137e-01 -2.43533701e-01
-1.14985394e+00 1.19343066e+00 1.11369260e-01 8.16258192e-01
-3.04486364e-01 3.53089482e-01 2.86855221e-01 2.82121181e-01
5.62089562e-01 8.08006108e-01 -2.95709252e-01 1.24880699e-02
-1.83702290e-01 3.18168938e-01 7.13808656e-01 6.26829326e-01
8.25476706e-01 -9.66550782e-02 -1.99028477e-01 9.55180407e-01
7.22129762e-01 4.21531886e-01 9.57098663e-01 -4.51990068e-01
1.27765954e-01 4.35701519e-01 1.46193460e-01 -1.30106580e+00
-1.40059546e-01 -2.63641000e-01 -5.49289942e-01 -1.98025212e-01
2.46966496e-01 -2.24847961e-02 -8.52509916e-01 1.90956688e+00
-1.72868237e-01 -3.27373147e-02 1.49695233e-01 6.55352235e-01
5.53762496e-01 9.25425887e-01 3.38513613e-01 3.24408635e-02
1.69937658e+00 -4.91028547e-01 -1.15925002e+00 -5.37191212e-01
7.58554101e-01 -7.98651099e-01 1.24944544e+00 -3.92725281e-02
-7.48945236e-01 -4.48614419e-01 -1.10164952e+00 -1.51716337e-01
-6.43778324e-01 -3.99601012e-01 7.34576941e-01 6.09276652e-01
-9.87722635e-01 3.27607632e-01 -6.27633691e-01 -4.00866926e-01
-4.82352301e-02 9.17586535e-02 -4.83755171e-01 -6.07432164e-02
-1.58355975e+00 1.32176733e+00 5.88033676e-01 -2.44767964e-01
-1.28695190e-01 -5.58249831e-01 -1.21457565e+00 1.85900569e-01
-2.63997108e-01 -8.47931087e-01 1.04329062e+00 -7.33518720e-01
-8.75085652e-01 7.48905540e-01 -5.27419031e-01 -3.91272604e-01
-2.11469412e-01 -1.12194836e-01 -5.21395087e-01 -1.59869045e-01
1.57786965e-01 5.80520034e-01 7.00216174e-01 -1.19481814e+00
-2.43777812e-01 -1.04367226e-01 5.88513501e-02 8.92916918e-02
-7.75638342e-01 -7.81777650e-02 -1.49033919e-01 -8.47841442e-01
1.59530323e-02 -6.60141110e-01 -3.20563614e-01 1.96686983e-02
-2.32687257e-02 -6.83354378e-01 4.44564164e-01 -2.25546196e-01
1.53765833e+00 -2.22255802e+00 2.29507595e-01 6.18019290e-02
4.63881940e-01 3.20271522e-01 -4.77830023e-01 7.57873356e-01
-3.03130835e-01 4.57019508e-01 -2.34135062e-01 -3.50339234e-01
4.24202293e-01 8.02037776e-01 -5.90714455e-01 4.42769021e-01
4.36692357e-01 9.33353603e-01 -1.09792209e+00 -2.05264464e-01
2.23286629e-01 6.92003489e-01 -6.01335645e-01 8.90030414e-02
-1.10746153e-01 -5.95288455e-01 -2.98515618e-01 -1.78699538e-01
6.58636093e-01 -1.11556798e-01 4.07977551e-01 -1.34215176e-01
2.29404062e-01 5.66959918e-01 -1.14295650e+00 1.39778078e+00
-6.61939919e-01 8.60630453e-01 -4.83318418e-01 -9.03706431e-01
7.57901371e-01 2.97001094e-01 1.61263570e-01 -1.88648701e-01
2.58014262e-01 -7.25064725e-02 5.06445169e-01 -4.69146371e-01
9.03603852e-01 -6.36324584e-01 -1.49728462e-01 7.80720770e-01
2.05534741e-01 -1.36057317e-01 2.71062940e-01 4.77537930e-01
9.58790123e-01 -5.65954864e-01 3.29053968e-01 -6.06092572e-01
2.58419603e-01 -2.88430065e-01 3.98175687e-01 4.84196633e-01
-2.49780547e-02 4.70024019e-01 7.67940938e-01 -4.02092546e-01
-1.06565237e+00 -1.48980904e+00 -3.78899455e-01 9.73710537e-01
2.01483831e-01 -8.50250125e-01 -2.96288192e-01 -3.78641784e-01
8.14414918e-02 1.25738156e+00 -8.74817014e-01 -4.74560440e-01
-1.29433066e-01 -8.46495986e-01 4.28301752e-01 5.26352227e-01
-3.59096020e-01 -7.91157365e-01 -2.25990683e-01 3.11959594e-01
-2.53735483e-02 -8.05649877e-01 -6.53738797e-01 1.36651322e-01
-5.62384009e-01 -8.51772189e-01 -4.52425361e-01 -8.20224881e-01
7.72111952e-01 4.36814636e-01 1.28902686e+00 -8.30485821e-02
-2.90025711e-01 4.39465076e-01 -5.23460984e-01 -4.78185296e-01
-4.62658584e-01 -1.79078132e-01 3.17470640e-01 -3.89862880e-02
9.90219116e-01 -4.36821073e-01 -2.87549824e-01 -1.11709662e-01
-1.49900997e+00 -3.12227339e-01 1.90767512e-01 8.99343133e-01
1.67276844e-01 1.08828999e-01 2.58078247e-01 -7.61872113e-01
1.25191844e+00 -7.00941980e-01 4.40697856e-02 1.35358036e-01
-6.16165757e-01 5.80881655e-01 3.53165448e-01 -6.84502304e-01
-6.25048280e-01 -5.64584970e-01 -2.74283379e-01 -8.34883228e-02
1.16868922e-03 6.08564198e-01 -7.47163072e-02 7.65947938e-01
4.15847987e-01 1.17506929e-01 1.34201854e-01 -3.73978585e-01
8.76078427e-01 6.05346143e-01 4.17082533e-02 -5.66616058e-01
7.65708864e-01 3.02007765e-01 -3.07969749e-01 -1.12555039e+00
-6.88379586e-01 -4.95102018e-01 -3.82377803e-01 3.18415195e-01
1.14118695e+00 -6.53952718e-01 -2.83900142e-01 -5.36706075e-02
-1.73909783e+00 3.51604521e-01 -4.57905293e-01 6.50882602e-01
7.38543645e-02 3.90444309e-01 -4.61402953e-01 -8.86818647e-01
9.33033600e-03 -1.05545688e+00 8.29621732e-01 -7.09769949e-02
-9.23957527e-01 -1.73781168e+00 3.90377522e-01 -3.07889789e-01
4.67928201e-01 -2.50101775e-01 1.30375433e+00 -1.07280338e+00
2.29261473e-01 -3.55513901e-01 -4.95626815e-02 5.28988123e-01
5.26810944e-01 9.36301947e-02 -7.98828006e-01 3.83560136e-02
-1.29285425e-01 2.30729848e-01 9.71746385e-01 7.82453045e-02
7.69418359e-01 -3.28440130e-01 -1.90709531e-01 1.50797457e-01
1.34114730e+00 -1.09753311e-01 6.69188857e-01 -7.20023457e-03
7.22054839e-01 7.22890496e-01 9.45217013e-02 2.22693160e-01
3.66272360e-01 4.03295934e-01 2.32693672e-01 2.76954442e-01
1.01325542e-01 -4.62677002e-01 5.48584104e-01 1.10709167e+00
2.64409721e-01 -5.79107106e-01 -9.24054325e-01 8.57187450e-01
-1.53884828e+00 -1.04306567e+00 -3.48638505e-01 1.93176532e+00
7.12687969e-01 1.74918890e-01 -2.15726569e-01 1.60317808e-01
6.04241729e-01 5.65883875e-01 5.88419475e-03 -9.46406782e-01
-1.85296789e-01 5.34485400e-01 2.85111666e-01 9.51067567e-01
-6.76787198e-01 9.14288700e-01 7.39393091e+00 6.24230504e-01
-6.96952879e-01 1.08842038e-01 9.13584456e-02 2.31853515e-01
-1.29368126e+00 -8.16147998e-02 -4.55088317e-01 4.24416602e-01
8.67497027e-01 -7.60775805e-01 -9.68354940e-02 5.14939308e-01
-5.74691035e-02 1.58978581e-01 -1.13563979e+00 7.42071211e-01
2.27004379e-01 -1.03809071e+00 5.38012147e-01 1.80432096e-01
4.78110760e-01 -1.98463321e-01 1.46235391e-01 2.16677919e-01
5.21717012e-01 -1.27432895e+00 3.65199238e-01 3.25582892e-01
4.22920316e-01 -7.49298036e-01 9.26374376e-01 1.81968719e-01
-1.06727719e+00 1.73119739e-01 -8.26112926e-01 -4.13405687e-01
2.36785740e-01 7.80879080e-01 -8.31597507e-01 2.01042876e-01
2.28293855e-02 6.79909348e-01 -3.62488925e-01 4.91918921e-01
-5.27076006e-01 4.88885313e-01 -8.17275941e-02 -4.10598844e-01
2.39499584e-01 -5.61259687e-01 6.00368261e-01 1.33817995e+00
3.07266653e-01 -1.12989776e-01 -1.59936473e-01 1.03228629e+00
2.96901329e-03 2.02899784e-01 -9.71725285e-01 -4.79165256e-01
4.70141709e-01 9.43100154e-01 -5.42381346e-01 -3.35310280e-01
-5.58327556e-01 1.04035580e+00 2.17960358e-01 3.05472672e-01
-6.22800887e-01 -5.79318762e-01 1.67693496e+00 2.28611063e-02
2.26932734e-01 -5.37961006e-01 -1.94632486e-01 -1.18415034e+00
-3.94434854e-02 -2.85203427e-01 1.04168482e-01 -7.37706065e-01
-1.84780967e+00 6.15207136e-01 1.42944574e-01 -7.70530403e-01
-3.10726017e-01 -1.09482872e+00 -8.91609371e-01 1.14644432e+00
-1.19261754e+00 -5.65653741e-01 4.04536694e-01 1.58245891e-01
6.07469082e-01 -1.02747813e-01 1.12847579e+00 1.21898048e-01
-4.06995714e-01 4.88547713e-01 -1.96042024e-02 2.82962829e-01
5.24757922e-01 -1.42173111e+00 6.79434597e-01 7.03136504e-01
6.59646332e-01 1.18542480e+00 1.26468778e+00 -3.86115670e-01
-1.43663061e+00 -8.92336249e-01 1.54682040e+00 -6.82318687e-01
1.24301624e+00 -3.80411059e-01 -1.05696070e+00 7.32952893e-01
5.57412744e-01 -4.49588895e-03 1.04822969e+00 3.40115935e-01
-7.59873450e-01 2.04497412e-01 -8.01845014e-01 8.32987070e-01
8.47627699e-01 -6.60009861e-01 -1.43650925e+00 3.40470910e-01
1.21799636e+00 2.93730825e-01 -6.41619325e-01 -1.01229563e-01
4.95204657e-01 -6.24859452e-01 9.67427015e-01 -1.18645728e+00
5.74617445e-01 -1.87334001e-01 -5.05854964e-01 -1.77413678e+00
-6.06184542e-01 -1.34814292e-01 8.03696439e-02 1.11443174e+00
6.02314532e-01 -9.77982581e-01 2.72691160e-01 7.47266412e-01
1.68767169e-01 -5.54659843e-01 -7.60356545e-01 -9.38656271e-01
4.63331789e-01 -9.11754906e-01 6.84863269e-01 9.09461558e-01
2.95089573e-01 6.35686278e-01 -1.79721434e-02 5.30622005e-02
2.89978296e-01 -3.31949323e-01 1.64939493e-01 -1.32408404e+00
-3.10797155e-01 -5.88575780e-01 -8.82243812e-01 -1.12714326e+00
6.60268188e-01 -1.12999225e+00 -1.29090101e-01 -1.59028649e+00
1.95867464e-01 -1.50419399e-01 -4.21327591e-01 1.91512629e-01
-4.17242825e-01 1.73766375e-01 1.39807358e-01 -2.02503249e-01
-2.33427510e-01 7.15856314e-01 8.99442613e-01 -1.54812858e-01
1.55757472e-01 -5.01786590e-01 -9.56528127e-01 7.32579529e-01
8.55354130e-01 -5.85106492e-01 -6.26030266e-01 -7.82751441e-01
6.90886736e-01 -6.63513839e-01 1.15592726e-01 -2.76181310e-01
-1.68556005e-01 -2.32751995e-01 -6.71897009e-02 -5.90216927e-02
3.07215333e-01 -8.68662953e-01 -3.24764773e-02 4.29341435e-01
-4.58194733e-01 5.86986125e-01 2.36236509e-02 5.80201626e-01
-1.74579516e-01 -4.71428126e-01 4.58857566e-01 6.22479096e-02
-4.16914165e-01 1.11429088e-01 -8.47799122e-01 2.48811871e-01
7.53613472e-01 -5.46002127e-02 2.63171978e-02 -2.75956959e-01
-4.86635685e-01 -4.01347727e-02 4.35848653e-01 8.00979018e-01
6.79238677e-01 -1.67988729e+00 -7.05586433e-01 3.27153087e-01
3.28856558e-01 -3.85889083e-01 -1.95319280e-01 2.44605139e-01
-3.54728848e-01 2.42580771e-01 5.29867783e-02 -2.60759979e-01
-1.00467038e+00 6.56500459e-01 1.32017031e-01 -3.22675295e-02
-3.30794632e-01 7.19117761e-01 2.87937880e-01 -2.84858435e-01
-1.42441005e-01 -4.12534148e-01 -3.82016927e-01 3.57091427e-01
7.66552269e-01 8.27727914e-02 -2.91186154e-01 -8.13941717e-01
-3.88529658e-01 4.65294808e-01 -1.49310738e-01 -4.39113885e-01
1.24717546e+00 -3.34570035e-02 -4.55908805e-01 9.96729970e-01
1.62291288e+00 1.20997084e-02 -3.55637014e-01 -1.04252972e-01
5.50098307e-02 -5.68551779e-01 -1.28701702e-01 -1.42599521e-02
-6.52252197e-01 9.78071988e-01 2.20886096e-01 6.57045841e-01
4.43983853e-01 2.45994672e-01 7.39432275e-01 2.80557930e-01
1.46088183e-01 -7.37684190e-01 -1.05304778e-01 7.38316178e-01
6.89772785e-01 -1.07382309e+00 -3.77803855e-02 -2.69278049e-01
-6.49858117e-01 1.11556149e+00 2.73752213e-01 -4.11806822e-01
9.72723365e-01 1.65728644e-01 6.20010868e-02 -1.55407339e-01
-9.73992348e-01 -3.66667420e-01 4.00184810e-01 6.27365530e-01
8.92905354e-01 3.25126171e-01 -8.68478119e-01 7.16944814e-01
-3.70786846e-01 -7.22419202e-01 5.72575331e-01 4.11492437e-01
-6.09551370e-01 -1.55636549e+00 -2.29804173e-01 3.92632872e-01
-2.67170846e-01 -3.91568273e-01 -2.73143768e-01 6.26604557e-01
7.09185824e-02 9.41934943e-01 5.32477021e-01 -4.58269328e-01
1.46757245e-01 3.69672686e-01 3.56162816e-01 -1.00274312e+00
-5.83238229e-02 -3.66640240e-01 -6.09303303e-02 -2.06323683e-01
-4.31881286e-02 -4.60748464e-01 -1.32314038e+00 -4.50323105e-01
-1.32299647e-01 4.63106006e-01 6.63677752e-01 1.05350518e+00
1.39056444e-01 5.94318807e-01 3.32605839e-01 -3.89402032e-01
-6.75767541e-01 -1.01821136e+00 -7.54704475e-01 6.99232399e-01
1.48034990e-01 -5.97796082e-01 -7.31916368e-01 -1.46187037e-01] | [10.473849296569824, 8.770050048828125] |
9b4b077f-6bdf-4987-98dc-d1886d260ce1 | a-perspective-on-objects-and-systematic | 1906.01035 | null | https://arxiv.org/abs/1906.01035v1 | https://arxiv.org/pdf/1906.01035v1.pdf | A Perspective on Objects and Systematic Generalization in Model-Based RL | In order to meet the diverse challenges in solving many real-world problems, an intelligent agent has to be able to dynamically construct a model of its environment. Objects facilitate the modular reuse of prior knowledge and the combinatorial construction of such models. In this work, we argue that dynamically bound features (objects) do not simply emerge in connectionist models of the world. We identify several requirements that need to be fulfilled in overcoming this limitation and highlight corresponding inductive biases. | ['Jürgen Schmidhuber', 'Sjoerd van Steenkiste', 'Klaus Greff'] | 2019-06-03 | null | null | null | null | ['systematic-generalization'] | ['reasoning'] | [ 2.65750587e-01 4.12437469e-01 -1.33977324e-01 -4.16932672e-01
1.43561319e-01 -6.52032375e-01 1.08241737e+00 2.34186620e-01
-5.85342646e-01 7.81681657e-01 9.67324376e-02 -6.81091845e-02
-7.64812529e-01 -8.61942351e-01 -5.66113949e-01 -4.66262400e-01
-2.52737850e-01 7.87494302e-01 3.97033691e-01 -5.38991392e-01
6.11396730e-01 8.01700115e-01 -1.66315401e+00 5.80175444e-02
6.40412509e-01 6.10461175e-01 8.61524999e-01 3.67355764e-01
-1.79174960e-01 6.58507645e-01 -1.96371064e-01 -3.17970812e-01
3.39974284e-01 -1.74467072e-01 -1.03150594e+00 9.17022750e-02
-1.33055896e-01 1.13096215e-01 -1.06580876e-01 9.76692140e-01
1.64615661e-02 3.60986024e-01 8.26272190e-01 -1.19464886e+00
-6.16755068e-01 9.61107135e-01 9.57680643e-02 4.56259698e-01
4.65980709e-01 1.23730615e-01 1.12543452e+00 -6.83482707e-01
8.54547083e-01 1.23289180e+00 2.69851685e-01 7.05461264e-01
-1.40858114e+00 -1.67282045e-01 8.66305292e-01 2.51951069e-01
-1.39924526e+00 -5.07501066e-01 1.06003261e+00 -3.06816071e-01
1.01257288e+00 3.10707211e-01 1.12312388e+00 1.08644199e+00
1.85080826e-01 5.82461715e-01 1.10701907e+00 -7.50495970e-01
5.75029433e-01 3.95251125e-01 1.56689622e-02 4.54689682e-01
5.97027898e-01 2.64756918e-01 -7.45419502e-01 -2.14866355e-01
9.31823134e-01 -3.32047373e-01 -1.33014828e-01 -8.27027500e-01
-1.21448219e+00 7.48484075e-01 6.32235765e-01 8.79457235e-01
-4.00199890e-01 2.50097811e-01 -7.27149099e-02 1.46184772e-01
-1.98879123e-01 1.41143072e+00 -5.59853077e-01 1.75972313e-01
-2.75756031e-01 2.10348517e-01 7.48933673e-01 9.43664789e-01
8.65166128e-01 -2.46433049e-01 6.21102393e-01 2.06660688e-01
5.64926744e-01 1.19866379e-01 5.96288323e-01 -7.72288561e-01
9.68515873e-02 6.28958583e-01 3.17128956e-01 -9.83177245e-01
-5.93333364e-01 -5.58229744e-01 -2.98743993e-01 -1.02365054e-01
1.96573764e-01 1.76389635e-01 -5.47937155e-01 1.98241580e+00
3.86284769e-01 -2.67653912e-01 2.30797797e-01 4.61745799e-01
3.55675757e-01 2.78856754e-01 4.28594530e-01 -3.41458479e-03
8.97741973e-01 -4.46221888e-01 -4.87241358e-01 -8.02209437e-01
5.87937832e-01 -2.76093900e-01 7.33292937e-01 2.94551224e-01
-9.84802604e-01 -4.65209872e-01 -1.29750466e+00 1.18377611e-01
-6.42240882e-01 -4.52251941e-01 1.38835454e+00 5.33262253e-01
-1.11580682e+00 4.23495948e-01 -7.13297367e-01 -7.44928181e-01
3.08895975e-01 7.20814347e-01 -4.94114876e-01 4.49680626e-01
-1.05801392e+00 1.54277170e+00 1.05849922e+00 3.57910573e-01
-8.31936002e-01 -1.08698919e-01 -8.28632951e-01 1.20895347e-02
3.59595507e-01 -9.22528565e-01 9.33928847e-01 -1.33334243e+00
-1.15821075e+00 1.05453324e+00 -3.05836536e-02 -4.49247867e-01
1.91014603e-01 -5.25750592e-02 -1.68869942e-01 1.42453304e-02
-3.26421857e-01 7.77695775e-01 5.18896997e-01 -1.83285689e+00
-7.57315993e-01 -3.34318370e-01 6.58693492e-01 5.00781000e-01
-3.20361257e-01 -1.64932728e-01 -1.31016821e-01 -2.11709321e-01
4.31312442e-01 -1.05405414e+00 -6.09141767e-01 -3.38436484e-01
-3.14945541e-02 -2.21212864e-01 2.77957171e-01 1.26304865e-01
8.80007088e-01 -2.13637853e+00 2.82758236e-01 4.47073579e-01
1.96158499e-01 -9.28245634e-02 -2.88555115e-01 4.79546428e-01
9.18232426e-02 2.50911683e-01 2.11902574e-01 -4.70998324e-02
2.38104686e-01 5.33271611e-01 -3.88815850e-01 3.40134621e-01
2.47382522e-01 1.03599036e+00 -1.02199340e+00 -4.89130586e-01
3.64721477e-01 1.74725816e-01 -5.38931489e-01 -4.44049053e-02
-4.75369334e-01 4.09271747e-01 -9.63380396e-01 2.39841402e-01
2.95026243e-01 -2.27266818e-01 6.68190479e-01 3.02467823e-01
-6.76553249e-02 5.71060240e-01 -1.30440056e+00 1.69567251e+00
-4.44650620e-01 4.72916216e-01 -9.94346589e-02 -1.08482063e+00
8.26649964e-01 2.56770462e-01 2.22923666e-01 -7.33207643e-01
2.72362351e-01 3.42136890e-01 5.81906319e-01 -5.20069599e-01
3.54215890e-01 -5.34503996e-01 6.06322251e-02 5.16090035e-01
6.22198060e-02 -3.84853572e-01 -8.18042317e-04 5.07404990e-02
7.71533430e-01 9.46127698e-02 4.10470575e-01 -7.62357175e-01
4.35467064e-01 9.20473877e-03 6.61038160e-01 9.42283452e-01
7.60654360e-02 1.07178688e-02 3.08675826e-01 -8.53790879e-01
-7.57938027e-01 -1.12095940e+00 -3.33446592e-01 1.03967714e+00
4.70532238e-01 -1.54883623e-01 -4.05411452e-01 -5.13130546e-01
-3.00534248e-01 8.69115174e-01 -9.86149132e-01 -3.18027794e-01
-5.72028518e-01 -8.66266072e-01 5.44488952e-02 4.23057914e-01
5.24175353e-02 -1.29830730e+00 -1.25336778e+00 4.03318882e-01
1.04426682e-01 -7.57825613e-01 4.65212435e-01 7.13162422e-01
-9.73772049e-01 -8.50703061e-01 3.01964045e-01 -9.24416721e-01
1.07674217e+00 1.91787884e-01 1.41818237e+00 3.65731776e-01
-6.28233254e-02 7.07859635e-01 -3.58068973e-01 -5.27032912e-01
-2.75205195e-01 3.15670729e-01 1.05685152e-01 -1.55389845e-01
4.65359837e-01 -9.85076547e-01 -4.01564032e-01 1.97130710e-01
-1.11105978e+00 1.97039485e-01 6.42945111e-01 7.19878674e-01
3.37541163e-01 3.34145904e-01 7.43243694e-01 -7.49317944e-01
5.01180232e-01 -6.45728350e-01 -5.74034810e-01 4.23365057e-01
-3.69492739e-01 2.73673713e-01 3.64741623e-01 -5.69975019e-01
-1.22389925e+00 1.92156002e-01 1.88082397e-01 4.77726489e-01
-2.62315482e-01 6.88641787e-01 -5.57401776e-01 -1.11139588e-01
7.68253267e-01 2.55987138e-01 -3.74053925e-01 -3.69838744e-01
4.29398447e-01 2.15128615e-01 -2.19164658e-02 -1.21514404e+00
8.14782679e-01 5.92068970e-01 1.84888005e-01 -5.88432193e-01
-7.78285861e-01 -2.55186092e-02 -1.10141647e+00 -9.00057182e-02
4.90817487e-01 -6.76474512e-01 -4.05487150e-01 -1.22508094e-01
-9.93966877e-01 -4.10216719e-01 -5.82528472e-01 3.51613432e-01
-8.19950700e-01 -1.47953525e-01 5.39030246e-02 -7.85301626e-01
3.39795589e-01 -8.67024839e-01 2.28257746e-01 3.46970707e-01
-5.20260811e-01 -1.26591587e+00 6.89286441e-02 -8.80565643e-02
4.04435426e-01 2.75764875e-02 1.01663733e+00 -9.38580394e-01
-7.80594230e-01 3.26012261e-02 1.88914180e-01 -2.42677405e-01
2.38062173e-01 -1.95229184e-02 -1.01407599e+00 -2.22896293e-01
3.93640697e-01 -3.01866084e-01 7.10713446e-01 1.81303233e-01
8.09638143e-01 -2.36606106e-01 -6.30536556e-01 3.35168093e-01
1.55059850e+00 5.68083227e-01 4.64084536e-01 6.67763472e-01
1.08234018e-01 1.04450321e+00 4.34530348e-01 3.13683003e-01
4.90771621e-01 3.75020087e-01 4.46466744e-01 2.90734589e-01
1.75044015e-01 -3.18455011e-01 -1.29126668e-01 5.11971831e-01
-2.01103821e-01 -8.29927176e-02 -1.04351819e+00 8.31773341e-01
-1.91379344e+00 -9.88519907e-01 2.15958253e-01 1.96488738e+00
9.28981304e-01 3.96577477e-01 -3.32593769e-01 -1.64501122e-04
4.47032064e-01 -7.76397809e-02 -5.61149120e-01 -3.28753382e-01
-1.75969854e-01 -2.81119235e-02 -1.04367152e-01 5.31323493e-01
-9.03288782e-01 1.05934870e+00 7.73658228e+00 2.52480239e-01
-9.82688665e-01 -1.17629021e-01 3.57474297e-01 9.86621678e-02
-6.56980336e-01 4.81073320e-01 -7.10762441e-01 9.93633866e-02
7.38924444e-01 -2.48386472e-01 5.02235770e-01 7.11843729e-01
-2.66108245e-01 -4.76785243e-01 -1.59071994e+00 4.75512177e-01
8.00556224e-03 -1.25457251e+00 2.12217912e-01 1.28949523e-01
6.99319482e-01 -2.20094644e-03 2.11079627e-01 2.09257737e-01
7.30861962e-01 -1.10499740e+00 8.71218443e-01 5.77620506e-01
-4.27791588e-02 -5.39218128e-01 3.23364615e-01 5.14778554e-01
-9.02430713e-01 -2.47241065e-01 -4.12490517e-01 -4.48921770e-01
-2.44483892e-02 3.65418226e-01 -7.78817892e-01 2.59638339e-01
4.13802266e-01 2.35985383e-01 -7.53059149e-01 1.07924187e+00
-3.52742523e-01 2.05976162e-02 -4.69993412e-01 -3.85210991e-01
2.89283246e-01 9.53537002e-02 4.07574385e-01 8.31504166e-01
6.96651712e-02 4.95083630e-02 2.09876835e-01 9.22151446e-01
2.77767420e-01 -7.12466100e-03 -7.98139274e-01 1.36652797e-01
5.61090350e-01 1.05520272e+00 -1.04048538e+00 -1.92586914e-01
-2.34158099e-01 3.72328192e-01 5.47755182e-01 1.11790895e-01
-5.55740416e-01 1.45958290e-01 5.23280084e-01 3.84040810e-02
1.11637719e-01 -5.87900698e-01 -4.54665393e-01 -1.17779660e+00
-3.79329510e-02 -6.58648252e-01 1.49402797e-01 -6.56651080e-01
-1.22907543e+00 5.06097913e-01 1.00566372e-01 -7.54307806e-01
-2.46348843e-01 -6.07572615e-01 -3.05900276e-01 5.28887033e-01
-1.49469066e+00 -1.33585799e+00 1.00551851e-01 6.71927333e-01
2.68797636e-01 9.58019197e-02 1.03308487e+00 -4.18061644e-01
-1.93537250e-01 -8.04133788e-02 -8.54219496e-02 -2.30009019e-01
1.47690386e-01 -1.14319146e+00 3.37348849e-01 7.01565564e-01
5.93540907e-01 1.21830344e+00 9.56117272e-01 -5.77934921e-01
-1.49676251e+00 -4.43398267e-01 1.02306747e+00 -9.39759970e-01
7.78315127e-01 -4.27823067e-01 -6.66708767e-01 1.04139316e+00
1.09922484e-01 -1.95104316e-01 8.20243597e-01 6.01320386e-01
-5.02278805e-01 -5.79040907e-02 -1.08242178e+00 8.02070856e-01
1.28660786e+00 -3.69512826e-01 -1.27198327e+00 7.64865428e-02
2.53589839e-01 1.29350945e-01 -4.81620610e-01 4.04500127e-01
4.92796183e-01 -8.25286686e-01 9.64775324e-01 -9.50260758e-01
-8.70095715e-02 -4.47177112e-01 -2.39815623e-01 -1.32822037e+00
-7.38236845e-01 -5.36133289e-01 9.81464386e-02 1.07528889e+00
6.35445654e-01 -9.30688202e-01 4.64238495e-01 1.32688463e+00
1.48977518e-01 -4.66604739e-01 -9.87866640e-01 -7.88753211e-01
2.06257522e-01 -4.72404391e-01 7.19531834e-01 1.04613030e+00
4.56376672e-01 2.85240322e-01 7.32239336e-02 3.67603600e-02
4.09851819e-01 2.37204790e-01 4.43286777e-01 -1.76233208e+00
-1.43265650e-01 -4.37681705e-01 -4.64365155e-01 -7.33546436e-01
3.08327556e-01 -7.47114420e-01 -1.74302757e-02 -1.54995370e+00
4.67598379e-01 -1.15834713e+00 -7.07578242e-01 4.20937538e-01
2.19155952e-01 -4.13615219e-02 3.60682815e-01 2.21822038e-01
-7.82888651e-01 4.77636755e-01 1.09222293e+00 7.96379298e-02
-1.58291638e-01 -3.41253668e-01 -1.22849941e+00 1.14500082e+00
1.24437225e+00 -5.72159231e-01 -6.00133777e-01 -7.30704188e-01
1.08719397e+00 -4.48954105e-01 2.86414891e-01 -1.03774333e+00
2.75895983e-01 -7.76750982e-01 7.49175370e-01 -1.01724871e-01
4.93228912e-01 -1.16225088e+00 3.78599554e-01 6.00824177e-01
-5.11333764e-01 -3.35824713e-02 5.29556051e-02 4.54281151e-01
4.67068963e-02 -6.68009698e-01 6.61192417e-01 -5.82554340e-01
-9.43054140e-01 -9.23060626e-02 -6.01690233e-01 -1.76396906e-01
1.19254935e+00 -2.91059583e-01 -2.05289170e-01 8.91237184e-02
-8.62666786e-01 1.04839876e-01 8.19137096e-01 5.70245624e-01
5.11597276e-01 -1.07826579e+00 -2.70550758e-01 1.70084476e-01
2.07106501e-01 -7.16346428e-02 4.57655489e-02 4.13185686e-01
-3.91616404e-01 5.89785993e-01 -5.73739111e-01 2.69614626e-03
-6.93070769e-01 7.96080709e-01 2.57758588e-01 -1.30519032e-01
-4.84829247e-01 8.89710128e-01 4.68090415e-01 -1.69374317e-01
-4.45970222e-02 -9.08901691e-02 -4.76198226e-01 1.66883901e-01
4.20072317e-01 -2.10983366e-01 -9.62185785e-02 -6.13903105e-01
-4.52797681e-01 3.25515002e-01 -1.88163534e-01 -4.27190721e-01
1.47667277e+00 -4.21368718e-01 -2.84396887e-01 5.25735617e-01
4.66458648e-01 6.53451979e-02 -1.15095615e+00 -1.34747520e-01
2.65561581e-01 -5.62711477e-01 -1.24162808e-01 -9.15268421e-01
-6.31452441e-01 4.78273541e-01 2.84901410e-01 4.54464972e-01
9.55746889e-01 3.32459807e-01 -2.54010875e-02 9.08984005e-01
1.02287269e+00 -1.36183596e+00 2.59128083e-02 4.85005379e-01
9.95937884e-01 -9.44601476e-01 9.46978256e-02 -2.73583621e-01
-4.05511349e-01 1.04980326e+00 5.45304060e-01 -2.19603464e-01
6.03208959e-01 7.23944157e-02 -1.58736765e-01 -2.97555327e-01
-1.10280728e+00 -3.76138270e-01 1.21584926e-02 9.67916250e-01
2.36740693e-01 -3.81009579e-02 -3.98079753e-01 3.91932994e-01
-3.27978164e-01 -3.97780538e-01 4.29359436e-01 1.07485712e+00
-5.85694373e-01 -1.29825032e+00 -3.65200490e-01 1.26522750e-01
-1.42416090e-01 -9.69838537e-03 -7.03569055e-01 8.70720804e-01
5.15420914e-01 1.05062962e+00 2.73115430e-02 -5.34765497e-02
-8.82626623e-02 1.28267348e-01 8.48404586e-01 -7.22858310e-01
-5.59871495e-01 -2.38121793e-01 9.37571973e-02 -3.86721015e-01
-7.14995205e-01 -8.08100283e-01 -1.25326335e+00 1.19106628e-01
-4.61797953e-01 2.37137452e-01 8.35352004e-01 9.61826921e-01
2.70030320e-01 3.76370311e-01 4.72256541e-01 -7.60394275e-01
-9.04941335e-02 -5.97588837e-01 -4.60284323e-01 3.91797394e-01
1.22559719e-01 -1.13475502e+00 -4.88184243e-01 1.34258866e-01] | [9.281251907348633, 6.710438251495361] |
f4effb82-9923-4b0c-9ed4-58a647c1b4b4 | collaborative-ranking-with-17-parameters | null | null | http://papers.nips.cc/paper/4829-collaborative-ranking-with-17-parameters | http://papers.nips.cc/paper/4829-collaborative-ranking-with-17-parameters.pdf | Collaborative Ranking With 17 Parameters | The primary application of collaborate filtering (CF) is to recommend a small set of items to a user, which entails ranking. Most approaches, however, formulate the CF problem as rating prediction, overlooking the ranking perspective. In this work we present a method for collaborative ranking that leverages the strengths of the two main CF approaches, neighborhood- and model-based. Our novel method is highly efficient, with only seventeen parameters to optimize and a single hyperparameter to tune, and beats the state-of-the-art collaborative ranking methods. We also show that parameters learned on one dataset yield excellent results on a very different dataset, without any retraining. | ['Richard S. Zemel', 'Maksims Volkovs'] | 2012-12-01 | null | null | null | neurips-2012-12 | ['collaborative-ranking'] | ['graphs'] | [-9.95578840e-02 -2.96297491e-01 -5.66091716e-01 -6.41884089e-01
-9.05346870e-01 -7.98873603e-01 5.42992592e-01 1.82782620e-01
-1.53100222e-01 7.23268986e-01 5.49094081e-01 -3.09733331e-01
-7.51602411e-01 -8.60640883e-01 -2.50720918e-01 -3.14812422e-01
-5.51198684e-02 7.72424519e-01 4.92808104e-01 -4.83088583e-01
7.34641135e-01 1.82782933e-01 -1.76258922e+00 7.55473495e-01
1.06217074e+00 1.10883164e+00 5.03197014e-02 3.61704409e-01
-4.64206301e-02 6.79892838e-01 -5.62284648e-01 -7.47460246e-01
2.99296051e-01 -1.78813815e-01 -7.30352044e-01 -3.30353796e-01
4.95753855e-01 -2.46411249e-01 -2.43613590e-02 6.82834268e-01
5.40038466e-01 6.39214456e-01 6.86301947e-01 -8.00777078e-01
-9.78293121e-01 7.48902261e-01 -7.68190771e-02 6.16725460e-02
7.15120971e-01 -8.41476500e-01 1.56907427e+00 -1.10575974e+00
4.12200153e-01 9.16453004e-01 7.61122406e-01 2.40550593e-01
-9.57388222e-01 -4.29165065e-01 5.19965351e-01 2.33902037e-01
-1.15661967e+00 -2.48666301e-01 5.43365538e-01 -4.05722469e-01
7.39642441e-01 3.65329564e-01 6.71004117e-01 6.13808751e-01
-1.60681561e-01 6.42861426e-01 1.17504346e+00 -3.84655178e-01
3.39090347e-01 3.38006765e-01 2.59246767e-01 3.05972099e-01
2.68187165e-01 2.72896171e-01 -5.80020964e-01 -6.66901290e-01
5.01171768e-01 5.46294510e-01 -3.10364403e-02 -4.11600143e-01
-9.56103206e-01 1.16630042e+00 1.21518180e-01 1.73695177e-01
-7.77950063e-02 -2.61576682e-01 1.25994891e-01 7.55569220e-01
8.27559114e-01 6.42071843e-01 -8.31313431e-01 5.21534458e-02
-8.94208968e-01 2.96096087e-01 1.00284064e+00 5.67170858e-01
3.43821347e-01 -4.82216001e-01 -3.08478355e-01 1.06343734e+00
5.35100996e-01 3.18834722e-01 3.56828094e-01 -9.11406696e-01
2.30220377e-01 3.75320703e-01 4.92922246e-01 -1.00114417e+00
-1.59811541e-01 -8.09844196e-01 -3.23404968e-01 2.19662823e-02
3.11937481e-01 -6.45616725e-02 -5.34850001e-01 1.31841445e+00
3.34858865e-01 5.68350665e-02 -1.97229549e-01 7.94858754e-01
8.32398713e-01 3.67046058e-01 -2.27415904e-01 -3.17823321e-01
8.36532116e-01 -1.40663528e+00 -4.17862654e-01 6.77370653e-02
5.51411450e-01 -1.05051100e+00 1.08349824e+00 9.28068519e-01
-1.16319215e+00 -6.03036106e-01 -8.33968639e-01 2.24143833e-01
-4.10093009e-01 1.06401801e-01 1.23158824e+00 8.21802676e-01
-1.16638148e+00 1.02478051e+00 -1.10952882e-02 -1.44956753e-01
-1.33626778e-02 7.11947143e-01 7.81855825e-03 -1.84561178e-01
-1.27068567e+00 9.67402697e-01 -4.05241139e-02 -1.70209184e-01
-5.52218199e-01 -7.28773117e-01 4.65098594e-04 9.29733962e-02
3.72034043e-01 -8.38860035e-01 1.22152269e+00 -8.01481068e-01
-1.78489065e+00 2.20150694e-01 -5.31536154e-02 -4.98911552e-02
4.25095558e-01 -7.00957000e-01 -8.46283555e-01 -2.70990849e-01
-1.68731630e-01 1.31853567e-02 7.16161907e-01 -1.34837961e+00
-8.79821897e-01 -1.68983936e-01 3.17874372e-01 2.88924515e-01
-7.08323419e-01 4.13946629e-01 -3.52744579e-01 -6.98413670e-01
-3.14574167e-02 -7.09532678e-01 -4.54438001e-01 -4.36730325e-01
1.55604467e-01 -3.73771191e-01 2.79052168e-01 -1.81384712e-01
1.75891054e+00 -1.60377407e+00 1.18288532e-01 6.06067121e-01
7.07669780e-02 2.90247738e-01 -2.75988728e-01 7.35163450e-01
2.27729335e-01 1.46221161e-01 4.52531487e-01 -2.02796936e-01
-4.94663827e-02 6.15787916e-02 -4.73871320e-01 2.10740939e-01
-5.99450052e-01 7.52641559e-01 -1.17128921e+00 -2.69002944e-01
-7.64917359e-02 3.95255029e-01 -9.84814405e-01 2.36962810e-01
-1.13119908e-01 2.88028151e-01 -4.78856325e-01 5.60039163e-01
4.06564862e-01 -4.43163157e-01 5.72464883e-01 -1.08600199e-01
-6.21049926e-02 5.14628172e-01 -1.33466399e+00 1.56728828e+00
-3.99748445e-01 -1.44382566e-01 -2.28077501e-01 -7.48809278e-01
9.98666763e-01 2.15721950e-01 6.46021903e-01 -3.86640370e-01
-1.05905123e-02 3.88535798e-01 -1.95951715e-01 -7.94453472e-02
5.62466383e-01 1.46384031e-01 8.09821859e-02 7.92826772e-01
-7.92283565e-02 2.75083572e-01 3.19478691e-01 3.38546783e-01
1.12707996e+00 -1.07867830e-01 1.58458486e-01 -2.19565809e-01
3.28480005e-01 -1.87267154e-01 4.21732306e-01 1.23398900e+00
3.01968306e-01 4.85912383e-01 -2.76638055e-03 -4.80645597e-01
-7.52012253e-01 -9.01455879e-01 1.62802096e-02 1.69525313e+00
1.40531391e-01 -8.66869330e-01 -5.93666956e-02 -9.92401779e-01
3.71208757e-01 4.02738780e-01 -6.16043210e-01 1.46327123e-01
-2.10555181e-01 -7.36473262e-01 -2.84808427e-01 5.36074400e-01
-1.55750647e-01 -6.49524570e-01 1.72415525e-01 2.91771382e-01
-3.34525406e-02 -3.99861038e-01 -5.70898950e-01 1.36860877e-01
-1.24231565e+00 -9.27718580e-01 -4.31322932e-01 -5.15097201e-01
5.86639822e-01 7.48267472e-01 1.59079039e+00 4.22616631e-01
4.27968651e-01 1.94175288e-01 -8.37866843e-01 -1.72241926e-01
1.07191093e-01 1.66134611e-01 1.94214866e-01 -4.09277417e-02
2.87264854e-01 -6.64360642e-01 -7.40578115e-01 9.37133670e-01
-4.83692110e-01 -4.62199152e-01 5.74604750e-01 8.18983555e-01
7.43798077e-01 1.13217056e-01 8.88382852e-01 -1.50439239e+00
9.74588513e-01 -6.96066201e-01 -3.09627324e-01 3.77191454e-01
-1.36828017e+00 -1.66327775e-01 7.68645108e-01 -6.96335554e-01
-8.65674496e-01 -6.28466532e-02 -8.72284174e-02 -9.33678597e-02
1.83811098e-01 7.16723859e-01 1.21704198e-01 -1.75737351e-01
6.74307525e-01 -3.65293443e-01 -4.19177532e-01 -9.59773302e-01
7.05119371e-01 5.32834172e-01 2.03904808e-01 -4.54340130e-01
7.63816357e-01 3.04372072e-01 -2.42836654e-01 1.07486799e-01
-1.50469637e+00 -8.55311513e-01 -6.14694357e-01 -2.65292317e-01
3.40739638e-02 -7.95739949e-01 -4.93043363e-01 -2.18342483e-01
-5.39128184e-01 -1.28733562e-02 -2.81843841e-01 6.37385786e-01
-3.25760752e-01 2.06497222e-01 -5.04439294e-01 -6.33581102e-01
-3.86804610e-01 -4.67821270e-01 5.57999194e-01 3.88076529e-02
-5.35100065e-02 -9.35081959e-01 4.80910748e-01 6.80274487e-01
7.87903070e-01 -3.72875661e-01 7.31421411e-01 -1.11944580e+00
-1.49001703e-01 -4.29268807e-01 -2.09893864e-02 2.11171910e-01
6.97408989e-02 4.88863047e-03 -6.43970132e-01 -5.38607001e-01
-2.87979066e-01 -2.62996495e-01 9.94331241e-01 2.22978503e-01
1.11044943e+00 -2.59076864e-01 -2.44762734e-01 4.44829404e-01
1.37749612e+00 2.91523159e-01 5.26633203e-01 1.85396448e-01
4.66464609e-01 3.62702847e-01 1.01332498e+00 4.98834193e-01
3.06429893e-01 7.90659904e-01 3.52547079e-01 -4.89420146e-02
1.87104821e-01 -3.83614391e-01 1.74482316e-01 1.09750664e+00
-5.49861133e-01 -3.09315413e-01 -2.62264282e-01 2.24582061e-01
-2.30691648e+00 -1.15068829e+00 -5.05109988e-02 2.55065155e+00
5.69417834e-01 5.57686165e-02 4.43152338e-01 8.91360044e-02
6.18241787e-01 -8.70096684e-02 -3.15981865e-01 -3.62856120e-01
1.00858830e-01 3.62199098e-01 2.48247802e-01 5.55341721e-01
-1.07626796e+00 8.17708969e-01 7.97998953e+00 7.54373372e-01
-4.86029714e-01 2.00558692e-01 3.56095880e-01 -4.46539700e-01
-5.56952357e-01 1.51267841e-01 -9.63973224e-01 3.84358078e-01
9.93611217e-01 -1.09659225e-01 6.85728788e-01 9.87851262e-01
1.30119160e-01 1.49245128e-01 -1.03098989e+00 6.78332090e-01
1.44951999e-01 -1.26214027e+00 1.50504440e-01 9.59616154e-02
1.18465841e+00 1.05951071e-01 1.32508889e-01 2.73415059e-01
7.25932658e-01 -9.09137607e-01 4.78468925e-01 7.81222224e-01
3.37495267e-01 -7.78957307e-01 8.78969550e-01 1.51054323e-01
-1.13846815e+00 -5.77283263e-01 -6.47845268e-01 -2.52374351e-01
8.67447481e-02 1.00634789e+00 -9.35949683e-02 5.98978341e-01
7.97824919e-01 8.39492738e-01 -5.83840191e-01 1.45627952e+00
-3.27850193e-01 6.71965957e-01 -9.74731743e-02 -1.88891694e-01
-2.10919589e-01 -3.84252250e-01 1.55650288e-01 9.48978186e-01
4.24216092e-01 1.42419294e-01 4.25880283e-01 1.09006437e-02
-1.28773242e-01 5.22976875e-01 -3.54633898e-01 2.92371422e-01
7.50680745e-01 1.34409666e+00 -4.08388168e-01 -3.06244612e-01
-6.07822657e-01 6.43051863e-01 5.96034110e-01 1.54669032e-01
-6.53115809e-01 -2.11322844e-01 4.03858215e-01 1.88741714e-01
6.84932768e-01 -5.17136790e-02 -1.86383978e-01 -1.29176056e+00
-8.16985667e-02 -9.18265283e-01 6.27528608e-01 -3.53480726e-01
-1.66937029e+00 4.98546213e-01 -5.39417341e-02 -1.48173010e+00
-2.19696537e-01 -5.25445819e-01 -5.48103333e-01 5.60719550e-01
-1.54176760e+00 -9.17643428e-01 -1.22336343e-01 6.07839823e-01
1.80230618e-01 -3.00224304e-01 8.67150009e-01 5.30890822e-01
-1.71485305e-01 7.06579268e-01 8.32029164e-01 -4.58254784e-01
1.16032541e+00 -1.33183289e+00 -6.72566891e-03 3.64255905e-01
3.57324749e-01 1.01134300e+00 4.37660366e-01 -5.21746099e-01
-1.29472721e+00 -8.46415281e-01 1.41166818e+00 -6.31268859e-01
6.32247150e-01 -1.71967506e-01 -5.71807861e-01 2.44136870e-01
-1.62778407e-01 2.09059510e-02 1.28786004e+00 9.56334352e-01
-4.78515565e-01 -2.88699389e-01 -1.08535814e+00 1.99767575e-01
1.30905998e+00 -4.24244612e-01 -5.41885078e-01 6.00885510e-01
3.87814671e-01 4.52642851e-02 -1.41751802e+00 3.84884834e-01
1.17939210e+00 -1.19938946e+00 1.24545538e+00 -9.30173814e-01
2.92478979e-01 -3.73666167e-01 -4.38655794e-01 -1.53404510e+00
-9.95555460e-01 -5.57301760e-01 -5.59948862e-01 1.05825150e+00
9.06394601e-01 -6.38393939e-01 9.03684855e-01 5.49013078e-01
-5.22936508e-02 -1.11631286e+00 -1.88623697e-01 -7.19937205e-01
-1.18417352e-01 -1.13436759e-01 8.39094102e-01 9.32749867e-01
1.02728859e-01 5.14538407e-01 -7.57250190e-01 -9.95802134e-02
3.82196844e-01 6.58642828e-01 6.21576309e-01 -1.94400501e+00
-7.61918187e-01 -3.12797874e-01 2.86019444e-01 -1.25929439e+00
-2.82706767e-01 -7.72251666e-01 -4.10737932e-01 -1.65945911e+00
3.76720548e-01 -8.77957821e-01 -1.02463782e+00 3.42378169e-01
-2.67488062e-01 5.52143633e-01 7.01595247e-02 5.70686996e-01
-1.09136558e+00 1.32429272e-01 1.08441663e+00 1.31434485e-01
-3.74462992e-01 5.37324846e-01 -1.30363321e+00 6.49686754e-01
7.61060238e-01 -7.03108847e-01 -7.87855566e-01 -4.02374744e-01
8.62623096e-01 -7.66466185e-03 -3.44654590e-01 -5.73039889e-01
4.51331019e-01 -3.68016243e-01 5.61377168e-01 -8.26359928e-01
9.25836563e-02 -6.37432635e-01 3.62512171e-01 1.78222544e-02
-6.87848508e-01 3.31450790e-01 -5.73815286e-01 7.11825132e-01
-1.67515427e-01 -3.38595897e-01 4.84689504e-01 -1.42936081e-01
-4.04243797e-01 3.62718403e-01 -2.75523849e-02 -9.26105380e-02
5.01248360e-01 -4.99221832e-02 -3.88075709e-01 -4.51711327e-01
-9.43656445e-01 1.15677081e-01 3.80035311e-01 4.91180599e-01
5.21843076e-01 -1.43480051e+00 -7.21159756e-01 -4.56122570e-02
1.48409083e-01 -6.93530977e-01 -1.40494704e-01 5.49542785e-01
8.19172934e-02 5.11806428e-01 1.13381945e-01 6.24322481e-02
-1.16335881e+00 7.57481515e-01 -1.79884676e-02 -7.81276584e-01
-2.37180084e-01 7.44609356e-01 -3.05899709e-01 -6.02493405e-01
3.32100332e-01 1.72953263e-01 -7.13864863e-01 3.62122715e-01
7.58564413e-01 5.90948105e-01 4.49266851e-01 -2.06914663e-01
-2.58871913e-01 5.71797788e-01 -4.47197109e-01 3.30000147e-02
1.33385992e+00 -1.62047446e-01 -8.29464197e-02 2.18228072e-01
8.78706157e-01 4.95598644e-01 -7.68215477e-01 -3.88591319e-01
-3.30143757e-02 -9.31934476e-01 3.32154840e-01 -1.38196528e+00
-1.12554157e+00 1.26257539e-01 4.22934532e-01 4.40851241e-01
1.05605185e+00 -1.12140499e-01 5.62289715e-01 6.82878315e-01
7.11346865e-01 -1.36550879e+00 2.71362793e-02 4.77889448e-01
6.11974239e-01 -1.17901039e+00 4.34552372e-01 -3.90956372e-01
-5.76135099e-01 8.94832671e-01 3.32440734e-01 -2.92341769e-01
1.05463731e+00 -5.66247925e-02 8.08354467e-02 -6.95407242e-02
-1.23215497e+00 -1.50995791e-01 7.00048566e-01 3.75783771e-01
7.97175884e-01 1.09440237e-01 -8.23947728e-01 7.82049477e-01
1.19562466e-02 9.70095471e-02 9.03749764e-02 9.35008943e-01
-6.68335438e-01 -1.75799358e+00 -6.52677640e-02 1.07505250e+00
-6.73157334e-01 -3.82923424e-01 -6.04530632e-01 2.53769249e-01
1.33089116e-02 1.50709689e+00 -3.54670107e-01 -9.09730434e-01
3.64768296e-01 -1.75561041e-01 3.48255485e-01 -7.72086442e-01
-1.20873868e+00 1.32500783e-01 3.89399737e-01 -6.24887705e-01
-5.55577993e-01 -6.61845088e-01 -6.31633341e-01 -4.41876411e-01
-1.04586768e+00 7.79495835e-01 4.90595907e-01 6.99602544e-01
5.99361420e-01 5.25611751e-02 1.22986519e+00 -6.78147495e-01
-7.34923303e-01 -7.82519937e-01 -8.66226077e-01 3.84867102e-01
-1.69410363e-01 -9.21444893e-01 -5.16273737e-01 -5.17308712e-01] | [10.006855010986328, 5.694890022277832] |
ea9ff778-4c55-44bd-a00e-aea561818afb | natural-language-processing-for-ehr-based | 1806.04820 | null | http://arxiv.org/abs/1806.04820v2 | http://arxiv.org/pdf/1806.04820v2.pdf | Natural Language Processing for EHR-Based Computational Phenotyping | This article reviews recent advances in applying natural language processing
(NLP) to Electronic Health Records (EHRs) for computational phenotyping.
NLP-based computational phenotyping has numerous applications including
diagnosis categorization, novel phenotype discovery, clinical trial screening,
pharmacogenomics, drug-drug interaction (DDI) and adverse drug event (ADE)
detection, as well as genome-wide and phenome-wide association studies.
Significant progress has been made in algorithm development and resource
construction for computational phenotyping. Among the surveyed methods,
well-designed keyword search and rule-based systems often achieve good
performance. However, the construction of keyword and rule lists requires
significant manual effort, which is difficult to scale. Supervised machine
learning models have been favored because they are capable of acquiring both
classification patterns and structures from data. Recently, deep learning and
unsupervised learning have received growing attention, with the former favored
for its performance and the latter for its ability to find novel phenotypes.
Integrating heterogeneous data sources have become increasingly important and
have shown promise in improving model performance. Often better performance is
achieved by combining multiple modalities of information. Despite these many
advances, challenges and opportunities remain for NLP-based computational
phenotyping, including better model interpretability and generalizability, and
proper characterization of feature relations in clinical narratives | ['Tristan Naumann', 'Yuan Luo', 'Zexian Zeng', 'Yu Deng', 'Xiaoyu Li'] | 2018-06-13 | null | null | null | null | ['computational-phenotyping'] | ['medical'] | [ 2.65866667e-01 -2.78313130e-01 -5.98808229e-01 -2.96584457e-01
-8.91401649e-01 -4.78424549e-01 2.52791137e-01 1.12018991e+00
-2.08922401e-01 9.97702479e-01 1.73689380e-01 -3.24067533e-01
-5.93055487e-01 -7.23570228e-01 -2.31966659e-01 -7.59073257e-01
-2.97215164e-01 9.16801929e-01 -5.09943128e-01 4.70748186e-01
-1.46718323e-01 4.76463944e-01 -1.23730254e+00 2.85042316e-01
1.15811098e+00 7.89336383e-01 1.45363748e-01 4.16142970e-01
-2.22925663e-01 5.47729731e-01 -4.50129747e-01 -3.22503299e-01
-2.54489630e-01 -3.65565658e-01 -5.32397568e-01 -2.16515556e-01
-2.41538882e-01 3.63163382e-01 5.24533950e-02 7.88631856e-01
8.90483320e-01 -3.03016543e-01 5.50499260e-01 -9.77971971e-01
-5.96529543e-01 3.80412221e-01 -3.44650745e-01 -3.03914715e-02
6.74643278e-01 2.88778961e-01 9.74867940e-01 -5.83336413e-01
7.59540498e-01 9.44305301e-01 7.27320135e-01 2.89953738e-01
-1.41586065e+00 -6.05751395e-01 -4.33820099e-01 4.56123888e-01
-1.71423292e+00 -1.89423636e-01 3.43430340e-01 -7.14875817e-01
1.36910737e+00 3.74010950e-01 7.42574096e-01 9.90395904e-01
3.59659880e-01 7.43479669e-01 7.72681594e-01 -2.59639770e-01
9.00288522e-02 4.72491980e-02 -1.99823491e-02 6.03154600e-01
3.32531393e-01 1.25561148e-01 -4.52917486e-01 -9.20815885e-01
2.36013964e-01 2.55137593e-01 -2.51202404e-01 5.05688824e-02
-1.20930862e+00 7.27792799e-01 -2.15363935e-01 1.99830920e-01
-6.94401443e-01 -3.38886827e-01 7.29542434e-01 1.67799234e-01
3.60265583e-01 8.83175671e-01 -8.66710961e-01 -1.33872405e-01
-9.00313735e-01 3.48731518e-01 7.20159471e-01 5.87996721e-01
2.31367439e-01 -2.25551397e-01 -2.25260600e-01 1.12397981e+00
2.04245020e-02 3.75733495e-01 7.64696658e-01 -3.57150614e-01
1.43698201e-01 9.11011994e-01 -7.19989315e-02 -9.76291180e-01
-9.55296457e-01 -2.65852898e-01 -1.18015635e+00 -4.77769345e-01
2.98287958e-01 -1.92142487e-01 -7.64375985e-01 1.56072390e+00
4.40520585e-01 1.66905120e-01 2.04510525e-01 4.76028681e-01
7.19824433e-01 3.67753237e-01 6.60886228e-01 -5.67536116e-01
1.62554276e+00 -1.19518027e-01 -6.81245744e-01 3.52394760e-01
9.74697292e-01 -5.90814352e-01 5.83156884e-01 5.67326248e-01
-6.40191257e-01 -6.32453039e-02 -6.71046615e-01 1.39502853e-01
-5.55595636e-01 1.70151174e-01 9.75925744e-01 4.86847848e-01
-6.01463318e-01 4.80849445e-01 -8.73071671e-01 -5.21874189e-01
8.88262451e-01 6.19159698e-01 -6.06818676e-01 -2.53855705e-01
-1.28807783e+00 7.88470805e-01 4.98791635e-01 -2.67578244e-01
-5.35358012e-01 -1.17235649e+00 -9.13421988e-01 1.72059625e-01
3.88007581e-01 -1.09102571e+00 6.18616462e-01 -2.16183066e-01
-1.36282766e+00 8.44478667e-01 -2.53437459e-01 -5.01368701e-01
-1.67215262e-02 1.57548293e-01 -7.80083179e-01 9.11106616e-02
7.82328174e-02 2.15443924e-01 1.66864216e-01 -2.45585725e-01
-5.92789412e-01 -7.60106206e-01 -7.94557989e-01 -3.13350372e-02
-1.53746247e-01 4.12294745e-01 -9.10436884e-02 -6.22281492e-01
-2.10976705e-01 -7.12025285e-01 -3.93981338e-01 -2.59052277e-01
-5.88771164e-01 -6.67831957e-01 4.70467985e-01 -6.09224081e-01
1.22209346e+00 -1.98962522e+00 -3.95788718e-03 1.26469076e-01
5.27032852e-01 4.99253035e-01 1.24737196e-01 6.23292983e-01
-2.04282761e-01 3.30503643e-01 1.04795434e-02 1.20642230e-01
-4.17303979e-01 -1.82228267e-01 3.57189104e-02 6.01006508e-01
5.61686575e-01 1.17180407e+00 -8.82656336e-01 -3.95620137e-01
2.44638264e-01 3.78403544e-01 -5.52843094e-01 1.27865016e-01
-4.88478601e-01 6.94695055e-01 -7.29429424e-01 1.01590037e+00
3.34021479e-01 -8.18717718e-01 7.44588077e-01 2.08779380e-01
1.30693734e-01 2.81577349e-01 -7.56639302e-01 1.29105949e+00
-1.65463552e-01 2.81736970e-01 -2.31092975e-01 -9.70998108e-01
7.79050171e-01 6.22775614e-01 8.64968538e-01 -5.96434653e-01
-4.62393537e-02 1.66232809e-01 2.74080604e-01 -1.05286729e+00
-2.13503093e-01 -3.36954027e-01 1.12691030e-01 2.05349833e-01
-3.26662868e-01 4.54265475e-01 1.70173407e-01 -1.42328262e-01
1.37073445e+00 -3.09384733e-01 1.04658723e+00 -7.83533826e-02
5.90952218e-01 3.51676643e-01 9.83930171e-01 4.48401242e-01
3.14852446e-02 2.91640759e-01 4.31570858e-01 -5.28463185e-01
-8.17226350e-01 -5.73423505e-01 -6.78406775e-01 5.70192337e-01
-2.86767215e-01 -5.41587949e-01 -5.94329163e-02 -3.44317943e-01
3.49926651e-01 2.80917078e-01 -4.31963116e-01 1.29088853e-02
-2.58225381e-01 -1.33634651e+00 9.99442220e-01 4.42566037e-01
1.17184103e-01 -1.02422631e+00 -1.91171467e-01 5.22100270e-01
3.05262748e-02 -9.44255292e-01 2.05440372e-02 2.06580088e-01
-7.40948856e-01 -1.62265670e+00 -5.76711476e-01 -6.10329390e-01
4.57239598e-01 -4.26956892e-01 7.45973110e-01 -3.24622020e-02
-8.56182694e-01 -1.29666135e-01 -2.19014838e-01 -7.37148583e-01
-3.84063542e-01 -9.39891338e-02 4.45908993e-01 -1.36690721e-01
8.09088886e-01 -5.11750221e-01 -4.71178621e-01 5.79228140e-02
-7.51804590e-01 -1.92532092e-01 8.45202029e-01 1.02792740e+00
9.71892655e-01 1.96014345e-01 1.10861075e+00 -1.21041596e+00
8.15147519e-01 -9.24914539e-01 -5.69840670e-01 3.78490090e-01
-1.06447875e+00 -1.68091338e-02 9.99428391e-01 -2.81220704e-01
-8.12915027e-01 1.18775144e-01 -3.21377337e-01 -7.17484131e-02
-7.15654910e-01 1.13997281e+00 -3.87658507e-01 3.29987496e-01
6.06088161e-01 3.81959349e-01 1.87531784e-01 -6.67646348e-01
-6.42081201e-02 8.66902530e-01 1.52199209e-01 -3.24000448e-01
4.32435051e-02 3.00477445e-01 1.97037503e-01 -8.98409605e-01
-4.79448855e-01 -6.37122273e-01 -2.61760205e-01 5.11060953e-01
9.80201364e-01 -8.17796350e-01 -1.04089975e+00 3.93022895e-01
-8.22758019e-01 4.74141724e-02 -2.15744898e-02 6.91896439e-01
-1.79250747e-01 4.43676829e-01 -5.38558483e-01 -5.40749967e-01
-5.68997204e-01 -1.17108202e+00 8.92316580e-01 9.04772654e-02
-7.66733170e-01 -1.04130030e+00 3.88948411e-01 2.58649260e-01
1.86244786e-01 4.58436936e-01 1.73772550e+00 -1.27820599e+00
-1.21100850e-01 -5.75130701e-01 -1.46436512e-01 -2.08171606e-01
3.49837095e-01 -1.02833629e-01 -6.86460495e-01 -5.96345915e-03
-4.61882859e-01 -2.92586356e-01 5.00201643e-01 5.40715754e-01
1.46428990e+00 -1.65833831e-01 -9.07684147e-01 7.92849123e-01
1.18848813e+00 7.23677814e-01 2.80242831e-01 -5.80311418e-02
6.25317514e-01 5.73812068e-01 3.52909058e-01 5.19492269e-01
2.68744826e-01 6.88883781e-01 -1.23065919e-01 -4.43416610e-02
7.95780867e-02 -2.43258044e-01 -2.71691084e-01 3.32105190e-01
1.70970917e-01 -3.30491751e-01 -1.21009564e+00 4.09414202e-01
-1.60491359e+00 -7.01006472e-01 -3.28791797e-01 2.04111981e+00
1.07740116e+00 -4.13996607e-01 1.36565924e-01 -2.60317046e-02
6.17749274e-01 -2.87563264e-01 -8.74214172e-01 -2.07088783e-01
-4.74516183e-01 2.93477803e-01 6.02447130e-02 9.01013147e-04
-8.53115797e-01 5.23175597e-01 6.82406378e+00 8.26323032e-01
-1.05086589e+00 -7.59441778e-02 8.98325503e-01 -3.82457748e-02
-1.28202990e-01 -2.76724100e-01 -7.45765567e-01 5.57948470e-01
9.38270688e-01 -2.51296729e-01 1.75174996e-01 7.53777623e-01
5.03204167e-01 -1.65857032e-01 -1.31982422e+00 1.31841683e+00
-2.56518096e-01 -1.88316858e+00 -5.24388440e-03 3.19719136e-01
5.04810989e-01 5.65877594e-02 -1.25996396e-01 -4.30543013e-02
1.66453496e-01 -1.23005116e+00 -2.09983185e-01 5.16477048e-01
1.27343440e+00 -6.36110842e-01 8.20181608e-01 2.58356422e-01
-9.34922159e-01 -1.08863302e-01 -3.05399895e-01 2.11681426e-02
9.32486057e-02 1.19711316e+00 -1.28204763e+00 6.56706333e-01
4.65079784e-01 9.19283092e-01 -2.96520323e-01 1.23472905e+00
-7.25416690e-02 6.85433924e-01 -2.28850186e-01 -1.77752078e-01
-3.19420129e-01 -9.02787447e-02 4.55744177e-01 1.14982748e+00
-2.26362702e-02 2.70510077e-01 3.52772355e-01 7.89811850e-01
-8.92191678e-02 4.42968160e-01 -5.27857363e-01 -6.34389043e-01
5.59156120e-01 1.06097627e+00 -5.22466063e-01 -5.10570109e-01
-3.16082150e-01 4.33708221e-01 2.79956788e-01 2.36941725e-01
-4.41124707e-01 -3.71427715e-01 1.02714372e+00 1.79020897e-01
-1.15292847e-01 2.37603694e-01 -3.88033301e-01 -9.99284506e-01
-3.78407240e-01 -1.16964853e+00 9.90301490e-01 -1.84415162e-01
-1.56786609e+00 4.36694980e-01 -1.04725957e-01 -9.64710593e-01
-3.37948322e-01 -5.12280583e-01 -3.42030138e-01 1.05578399e+00
-1.29120815e+00 -7.10069418e-01 1.49150357e-01 4.02965575e-01
2.84903258e-01 -4.76657510e-01 1.22064090e+00 5.81289589e-01
-1.08075774e+00 7.23359764e-01 3.02583784e-01 -7.79063851e-02
7.75610149e-01 -9.54744339e-01 8.22092593e-03 3.10091555e-01
-2.23335624e-01 6.82902753e-01 3.64431471e-01 -9.85375762e-01
-1.55996656e+00 -1.23046827e+00 1.17618966e+00 -4.37690258e-01
4.60856736e-01 -9.08943117e-02 -9.08119977e-01 3.17320853e-01
-3.26502770e-01 -1.49393901e-01 1.53989601e+00 3.93539578e-01
-3.26267004e-01 1.09997004e-01 -1.30244851e+00 4.09566939e-01
6.31669998e-01 -3.98674935e-01 -2.14597940e-01 6.54267967e-01
2.87304491e-01 -1.43742152e-02 -1.13046801e+00 5.19394040e-01
5.72823584e-01 -4.05632257e-01 8.26991081e-01 -1.13388193e+00
2.58527100e-01 -3.61557722e-01 2.81440496e-01 -1.13996589e+00
-5.93047261e-01 -5.85548878e-01 -1.05236575e-01 7.29329228e-01
7.47443914e-01 -8.82849038e-01 7.45862901e-01 7.57788658e-01
1.35607645e-01 -1.23505938e+00 -6.27672613e-01 -4.39723462e-01
-2.64094710e-01 -4.87540841e-01 8.53738964e-01 1.19417644e+00
6.20975435e-01 3.93633544e-01 -2.68841684e-01 1.58894226e-01
2.05028340e-01 3.53010088e-01 5.36005080e-01 -1.47264028e+00
-4.93596226e-01 -4.94359165e-01 -6.12356722e-01 -4.41575140e-01
5.70842288e-02 -1.14728737e+00 -4.54736710e-01 -1.55254388e+00
5.49846113e-01 -5.38830280e-01 -2.28541210e-01 6.71774387e-01
-4.18356806e-01 -6.21231087e-02 -4.79384780e-01 1.16524667e-01
-4.35918719e-01 1.63609371e-01 8.54593873e-01 -1.42632902e-01
-5.10098100e-01 -8.86463001e-02 -8.16079259e-01 6.16049767e-01
9.26288784e-01 -6.21638596e-01 -2.43192911e-01 2.90585011e-02
4.41397548e-01 3.52860093e-01 1.42547879e-02 -4.23532635e-01
1.23775743e-01 -2.81813979e-01 5.97135663e-01 -5.03133833e-01
-3.46647017e-02 -3.22223336e-01 6.32007420e-01 7.30595887e-01
-3.71258557e-01 7.68990144e-02 1.00363843e-01 8.87432337e-01
-3.12503904e-01 3.01472336e-01 6.73932850e-01 -4.60699126e-02
-4.20360982e-01 5.27629793e-01 -6.67073786e-01 -1.26064420e-01
1.05229640e+00 -9.49399546e-02 -2.10189044e-01 1.55731179e-02
-9.31227803e-01 3.61737072e-01 1.14878364e-01 2.73767501e-01
5.75203657e-01 -8.99835646e-01 -9.10358727e-01 3.08427423e-01
4.93554920e-01 -1.33544445e-01 3.92097712e-01 1.29812157e+00
-5.40237904e-01 9.17500973e-01 1.67867661e-01 -4.72376287e-01
-1.28231311e+00 8.31126153e-01 5.80144823e-02 -6.84503138e-01
-7.02031076e-01 7.22427666e-01 2.22067207e-01 -1.75052553e-01
1.98990345e-01 -1.22822374e-01 -2.65609205e-01 9.39891860e-03
7.42650628e-01 3.38473499e-01 1.62471652e-01 -2.69559771e-01
-7.06242800e-01 1.36179209e-01 -2.29106426e-01 6.95201814e-01
1.25957620e+00 2.85683095e-01 -3.43997002e-01 1.10025212e-01
1.16045189e+00 -1.55503526e-02 -3.13727230e-01 -7.67847225e-02
2.63465106e-01 -1.59100100e-01 -2.34058216e-01 -1.18558633e+00
-6.69863224e-01 5.56939900e-01 2.27193117e-01 1.42910525e-01
1.07049811e+00 1.27692088e-01 7.25414813e-01 3.63638610e-01
2.87944049e-01 -8.03025663e-01 -6.52880728e-01 2.28414685e-01
3.61131221e-01 -1.26582551e+00 1.24477006e-01 -3.84264886e-01
-3.87692004e-01 8.90143514e-01 1.95527554e-01 4.29935992e-01
7.57992923e-01 3.33589494e-01 -1.28447697e-01 -5.61924934e-01
-9.42733884e-01 4.99359369e-02 2.09744647e-01 6.79726303e-01
6.69886053e-01 4.68328029e-01 -7.29053199e-01 9.73135352e-01
4.24650870e-02 1.25167519e-01 -5.42259812e-02 5.29693961e-01
1.04232758e-01 -1.49718642e+00 -2.02983186e-01 1.18073153e+00
-9.19743836e-01 -4.14973974e-01 -5.32472968e-01 5.53743362e-01
2.56783932e-01 8.96501541e-01 -2.97988951e-01 -2.18809977e-01
7.40626231e-02 4.79655266e-01 1.20518886e-01 -9.18047726e-01
-5.62657297e-01 1.72660530e-01 3.27198476e-01 -5.18283427e-01
-1.47723123e-01 -7.07926869e-01 -1.32123673e+00 1.57537665e-02
-3.59639466e-01 3.74837786e-01 4.43176508e-01 1.06226146e+00
1.11839867e+00 4.16072845e-01 2.83937782e-01 9.63983983e-02
-2.12775856e-01 -8.21588278e-01 -8.06646049e-01 1.60307840e-01
-8.44426733e-03 -4.94879037e-01 1.92424908e-01 6.20058738e-02] | [6.236780643463135, 5.832752227783203] |
fbbd9eda-2de0-4084-a4bc-861601cdc66f | flexible-modal-deception-detection-with-audio | 2302.05727 | null | https://arxiv.org/abs/2302.05727v1 | https://arxiv.org/pdf/2302.05727v1.pdf | Flexible-modal Deception Detection with Audio-Visual Adapter | Detecting deception by human behaviors is vital in many fields such as custom security and multimedia anti-fraud. Recently, audio-visual deception detection attracts more attention due to its better performance than using only a single modality. However, in real-world multi-modal settings, the integrity of data can be an issue (e.g., sometimes only partial modalities are available). The missing modality might lead to a decrease in performance, but the model still learns the features of the missed modality. In this paper, to further improve the performance and overcome the missing modality problem, we propose a novel Transformer-based framework with an Audio-Visual Adapter (AVA) to fuse temporal features across two modalities efficiently. Extensive experiments conducted on two benchmark datasets demonstrate that the proposed method can achieve superior performance compared with other multi-modal fusion methods under flexible-modal (multiple and missing modalities) settings. | ['Alex Kot', 'Adams Wai-Kin Kong', 'Bingquan Shen', 'Xiaobao Guo', 'Nithish Muthuchamy Selvaraj', 'Zitong Yu', 'Zhaoxu Li'] | 2023-02-11 | null | null | null | null | ['deception-detection'] | ['miscellaneous'] | [ 2.18587145e-01 -7.13537693e-01 -1.42376825e-01 -2.52301186e-01
-1.06031477e+00 -4.93815631e-01 4.84439880e-01 -1.07304111e-01
-1.28098503e-01 5.91833651e-01 2.88419396e-01 9.32629278e-04
-2.45620571e-02 -3.69586736e-01 -4.29395646e-01 -8.49815547e-01
4.79348242e-01 -2.26982743e-01 3.45010459e-01 1.01433702e-01
3.15268457e-01 1.76761061e-01 -1.70146585e+00 5.70379913e-01
8.51499438e-01 1.30844212e+00 -3.65929335e-01 2.42771268e-01
2.73845643e-01 7.89715111e-01 -5.44615149e-01 -7.76888847e-01
1.53968155e-01 -4.02376562e-01 -4.56613570e-01 2.13331699e-01
3.01472932e-01 -6.80629432e-01 -5.61995268e-01 1.18373311e+00
4.21404630e-01 -3.12352609e-02 3.84090215e-01 -1.85003579e+00
-3.41579854e-01 1.66430146e-01 -1.06135380e+00 1.94023132e-01
8.10722291e-01 8.34424868e-02 6.26118898e-01 -8.74063790e-01
8.46708417e-02 1.34276342e+00 4.89903718e-01 4.78802770e-01
-8.73322546e-01 -9.67687845e-01 7.35501274e-02 8.52645218e-01
-1.59595513e+00 -7.52878308e-01 8.54190111e-01 -1.61179379e-01
3.78586531e-01 3.76639068e-01 4.52684850e-01 1.39845395e+00
8.96694064e-02 1.06141329e+00 1.27394295e+00 -5.10450602e-02
-1.24561824e-01 2.47662008e-01 -7.78328255e-02 6.13102436e-01
1.41687125e-01 1.89238250e-01 -1.05415750e+00 -5.89244306e-01
2.41095781e-01 5.37718475e-01 -4.78572756e-01 -2.63787001e-01
-1.14408946e+00 6.50716007e-01 8.10340941e-02 1.85914978e-01
-3.31680804e-01 -1.66922495e-01 5.61780572e-01 2.73996294e-01
9.79503393e-02 -3.08093011e-01 1.51038006e-01 -3.46847355e-01
-1.02935779e+00 1.81233525e-01 3.33363682e-01 5.37114501e-01
4.04180169e-01 -2.80951411e-02 -1.17044389e-01 7.53126204e-01
4.08887833e-01 5.84950268e-01 4.58104342e-01 -9.19121385e-01
5.50792396e-01 6.49130106e-01 1.73725903e-01 -1.34206450e+00
-1.71724021e-01 -6.07122704e-02 -1.08475769e+00 -1.35366339e-03
4.98854756e-01 1.53293505e-01 -5.91203153e-01 1.65138245e+00
4.78683054e-01 4.87224877e-01 -8.52617025e-02 1.21711886e+00
6.97562814e-01 5.20544887e-01 -2.88995989e-02 -4.58602816e-01
1.44317782e+00 -5.05272627e-01 -8.53270590e-01 -6.67716889e-03
1.93298593e-01 -8.02563369e-01 7.67650723e-01 7.01193631e-01
-7.58554757e-01 -2.25721046e-01 -9.06707346e-01 1.80539936e-01
-7.07155690e-02 -1.39373794e-01 4.47332114e-01 7.87125289e-01
-4.67593193e-01 -3.74929234e-02 -6.12431943e-01 -2.43464813e-01
5.62009811e-01 8.18253979e-02 -6.97432041e-01 -5.32720864e-01
-1.31474936e+00 5.21324754e-01 2.48921528e-01 1.96722969e-01
-9.79374409e-01 -4.18070376e-01 -7.66397357e-01 -3.25792134e-02
6.03971481e-01 -6.01392865e-01 9.57835436e-01 -9.38034236e-01
-1.07595694e+00 5.28585136e-01 -2.33628556e-01 -2.03396350e-01
6.59314156e-01 -1.60676479e-01 -8.60158622e-01 4.41297650e-01
-9.79564339e-02 2.13319048e-01 1.40134418e+00 -1.43659484e+00
-9.31421638e-01 -6.71426058e-01 1.58976108e-01 1.50183991e-01
-6.42597616e-01 1.74885556e-01 -1.73111543e-01 -5.11979580e-01
1.08568572e-01 -7.48944104e-01 4.39468920e-01 1.05762199e-01
-4.40708488e-01 1.11654997e-02 1.31449020e+00 -8.38204205e-01
1.30328774e+00 -2.41721559e+00 1.03081569e-01 2.01440468e-01
2.48069212e-01 3.90871644e-01 1.71312168e-01 4.48483825e-01
2.28363454e-01 -1.04025833e-01 -1.99672744e-01 -4.86484140e-01
-7.61392489e-02 1.15054831e-01 -2.79217809e-01 7.40420520e-01
-1.36267573e-01 5.71653843e-01 -6.70914412e-01 -7.19051480e-01
2.26057172e-01 6.04786873e-01 -2.21123680e-01 1.43732480e-03
4.46413010e-01 4.76683468e-01 -3.97294670e-01 1.28373349e+00
9.43352342e-01 -9.59234461e-02 -9.64549333e-02 -3.26021403e-01
2.72194922e-01 -2.73374707e-01 -1.33429050e+00 1.60604155e+00
-2.30134547e-01 3.87813807e-01 3.08064222e-01 -8.65317881e-01
5.15818894e-01 5.49894452e-01 4.15655255e-01 -7.54944682e-01
2.15614587e-01 1.47556111e-01 -1.33490920e-01 -7.18517661e-01
5.03445029e-01 -3.25759560e-01 -2.22200587e-01 3.87579083e-01
-1.79127499e-01 4.33025748e-01 -2.31727764e-01 3.07802379e-01
1.07319164e+00 -1.56584293e-01 2.03914627e-01 4.97780144e-01
8.27135682e-01 -2.16980398e-01 8.35723042e-01 5.22977114e-01
-6.16008699e-01 6.13648713e-01 3.53916466e-01 -1.33879229e-01
-5.03632128e-01 -9.19508874e-01 2.08756290e-02 9.11791980e-01
5.66619694e-01 -3.74512017e-01 -3.90166223e-01 -9.01277840e-01
7.40797296e-02 5.14932215e-01 -3.63119304e-01 -4.52381104e-01
-2.48242885e-01 -6.74685001e-01 9.35483634e-01 2.47813419e-01
7.08295941e-01 -6.03444278e-01 -6.77936137e-01 -1.18139826e-01
-8.70306671e-01 -1.27040935e+00 -4.97045755e-01 -7.12293863e-01
-6.54987156e-01 -1.23013580e+00 -5.79316854e-01 -3.25224429e-01
3.96436334e-01 8.12646568e-01 4.60706234e-01 2.92778611e-01
-9.31928400e-03 5.77086210e-01 -4.27119374e-01 -2.05177307e-01
-2.64864117e-01 -3.78181219e-01 2.32269868e-01 7.87754774e-01
3.76317203e-01 -4.96601731e-01 -5.73816419e-01 4.87679660e-01
-1.20801985e+00 -1.04480155e-01 4.26408172e-01 9.97762561e-01
2.22163379e-01 3.24726790e-01 7.47812450e-01 -3.49377930e-01
6.47152007e-01 -5.28195202e-01 -2.37784415e-01 3.26765627e-01
-3.04707378e-01 -4.17514920e-01 5.25852561e-01 -6.82264566e-01
-1.16606581e+00 -6.23090677e-02 2.57577032e-01 -7.80833662e-01
-2.29622439e-01 5.52314103e-01 -4.44944352e-01 -2.66771525e-01
1.52035490e-01 5.90324700e-01 1.42549306e-01 -4.43777949e-01
-1.38889566e-01 9.58593428e-01 5.23486197e-01 -2.54937023e-01
8.62795293e-01 6.21332109e-01 9.01337266e-02 -5.01072466e-01
-4.90188003e-01 -5.16143858e-01 -2.00365290e-01 -4.41567987e-01
1.78030998e-01 -1.02356458e+00 -1.12374413e+00 8.79647553e-01
-1.04338062e+00 6.05717480e-01 3.34387928e-01 5.78484535e-01
-2.17821106e-01 9.89527047e-01 -4.36638564e-01 -1.14089429e+00
-2.20850497e-01 -1.18105805e+00 1.08701336e+00 3.53382856e-01
1.62724197e-01 -5.79814434e-01 -2.86671788e-01 9.07324076e-01
2.96216637e-01 2.49713317e-01 5.21836519e-01 -4.77808952e-01
-4.73931104e-01 -5.85768342e-01 -3.60471129e-01 2.73956865e-01
1.41574472e-01 -2.11016148e-01 -1.10359883e+00 -3.84664506e-01
1.67160630e-01 -4.86203164e-01 7.74528563e-01 9.20705963e-03
1.10103679e+00 -3.42205375e-01 -3.53740394e-01 3.05116534e-01
1.12658846e+00 1.15398966e-01 5.55522025e-01 5.76209724e-02
7.44777203e-01 4.71283138e-01 8.93429756e-01 8.08386683e-01
5.93944907e-01 7.11393237e-01 7.48748302e-01 2.50571191e-01
1.72851831e-01 -1.80560187e-01 6.79596126e-01 5.18195093e-01
1.16251402e-01 -3.32124829e-01 -6.53658211e-01 5.47400177e-01
-1.99168479e+00 -1.30321550e+00 -7.31373057e-02 2.45834255e+00
5.16912103e-01 -1.48407385e-01 4.96862769e-01 5.94494462e-01
8.53509367e-01 1.16092116e-01 -6.15336835e-01 -1.57645587e-02
-2.69170284e-01 -5.25373757e-01 1.00235857e-01 1.44923583e-01
-9.57375824e-01 2.72547573e-01 5.54556465e+00 1.17539191e+00
-1.02114511e+00 4.26559687e-01 2.94268191e-01 -3.24492455e-01
-1.29677653e-01 -1.60532176e-01 -5.19376040e-01 8.60263050e-01
7.04027474e-01 -1.59048244e-01 5.26591599e-01 2.99321860e-01
2.55162090e-01 -3.71367484e-01 -9.49038327e-01 1.53285301e+00
5.58612227e-01 -7.11377501e-01 1.05384938e-01 2.47103304e-01
1.98941886e-01 -5.15520334e-01 1.54746562e-01 1.00182891e-01
-3.72075707e-01 -9.40371692e-01 7.62823105e-01 6.35433614e-01
7.49038935e-01 -1.00941646e+00 8.28865469e-01 7.74940014e-01
-1.09194672e+00 -3.35365564e-01 -8.13400894e-02 1.09018199e-01
3.31665635e-01 5.60306370e-01 -2.48019308e-01 9.24469233e-01
8.41325223e-01 6.78837299e-01 -5.35029471e-01 1.18478489e+00
1.33419484e-01 5.08947074e-01 -3.76972079e-01 2.57474631e-01
-1.11378148e-01 1.06461391e-01 8.18359733e-01 8.25762510e-01
6.06550694e-01 1.70123994e-01 1.40887603e-01 4.93397713e-01
-3.22028473e-02 -6.50471598e-02 -6.03933811e-01 4.63855453e-02
5.32727122e-01 1.21340132e+00 -1.10896371e-01 -1.26354158e-01
-6.70749784e-01 1.05241299e+00 -1.97949886e-01 1.94411099e-01
-1.10125697e+00 -1.90893069e-01 5.87216496e-01 9.47020203e-02
5.74670881e-02 6.57636523e-02 2.59749442e-02 -1.53184867e+00
1.96767852e-01 -1.21318841e+00 8.96528065e-01 -7.43953347e-01
-1.37048101e+00 1.75583839e-01 7.25613385e-02 -1.63439572e+00
-2.12182745e-01 -1.35779470e-01 -2.43588999e-01 5.73012650e-01
-1.52216315e+00 -1.54577792e+00 -4.14821297e-01 1.14724600e+00
3.01971436e-01 -2.00076222e-01 5.10367155e-01 6.23575032e-01
-5.30377984e-01 9.60316598e-01 -8.00703093e-02 -7.56253973e-02
9.91239011e-01 -5.48934996e-01 -6.14143014e-01 9.55833495e-01
-2.83141825e-02 3.72895658e-01 6.91284657e-01 -4.45674360e-01
-1.63361502e+00 -6.82600081e-01 5.56797147e-01 -2.76005954e-01
4.93330479e-01 -7.29809478e-02 -1.10364544e+00 3.65229428e-01
1.84970096e-01 2.17597093e-02 8.50293398e-01 -1.75612494e-01
-6.13536596e-01 -3.62431943e-01 -1.54524052e+00 2.53407687e-01
7.75028467e-01 -7.29851544e-01 -4.46068645e-01 -3.38708423e-02
2.03614041e-01 -3.39288443e-01 -7.61295378e-01 3.91732603e-01
8.87429178e-01 -1.24686277e+00 9.90350008e-01 -2.39951730e-01
1.62356079e-01 -5.30735433e-01 -3.63247365e-01 -1.08840251e+00
1.01156950e-01 -4.63278890e-01 -5.91874599e-01 1.48211288e+00
-2.41749689e-01 -7.38236129e-01 4.60286111e-01 5.62535882e-01
2.66251951e-01 -2.44320959e-01 -1.47146165e+00 -7.16024697e-01
-4.94714171e-01 -4.02748853e-01 7.28960991e-01 9.53495324e-01
1.92912638e-01 2.58240080e-03 -1.07603002e+00 3.87767494e-01
7.86443591e-01 3.18315536e-01 6.42464638e-01 -1.25966918e+00
-1.25495493e-01 -3.14012468e-02 -7.18409777e-01 -8.11466098e-01
7.86200315e-02 -4.26079035e-01 -1.89589232e-01 -9.64006424e-01
7.05803812e-01 3.78618091e-02 -6.44341409e-01 6.39787614e-01
-1.68272913e-01 4.98857647e-01 2.86748499e-01 3.01627785e-01
-7.82971561e-01 6.58432186e-01 9.44853067e-01 -3.28364164e-01
9.73703340e-02 8.29452127e-02 -7.89565206e-01 7.17452824e-01
5.55667222e-01 -5.94284654e-01 -2.96074927e-01 -2.05845565e-01
5.57615459e-02 5.47199368e-01 8.84874284e-01 -1.01805699e+00
4.72767055e-01 -3.38753879e-01 3.04822385e-01 -6.20337546e-01
6.79737866e-01 -1.17595005e+00 2.25527704e-01 2.04799801e-01
2.38142721e-02 -3.99296954e-02 7.70028457e-02 9.22774196e-01
-5.96580923e-01 2.11658058e-04 7.28855193e-01 8.16263184e-02
-6.19156957e-01 1.59254730e-01 -2.61210591e-01 -3.03864151e-01
1.01480234e+00 -4.30286825e-01 -6.15580261e-01 -7.42308319e-01
-3.86938661e-01 2.95085162e-01 5.96204638e-01 6.42710149e-01
9.57258761e-01 -1.67045021e+00 -6.44105494e-01 9.26347673e-02
3.63597989e-01 -5.65570593e-01 7.26660430e-01 1.33844602e+00
2.84100711e-01 1.21312067e-01 -1.81347758e-01 -6.08889997e-01
-1.85422492e+00 6.30913973e-01 2.09986717e-01 1.53732821e-04
-2.11450770e-01 2.38874495e-01 6.47644103e-02 3.34888399e-02
2.22987756e-01 3.94623816e-01 -1.58185601e-01 1.89403117e-01
8.32006872e-01 6.32721186e-01 7.51148835e-02 -1.29273403e+00
-7.35048175e-01 3.87829751e-01 -3.70954024e-03 -8.12111348e-02
9.47934151e-01 -6.59389436e-01 3.49410288e-02 4.37821954e-01
1.03589249e+00 7.27564618e-02 -8.90331388e-01 -5.19300997e-01
-5.14829934e-01 -1.09146690e+00 1.81227386e-01 -7.33671427e-01
-1.21330798e+00 9.41999972e-01 7.09549129e-01 2.51255929e-01
1.57129228e+00 -3.33694339e-01 1.09735906e+00 2.80561727e-02
5.77408791e-01 -9.91129339e-01 1.59446418e-01 -8.03459361e-02
7.10003853e-01 -1.48996437e+00 9.64952633e-02 -4.18190092e-01
-8.31379414e-01 9.63945508e-01 4.48105484e-01 4.42555368e-01
5.44747293e-01 -1.68545172e-01 -1.47926703e-01 -5.46884276e-02
-6.17015183e-01 1.33974208e-02 2.87227601e-01 6.11770868e-01
-1.26492977e-01 -2.58677211e-02 -2.08506569e-01 7.43262887e-01
3.56303751e-01 1.30111694e-01 5.67153156e-01 1.07047260e+00
-1.45344868e-01 -9.60892558e-01 -9.20032561e-01 4.72525179e-01
-7.23440051e-01 2.15493217e-01 -3.32848251e-01 5.21199882e-01
1.77122578e-01 1.55192745e+00 -2.92352647e-01 -6.62162125e-01
2.38943547e-01 7.33547732e-02 3.71823400e-01 -4.75405082e-02
-4.64155585e-01 2.00411618e-01 -8.68702456e-02 -7.35906839e-01
-8.03552628e-01 -9.74191785e-01 -8.33312511e-01 -7.88476408e-01
-5.75322568e-01 -9.36320573e-02 2.76652396e-01 1.05640090e+00
4.39857304e-01 1.05519906e-01 8.09260964e-01 -5.83234429e-01
-5.85136473e-01 -6.55565381e-01 -7.11606562e-01 5.93941152e-01
6.09622717e-01 -9.68671501e-01 -5.65903783e-01 -3.13341478e-03] | [13.18422794342041, 1.5161834955215454] |
503f24a9-d42d-45af-907b-9eeb8bb08f6b | cobit-a-contrastive-bi-directional-image-text | 2303.13455 | null | https://arxiv.org/abs/2303.13455v1 | https://arxiv.org/pdf/2303.13455v1.pdf | CoBIT: A Contrastive Bi-directional Image-Text Generation Model | The field of vision and language has witnessed a proliferation of pre-trained foundation models. Most existing methods are independently pre-trained with contrastive objective like CLIP, image-to-text generative objective like PaLI, or text-to-image generative objective like Parti. However, the three objectives can be pre-trained on the same data, image-text pairs, and intuitively they complement each other as contrasting provides global alignment capacity and generation grants fine-grained understanding. In this work, we present a Contrastive Bi-directional Image-Text generation model (CoBIT), which attempts to unify the three pre-training objectives in one framework. Specifically, CoBIT employs a novel unicoder-decoder structure, consisting of an image unicoder, a text unicoder and a cross-modal decoder. The image/text unicoders can switch between encoding and decoding in different tasks, enabling flexibility and shared knowledge that benefits both image-to-text and text-to-image generations. CoBIT achieves superior performance in image understanding, image-text understanding (Retrieval, Captioning, VQA, SNLI-VE) and text-based content creation, particularly in zero-shot scenarios. For instance, 82.7% in zero-shot ImageNet classification, 9.37 FID score in zero-shot text-to-image generation and 44.8 CIDEr in zero-shot captioning. | ['Jiahui Yu', 'Jason Baldridge', 'Kai-Wei Chang', 'Zhecan Wang', 'Mandy Guo', 'Haoxuan You'] | 2023-03-23 | null | null | null | null | ['zero-shot-text-to-image-generation'] | ['natural-language-processing'] | [ 6.50148690e-01 3.41732979e-01 6.39352500e-02 -3.79792154e-01
-9.63384628e-01 -3.74557137e-01 1.14826107e+00 -3.91345620e-01
-2.68085122e-01 7.04147577e-01 3.59257817e-01 -8.31952468e-02
1.30438209e-01 -7.68041015e-01 -9.68740463e-01 -6.18745863e-01
5.87283373e-01 7.64896631e-01 1.73521936e-02 -3.47182691e-01
6.79246560e-02 -2.84462988e-01 -1.78330445e+00 5.99729955e-01
9.59409177e-01 1.10932839e+00 7.63996124e-01 1.14943564e+00
-4.35578823e-01 9.62064981e-01 -5.18298268e-01 -7.84595847e-01
5.48733622e-02 -7.63233542e-01 -8.13956261e-01 2.81586736e-01
4.97885555e-01 -5.77227712e-01 -4.50389743e-01 8.68448913e-01
8.47802699e-01 -8.10923278e-02 8.98065329e-01 -1.50428295e+00
-1.37163675e+00 7.38308966e-01 -4.51456815e-01 -2.00785726e-01
3.63751173e-01 3.20598990e-01 9.39636946e-01 -9.68481064e-01
8.50242198e-01 1.45077455e+00 2.70980656e-01 7.84561098e-01
-1.06668699e+00 -3.85227412e-01 -4.59858149e-01 1.32905900e-01
-1.11566925e+00 -5.06070137e-01 3.41741502e-01 -5.56029320e-01
1.01503074e+00 1.71960533e-01 5.55847406e-01 1.39856827e+00
2.96082556e-01 9.74353135e-01 9.03962553e-01 -4.27498430e-01
-9.78763625e-02 1.26758501e-01 -3.74466538e-01 3.55278224e-01
-2.94879600e-02 7.92759284e-02 -6.99658632e-01 4.48303550e-01
7.76340842e-01 -9.57977101e-02 -1.57252133e-01 4.89569046e-02
-1.31269169e+00 1.01725960e+00 2.50082135e-01 1.65743709e-01
-3.46036881e-01 4.25269425e-01 4.27680522e-01 2.78497308e-01
1.75125778e-01 3.79594326e-01 -6.27226904e-02 -2.40334928e-01
-1.01957941e+00 1.15831882e-01 6.50798440e-01 1.11695206e+00
7.15735972e-01 4.26143795e-01 -8.38767290e-01 8.88044596e-01
2.89232492e-01 9.90235627e-01 8.30606341e-01 -7.73883581e-01
3.99853796e-01 7.73448423e-02 -2.72680312e-01 -5.81830740e-01
1.19079866e-01 -2.15246841e-01 -9.92712140e-01 -1.22955404e-01
-3.43318582e-01 -1.66982889e-01 -1.69077468e+00 1.65115678e+00
-9.77927148e-02 9.68035087e-02 4.76550728e-01 7.39514410e-01
1.27790856e+00 1.18148279e+00 2.46863872e-01 2.28228644e-02
1.48774087e+00 -1.31891036e+00 -8.00732255e-01 -4.12736565e-01
3.90028447e-01 -1.10298836e+00 9.54512477e-01 -2.60195788e-03
-1.27213395e+00 -8.60906065e-01 -8.53973806e-01 -3.70726377e-01
-4.85695839e-01 -3.84223624e-03 2.38606930e-01 5.87581098e-01
-1.29069400e+00 -4.34742123e-02 -1.46892950e-01 -4.05010641e-01
2.68419623e-01 -1.40308887e-02 -2.48335868e-01 -3.12857747e-01
-1.29936409e+00 8.36453557e-01 8.73692811e-01 -4.84154493e-01
-1.38803113e+00 -5.80135465e-01 -9.74126577e-01 1.47242457e-01
2.94910073e-01 -1.32748461e+00 1.29753757e+00 -1.08699739e+00
-1.49738121e+00 1.03065741e+00 6.81181252e-02 -7.19249487e-01
4.13319111e-01 -1.74923480e-01 -2.68398196e-01 2.74765432e-01
2.92621940e-01 1.54879427e+00 1.26516807e+00 -1.38736343e+00
-5.08796036e-01 -7.60774240e-02 6.36550272e-03 5.06986380e-01
-1.32749170e-01 -9.41619575e-02 -8.00860405e-01 -7.50751317e-01
-3.93487036e-01 -5.63452423e-01 6.24005273e-02 -1.92447290e-01
-4.38542843e-01 1.01751154e-02 9.86764908e-01 -7.44546473e-01
8.99077952e-01 -1.91558969e+00 1.67895079e-01 -5.05921245e-01
1.37452900e-01 4.88866210e-01 -5.97381651e-01 7.64528632e-01
-4.53861356e-02 1.85949802e-01 -2.77592897e-01 -5.16715288e-01
1.49453923e-01 4.90336835e-01 -4.97065187e-01 -3.09131324e-01
4.14984018e-01 1.62489998e+00 -7.80115426e-01 -8.28393757e-01
5.20631135e-01 5.26325703e-01 -5.12914002e-01 3.65341932e-01
-4.17644173e-01 1.87397316e-01 -2.71881014e-01 5.42407155e-01
4.83187675e-01 -4.75839347e-01 -1.37899742e-01 -1.51752442e-01
1.58932745e-01 -3.59150708e-01 -7.83035398e-01 1.97666216e+00
-5.39344847e-01 8.50627661e-01 -3.75513703e-01 -9.05365407e-01
7.84199417e-01 4.98777717e-01 4.40287948e-01 -1.16702378e+00
3.52655500e-01 4.92846519e-02 -3.19508612e-01 -6.95213079e-01
9.09331858e-01 -2.11247012e-01 -1.28126457e-01 4.37864393e-01
6.67595625e-01 -4.70635831e-01 3.96370292e-01 5.33875823e-01
6.13384485e-01 -2.19816063e-03 1.73408478e-01 2.47403562e-01
2.92412013e-01 -8.96339118e-02 -1.60188019e-01 9.88172650e-01
2.47056961e-01 1.21469748e+00 3.02923083e-01 6.15429878e-02
-1.55514252e+00 -1.24590981e+00 -7.58290812e-02 1.15841782e+00
3.21993619e-01 -3.47664297e-01 -8.60093772e-01 -3.28129709e-01
-3.93558919e-01 1.00538266e+00 -5.62125862e-01 -3.04313034e-01
-3.94662982e-03 -4.68392700e-01 7.35247195e-01 3.64341199e-01
7.51475990e-01 -1.19052827e+00 -5.56572855e-01 2.62154996e-01
-6.23063505e-01 -1.40288806e+00 -7.04918504e-01 -1.47832185e-01
-4.79178309e-01 -5.74312747e-01 -1.16872990e+00 -8.79461050e-01
4.79125649e-01 3.53685260e-01 1.29975343e+00 -1.55858070e-01
-6.28976583e-01 7.59127319e-01 -7.80406117e-01 -5.61408758e-01
-6.51493013e-01 -2.70075589e-01 -4.89503235e-01 1.16280764e-01
-1.40002787e-01 -2.90837348e-01 -4.90626127e-01 1.48723394e-01
-1.51513052e+00 6.83839262e-01 9.11305130e-01 1.21167278e+00
5.42100370e-01 -4.57615435e-01 4.58738744e-01 -5.62891960e-01
7.30242729e-01 -5.16015053e-01 -1.64602816e-01 7.12310553e-01
-4.84346807e-01 1.36920556e-01 2.86691457e-01 -4.18660790e-01
-1.30347896e+00 -8.06866866e-03 -2.74160445e-01 -7.65539527e-01
-2.38634553e-02 5.23166537e-01 -4.73585837e-02 2.22591266e-01
7.66206086e-01 1.00660276e+00 -3.66913974e-02 -3.60737294e-02
9.90146101e-01 9.17382538e-01 7.95468330e-01 -4.20273453e-01
6.94575965e-01 2.44338959e-01 -5.15375793e-01 -8.36046219e-01
-6.52578592e-01 -3.36227268e-01 -3.76478314e-01 -3.77527654e-01
1.49383211e+00 -1.12152791e+00 -2.16685578e-01 6.81879580e-01
-1.42056465e+00 -2.65108466e-01 -5.90045929e-01 1.93447441e-01
-9.44733858e-01 3.63100052e-01 -3.05803210e-01 -6.43528938e-01
-9.56665695e-01 -1.35291147e+00 1.58324599e+00 4.17696625e-01
1.76680848e-01 -8.51148725e-01 -9.43653584e-02 6.14015043e-01
4.63777661e-01 -2.04767119e-02 5.91889083e-01 -5.67170382e-01
-5.72998643e-01 -1.63781941e-02 -6.67107522e-01 5.59388399e-01
-2.84119755e-01 -2.71718651e-01 -9.14979517e-01 -7.78711140e-02
-2.48843908e-01 -8.97609949e-01 9.30785298e-01 5.32867670e-01
9.83840942e-01 -3.03220868e-01 9.51456837e-03 7.40284920e-01
1.59794760e+00 4.15420860e-01 1.34316003e+00 4.42730598e-02
7.17429161e-01 3.45384628e-01 3.96122128e-01 4.24214721e-01
6.36769414e-01 5.88050842e-01 3.73146892e-01 -2.78011531e-01
-7.20888197e-01 -4.07737255e-01 5.35735011e-01 8.55055332e-01
4.04281281e-02 -8.65680456e-01 -7.74523020e-01 6.10998869e-01
-1.87297952e+00 -1.06645870e+00 -6.05823845e-02 1.74946642e+00
8.08123291e-01 -2.77414501e-01 -3.02465439e-01 -4.22130466e-01
6.88225508e-01 2.57209092e-01 -4.89035279e-01 -4.39916968e-01
-3.77270848e-01 2.15849549e-01 3.99819732e-01 2.74059772e-01
-6.91305280e-01 1.25996506e+00 6.18634462e+00 1.25559998e+00
-1.04245567e+00 3.76407474e-01 7.54671037e-01 5.24081104e-02
-4.02300775e-01 -1.00391231e-01 -7.23610461e-01 4.58392888e-01
9.54204023e-01 -4.74381328e-01 2.97294140e-01 8.24671030e-01
-1.88770533e-01 -4.49691378e-02 -8.28183711e-01 1.51371515e+00
7.39121854e-01 -1.58827066e+00 8.24353218e-01 3.15537900e-02
1.02024078e+00 1.07344538e-01 2.14407042e-01 4.27906066e-01
3.13164204e-01 -1.14691508e+00 9.36661005e-01 6.46525264e-01
1.32879508e+00 -3.06097239e-01 6.46852732e-01 1.17425628e-01
-1.01092958e+00 9.40836370e-02 -2.60684729e-01 5.13840377e-01
6.25522316e-01 2.35365123e-01 -7.51276314e-01 8.89511943e-01
4.72095609e-01 5.33813655e-01 -3.88122827e-01 7.82400787e-01
-1.55995684e-02 1.66590050e-01 8.18633661e-02 1.68057904e-01
5.38685262e-01 -5.11931814e-03 5.03348589e-01 1.18314183e+00
6.74974978e-01 4.94121127e-02 1.53784916e-01 9.14005399e-01
-2.17425182e-01 2.73768082e-02 -5.69412827e-01 -4.84189719e-01
6.51776269e-02 1.12909627e+00 -5.36559403e-01 -7.96325028e-01
-3.21511894e-01 1.41408491e+00 -2.45745122e-01 4.40599531e-01
-1.10583687e+00 -5.39043665e-01 3.29576761e-01 -1.57592267e-01
4.01574731e-01 -1.25490827e-02 -5.47182113e-02 -1.31405807e+00
-3.06233823e-01 -9.56401825e-01 1.55937135e-01 -1.54078341e+00
-1.08184910e+00 7.71379054e-01 1.03210025e-01 -9.22014654e-01
-6.93652153e-01 -5.26060104e-01 -5.33879638e-01 7.50346124e-01
-1.53615785e+00 -1.56475961e+00 -5.34506023e-01 8.30897748e-01
1.21092248e+00 -2.68391639e-01 5.51666856e-01 3.87667507e-01
-2.20496267e-01 6.02944672e-01 1.36650324e-01 3.05733643e-02
6.84490323e-01 -9.94496047e-01 5.10303259e-01 9.06366110e-01
3.50405723e-01 -1.37477124e-03 5.50046921e-01 -7.10933030e-01
-1.39914143e+00 -1.16785645e+00 6.63309813e-01 -2.74979234e-01
4.58823800e-01 -2.26089612e-01 -5.99722922e-01 5.43663740e-01
6.61957920e-01 -5.05131423e-01 4.10704017e-01 -6.00040853e-01
-2.26232082e-01 1.51152298e-01 -7.26561785e-01 6.70665026e-01
9.55934227e-01 -5.89063644e-01 -5.91164708e-01 4.54122961e-01
1.07865036e+00 -4.79261070e-01 -7.80197799e-01 1.51595548e-01
5.47750652e-01 -8.56074035e-01 1.20527208e+00 -3.54889512e-01
1.10651457e+00 1.65629841e-03 -3.77935678e-01 -1.07767427e+00
-9.03853998e-02 -5.72033644e-01 6.12850636e-02 1.23299050e+00
3.50186884e-01 -4.09894198e-01 3.39677364e-01 2.44226247e-01
-5.12919009e-01 -4.90800977e-01 -6.64177239e-01 -7.78485179e-01
-2.48805106e-01 -5.19665718e-01 5.28402090e-01 6.66926682e-01
-5.37791491e-01 7.92591870e-01 -1.07068253e+00 -4.05735701e-01
5.23938417e-01 -2.41590559e-01 9.91531134e-01 -8.83772552e-01
-4.44366485e-01 -5.02994955e-01 -3.63065839e-01 -1.15518236e+00
-2.85021186e-01 -1.02126181e+00 2.39382952e-01 -1.98872948e+00
5.35440981e-01 8.16384181e-02 2.07736224e-01 5.77312827e-01
-1.80935577e-01 4.09373432e-01 6.60165429e-01 2.27612227e-01
-7.36581087e-01 9.52634156e-01 1.66775835e+00 -3.06360841e-01
3.93109582e-02 -6.34102941e-01 -6.38370752e-01 2.57546067e-01
5.09476066e-01 -3.79149169e-01 -7.27243960e-01 -7.21700251e-01
6.72469065e-02 3.12331676e-01 5.05537510e-01 -1.00348806e+00
2.56013185e-01 -4.35247757e-02 2.89337695e-01 -6.74227178e-01
6.23860240e-01 -3.37349594e-01 3.54833245e-01 3.76395404e-01
-2.85079837e-01 -1.94214031e-01 2.12776624e-02 5.99274576e-01
-5.26686430e-01 -5.31375408e-01 6.28790081e-01 -2.77875692e-01
-1.08935654e+00 4.07804132e-01 -1.80613980e-01 2.50209361e-01
1.16205013e+00 -5.09755433e-01 -5.96195102e-01 -8.26659799e-01
-4.66194153e-01 2.19214156e-01 1.77266866e-01 7.88232088e-01
9.59021688e-01 -1.22170091e+00 -1.08244073e+00 1.01267785e-01
3.85060430e-01 -3.00375193e-01 7.28031814e-01 6.05859756e-01
-4.42645431e-01 6.70915365e-01 -5.18644989e-01 -8.36693227e-01
-1.11651099e+00 3.43598843e-01 6.73840716e-02 -3.26401085e-01
-4.53815132e-01 8.79895210e-01 7.23490059e-01 -6.45582154e-02
-5.08968830e-02 2.66736329e-01 2.01937649e-02 2.02158108e-01
6.49213076e-01 1.65147465e-02 -3.20067436e-01 -8.41483653e-01
3.61602426e-01 6.41383708e-01 -1.68584108e-01 -4.43088889e-01
1.07322419e+00 -3.44517320e-01 -3.20178792e-02 1.88363791e-01
1.22099257e+00 -8.82743597e-01 -1.05876493e+00 -1.71166912e-01
-5.89457810e-01 -2.69504339e-01 1.34091720e-01 -1.01823688e+00
-9.62988555e-01 1.09056020e+00 6.65739477e-01 9.24189761e-02
1.27901721e+00 2.24568903e-01 1.22230291e+00 2.19841927e-01
5.53519577e-02 -1.07582641e+00 7.62485504e-01 7.43619680e-01
1.07292867e+00 -1.41122103e+00 -3.24415803e-01 -2.54662298e-02
-1.21410573e+00 9.28830028e-01 5.40128231e-01 4.02384132e-01
1.27323210e-01 -4.90379520e-02 -1.61513984e-01 -2.14693367e-01
-9.31595683e-01 -6.11598134e-01 6.66658580e-01 9.19698060e-01
1.36261165e-01 -8.49422514e-02 -9.23922881e-02 3.26871961e-01
-2.99619913e-01 1.28117427e-01 4.19309855e-01 6.68532491e-01
-5.06928980e-01 -9.54273462e-01 -2.61540562e-01 4.91204143e-01
-2.31067359e-01 -6.37328684e-01 -1.53532162e-01 6.80780113e-01
9.29790288e-02 8.20033312e-01 3.01596105e-01 -4.39912438e-01
-2.69430056e-02 1.83751240e-01 4.18327004e-01 -6.42619610e-01
-3.75435919e-01 2.04140604e-01 3.01975477e-02 -2.04308704e-01
-3.44831765e-01 -1.44801959e-01 -9.29706156e-01 -1.02881245e-01
-3.46917033e-01 -1.96755230e-01 9.92889822e-01 8.94066393e-01
5.59238136e-01 7.51782060e-01 3.57105106e-01 -8.44459534e-01
-1.48869455e-01 -1.02426064e+00 -3.41556907e-01 4.94131505e-01
-4.91270842e-03 -3.75906050e-01 -9.27290786e-03 7.82541811e-01] | [11.037555694580078, 1.0820796489715576] |
5638c3da-322a-4f0f-8ffa-e9893451dc89 | continuous-copy-paste-for-one-stage-multi | null | null | http://openaccess.thecvf.com//content/ICCV2021/html/Xu_Continuous_Copy-Paste_for_One-Stage_Multi-Object_Tracking_and_Segmentation_ICCV_2021_paper.html | http://openaccess.thecvf.com//content/ICCV2021/papers/Xu_Continuous_Copy-Paste_for_One-Stage_Multi-Object_Tracking_and_Segmentation_ICCV_2021_paper.pdf | Continuous Copy-Paste for One-Stage Multi-Object Tracking and Segmentation | Current one-step multi-object tracking and segmentation (MOTS) methods lag behind recent two-step methods. By separating the instance segmentation stage from the tracking stage, two-step methods can exploit non-video datasets as extra data for training instance segmentation. Moreover, instances belonging to different IDs on different frames, rather than limited numbers of instances in raw consecutive frames, can be gathered to allow more effective hard example mining in the training of trackers. In this paper, we bridge this gap by presenting a novel data augmentation strategy named continuous copy-paste (CCP). Our intuition behind CCP is to fully exploit the pixel-wise annotations provided by MOTS to actively increase the number of instances as well as unique instance IDs in training. Without any modifications to frameworks, current MOTS methods achieve significant performance gains when trained with CCP. Based on CCP, we propose the first effective one-stage online MOTS method named CCPNet, which generates instance masks as well as the tracking results in one shot. Our CCPNet surpasses all state-of-the-art methods by large margins (3.8% higher sMOTSA and 4.1% higher MOTSA for pedestrians on the KITTI MOTS Validation) and ranks 1st on the KITTI MOTS leaderboard. Evaluations across three datasets also demonstrate the effectiveness of both CCP and CCPNet. Our codes are publicly available at: https://github.com/detectRecog/CCP. | ['Liusheng Huang', 'Zhi Chen', 'Wei Yang', 'Zhenbo Shi', 'Ajin Meng', 'Zhenbo Xu'] | 2021-01-01 | null | null | null | iccv-2021-1 | ['multi-object-tracking-and-segmentation'] | ['computer-vision'] | [ 3.49510223e-01 -4.26810933e-03 -2.70565063e-01 -2.21829981e-01
-8.67669165e-01 -5.28106451e-01 5.52270651e-01 -1.70060679e-01
-6.25908911e-01 7.56151438e-01 -2.50512958e-01 -8.95624980e-02
3.07114482e-01 -4.55112606e-01 -9.96811152e-01 -5.69112360e-01
7.87706003e-02 5.90624094e-01 1.04833066e+00 2.79181208e-02
-1.93302050e-01 1.36442566e-02 -1.74305308e+00 2.69242793e-01
7.94854045e-01 9.40804601e-01 3.36920977e-01 6.85012281e-01
-7.80881122e-02 4.97257173e-01 -6.22410774e-01 -3.93435329e-01
6.69781685e-01 -1.47954643e-01 -3.82581770e-01 5.54496013e-02
7.45848119e-01 -3.30994397e-01 -2.16274843e-01 9.07644928e-01
4.55009818e-01 7.24441335e-02 2.90311337e-01 -1.53134978e+00
-7.07592741e-02 6.74189150e-01 -1.04265869e+00 3.73335868e-01
-1.17254443e-02 6.41826332e-01 7.83928931e-01 -8.02228689e-01
4.84806836e-01 1.02467573e+00 8.99081230e-01 6.85682654e-01
-1.08698940e+00 -9.13487792e-01 2.60140270e-01 1.81823865e-01
-1.24486887e+00 -4.60339874e-01 4.95999038e-01 -4.95419949e-01
5.81323802e-01 4.29099560e-01 8.42741072e-01 9.95614767e-01
-2.20235184e-01 1.29325640e+00 1.19818211e+00 -1.05486020e-01
1.16942741e-01 1.59444436e-01 2.10803404e-01 6.35146201e-01
5.30589461e-01 3.10669571e-01 -5.93316317e-01 1.81383684e-01
5.65968454e-01 -1.17358163e-01 7.24335983e-02 -2.30468750e-01
-1.21396291e+00 5.10237396e-01 4.13960218e-01 -1.44605208e-02
-2.90580541e-01 2.52246171e-01 3.11327636e-01 -1.13119207e-01
4.32082981e-01 1.73990712e-01 -5.72573960e-01 -3.01580638e-01
-1.25645411e+00 6.84541091e-02 4.89660949e-01 1.01808929e+00
6.68977976e-01 -3.60836685e-02 -4.82437789e-01 5.43037176e-01
2.69436181e-01 7.85144031e-01 1.65385500e-01 -9.71089900e-01
5.28670192e-01 6.07834518e-01 8.39843899e-02 -4.89395499e-01
-2.72113800e-01 -6.41020954e-01 -3.72639179e-01 2.43939921e-01
7.06714809e-01 -2.97555029e-01 -1.24201417e+00 1.56699538e+00
7.74864018e-01 5.99538147e-01 -2.70996183e-01 9.06072021e-01
7.84293771e-01 5.03600597e-01 1.69481635e-01 8.38565826e-02
1.59101570e+00 -1.09307027e+00 -4.28792924e-01 -5.99718511e-01
6.72228694e-01 -4.95026380e-01 7.16688991e-01 2.02414155e-01
-8.26233089e-01 -6.88228488e-01 -9.98932362e-01 2.77492344e-01
-2.99613476e-01 3.55084717e-01 6.26932800e-01 8.33202004e-01
-8.75348806e-01 3.86305779e-01 -1.11493933e+00 -3.13206345e-01
9.30465162e-01 4.57381159e-01 -1.38939112e-01 -6.55928329e-02
-7.89181650e-01 4.62001026e-01 5.12627304e-01 5.97531721e-02
-9.21606362e-01 -1.00850403e+00 -6.16674185e-01 -2.58151233e-01
9.05057371e-01 -5.20968378e-01 1.25824189e+00 -8.02260697e-01
-1.18305969e+00 7.19359636e-01 -2.42849201e-01 -8.56309831e-01
7.52408206e-01 -3.99012208e-01 -1.90806389e-01 6.33838400e-02
4.00831431e-01 1.25737572e+00 8.86697829e-01 -1.19914734e+00
-1.16874778e+00 2.37836805e-03 -3.92680168e-02 -8.21689442e-02
-1.64087191e-01 -8.15615207e-02 -8.73873413e-01 -3.43670100e-01
-1.90049559e-01 -1.08980465e+00 -2.62760013e-01 1.06625699e-01
-5.67348242e-01 -2.48568341e-01 1.08976841e+00 -4.85300392e-01
9.37423289e-01 -2.07044911e+00 -1.11399963e-01 -7.61955008e-02
3.81887704e-01 7.31320798e-01 -1.56157479e-01 -9.81288180e-02
3.27807575e-01 -1.05259396e-01 -2.10534126e-01 -6.48196697e-01
-3.96811329e-02 3.20992142e-01 9.21005458e-02 2.97969818e-01
4.47346210e-01 1.07855427e+00 -1.05453861e+00 -6.96219146e-01
4.11622375e-01 3.48075986e-01 -4.95581657e-01 -1.15866378e-01
-4.15249884e-01 7.80595839e-01 -3.91592085e-01 9.22036588e-01
7.96433985e-01 -3.36084396e-01 -1.19230151e-01 -1.58304214e-01
-3.69543910e-01 7.71535113e-02 -1.30773270e+00 1.40302444e+00
2.92581841e-02 8.18403780e-01 -1.00593738e-01 -7.10388005e-01
5.17771721e-01 -3.16487178e-02 6.26305640e-01 -8.23128879e-01
1.23328276e-01 1.77530289e-01 2.86672004e-02 -1.90555468e-01
6.47605479e-01 2.76805997e-01 1.40376352e-02 2.04256792e-02
-1.48211373e-02 5.24612069e-01 6.61855578e-01 3.59898388e-01
1.18046248e+00 4.46822971e-01 -1.67301800e-02 -8.78448635e-02
3.05773884e-01 2.64300942e-01 9.46793735e-01 1.10942483e+00
-5.33901036e-01 5.44639826e-01 2.86836952e-01 -2.66651243e-01
-8.51622760e-01 -8.94520998e-01 -9.22786891e-02 1.02391589e+00
3.50971609e-01 -4.07736957e-01 -8.39992881e-01 -8.03617001e-01
4.44382168e-02 3.29759330e-01 -6.12265110e-01 2.24423110e-01
-5.71007192e-01 -9.38788176e-01 7.48734057e-01 7.60071516e-01
7.40258932e-01 -9.54213262e-01 -9.44565237e-01 2.24526092e-01
-1.98355168e-01 -1.65451121e+00 -4.89087254e-01 1.08627811e-01
-6.26219988e-01 -1.15643919e+00 -6.43758059e-01 -2.40572885e-01
5.24251878e-01 5.46146274e-01 9.72130835e-01 1.73147678e-01
-4.97987866e-01 3.12661737e-01 -5.20034373e-01 -5.81494749e-01
-2.30048597e-01 2.37293154e-01 -7.76476087e-03 1.29220441e-01
4.50316638e-01 -3.09459955e-01 -7.58270085e-01 5.18612981e-01
-5.93621135e-01 4.20384169e-01 8.71032000e-01 5.89583039e-01
8.36678386e-01 -1.84796974e-01 3.17623407e-01 -8.24923515e-01
-3.55788946e-01 -4.46025223e-01 -9.72811282e-01 -1.64558347e-02
-4.27066058e-01 -2.80331802e-02 1.90190241e-01 -6.04641795e-01
-7.41764426e-01 4.78887945e-01 -6.19968250e-02 -5.59108794e-01
-2.84548849e-01 -8.61089379e-02 -6.21077903e-02 -1.88531160e-01
4.49856997e-01 1.09867096e-01 -1.17083043e-01 -5.21166325e-01
3.64521861e-01 3.02911609e-01 6.38899505e-01 -3.98206949e-01
1.12130392e+00 6.35476828e-01 -1.00482784e-01 -7.62371004e-01
-8.89287651e-01 -8.68640661e-01 -6.63472116e-01 -5.43213010e-01
1.03149807e+00 -1.06071734e+00 -6.82554841e-01 5.35795033e-01
-8.36478353e-01 -5.66537559e-01 -4.96287167e-01 3.39603454e-01
-5.73786013e-02 1.99782878e-01 -2.73014009e-01 -1.03706324e+00
-2.06147149e-01 -1.05258846e+00 1.31573510e+00 6.25625789e-01
1.66868389e-01 -5.81434548e-01 -2.32656330e-01 5.68014383e-01
1.81587175e-01 3.98158580e-01 -2.01436207e-01 -6.92893326e-01
-1.11763275e+00 -2.31360868e-01 -3.41806561e-01 1.78474605e-01
-1.79672122e-01 8.21385626e-03 -1.00808442e+00 -3.24508935e-01
-5.81803501e-01 -1.40421882e-01 1.17551184e+00 6.12387538e-01
8.48118663e-01 -4.74914238e-02 -6.30317867e-01 6.74347937e-01
1.24631464e+00 6.35675415e-02 5.12526810e-01 4.91464615e-01
9.65725899e-01 2.75738597e-01 8.72883201e-01 3.64376754e-01
3.68622422e-01 9.20406580e-01 5.65347672e-01 -1.56468809e-01
-5.64849734e-01 -1.96046576e-01 5.20543098e-01 2.59913713e-01
-1.83719397e-01 -1.80763140e-01 -8.04168820e-01 7.37128258e-01
-1.93652380e+00 -9.82338130e-01 -5.83685517e-01 2.24527764e+00
6.13525808e-01 5.18375337e-01 6.02202773e-01 6.02100715e-02
8.53136003e-01 7.97628388e-02 -6.56052947e-01 3.92906010e-01
-1.47086906e-03 1.89957678e-01 1.08893788e+00 3.03435862e-01
-1.39234722e+00 1.02562928e+00 4.59602880e+00 9.38995361e-01
-7.09373891e-01 4.10851777e-01 3.64905238e-01 -4.17064369e-01
3.74503225e-01 9.60617214e-02 -1.47736537e+00 8.11368942e-01
9.20948982e-01 4.37314630e-01 1.11797996e-01 8.08640778e-01
1.86016396e-01 -3.30368340e-01 -8.47152948e-01 8.41613352e-01
-3.61823827e-01 -1.40837240e+00 -3.01057875e-01 2.56468862e-01
7.20643938e-01 4.34093088e-01 -5.12875132e-02 4.96997416e-01
3.44141811e-01 -4.82701927e-01 9.03039813e-01 1.20443650e-01
6.38558507e-01 -3.49921018e-01 7.12981164e-01 2.41756991e-01
-1.68529999e+00 -1.20391130e-01 -2.52557904e-01 1.14466742e-01
4.57741112e-01 5.84993720e-01 -7.94611454e-01 8.66311252e-01
8.97333264e-01 7.96751499e-01 -9.71734822e-01 1.54429746e+00
-1.46675929e-01 9.72877800e-01 -7.91085720e-01 9.16878507e-02
3.74185652e-01 1.23917453e-01 8.82294595e-01 1.30374610e+00
5.35471700e-02 -5.93815893e-02 4.66856867e-01 7.46670544e-01
1.19829457e-02 -4.07212138e-01 -1.26128137e-01 1.36611491e-01
5.99544942e-01 1.53065491e+00 -1.04647422e+00 -5.54411888e-01
-4.62072670e-01 6.83799565e-01 -6.58731237e-02 9.10356641e-02
-1.31265664e+00 -3.81094813e-02 6.24451041e-01 3.85274947e-01
8.14231873e-01 -8.99397582e-02 -1.76313475e-01 -8.76702845e-01
1.56520635e-01 -6.78350151e-01 3.79932612e-01 -3.13562423e-01
-9.69953418e-01 4.59742337e-01 1.64137259e-01 -1.43984139e+00
2.87682749e-02 -5.90385139e-01 -4.64413345e-01 4.11098868e-01
-1.74071777e+00 -1.34662151e+00 -5.68564534e-01 4.05286431e-01
7.43195772e-01 4.00556624e-02 8.91487449e-02 6.93644822e-01
-8.45187068e-01 7.50344753e-01 -1.23357654e-01 3.79116774e-01
5.21138370e-01 -1.21936142e+00 6.40339077e-01 1.24161983e+00
2.59660929e-01 1.80419385e-01 7.06270278e-01 -8.03616047e-01
-1.13186038e+00 -1.40245056e+00 2.87726849e-01 -6.55646145e-01
5.62366784e-01 -3.94378543e-01 -7.45325446e-01 6.64753914e-01
-1.66506190e-02 3.81969273e-01 3.02588999e-01 -5.96593656e-02
-1.46513760e-01 -7.69962445e-02 -9.81608391e-01 4.92870301e-01
1.35040426e+00 6.57347962e-02 -2.14806870e-01 4.02937859e-01
6.55268788e-01 -7.26727545e-01 -5.87607622e-01 4.09343064e-01
5.08718431e-01 -9.22303021e-01 1.06630504e+00 -1.71997100e-01
4.09620292e-02 -7.31926203e-01 -1.84845608e-02 -7.22019374e-01
-3.21682729e-02 -8.06046486e-01 -4.03737813e-01 1.35752082e+00
3.83978218e-01 -6.29907608e-01 1.01904655e+00 4.57356811e-01
-3.45395386e-01 -6.15940630e-01 -1.17412674e+00 -1.12664807e+00
-3.80699277e-01 -4.62147146e-01 4.23073173e-01 5.07259250e-01
-6.82055831e-01 6.06330447e-02 -4.46407884e-01 4.37387854e-01
1.09803307e+00 -1.12355053e-01 1.22114134e+00 -1.30506790e+00
-5.37708998e-01 -1.86602220e-01 -5.65714121e-01 -1.31146038e+00
-3.20762634e-01 -7.13314176e-01 2.99457252e-01 -1.41417468e+00
3.31027627e-01 -7.01630533e-01 -2.50064433e-01 8.46693039e-01
-4.82604384e-01 6.54141247e-01 5.77440202e-01 3.02457720e-01
-1.10346329e+00 2.95859724e-01 9.68140125e-01 -6.10388853e-02
-3.31135631e-01 1.88854203e-01 -4.92401809e-01 6.29096687e-01
7.43151486e-01 -8.01360130e-01 -2.25307703e-01 -2.37459570e-01
-3.35492194e-01 -4.24451709e-01 8.19866955e-01 -1.58862996e+00
3.54687840e-01 5.02324402e-02 4.88850921e-01 -1.02954936e+00
4.09144878e-01 -6.43613577e-01 2.41810232e-01 5.77270865e-01
1.29780293e-01 -2.20771462e-01 5.15811265e-01 7.52576888e-01
2.41579294e-01 -3.07423621e-02 7.42890358e-01 6.18474698e-03
-1.14591002e+00 4.26268727e-01 -2.65645031e-02 2.07533285e-01
1.15698397e+00 -6.35887563e-01 -4.51510519e-01 1.85652032e-01
-4.08772558e-01 5.77012539e-01 2.61538953e-01 5.14483333e-01
3.09079319e-01 -1.07394242e+00 -6.79881573e-01 1.60360813e-01
8.20381716e-02 1.95136726e-01 2.76174247e-01 1.36352253e+00
-1.85317043e-02 3.01309079e-01 -8.74275789e-02 -1.07288074e+00
-1.54997563e+00 2.14485690e-01 2.11012676e-01 -2.95623392e-01
-9.02143538e-01 9.10810471e-01 2.71695226e-01 -1.32992283e-01
1.70030877e-01 -3.72079015e-01 1.38856739e-01 -2.61180159e-02
6.38457835e-01 3.45209599e-01 -1.49936423e-01 -5.84000885e-01
-5.34679353e-01 3.55271608e-01 -3.46146822e-01 -1.06084906e-01
1.11101401e+00 8.82721096e-02 5.42816937e-01 1.51603132e-01
6.77791655e-01 -1.41642645e-01 -2.05807853e+00 -1.49878561e-01
1.39973491e-01 -6.21199965e-01 -8.69363248e-02 -8.57851505e-01
-1.33740139e+00 4.20817763e-01 8.04183185e-01 8.16594362e-02
9.30515051e-01 2.21897230e-01 1.10202050e+00 -5.74865974e-02
4.23123360e-01 -1.09545112e+00 3.13252844e-02 2.01136589e-01
1.58604994e-01 -1.47039139e+00 3.36119160e-02 -5.15774906e-01
-6.36895239e-01 6.02608860e-01 8.25657547e-01 7.04164430e-02
1.97841451e-01 4.47155118e-01 -3.45164128e-02 -1.09049208e-01
-5.45662224e-01 -7.36221552e-01 2.44693995e-01 7.71257818e-01
-2.53725648e-02 4.91679348e-02 -6.99751452e-02 3.58017623e-01
1.00380033e-01 9.74142700e-02 3.29471290e-01 1.04233909e+00
-6.01617634e-01 -1.13168645e+00 -5.37208617e-01 5.25134802e-01
-3.59470725e-01 6.21840954e-02 -2.20866501e-01 9.74310517e-01
3.97073209e-01 9.53016579e-01 -1.02801584e-01 -5.01486719e-01
3.30263257e-01 -1.11198053e-01 3.52310926e-01 -3.84891152e-01
-5.91703057e-01 2.33440828e-02 1.12481914e-01 -7.22419262e-01
-6.93368733e-01 -1.02750123e+00 -1.29033661e+00 -2.70722032e-01
-5.49798667e-01 -1.47820190e-01 6.10569656e-01 1.05514801e+00
5.41581571e-01 8.62723053e-01 1.44334167e-01 -1.15056169e+00
-2.03904226e-01 -8.22674215e-01 -1.30926400e-01 2.77295768e-01
3.16079497e-01 -9.70585704e-01 -1.25256807e-01 1.37213826e-01] | [6.546874046325684, -1.9836654663085938] |
bcf0a889-d8f1-422d-9e8c-2795103bde11 | digital-twin-based-3d-map-management-for-edge | 2305.16571 | null | https://arxiv.org/abs/2305.16571v1 | https://arxiv.org/pdf/2305.16571v1.pdf | Digital Twin-Based 3D Map Management for Edge-Assisted Mobile Augmented Reality | In this paper, we design a 3D map management scheme for edge-assisted mobile augmented reality (MAR) to support the pose estimation of individual MAR device, which uploads camera frames to an edge server. Our objective is to minimize the pose estimation uncertainty of the MAR device by periodically selecting a proper set of camera frames for uploading to update the 3D map. To address the challenges of the dynamic uplink data rate and the time-varying pose of the MAR device, we propose a digital twin (DT)-based approach to 3D map management. First, a DT is created for the MAR device, which emulates 3D map management based on predicting subsequent camera frames. Second, a model-based reinforcement learning (MBRL) algorithm is developed, utilizing the data collected from both the actual and the emulated data to manage the 3D map. With extensive emulated data provided by the DT, the MBRL algorithm can quickly provide an adaptive map management policy in a highly dynamic environment. Simulation results demonstrate that the proposed DT-based 3D map management outperforms benchmark schemes by achieving lower pose estimation uncertainty and higher data efficiency in dynamic environments. | ['Weihua Zhuang', 'Xuemin Shen', 'Nan Cheng', 'Mushu Li', 'Jie Gao', 'Conghao Zhou'] | 2023-05-26 | null | null | null | null | ['pose-estimation', 'model-based-reinforcement-learning'] | ['computer-vision', 'reasoning'] | [-2.76301175e-01 -5.17028496e-02 -2.05583468e-01 -1.96464770e-02
-8.83323193e-01 -4.02669191e-01 -3.12393606e-02 -3.10472101e-01
-2.18113020e-01 6.43014014e-01 1.22389689e-01 -3.73246878e-01
-4.90104184e-02 -7.84083366e-01 -1.00684893e+00 -5.83722889e-01
-2.66476721e-01 5.18986046e-01 4.63974476e-01 1.13922618e-01
8.31717923e-02 5.30902088e-01 -1.25168300e+00 -5.95184207e-01
5.06442070e-01 1.46440649e+00 7.25649893e-01 9.55766976e-01
3.25114757e-01 4.85122561e-01 -5.08581519e-01 1.60147175e-01
2.54700094e-01 7.27883726e-02 -2.68231109e-02 3.21256518e-01
-1.86959326e-01 -9.70134854e-01 -5.51881313e-01 5.14271975e-01
1.05000067e+00 2.02574819e-01 -9.35607851e-02 -1.33872604e+00
2.52576560e-01 1.04447156e-01 -6.78806722e-01 2.92248636e-01
8.23775709e-01 -2.03945994e-01 2.14261472e-01 -8.14297497e-01
6.20304644e-01 6.29352689e-01 7.11050093e-01 2.97177851e-01
-6.19262457e-01 -5.06339014e-01 3.22422087e-01 2.45929182e-01
-1.83403826e+00 -5.81710756e-01 6.97625637e-01 -2.93784682e-02
5.13845205e-01 2.37568945e-01 8.89117837e-01 4.70460922e-01
2.48424515e-01 6.43740058e-01 6.43330336e-01 -4.17958200e-01
6.29111528e-01 -8.52103382e-02 -3.98035437e-01 4.38785732e-01
1.43377502e-02 2.61626709e-02 -9.16442275e-01 -4.34597880e-01
9.90364015e-01 -1.58099487e-01 -3.50625485e-01 -7.71450162e-01
-1.30051959e+00 6.56171069e-02 1.17520422e-01 -3.34101498e-01
-8.31664801e-01 4.76747125e-01 1.14189900e-01 1.08316645e-01
2.84776717e-01 -2.81329513e-01 -4.94315863e-01 -6.38854384e-01
-6.62033141e-01 -7.06078708e-02 6.63382530e-01 1.63037586e+00
6.58093750e-01 -2.67367810e-02 9.81026590e-02 3.93879533e-01
6.07508957e-01 9.67146873e-01 1.06504709e-01 -1.21022236e+00
6.00313365e-01 3.41390446e-02 5.60553670e-01 -1.02530336e+00
-3.35093349e-01 -2.01472238e-01 -5.62015295e-01 -2.76657701e-01
-2.19419748e-01 -6.13547504e-01 -4.98464614e-01 1.53199744e+00
1.00843334e+00 9.97221589e-01 2.23208051e-02 1.03892612e+00
2.86674857e-01 8.31507981e-01 -5.00218868e-01 -7.09088445e-01
8.99283290e-01 -3.04988295e-01 -8.37841511e-01 -3.94205749e-02
5.23805737e-01 -5.22367537e-01 3.47071499e-01 2.04732299e-01
-1.20070505e+00 -2.60270447e-01 -1.31274962e+00 5.79716623e-01
1.17491201e-01 2.16898486e-01 7.88006186e-02 7.77082980e-01
-1.36999035e+00 -1.00772038e-01 -9.14817750e-01 -4.08073664e-01
2.85049230e-02 8.60103607e-01 7.12721422e-02 -1.46303535e-01
-9.49651837e-01 6.52532756e-01 2.34547600e-01 4.51449901e-02
-7.97468066e-01 -5.13290286e-01 -6.04858518e-01 -1.06581420e-01
7.16428220e-01 -8.00320625e-01 1.36261249e+00 -2.01967329e-01
-1.96900868e+00 4.35936272e-01 -2.63970286e-01 -3.31998616e-01
3.20750505e-01 -1.67473629e-01 -4.94490385e-01 2.22172126e-01
1.22318380e-02 3.27236325e-01 7.39982665e-01 -1.44666207e+00
-8.64874840e-01 -5.18347800e-01 1.66919321e-01 8.02876353e-01
-1.17834240e-01 -4.33375388e-01 -1.17743421e+00 -1.84400499e-01
5.15998960e-01 -1.22600341e+00 -2.70988226e-01 -1.29238635e-01
-1.36482641e-01 4.73067701e-01 9.20832276e-01 -4.20019120e-01
1.26850748e+00 -2.19971323e+00 9.56334081e-03 4.84702587e-01
7.06596971e-02 -1.33926988e-01 4.38649267e-01 1.91703558e-01
5.80521762e-01 -3.36324722e-01 5.87627530e-01 -5.76307774e-01
-3.19733977e-01 4.16230708e-01 -1.18597016e-01 3.33260715e-01
-6.98006809e-01 4.46580023e-01 -9.09165561e-01 -4.95657593e-01
5.40465176e-01 4.89529252e-01 -6.02241755e-01 3.45586985e-01
1.35596290e-01 7.03351736e-01 -7.28570759e-01 6.62273645e-01
1.14620984e+00 -3.39363486e-01 4.35161531e-01 -1.07131891e-01
4.17456590e-03 2.39070114e-02 -1.55941546e+00 1.83679593e+00
-7.40528286e-01 2.95826077e-01 4.88021970e-01 -4.66778487e-01
8.10844541e-01 4.74810660e-01 9.36227083e-01 -7.32891858e-01
1.79953158e-01 2.45027721e-01 -7.56180406e-01 -2.26548165e-01
9.12923932e-01 4.34520096e-01 -3.53698760e-01 6.59748614e-01
-4.21172380e-01 -1.47793489e-02 -5.32811046e-01 2.22308919e-01
1.31397116e+00 1.78170711e-01 2.81079024e-01 2.14512900e-01
5.79749644e-01 -3.12211454e-01 8.26653063e-01 7.27254987e-01
-3.28191280e-01 2.39621386e-01 -1.86697930e-01 -3.28560024e-01
-7.84997582e-01 -1.19303513e+00 1.48531020e-01 6.73102319e-01
1.02764118e+00 -2.28209585e-01 -5.88767409e-01 -3.93788964e-01
8.42882395e-02 3.15003753e-01 -1.44003779e-02 1.03784502e-02
-5.44986248e-01 -5.57930887e-01 -8.86517614e-02 3.49351019e-01
7.85956621e-01 -8.52895454e-02 -5.97456753e-01 3.60994548e-01
-3.53284299e-01 -1.50577295e+00 -8.24278295e-01 -1.51921824e-01
-7.97955036e-01 -7.38132834e-01 -3.41850907e-01 -4.71467316e-01
7.81380117e-01 9.23726201e-01 6.81860030e-01 -6.55975863e-02
5.10634482e-01 1.01238394e+00 -4.43463326e-01 -2.57136583e-01
-9.63897854e-02 -3.41410283e-03 4.77776468e-01 1.30606025e-01
2.42050197e-02 -6.17913008e-01 -6.71920180e-01 9.91750181e-01
-2.93315530e-01 1.16280504e-01 2.62714922e-01 4.70579475e-01
1.08946919e+00 2.18792289e-01 5.09550750e-01 -2.99578011e-01
2.31756061e-01 -6.74544156e-01 -8.88442039e-01 9.21641570e-03
-5.62686980e-01 -3.60611916e-01 1.43793285e-01 -3.60362142e-01
-9.69465256e-01 4.43240374e-01 5.81388287e-02 -5.07660985e-01
3.95855337e-01 3.29806328e-01 -4.52473193e-01 -2.99127996e-01
6.99605420e-02 3.28312099e-01 -8.35425314e-03 -5.80528192e-02
8.71599540e-02 1.13482809e+00 4.71752435e-01 -4.20486122e-01
8.81439984e-01 5.48872471e-01 5.92060350e-02 -7.08960056e-01
-3.74190986e-01 -6.27613366e-01 -2.80709952e-01 -7.41998017e-01
5.87060869e-01 -1.66541290e+00 -8.90051425e-01 3.52399319e-01
-9.73382533e-01 -2.39602655e-01 5.36193773e-02 7.29490101e-01
-9.86426592e-01 2.26375461e-01 -2.53042340e-01 -9.63990867e-01
-1.70177251e-01 -1.30616975e+00 1.38358998e+00 2.75236011e-01
1.91606760e-01 -6.25858009e-01 -1.19264927e-02 4.74165231e-01
3.26887488e-01 7.42580369e-02 1.76095262e-01 1.09104939e-01
-1.34339845e+00 -5.30038953e-01 6.17446862e-02 -4.78613108e-01
3.38059887e-02 -5.11657298e-01 -8.09856117e-01 -4.71639097e-01
-6.49392456e-02 2.77208924e-01 -2.74990678e-01 6.68455958e-01
9.76837218e-01 -4.74760495e-02 -6.46048486e-01 7.49341249e-01
1.18909538e+00 4.43366110e-01 5.76258540e-01 4.49611276e-01
5.72779119e-01 -2.88819671e-01 1.20839286e+00 1.01339662e+00
1.09518588e+00 1.35051084e+00 7.80283391e-01 3.40317637e-01
2.90584177e-01 -3.83749336e-01 4.67809260e-01 9.18720424e-01
2.06545219e-01 -5.59688568e-01 -3.64283472e-01 -9.58045654e-04
-2.23014569e+00 -5.64199924e-01 1.95397094e-01 2.69810438e+00
2.46313110e-01 1.49539813e-01 9.82289389e-02 2.97916196e-02
1.08460081e+00 1.06645755e-01 -6.64223850e-01 8.62968937e-02
3.91633987e-01 -6.37484729e-01 9.07159448e-01 5.11012077e-01
-7.96730399e-01 6.04970276e-01 5.90853119e+00 4.47893679e-01
-9.76700604e-01 1.95444137e-01 2.65866131e-01 -8.00470188e-02
-3.97437736e-02 -4.18047793e-02 -8.10927689e-01 8.63816440e-01
1.08898473e+00 -2.20817760e-01 4.81061459e-01 9.62216675e-01
8.08623075e-01 -6.44544244e-01 -7.16345370e-01 1.34841216e+00
-2.37598829e-02 -1.53790712e+00 -4.65542972e-01 4.39379692e-01
5.83948493e-01 2.15199888e-01 -1.70132652e-01 6.02252632e-02
6.50748238e-02 9.85533521e-02 7.39653468e-01 5.21374702e-01
8.36658359e-01 -9.88248169e-01 7.14727938e-01 5.53473890e-01
-1.58156812e+00 -2.35083580e-01 -2.04674855e-01 1.14504196e-01
6.31544769e-01 6.70003414e-01 -9.62704003e-01 5.87531328e-01
6.36204779e-01 4.52080578e-01 4.47184667e-02 1.17796755e+00
7.38357082e-02 1.02345742e-01 -6.14298522e-01 8.77753869e-02
-3.46424639e-01 2.62480957e-04 7.97992885e-01 3.18664134e-01
7.88066089e-01 4.30011988e-01 5.10389507e-01 8.80126655e-02
-1.50910551e-02 -1.90997198e-01 -6.13420308e-01 6.33703768e-01
1.27123475e+00 9.58935559e-01 -5.40185869e-01 -2.15778202e-01
-3.38262469e-01 1.01296949e+00 -2.34929740e-01 2.75044918e-01
-9.72869456e-01 1.56200230e-01 6.22353554e-01 2.87200123e-01
3.38033378e-01 -6.56534612e-01 5.95252104e-02 -9.30195212e-01
2.19603688e-01 -2.86300451e-01 2.22728044e-01 -1.13162172e+00
-4.49697763e-01 2.67680764e-01 -1.13149092e-01 -1.79006946e+00
-4.18826550e-01 -1.72681324e-02 -4.33895200e-01 4.96712923e-01
-1.42439497e+00 -8.24689507e-01 -5.74480832e-01 6.07201815e-01
2.91261077e-01 -1.00719385e-01 6.67445838e-01 6.41661525e-01
-5.08266211e-01 4.46307927e-01 2.91619807e-01 -3.85369331e-01
3.56661677e-01 -8.52171421e-01 2.77335644e-01 7.53988564e-01
-3.07780325e-01 -4.32356670e-02 4.89164233e-01 -8.50842237e-01
-2.28497124e+00 -8.88803244e-01 3.83525670e-01 -3.38998020e-01
2.07833961e-01 -2.80308157e-01 -1.96210220e-01 6.12771928e-01
-4.18710411e-01 3.79604071e-01 5.29104769e-01 -3.44317824e-01
5.69432855e-01 -3.92691463e-01 -1.25219905e+00 4.41980660e-01
1.00953984e+00 -3.98734570e-01 2.89514512e-01 1.24030478e-01
7.44630635e-01 -1.28082895e+00 -9.38955009e-01 4.17210788e-01
5.71970940e-01 -6.94779813e-01 9.65865791e-01 5.05003154e-01
-7.35151827e-01 -7.44051754e-01 -5.54438055e-01 -1.14713752e+00
1.14577964e-01 -1.03725457e+00 -7.19265878e-01 9.46270227e-01
1.67183146e-01 -2.26785719e-01 1.21347213e+00 6.86620891e-01
-1.43855676e-01 -5.04026890e-01 -1.50354660e+00 -6.34648323e-01
-1.09758162e+00 -7.09447742e-01 8.21518123e-01 3.81477207e-01
-1.24073945e-01 1.57098800e-01 -7.52915025e-01 1.01136923e+00
5.49029648e-01 -2.98092037e-01 1.26459944e+00 -7.39157021e-01
-3.89925629e-01 5.02856314e-01 -6.47482574e-01 -1.63890541e+00
-2.56850392e-01 -2.98478276e-01 1.85669616e-01 -1.34051621e+00
-1.92047447e-01 -8.23858380e-01 -2.19946150e-02 -2.05007300e-01
-7.15772156e-03 -4.38883118e-02 1.25887334e-01 2.83879817e-01
-1.13718081e+00 4.97905612e-01 9.20143723e-01 3.04101795e-01
-7.63574421e-01 6.90254152e-01 -1.79750428e-01 4.77365673e-01
6.01682723e-01 -2.61888862e-01 -7.30558574e-01 -7.19755828e-01
3.37389320e-01 8.46115589e-01 9.71536934e-02 -1.25261402e+00
5.13608098e-01 3.84686887e-02 3.55413944e-01 -1.16798162e+00
6.84186161e-01 -1.49464691e+00 4.89969343e-01 2.94136018e-01
4.15523797e-01 2.36317411e-01 1.63290963e-01 9.48749065e-01
2.85121351e-01 3.47945571e-01 2.94903159e-01 2.92548656e-01
-8.85576904e-01 6.83917522e-01 -6.09588504e-01 -4.33946788e-01
1.46575701e+00 -5.98473430e-01 8.40799883e-02 -1.07904625e+00
-6.66341484e-01 6.51168883e-01 5.37462950e-01 2.64719158e-01
7.49426901e-01 -1.58934140e+00 1.17233414e-02 1.81548223e-01
1.36665776e-01 2.34079391e-01 4.56331074e-01 8.94748449e-01
-5.70232213e-01 -8.88225855e-04 1.86136588e-01 -9.32969511e-01
-1.10185945e+00 3.09643716e-01 4.14019018e-01 -6.16594143e-02
-3.94261986e-01 3.66173595e-01 -4.27040845e-01 -2.66819000e-01
3.49170089e-01 7.77804255e-02 1.99430004e-01 -2.68709958e-01
5.12019336e-01 5.41225731e-01 1.16252318e-01 -6.76469862e-01
-3.50320458e-01 6.86185181e-01 1.65662631e-01 -2.61164606e-01
1.16953039e+00 -1.26702714e+00 5.25330245e-01 1.01812690e-01
8.25462818e-01 1.12466753e-01 -1.62925529e+00 -4.49981809e-01
-3.54484648e-01 -7.92716205e-01 5.84617555e-01 -3.66867274e-01
-1.04325843e+00 1.85669698e-02 9.38959360e-01 -1.17800623e-01
1.23892355e+00 -1.95583537e-01 1.00902259e+00 3.33424121e-01
1.29259956e+00 -1.38415527e+00 -3.95097360e-02 3.27440977e-01
2.01976612e-01 -1.13345122e+00 1.10382602e-01 -4.02356267e-01
-5.37981689e-01 7.56326318e-01 7.36214161e-01 3.04178357e-01
9.10224736e-01 5.86999476e-01 -5.10234497e-02 -1.00343227e-02
-5.09209037e-01 -5.45984097e-02 -4.24347401e-01 8.90901744e-01
-4.36569154e-01 1.46886706e-01 1.95288360e-01 2.78346211e-01
1.31947517e-01 1.55319288e-01 7.68321931e-01 1.32706594e+00
-5.59969127e-01 -1.01091695e+00 -7.31858134e-01 3.05217505e-01
-4.65849973e-03 4.62773114e-01 2.68332094e-01 4.90120143e-01
2.12635137e-02 7.86644101e-01 2.38910630e-01 -7.51882911e-01
2.23218039e-01 -4.04143929e-01 2.87328094e-01 -2.87591636e-01
2.32314348e-01 2.92679816e-01 1.24075204e-01 -7.35999763e-01
-2.73855746e-01 -5.79081118e-01 -1.36250722e+00 -5.83919883e-01
-4.83133942e-01 1.07858062e-01 1.07394493e+00 8.01573992e-01
1.02585816e+00 3.00497770e-01 1.41378784e+00 -1.29264116e+00
-1.80865943e-01 -2.86770731e-01 -6.79510832e-01 -4.70208287e-01
3.52739394e-01 -7.00037241e-01 -1.48462921e-01 -2.45369449e-01] | [7.201964855194092, -1.6411545276641846] |
b5770049-4f9f-4e9f-a70d-69c050097ab2 | multi-level-wavelet-cnn-for-image-restoration | 1805.07071 | null | http://arxiv.org/abs/1805.07071v2 | http://arxiv.org/pdf/1805.07071v2.pdf | Multi-level Wavelet-CNN for Image Restoration | The tradeoff between receptive field size and efficiency is a crucial issue
in low level vision. Plain convolutional networks (CNNs) generally enlarge the
receptive field at the expense of computational cost. Recently, dilated
filtering has been adopted to address this issue. But it suffers from gridding
effect, and the resulting receptive field is only a sparse sampling of input
image with checkerboard patterns. In this paper, we present a novel multi-level
wavelet CNN (MWCNN) model for better tradeoff between receptive field size and
computational efficiency. With the modified U-Net architecture, wavelet
transform is introduced to reduce the size of feature maps in the contracting
subnetwork. Furthermore, another convolutional layer is further used to
decrease the channels of feature maps. In the expanding subnetwork, inverse
wavelet transform is then deployed to reconstruct the high resolution feature
maps. Our MWCNN can also be explained as the generalization of dilated
filtering and subsampling, and can be applied to many image restoration tasks.
The experimental results clearly show the effectiveness of MWCNN for image
denoising, single image super-resolution, and JPEG image artifacts removal. | ['WangMeng Zuo', 'Pengju Liu', 'Liang Lin', 'Kai Zhang', 'Hongzhi Zhang'] | 2018-05-18 | null | null | null | null | ['jpeg-artifact-correction'] | ['computer-vision'] | [ 4.17535841e-01 -2.02358723e-01 2.40877017e-01 -1.29451752e-01
-6.20968230e-02 -1.12059796e-02 1.27661929e-01 -3.83398294e-01
-5.61469793e-01 5.23097992e-01 3.47866476e-01 1.24310814e-01
-5.27711697e-02 -1.15307128e+00 -6.04997098e-01 -1.00111938e+00
2.60053873e-01 -7.56878912e-01 6.57275558e-01 -3.96024168e-01
2.59694397e-01 4.25758958e-01 -1.42078793e+00 3.67387742e-01
8.72875392e-01 1.11369574e+00 7.52507567e-01 2.31089547e-01
-1.03118777e-01 8.23127270e-01 -3.55420172e-01 -9.04895440e-02
4.65915501e-01 -3.31013143e-01 -2.90020704e-01 1.14348762e-01
2.68781155e-01 -6.41712010e-01 -7.26797223e-01 1.52660263e+00
5.64320207e-01 2.10917860e-01 6.34974986e-02 -5.61208487e-01
-8.86335552e-01 4.56740767e-01 -7.05811024e-01 3.99405509e-01
-2.47444391e-01 -1.03018582e-01 4.32982683e-01 -9.38969195e-01
3.48392904e-01 1.41211581e+00 7.23170578e-01 4.23043370e-01
-1.15911376e+00 -7.94768870e-01 8.45678374e-02 2.46415198e-01
-1.34595501e+00 -3.24992806e-01 8.85145247e-01 6.07453585e-02
6.88055933e-01 1.88841477e-01 6.85573041e-01 6.91028178e-01
2.21987888e-01 3.08746338e-01 1.29761195e+00 -2.88889140e-01
1.65381283e-01 -2.00244203e-01 -2.27841482e-01 5.65048158e-01
2.82869965e-01 4.95998152e-02 -4.09372896e-01 2.78448164e-01
1.71895766e+00 4.90557820e-01 -7.78664768e-01 1.32773370e-01
-1.04211366e+00 6.47345364e-01 8.98006499e-01 4.96918201e-01
-4.38496530e-01 1.07788511e-01 3.13387364e-01 4.38567430e-01
3.48162413e-01 2.59698510e-01 -3.54974180e-01 3.14001590e-01
-7.81746864e-01 1.37655675e-01 2.09778532e-01 7.24931955e-01
7.73922563e-01 2.83718944e-01 -6.42264634e-02 1.07518160e+00
-1.48984725e-02 2.81362399e-03 7.46337354e-01 -1.28785729e+00
3.22892785e-01 5.81399202e-01 -7.78180808e-02 -1.11444795e+00
-2.19679743e-01 -6.33392036e-01 -1.71206343e+00 5.07350743e-01
9.83715504e-02 1.63082421e-01 -8.58921587e-01 1.40463495e+00
1.64801683e-02 2.41280735e-01 1.45135388e-01 1.07194889e+00
9.23133135e-01 8.31853092e-01 -2.15062842e-01 -3.45952660e-01
1.44783926e+00 -8.21053922e-01 -8.36432397e-01 -1.75511599e-01
1.09739248e-02 -7.97788858e-01 9.85670328e-01 3.52191538e-01
-1.19598162e+00 -1.12950420e+00 -1.12615252e+00 -4.43488151e-01
-8.83388985e-03 3.02175045e-01 4.69987780e-01 2.15041667e-01
-1.10781312e+00 8.81620288e-01 -8.08648705e-01 2.37210952e-02
6.38595462e-01 2.17086330e-01 -4.07842577e-01 -3.54573309e-01
-1.18537760e+00 6.47993505e-01 4.18023050e-01 4.83893186e-01
-4.13433999e-01 -5.48572779e-01 -6.85873032e-01 3.78457069e-01
6.84109330e-02 -3.80276084e-01 6.96633577e-01 -7.66763389e-01
-1.30549061e+00 2.05332965e-01 8.58467445e-02 -2.94887006e-01
2.85422236e-01 6.77692816e-02 -3.77135128e-01 1.97635606e-01
1.60878059e-04 6.18104458e-01 1.20926833e+00 -9.50338900e-01
-6.37764931e-01 -3.16105843e-01 1.67823373e-03 1.82235137e-01
-4.66389239e-01 -2.35655740e-01 -2.78390527e-01 -1.14693081e+00
6.66994452e-01 -2.20492512e-01 -4.15325016e-01 1.06532842e-01
1.59807920e-01 -5.22692315e-02 8.11771035e-01 -8.67806017e-01
1.24191844e+00 -2.50728345e+00 1.14474595e-01 -7.22244456e-02
3.86800021e-01 2.83092111e-01 -2.44506493e-01 1.25631422e-01
-1.54345617e-01 4.16160412e-02 -2.55914956e-01 1.47918940e-01
-6.73383594e-01 3.11885238e-01 -3.39251399e-01 4.01174009e-01
3.89165819e-01 6.06284976e-01 -6.17474616e-01 -3.11423779e-01
2.36418575e-01 7.88695455e-01 -6.95590734e-01 -8.67314413e-02
3.05578917e-01 5.36702991e-01 -4.67623889e-01 5.05546629e-01
1.20342076e+00 -5.25273904e-02 -1.46121100e-01 -7.16748059e-01
-5.40542781e-01 -2.10779071e-01 -1.28923810e+00 1.66741800e+00
-6.02127254e-01 5.20089149e-01 3.87696147e-01 -1.14106762e+00
1.03806329e+00 8.45154561e-03 1.76691800e-01 -9.21848357e-01
2.04412222e-01 1.90625608e-01 1.52757883e-01 -6.00665629e-01
2.14822203e-01 -1.86266795e-01 4.37574148e-01 -1.43653631e-01
1.03491463e-03 1.47210479e-01 5.41139999e-03 -3.24547738e-01
9.90228593e-01 8.08387175e-02 2.47008815e-01 -3.86896163e-01
6.72997355e-01 -3.17161649e-01 8.35747957e-01 5.61619818e-01
9.45835561e-02 7.62879789e-01 1.86621979e-01 -8.55248630e-01
-1.23403287e+00 -6.47469640e-01 -2.51102686e-01 7.38344073e-01
4.16861892e-01 -1.74792364e-01 -7.87000716e-01 -9.26440861e-03
-3.84484798e-01 7.06113055e-02 -4.16028529e-01 -2.22435758e-01
-9.39059377e-01 -6.67070329e-01 2.45668709e-01 6.00185692e-01
1.51159859e+00 -1.13643134e+00 -8.50901604e-01 3.63765448e-01
-3.44930738e-01 -9.85083401e-01 -4.10003096e-01 9.33949947e-02
-1.23770225e+00 -8.71670008e-01 -8.41116369e-01 -1.25803459e+00
9.33462441e-01 6.65587664e-01 5.08837581e-01 2.65087426e-01
-2.93594450e-01 -4.21928793e-01 -4.98332262e-01 3.64463851e-02
-5.30061759e-02 -2.55975038e-01 -1.03091642e-01 7.04298913e-02
6.65645907e-03 -9.03752208e-01 -1.07463145e+00 2.21200898e-01
-1.32231760e+00 2.05144092e-01 1.01656151e+00 1.26528323e+00
5.97093940e-01 8.02414060e-01 3.58636558e-01 -5.64139545e-01
6.44356608e-01 1.74849913e-01 -6.06267333e-01 -1.05432630e-01
-2.69219011e-01 -2.65280642e-02 1.04053688e+00 -6.45730138e-01
-1.20950961e+00 -8.52078944e-02 -4.45577711e-01 -4.80035901e-01
1.43294856e-02 3.69182557e-01 -1.56128064e-01 -3.76403302e-01
4.98993874e-01 5.82287669e-01 1.62898123e-01 -8.69237006e-01
-7.12697208e-03 5.09408712e-01 5.73818982e-01 -1.87492538e-02
7.14903831e-01 7.36051679e-01 1.16700731e-01 -8.82095158e-01
-5.93891740e-01 -8.13836977e-02 -3.87410343e-01 7.60978386e-02
8.03065777e-01 -1.13788331e+00 -7.71299541e-01 6.17887795e-01
-1.17027450e+00 -1.20415293e-01 -3.26646894e-01 5.54009438e-01
-1.68912426e-01 4.14132029e-01 -8.95630240e-01 -3.52647126e-01
-3.35047096e-01 -1.10221326e+00 6.83715522e-01 4.48503375e-01
4.61232513e-01 -4.60534215e-01 -6.40007615e-01 -1.53370261e-01
7.10000694e-01 -5.75259281e-03 1.01017773e+00 2.07566857e-01
-5.33641517e-01 -1.33579910e-01 -5.97180784e-01 8.20861816e-01
1.23160705e-01 -5.31943023e-01 -8.35018814e-01 -5.28499246e-01
6.38822615e-01 -9.84898061e-02 1.27451956e+00 7.15167940e-01
1.59842873e+00 -5.24207950e-01 5.04399277e-02 9.89556611e-01
1.79334652e+00 1.19177133e-01 1.01839495e+00 4.39852059e-01
4.97455776e-01 5.62387705e-01 3.88064355e-01 3.14723909e-01
-8.71613771e-02 2.23482981e-01 6.02411747e-01 -4.05044407e-01
-4.77730095e-01 -1.26275718e-01 1.23588927e-01 8.71161163e-01
-6.24788582e-01 2.76103437e-01 -2.90030569e-01 3.13663751e-01
-1.57839131e+00 -1.04309428e+00 7.73030892e-02 1.91679287e+00
6.46755874e-01 9.02483165e-02 -4.32052761e-01 3.21466357e-01
9.08120990e-01 2.58825868e-01 -4.95240480e-01 1.05918683e-01
-2.59855360e-01 3.92913669e-01 6.76388443e-01 3.79525095e-01
-9.06108320e-01 6.96220994e-01 5.48619556e+00 1.17922020e+00
-1.10448682e+00 1.53808042e-01 6.68335438e-01 1.67368546e-01
7.74059147e-02 -8.03681612e-02 -6.49876654e-01 4.68288809e-01
5.60235493e-02 2.83814758e-01 7.00218260e-01 6.46165550e-01
2.92700112e-01 1.18047083e-02 -4.71494198e-01 1.18633127e+00
-2.76175797e-01 -1.40656126e+00 1.52963713e-01 1.25678889e-02
6.59887552e-01 -2.26495445e-01 -3.26190926e-02 1.64190099e-01
-1.66288868e-01 -9.96522248e-01 4.25619334e-01 3.73739749e-01
1.04032648e+00 -8.53876472e-01 8.78207326e-01 4.21557337e-01
-1.35001969e+00 -5.76093674e-01 -1.19390237e+00 -2.74119258e-01
-1.69914097e-01 5.74667811e-01 4.82496880e-02 3.95012617e-01
1.20841241e+00 6.67430222e-01 -2.52790242e-01 8.01907897e-01
3.78169157e-02 2.36503124e-01 -1.21924356e-01 2.79906571e-01
1.87866181e-01 -4.79459316e-01 2.43772864e-01 8.12588215e-01
6.58892035e-01 4.84326839e-01 3.46662924e-02 7.50702024e-01
-2.45682970e-01 -1.06115438e-01 -4.29808944e-01 4.93822008e-01
4.16977763e-01 1.36182106e+00 -7.81761527e-01 -3.31172854e-01
-6.55782521e-01 1.08882129e+00 7.80637935e-02 5.74276805e-01
-4.36053425e-01 -6.58954322e-01 4.78549808e-01 3.39421928e-01
5.91786504e-01 -7.97936618e-02 -1.55542016e-01 -1.17314601e+00
1.22220136e-01 -7.53663480e-01 -6.82541803e-02 -6.70151591e-01
-9.38317478e-01 7.67518759e-01 -3.41779172e-01 -1.61885834e+00
5.67664623e-01 -5.47325611e-01 -5.95367134e-01 1.05277908e+00
-1.90029669e+00 -8.80422473e-01 -5.71894765e-01 6.92183733e-01
8.22961032e-01 -6.46301284e-02 5.42443633e-01 3.56712371e-01
-3.46119314e-01 1.10894546e-01 1.49194017e-01 2.63409019e-01
3.86139005e-01 -8.03048968e-01 2.05017194e-01 9.91118252e-01
-6.40480876e-01 7.61071324e-01 5.15623689e-01 -5.75184524e-01
-1.24271643e+00 -1.39658761e+00 4.53964531e-01 5.58276236e-01
2.84187883e-01 -1.78415149e-01 -1.11652100e+00 3.25603634e-01
1.93612248e-01 4.35676992e-01 1.77427575e-01 -4.90560263e-01
-2.34399334e-01 -5.85550427e-01 -1.21392846e+00 6.21503830e-01
1.01859272e+00 -1.54888615e-01 -3.52761388e-01 1.01242382e-02
7.81085908e-01 -3.39564204e-01 -1.00043666e+00 5.49127102e-01
5.13077140e-01 -1.22480965e+00 1.19042778e+00 1.73103791e-02
7.29761481e-01 -4.32098597e-01 -1.09183744e-01 -1.28528821e+00
-9.40034688e-01 -3.06152582e-01 2.69169569e-01 9.68028784e-01
-2.34203115e-01 -6.89590514e-01 5.20719528e-01 -1.20250054e-01
-1.74737900e-01 -7.70439863e-01 -1.01009095e+00 -5.20354629e-01
-1.21260509e-01 8.67473930e-02 4.58142936e-01 6.75130010e-01
-4.43032533e-01 2.15611085e-01 -4.13527101e-01 2.55813509e-01
6.95920289e-01 3.23109329e-02 1.86702639e-01 -1.18880284e+00
-1.47105530e-01 -4.58655596e-01 -3.27643096e-01 -1.35813677e+00
-3.46523464e-01 -5.57294548e-01 -9.88531299e-03 -1.48266351e+00
1.29548237e-02 -1.02130964e-01 -3.88044119e-01 3.32703114e-01
-7.57761821e-02 6.58721089e-01 9.53731611e-02 4.90636826e-01
1.01756342e-01 5.22667170e-01 1.85157931e+00 -2.71040224e-03
-2.09774479e-01 -5.02818413e-02 -6.65420592e-01 9.45839524e-01
8.59876037e-01 -9.06777680e-02 -3.71028364e-01 -6.65799558e-01
1.10839307e-01 1.68799490e-01 5.82010746e-01 -1.11093426e+00
3.80170256e-01 1.10033199e-01 9.47869420e-01 -4.61898178e-01
2.63452083e-01 -9.73054647e-01 2.34794407e-03 8.23456943e-01
-1.72711596e-01 5.35457060e-02 -3.04437671e-02 5.57802618e-01
-6.06803417e-01 -1.88257664e-01 1.14010715e+00 -5.83469987e-01
-7.92917907e-01 3.65707129e-01 -2.71871477e-01 -5.36866069e-01
6.66073024e-01 -4.77246732e-01 -2.61117786e-01 -4.68210177e-03
-5.13888180e-01 -8.68322477e-02 3.25049728e-01 1.00336529e-01
1.13190997e+00 -1.39241576e+00 -6.63512588e-01 6.73375487e-01
-3.76637667e-01 3.62373739e-01 6.55493736e-01 8.20847154e-01
-8.29149485e-01 -5.34909870e-03 -6.63443506e-01 -2.57054776e-01
-1.12241364e+00 5.02678514e-01 2.66559750e-01 -1.69314682e-01
-1.07412016e+00 8.45607162e-01 4.51448411e-01 7.61863217e-03
1.94325060e-01 -4.90664184e-01 -5.30485153e-01 -2.18793347e-01
8.69877458e-01 5.58674097e-01 -1.29436284e-01 -2.88414836e-01
-1.30737079e-02 8.50125670e-01 -1.21570624e-01 2.70047814e-01
1.57109046e+00 -4.07420695e-01 -6.81708276e-01 -2.19953656e-01
1.06320202e+00 -1.13808885e-01 -1.44958305e+00 -3.16257209e-01
-5.59956431e-01 -6.09304845e-01 4.56576049e-01 -2.38433510e-01
-1.35543704e+00 9.53105927e-01 7.30223477e-01 3.38645488e-01
1.77879751e+00 -4.40342128e-01 8.40672076e-01 1.79448366e-01
1.60591230e-01 -9.99730706e-01 1.10783018e-01 2.02990606e-01
1.07771087e+00 -1.08713031e+00 8.69329199e-02 -6.22381270e-01
-2.05194920e-01 1.49029768e+00 8.16907763e-01 -6.55268908e-01
6.23814881e-01 3.41677159e-01 -2.13081136e-01 4.33797128e-02
-3.02836448e-01 -1.02026604e-01 -2.88551241e-01 5.01142621e-01
3.03498656e-01 -2.78790355e-01 -5.13995409e-01 5.73576987e-01
-1.10306032e-02 7.10485056e-02 5.57038426e-01 6.61614180e-01
-8.32673907e-01 -7.55941033e-01 -4.28671092e-01 4.08932358e-01
-5.73222101e-01 -3.54791015e-01 4.70007986e-01 4.98527139e-01
5.87462008e-01 8.19331706e-01 2.30407059e-01 -3.66994232e-01
3.90184253e-01 -5.97213805e-01 5.16443908e-01 -3.48696709e-01
-3.07374269e-01 5.63180387e-01 -6.31727695e-01 -5.12682855e-01
-5.27342737e-01 -6.71659186e-02 -9.95063901e-01 -2.59033948e-01
-1.73326582e-01 -7.62932971e-02 3.57383490e-01 4.41120744e-01
1.70198187e-01 9.32959259e-01 6.89545393e-01 -8.44450235e-01
-6.22203410e-01 -1.25519502e+00 -7.53856480e-01 1.83254391e-01
5.68344951e-01 -2.61290967e-01 -2.57797092e-01 1.79193124e-01] | [11.232606887817383, -2.1629555225372314] |
eaafaa96-4d5a-475b-b6ae-b673bdd66665 | lorentz-equivariant-model-for-knowledge | 2302.04545 | null | https://arxiv.org/abs/2302.04545v2 | https://arxiv.org/pdf/2302.04545v2.pdf | Lorentz Equivariant Model for Knowledge-Enhanced Hyperbolic Collaborative Filtering | Introducing prior auxiliary information from the knowledge graph (KG) to assist the user-item graph can improve the comprehensive performance of the recommender system. Many recent studies show that the ensemble properties of hyperbolic spaces fit the scale-free and hierarchical characteristics exhibited in the above two types of graphs well. However, existing hyperbolic methods ignore the consideration of equivariance, thus they cannot generalize symmetric features under given transformations, which seriously limits the capability of the model. Moreover, they cannot balance preserving the heterogeneity and mining the high-order entity information to users across two graphs. To fill these gaps, we propose a rigorously Lorentz group equivariant knowledge-enhanced collaborative filtering model (LECF). Innovatively, we jointly update the attribute embeddings (containing the high-order entity signals from the KG) and hyperbolic embeddings (the distance between hyperbolic embeddings reveals the recommendation tendency) by the LECF layer with Lorentz Equivariant Transformation. Moreover, we propose Hyperbolic Sparse Attention Mechanism to sample the most informative neighbor nodes. Lorentz equivariance is strictly maintained throughout the entire model, and enforcing equivariance is proven necessary experimentally. Extensive experiments on three real-world benchmarks demonstrate that LECF remarkably outperforms state-of-the-art methods. | ['Jin Huang', 'Jing Xiao', 'Ruzhong Xie', 'Weihao Yu', 'Bosong Huang'] | 2023-02-09 | null | null | null | null | ['collaborative-filtering'] | ['miscellaneous'] | [-5.70929706e-01 6.77820593e-02 -4.82707731e-02 -4.14127797e-01
-1.02680244e-01 -7.02078938e-01 3.99670601e-01 -1.05163135e-01
-4.58696000e-02 2.26623744e-01 5.71366668e-01 1.43620549e-02
-9.26000655e-01 -9.06631708e-01 -5.50740957e-01 -1.04739368e+00
-7.87357092e-02 2.80034751e-01 1.60269737e-01 -5.28335392e-01
1.05671704e-01 2.23040402e-01 -1.13891828e+00 -7.82833472e-02
1.15351391e+00 8.66553903e-01 -7.58741722e-02 7.14950264e-02
1.66832879e-01 2.06024319e-01 1.66094467e-01 -7.26078749e-01
3.34621847e-01 -2.38939658e-01 -5.84228754e-01 -2.87341684e-01
4.07735020e-01 -6.56311065e-02 -8.35779309e-01 1.35127378e+00
5.01360953e-01 6.02479398e-01 6.79023087e-01 -1.21402442e+00
-1.63751459e+00 1.01638734e+00 -2.39140138e-01 8.90024602e-02
9.86953601e-02 -3.31456572e-01 1.46155763e+00 -1.27985322e+00
5.63039839e-01 1.06521845e+00 9.78156567e-01 -4.10807841e-02
-9.65796173e-01 -6.09184563e-01 5.11751115e-01 4.28209901e-01
-1.79060268e+00 7.92731643e-02 1.09164000e+00 -1.72543928e-01
2.36294612e-01 4.48749840e-01 8.36356640e-01 1.08563495e+00
-1.84249878e-02 4.68358636e-01 8.50734651e-01 2.40461618e-01
1.41852619e-02 2.24967912e-01 4.39102411e-01 9.85511005e-01
6.01495385e-01 -1.93095431e-01 -5.93855858e-01 -2.72050768e-01
7.55205691e-01 3.28898758e-01 -5.11186957e-01 -6.79736853e-01
-1.31771243e+00 9.08110678e-01 8.27370286e-01 2.39239588e-01
-2.88538218e-01 -3.97734016e-01 7.11098984e-02 2.61513293e-01
2.13790670e-01 6.27435386e-01 -2.94918269e-01 5.76407909e-01
-3.47074866e-01 -3.27736586e-02 7.02580929e-01 1.45280790e+00
8.23416471e-01 -7.07518905e-02 -2.09361017e-01 5.48947334e-01
6.82094157e-01 5.28270483e-01 2.11363912e-01 -6.27212882e-01
9.48794708e-02 6.52824283e-01 -1.85507402e-01 -1.63846135e+00
-5.80216467e-01 -1.24820876e+00 -1.13569427e+00 -7.55599976e-01
3.60004485e-01 4.34168540e-02 -9.30554122e-02 1.65308738e+00
5.99985421e-01 4.70148027e-01 -1.03042975e-01 1.11243379e+00
1.05403459e+00 6.23976588e-01 -2.11404130e-01 -2.38015920e-01
1.50493038e+00 -8.70725393e-01 -7.67277658e-01 5.38512707e-01
5.53137124e-01 -6.43747687e-01 1.24761403e+00 2.82419205e-01
-8.23325396e-01 -5.08943439e-01 -1.11088765e+00 -3.11642021e-01
-5.29680550e-01 6.56836033e-02 9.05335784e-01 6.00095868e-01
-6.95679247e-01 5.86903751e-01 -3.37754071e-01 -1.72037929e-01
9.26371813e-02 3.47453594e-01 -3.25893879e-01 -1.89556420e-01
-1.60733199e+00 1.02368720e-01 3.73220854e-02 3.81873637e-01
-1.77027643e-01 -1.17800689e+00 -6.45706475e-01 1.67968258e-01
3.33875418e-01 -8.32762003e-01 5.36611617e-01 -2.89930761e-01
-1.39862299e+00 1.42425105e-01 3.70422125e-01 1.09542802e-01
3.46111685e-01 -1.21205509e-01 -7.91391790e-01 -1.44003227e-01
-8.61884728e-02 -6.13862984e-02 6.72298074e-01 -1.17513323e+00
-4.81875896e-01 -6.12619698e-01 3.51004273e-01 3.71787131e-01
-8.45804572e-01 -5.47582507e-01 -6.80534482e-01 -8.47472906e-01
5.09123385e-01 -1.07339513e+00 1.20955192e-01 -1.66085690e-01
-2.83998340e-01 -4.30753559e-01 7.49341071e-01 -5.53171933e-01
1.51387823e+00 -2.30606937e+00 3.02581847e-01 6.20185614e-01
3.75972688e-01 -1.95441946e-01 -9.26257372e-02 4.54972982e-01
2.23589405e-01 -1.05503894e-01 2.45520845e-01 -8.05604905e-02
3.83571506e-01 2.74514049e-01 -4.33831841e-01 8.34057868e-01
-3.18181366e-01 8.88731837e-01 -1.07730579e+00 -3.07627410e-01
-1.28455922e-01 7.57989228e-01 -9.77500558e-01 -9.14588124e-02
2.25385115e-01 3.54788512e-01 -6.80183291e-01 2.93699473e-01
9.83549535e-01 -5.93652487e-01 3.11383843e-01 -1.06178606e+00
-7.14657158e-02 1.71013936e-01 -1.63885069e+00 1.84496570e+00
-5.78063019e-02 -1.68314800e-01 -1.32692968e-02 -9.03858721e-01
9.06745493e-01 -1.18425742e-01 5.29351711e-01 -6.30131125e-01
-1.94687899e-02 5.91114676e-03 -3.79545502e-02 -3.53833765e-01
7.43738472e-01 2.28436619e-01 -1.12954803e-01 1.31711692e-01
2.73452729e-01 5.95145941e-01 -2.00396448e-01 5.56055069e-01
6.54751778e-01 -1.08396180e-01 -1.46680400e-01 -8.53154421e-01
6.36754274e-01 -5.14612436e-01 7.73162007e-01 6.81083083e-01
1.58052161e-01 4.51223999e-01 2.20661938e-01 -2.96708792e-01
-8.96036267e-01 -1.25537980e+00 -4.30790424e-01 1.35384822e+00
7.45348394e-01 -8.28724504e-01 -4.67977941e-01 -8.66788149e-01
1.28596485e-01 5.91479838e-01 -9.16988075e-01 -5.68573475e-01
-4.62917268e-01 -1.01185095e+00 2.45727569e-01 4.75915790e-01
5.38547218e-01 -2.63799310e-01 6.64773107e-01 3.64925079e-02
-4.53905612e-02 -9.96329784e-01 -1.12127638e+00 -1.69305727e-01
-4.14688885e-01 -1.07647753e+00 -4.20489132e-01 -9.14744198e-01
7.24461198e-01 6.37543619e-01 6.46398783e-01 -2.47137696e-02
1.85548291e-01 5.79648435e-01 -6.49122655e-01 1.02371104e-01
3.86994183e-01 3.33098620e-01 4.09509361e-01 5.15386641e-01
2.15632007e-01 -7.52993107e-01 -9.55465496e-01 6.99793994e-01
-5.86978316e-01 -1.16190039e-01 4.85488832e-01 7.14390934e-01
7.45764136e-01 3.79622847e-01 5.95736086e-01 -9.96095598e-01
5.29777050e-01 -7.35075474e-01 -2.92773277e-01 2.66442627e-01
-9.00673389e-01 8.51310045e-02 1.07930386e+00 -5.97661257e-01
-8.80185962e-01 -4.09916759e-01 8.65275785e-02 -2.55896896e-01
3.65978569e-01 5.76961100e-01 -4.85382706e-01 -2.84749627e-01
2.61108905e-01 2.31280997e-01 -3.54346544e-01 -7.28124976e-01
7.16218710e-01 3.59059662e-01 5.57147264e-01 -5.76899409e-01
1.04810369e+00 6.24858856e-01 1.81431502e-01 -7.59031475e-01
-1.14250278e+00 -4.82364029e-01 -5.99883556e-01 4.66612279e-02
6.67125344e-01 -8.60824883e-01 -9.74969685e-01 -5.77668585e-02
-6.12387896e-01 3.36915433e-01 -3.41269553e-01 8.33931804e-01
-1.46434098e-01 5.76244652e-01 -5.15722752e-01 -5.59400141e-01
-2.53122330e-01 -7.58563340e-01 9.01273489e-01 2.16904566e-01
2.03241840e-01 -1.18081725e+00 1.40705351e-02 2.95250267e-01
3.59755039e-01 -2.73036659e-01 1.09609258e+00 -8.17964196e-01
-6.12886250e-01 1.80061925e-02 -2.26510465e-01 -1.28832594e-01
-7.67773241e-02 -2.03855649e-01 -6.53111577e-01 -4.14724469e-01
-1.33040354e-01 3.16920429e-01 7.99679577e-01 1.19950613e-02
1.25942111e+00 -6.02381408e-01 -1.58728719e-01 9.99942303e-01
1.12762785e+00 -3.48179281e-01 3.66097957e-01 -4.40984853e-02
1.25275934e+00 3.99021238e-01 3.35354596e-01 4.48298365e-01
9.11409974e-01 6.95427954e-01 2.26854309e-01 1.48585379e-01
-7.09081218e-02 -7.04277992e-01 3.57902616e-01 1.78151298e+00
-3.79478425e-01 -3.71598219e-03 -1.93522453e-01 1.85049385e-01
-1.98880315e+00 -1.01205826e+00 -5.63306451e-01 2.16158342e+00
4.69987661e-01 -2.99727142e-01 1.29244521e-01 -1.73649594e-01
6.64948523e-01 7.03275427e-02 -4.59521532e-01 3.38667572e-01
-4.02217209e-01 -2.02274024e-01 5.07397532e-01 4.80929017e-01
-1.10331976e+00 6.54876113e-01 4.92767096e+00 1.03475344e+00
-6.19002938e-01 2.02724472e-01 -2.66242027e-01 1.26875460e-01
-8.80767226e-01 -2.88689439e-03 -9.77746069e-01 4.59773421e-01
4.37248826e-01 -2.99104273e-01 6.18210077e-01 7.13784397e-01
-8.76130089e-02 8.43797445e-01 -8.10252845e-01 9.16199982e-01
2.62383819e-01 -1.24520135e+00 3.12594473e-01 1.82099983e-01
8.85945320e-01 -7.68781751e-02 4.52003092e-01 6.12885654e-01
2.74846166e-01 -5.32930970e-01 4.90790516e-01 1.02677250e+00
4.10778821e-01 -8.54745984e-01 6.86615229e-01 7.70467967e-02
-1.53721952e+00 -6.78102076e-02 -7.37094939e-01 4.46209043e-01
-3.30855623e-02 6.34318352e-01 -2.94711858e-01 1.23699284e+00
8.18486035e-01 1.13869917e+00 -6.90914333e-01 8.94377708e-01
-1.24300513e-02 6.46159172e-01 -3.11869234e-01 -7.36170858e-02
2.40577787e-01 -9.72388029e-01 6.66327357e-01 9.47137058e-01
5.87200403e-01 4.04189259e-01 2.15090945e-01 7.87060559e-01
-2.36472636e-01 6.48375690e-01 -3.77242237e-01 7.27156922e-02
5.06657600e-01 1.54119575e+00 -5.15939415e-01 -2.42899824e-02
-6.87036335e-01 9.21122253e-01 2.61527002e-01 6.43902302e-01
-8.91501129e-01 -3.29364538e-01 5.92192411e-01 2.63069749e-01
4.14332122e-01 -3.73103946e-01 1.08121507e-01 -1.41165948e+00
6.53211325e-02 -5.80520689e-01 7.22468138e-01 -4.74214822e-01
-1.86420202e+00 2.17902139e-01 -3.19582999e-01 -1.10920763e+00
7.06450641e-01 -5.75097561e-01 -5.17984927e-01 5.02772391e-01
-1.29013395e+00 -1.37619627e+00 -5.37237167e-01 9.43270445e-01
-3.94573584e-02 -2.23911136e-01 8.01762164e-01 7.21918166e-01
-6.94821000e-01 1.00882781e+00 5.88035345e-01 8.94095078e-02
8.85305464e-01 -1.37581384e+00 -6.88230246e-02 4.32231963e-01
3.75190645e-01 1.30676341e+00 4.33399796e-01 -6.12324715e-01
-2.23695207e+00 -1.34484315e+00 3.69141787e-01 -4.98925477e-01
9.90837991e-01 -6.44476175e-01 -1.13756025e+00 6.05581880e-01
-5.55053242e-02 2.70817310e-01 9.90792096e-01 6.49519324e-01
-7.37074733e-01 -3.47366095e-01 -7.17313409e-01 6.70315862e-01
1.73839736e+00 -6.40331268e-01 -4.20184255e-01 5.08797765e-01
9.06340420e-01 5.26176281e-02 -1.44365668e+00 5.42238474e-01
6.13547087e-01 -6.78683460e-01 1.10767221e+00 -9.68866825e-01
-2.86268085e-01 -6.43185616e-01 -4.53208774e-01 -1.26618946e+00
-1.17062902e+00 -8.04152787e-01 -4.29326832e-01 1.28445327e+00
2.63084501e-01 -7.12065279e-01 5.30064046e-01 1.71405181e-01
-3.79311025e-01 -6.47117376e-01 -5.94077766e-01 -8.43384266e-01
1.26434267e-01 -6.46100342e-02 8.51880252e-01 1.57527959e+00
6.85074180e-02 6.65268123e-01 -4.63949293e-01 7.24190891e-01
7.62903214e-01 3.65624487e-01 5.59732318e-01 -1.63499510e+00
-3.19677413e-01 -5.14228702e-01 -4.13009882e-01 -1.16848588e+00
8.82734656e-02 -1.53596187e+00 -6.53629959e-01 -1.31326282e+00
3.05140942e-01 -6.48480237e-01 -5.57421923e-01 -4.67877695e-03
-1.45152047e-01 1.44817814e-01 -1.70356035e-01 2.91771412e-01
-1.02650785e+00 1.01969409e+00 1.43347836e+00 7.26724565e-02
1.68215968e-02 -1.61291257e-01 -1.25942862e+00 6.68986559e-01
3.29917938e-01 -1.82081535e-01 -7.08437026e-01 -2.29479894e-01
8.00863743e-01 -6.18063390e-01 2.79478610e-01 -5.87558985e-01
6.65636420e-01 4.32680286e-02 4.20143932e-01 -3.21648061e-01
1.15821309e-01 -9.46230650e-01 2.66674399e-01 -1.58354864e-01
-2.68476844e-01 -1.59350056e-02 -4.73857135e-01 1.12773061e+00
2.58013487e-01 8.17412287e-02 5.19983232e-01 4.05585289e-01
-4.23322976e-01 9.58676696e-01 3.23341459e-01 1.75965443e-01
7.01163530e-01 2.71880120e-01 -4.45066154e-01 -1.70869991e-01
-8.07783484e-01 2.44583562e-01 3.48847419e-01 5.24319291e-01
2.54752666e-01 -1.90733945e+00 -6.31344318e-01 4.80163187e-01
3.42902482e-01 -2.03449205e-01 6.75768495e-01 1.28135586e+00
2.05987558e-01 9.35992822e-02 2.18280464e-01 -2.75481880e-01
-7.26981401e-01 9.10111308e-01 1.65066212e-01 3.62377763e-02
-1.00687468e+00 5.45936882e-01 4.18147296e-01 -6.88107431e-01
6.63055703e-02 -4.85061258e-02 -3.77213210e-01 2.53327668e-01
4.21187729e-01 5.70209742e-01 -2.56048120e-03 -7.06719697e-01
-2.74593174e-01 8.08195770e-01 -2.81384140e-01 2.98849374e-01
1.30619299e+00 -3.45394403e-01 -7.98558518e-02 2.88905680e-01
1.33785415e+00 7.02418566e-01 -8.75762582e-01 -5.67157865e-01
-4.50001031e-01 -3.31609368e-01 1.73281774e-01 -2.00981632e-01
-1.07952881e+00 5.05196631e-01 2.27652505e-01 5.67887545e-01
7.08289921e-01 -5.52146882e-02 6.16678119e-01 6.56075478e-01
3.50208700e-01 -1.16470945e+00 -8.13804343e-02 5.17531812e-01
8.46016467e-01 -6.97692990e-01 8.55863467e-02 -8.18886518e-01
-5.58963239e-01 8.49900603e-01 4.72525656e-01 -4.06352818e-01
1.28639412e+00 -3.10452402e-01 -4.96673822e-01 -3.95200849e-01
-4.05042768e-01 -4.40355465e-02 8.15810740e-01 4.51086253e-01
2.95051664e-01 2.28508376e-02 -4.16268736e-01 1.28587973e+00
-5.04482925e-01 -7.25893378e-01 1.65875882e-01 1.55793622e-01
-2.47162357e-01 -6.83279812e-01 9.32559222e-02 2.88859427e-01
-8.67054313e-02 -1.29689619e-01 -3.12781632e-01 8.32272291e-01
1.18326083e-01 6.77421272e-01 -1.54745579e-01 -6.63855076e-01
5.35468400e-01 -1.15954138e-01 3.62709582e-01 -1.45886138e-01
-3.29610080e-01 2.78320372e-01 -1.58403143e-01 -4.13911849e-01
-1.21316776e-01 -6.77132130e-01 -1.18180835e+00 -5.90961277e-01
-4.50126976e-01 6.84823394e-01 2.58764565e-01 6.60752296e-01
7.31803417e-01 5.03952265e-01 8.49047303e-01 -2.34033480e-01
-8.41623724e-01 -7.73097336e-01 -1.22934890e+00 6.48310721e-01
-6.83122501e-02 -9.72095549e-01 -5.24270594e-01 -3.23876202e-01] | [10.239176750183105, 5.655519962310791] |
508da9a9-e9c9-4f4b-a871-ff2c5f5cfe78 | named-entities-troubling-your-neural-methods | 1804.09540 | null | https://arxiv.org/abs/1804.09540v2 | https://arxiv.org/pdf/1804.09540v2.pdf | NE-Table: A Neural key-value table for Named Entities | Many Natural Language Processing (NLP) tasks depend on using Named Entities (NEs) that are contained in texts and in external knowledge sources. While this is easy for humans, the present neural methods that rely on learned word embeddings may not perform well for these NLP tasks, especially in the presence of Out-Of-Vocabulary (OOV) or rare NEs. In this paper, we propose a solution for this problem, and present empirical evaluations on: a) a structured Question-Answering task, b) three related Goal-Oriented dialog tasks, and c) a Reading-Comprehension task, which show that the proposed method can be effective in dealing with both in-vocabulary and OOV NEs. We create extended versions of dialog bAbI tasks 1,2 and 4 and OOV versions of the CBT test set available at - https://github.com/IBM/ne-table-datasets. | ['Lazaros Polymenakos', 'Satinder Singh', 'Xiaoxiao Guo', 'Mo Yu', 'Jatin Ganhotra', 'Janarthanan Rajendran'] | 2018-04-22 | ne-table-a-neural-key-value-table-for-named | https://aclanthology.org/R19-1114 | https://aclanthology.org/R19-1114.pdf | ranlp-2019-9 | ['goal-oriented-dialog'] | ['natural-language-processing'] | [-2.60225922e-01 2.42748469e-01 7.20996633e-02 -4.58743185e-01
-5.22174895e-01 -6.90110147e-01 7.67657101e-01 6.69473648e-01
-9.81352687e-01 9.88208294e-01 5.13840199e-01 -4.34444994e-01
-6.29695430e-02 -7.55143583e-01 -2.55818188e-01 -1.24131411e-01
9.85489339e-02 9.58593071e-01 5.45373023e-01 -7.15899885e-01
2.45164782e-01 1.26783699e-01 -1.10611355e+00 1.08507320e-01
9.90382195e-01 6.91537321e-01 2.61836708e-01 6.33333147e-01
-7.51362801e-01 9.01763260e-01 -6.51733577e-01 -6.43244863e-01
-7.34482286e-03 -1.87163323e-01 -1.19399428e+00 -2.94014722e-01
3.23335618e-01 -3.18750560e-01 -3.31310719e-01 7.54562557e-01
5.15318811e-01 6.55459464e-01 7.93151855e-01 -1.15099573e+00
-9.14284170e-01 6.40698195e-01 8.52526538e-03 3.54002953e-01
5.57852030e-01 -1.25970840e-02 1.32321310e+00 -1.01123691e+00
7.59508550e-01 1.38236070e+00 4.44908381e-01 9.10665035e-01
-1.06615162e+00 -3.05499345e-01 -2.79708430e-02 2.61375189e-01
-9.08279002e-01 -4.61511761e-01 5.76397955e-01 -2.77780950e-01
1.30821347e+00 1.91271901e-02 3.29074003e-02 1.47237897e+00
2.86855437e-02 8.23346853e-01 9.69545424e-01 -5.38075268e-01
3.64060670e-01 4.86930758e-01 1.06812406e+00 3.76633853e-01
1.07103914e-01 -6.95631281e-02 -6.81235790e-01 -2.27500901e-01
3.70895356e-01 -5.13818383e-01 -5.84126234e-01 -3.23011428e-01
-1.18305957e+00 1.32319212e+00 2.41797194e-01 7.23293841e-01
-4.25080031e-01 -2.35176295e-01 4.85538274e-01 5.00508964e-01
4.20653880e-01 8.24498832e-01 -8.54765356e-01 -2.82762378e-01
-4.82102185e-01 2.71237671e-01 1.52035093e+00 9.58324075e-01
6.06497109e-01 -1.23584397e-01 -3.86809409e-01 1.13963234e+00
2.67014027e-01 1.88580200e-01 9.06590700e-01 -6.95036590e-01
8.01743805e-01 6.38954997e-01 4.25327212e-01 -8.30479860e-01
-6.68467224e-01 2.44897991e-01 -5.54019392e-01 -2.08067372e-01
9.00769651e-01 -2.99544722e-01 -6.02809191e-01 1.93019319e+00
4.60865647e-01 -3.35413158e-01 3.77551168e-01 7.06269324e-01
1.42244709e+00 7.95627773e-01 1.91096485e-01 -8.31173584e-02
1.64632905e+00 -1.05070186e+00 -1.31052959e+00 -4.32133496e-01
5.73285639e-01 -4.36230034e-01 1.21411860e+00 2.79846732e-02
-8.48556519e-01 -4.97716606e-01 -7.72694767e-01 -4.65924412e-01
-9.03454840e-01 -1.93242654e-02 2.54989773e-01 6.29600585e-01
-9.84276891e-01 2.47569174e-01 -4.18606371e-01 -7.12903321e-01
1.59859478e-01 1.52185649e-01 -4.91360933e-01 -5.76751046e-02
-1.64662445e+00 1.35526741e+00 6.30878925e-01 3.56771052e-02
-7.54414141e-01 -5.53892314e-01 -1.05084932e+00 2.78764188e-01
7.05150962e-01 -3.64817172e-01 1.38983846e+00 -6.39898956e-01
-1.39627194e+00 8.60323727e-01 -2.16817871e-01 -4.82963800e-01
2.75889307e-01 -5.16804576e-01 -3.11583787e-01 -4.74375021e-03
-8.03762898e-02 6.88284457e-01 5.06079912e-01 -1.05910349e+00
-2.40717590e-01 -4.70027298e-01 3.88894111e-01 2.40106076e-01
-4.11258578e-01 7.01936558e-02 7.49939978e-02 -4.10707712e-01
-3.53287250e-01 -6.74352646e-01 -7.78130591e-02 -1.07733428e-01
-4.98581827e-01 -8.49254251e-01 4.61664885e-01 -9.13770139e-01
1.02646291e+00 -1.88631165e+00 9.90293175e-02 -2.10707188e-01
2.58430451e-01 5.64538419e-01 -2.87613511e-01 8.59817147e-01
1.22606605e-01 2.08036602e-01 -1.20122746e-01 -2.00045019e-01
2.37649277e-01 2.65662104e-01 -2.70023763e-01 -2.77471896e-02
2.27835014e-01 9.92677987e-01 -8.08384120e-01 -3.49891156e-01
1.45094737e-01 5.14116026e-02 -2.84638673e-01 5.64028978e-01
-5.14727294e-01 3.38573247e-01 -4.44929481e-01 2.56702542e-01
3.06071401e-01 -1.33585975e-01 1.05070911e-01 8.85649547e-02
-8.63075815e-03 6.70958757e-01 -1.09651434e+00 1.62956882e+00
-5.94622135e-01 7.75037348e-01 8.43949392e-02 -8.81700397e-01
9.73572016e-01 5.69541872e-01 -1.55386448e-01 -5.62880814e-01
4.07215983e-01 -5.10021746e-02 1.93246901e-01 -6.81514025e-01
6.89651132e-01 -1.75522774e-01 4.48593125e-03 4.12726253e-01
6.97785676e-01 -8.83040056e-02 4.68933195e-01 3.39513808e-01
1.19086909e+00 -2.99938887e-01 7.40501821e-01 -4.75193560e-01
6.97108269e-01 7.07501099e-02 3.25123668e-01 8.46423924e-01
-6.99872434e-01 3.56454819e-01 7.85986304e-01 -2.32101813e-01
-8.70689154e-01 -1.00323737e+00 -2.21851021e-01 1.37889242e+00
8.30378570e-03 -4.92247045e-01 -6.98944151e-01 -7.93225169e-01
-7.03219026e-02 1.38269866e+00 -5.94124973e-01 1.94161892e-01
-3.81873012e-01 -3.16490203e-01 5.04722655e-01 5.34942687e-01
5.98781049e-01 -1.52371418e+00 -4.85049248e-01 4.41314131e-01
-2.69130260e-01 -1.38436759e+00 -3.31016570e-01 2.55436361e-01
-6.47412539e-01 -1.10420763e+00 -6.14781260e-01 -8.85060608e-01
7.31360586e-03 -5.92432283e-02 1.40308046e+00 -3.78054380e-02
7.43771419e-02 6.68267310e-01 -7.02862918e-01 -5.12615561e-01
-5.25101185e-01 2.09840566e-01 1.10505344e-02 -2.67417371e-01
8.92517269e-01 -2.67608017e-01 -6.00079708e-02 3.54582578e-01
-9.04810071e-01 -2.02695742e-01 4.22503315e-02 1.10380602e+00
-8.26214999e-02 -3.34643185e-01 9.67040002e-01 -9.74259913e-01
1.16374362e+00 -6.67190194e-01 -5.10654807e-01 6.23458326e-01
-1.10472433e-01 1.08113050e-01 6.37753308e-01 -4.79075521e-01
-1.36917245e+00 -2.47760728e-01 -4.56019938e-01 -2.78931968e-02
-6.13222420e-01 5.20200193e-01 -3.25955063e-01 4.75611210e-01
8.98482621e-01 3.63687240e-02 -1.25918865e-01 -5.35574615e-01
6.29802048e-01 8.57229114e-01 5.20548522e-02 -5.27226210e-01
5.96992850e-01 -6.09559864e-02 -6.68246329e-01 -1.33293390e+00
-9.00319099e-01 -6.70172572e-01 -6.44852936e-01 7.01850355e-02
1.28075528e+00 -7.05332339e-01 -4.65712190e-01 2.01954380e-01
-1.49199557e+00 -4.63661104e-01 -2.57890791e-01 4.10767287e-01
-3.53085756e-01 3.77146989e-01 -6.71158016e-01 -7.96230793e-01
-3.39914739e-01 -1.00158060e+00 6.71366155e-01 3.79931957e-01
-5.59579194e-01 -1.19708002e+00 3.19868654e-01 6.16426826e-01
6.59512937e-01 -3.48813057e-01 1.16333926e+00 -1.78597581e+00
-1.28472447e-01 -2.23526340e-02 -1.40526354e-01 5.05648315e-01
2.83471611e-03 -4.89671856e-01 -1.13609886e+00 2.23675910e-02
2.12290511e-01 -8.52867544e-01 6.79339230e-01 2.70530079e-02
4.69715536e-01 -2.14976400e-01 -3.80056128e-02 -2.36848086e-01
1.04101133e+00 2.07187116e-01 4.36114401e-01 1.40902668e-01
3.45893830e-01 1.10377991e+00 5.11655271e-01 2.98321873e-01
5.37346303e-01 6.37822807e-01 1.64727956e-01 4.37059551e-01
3.31692323e-02 -1.42207503e-01 2.38875076e-01 9.50150192e-01
2.78791189e-01 -7.44943619e-01 -1.30216026e+00 8.43412817e-01
-1.77160776e+00 -6.71004593e-01 -2.62750447e-01 1.72614348e+00
1.07693172e+00 -1.76454291e-01 -1.22927144e-01 -2.79550463e-01
6.32791758e-01 3.09700668e-01 -3.85537952e-01 -5.10984123e-01
-8.69525895e-02 4.63961601e-01 -6.43200008e-03 5.62417030e-01
-1.14996481e+00 1.28848433e+00 5.44980764e+00 7.06798613e-01
-4.48721290e-01 5.61904252e-01 2.36200675e-01 5.10883868e-01
-2.23949596e-01 -1.37075603e-01 -9.32618022e-01 2.21061647e-01
1.23151410e+00 -5.80366589e-02 1.68630153e-01 8.15916002e-01
-1.72851846e-01 -1.87162548e-01 -1.25248456e+00 7.00299799e-01
2.12525025e-01 -1.00477469e+00 2.89657302e-02 -3.37819576e-01
2.95475185e-01 7.33842254e-02 -5.21535039e-01 8.44570696e-01
3.90302956e-01 -1.00232005e+00 1.32876873e-01 1.47846818e-01
2.80164838e-01 -2.32412159e-01 1.07841325e+00 6.24231637e-01
-7.29186535e-01 -3.19608301e-02 -6.38836324e-01 1.64320111e-01
3.24720681e-01 3.10340315e-01 -7.40299761e-01 4.23217326e-01
6.68570876e-01 1.67729631e-01 -5.99651873e-01 7.15250313e-01
-6.68354630e-01 7.45957553e-01 -2.75203377e-01 -6.28816605e-01
3.14678162e-01 -1.00398630e-01 5.91195703e-01 1.06956172e+00
-1.43866435e-01 2.50227720e-01 -5.37232235e-02 9.73991573e-01
-2.86838233e-01 5.09005129e-01 -7.97304928e-01 -2.96196878e-01
4.72629458e-01 1.14027584e+00 -5.17358005e-01 -2.94520497e-01
-7.10191309e-01 7.96555698e-01 7.25206494e-01 4.47718889e-01
-5.94961703e-01 -4.62110430e-01 4.67557400e-01 -1.70809045e-01
2.92347610e-01 -4.13070023e-01 8.13844427e-02 -1.46166694e+00
1.24821358e-03 -1.05214858e+00 4.04132724e-01 -7.24358976e-01
-1.70727634e+00 7.70262539e-01 5.85848689e-02 -5.92311084e-01
-3.56323421e-01 -9.65102911e-01 -6.08014584e-01 7.49811649e-01
-1.45007074e+00 -8.34396720e-01 -2.95170367e-01 6.23312652e-01
9.55542445e-01 -3.47894937e-01 1.11155045e+00 1.18642487e-01
-4.22340631e-01 3.89211267e-01 1.00877129e-01 4.12744671e-01
9.47047889e-01 -1.40394104e+00 1.53078526e-01 3.26673836e-01
4.42400873e-01 6.82176054e-01 6.69661999e-01 -4.66022372e-01
-1.13710427e+00 -6.65225029e-01 1.40934968e+00 -7.52221942e-01
9.83972967e-01 -6.93476498e-01 -1.20810235e+00 9.02214706e-01
8.50211918e-01 -4.06868488e-01 9.20100212e-01 3.97814274e-01
-3.69639724e-01 4.14814860e-01 -1.22679198e+00 6.21748388e-01
8.67374539e-01 -3.86519551e-01 -1.60024536e+00 6.48541868e-01
8.29659104e-01 -3.26625973e-01 -7.91939199e-01 2.55220607e-02
8.11376870e-02 -6.86652720e-01 8.50422502e-01 -1.08346641e+00
3.80823761e-01 1.72111824e-01 -3.44876617e-01 -1.56301713e+00
9.77675617e-02 -2.85505652e-01 7.19440728e-02 1.52526069e+00
6.73073471e-01 -7.63914049e-01 3.07465225e-01 8.40808511e-01
6.92918971e-02 -4.64901924e-01 -1.33783233e+00 -7.81852901e-01
5.33360779e-01 -3.44346106e-01 2.10811704e-01 1.11739993e+00
1.23736128e-01 8.92168105e-01 -1.26654848e-01 -6.29201010e-02
1.35552287e-01 -2.05341294e-01 6.69175029e-01 -1.28473604e+00
-3.42739969e-02 -1.79302260e-01 -1.70110613e-01 -1.05870223e+00
6.40102923e-01 -7.15374947e-01 1.28650412e-01 -1.54046929e+00
-9.90147740e-02 -1.23517357e-01 4.36074510e-02 4.32367235e-01
-3.06555927e-01 -4.13840175e-01 3.41539234e-01 -1.08785564e-02
-5.31226695e-01 1.05605876e+00 7.86494911e-01 -1.45477682e-01
-1.83044046e-01 -3.13365042e-01 -2.91347295e-01 6.71722591e-01
1.00603080e+00 -4.86640424e-01 -3.58961701e-01 -3.03891778e-01
5.85202537e-02 2.00290352e-01 1.60776213e-01 -7.81088233e-01
2.69958675e-01 5.89496046e-02 -4.78887707e-02 -3.91553730e-01
5.92316687e-01 -7.46883571e-01 -6.97108030e-01 1.05085619e-01
-6.63957596e-01 9.12091881e-02 3.04606766e-01 5.09828389e-01
-3.71262997e-01 -9.40752029e-01 5.98386705e-01 -2.04354733e-01
-8.79532874e-01 -4.70883362e-02 -6.11241817e-01 6.77639306e-01
9.01919603e-01 2.08257079e-01 -6.95390999e-01 -7.12130964e-01
-8.09943974e-01 6.21040404e-01 -1.60311200e-02 7.49937832e-01
5.31003177e-01 -8.97797048e-01 -6.86385512e-01 -1.03469610e-01
2.70286888e-01 -2.87770540e-01 2.30391845e-01 6.48671269e-01
-4.83499020e-01 6.06283903e-01 -1.79742813e-01 -1.36851460e-01
-1.00865710e+00 5.84964573e-01 2.48529494e-01 -5.97545803e-01
-3.94790024e-01 8.50971997e-01 3.13247979e-01 -1.07036245e+00
4.03129369e-01 -1.59390807e-01 -7.95162737e-01 3.97486091e-01
5.63201427e-01 8.94655436e-02 -7.41780251e-02 -5.19569695e-01
-3.03400427e-01 2.03149952e-02 -2.25357637e-01 -2.94167459e-01
1.17915082e+00 -1.54052451e-01 8.52931514e-02 7.97435820e-01
9.91593421e-01 -6.86516985e-02 -4.11510289e-01 -4.30891871e-01
4.75477070e-01 -9.60183740e-02 -2.62616187e-01 -8.21461439e-01
-6.18686497e-01 1.14215744e+00 3.79247755e-01 4.54665571e-01
5.43638349e-01 6.51728958e-02 8.11648071e-01 9.55562294e-01
2.96224684e-01 -1.13874960e+00 -2.04811804e-03 1.03392780e+00
1.08320427e+00 -1.45574164e+00 -4.08005506e-01 -2.83075303e-01
-9.64852095e-01 1.14550352e+00 9.04327571e-01 -5.63695356e-02
7.83547938e-01 -2.22782835e-01 2.13652089e-01 -3.05242509e-01
-9.07343984e-01 -3.70211363e-01 2.38456596e-02 6.61682963e-01
5.90150595e-01 -2.44223312e-01 -6.85821950e-01 8.80184472e-01
-1.59291089e-01 -2.34577596e-01 6.29974127e-01 9.78204370e-01
-3.72813553e-01 -1.02403724e+00 -6.39818385e-02 4.63281870e-01
-3.07330668e-01 -3.07711154e-01 -7.83995986e-01 1.19156408e+00
-3.71046007e-01 1.38203120e+00 -5.31938709e-02 -2.40915250e-02
5.03355265e-01 7.67023802e-01 5.69135621e-02 -7.90711701e-01
-8.13202381e-01 -5.51322520e-01 6.81783617e-01 -5.41344821e-01
-4.69794393e-01 -2.72085786e-01 -9.98651743e-01 -9.20552686e-02
-6.06545746e-01 3.35467339e-01 4.69520688e-01 9.25314963e-01
1.80649564e-01 2.90901691e-01 -1.46956434e-02 -3.03877205e-01
-8.96666825e-01 -1.48121643e+00 -7.65291512e-01 6.29182041e-01
2.55472884e-02 -7.86942005e-01 -6.46938145e-01 -2.82349497e-01] | [12.571678161621094, 7.991552352905273] |
d0b67964-5d49-4c27-8142-2d9152980ed7 | two-stage-single-image-reflection-removal | 2012.00945 | null | https://arxiv.org/abs/2012.00945v2 | https://arxiv.org/pdf/2012.00945v2.pdf | Two-Stage Single Image Reflection Removal with Reflection-Aware Guidance | Removing undesired reflection from an image captured through a glass surface is a very challenging problem with many practical application scenarios. For improving reflection removal, cascaded deep models have been usually adopted to estimate the transmission in a progressive manner. However, most existing methods are still limited in exploiting the result in prior stage for guiding transmission estimation. In this paper, we present a novel two-stage network with reflection-aware guidance (RAGNet) for single image reflection removal (SIRR). To be specific, the reflection layer is firstly estimated due to that it generally is much simpler and is relatively easier to estimate. Reflectionaware guidance (RAG) module is then elaborated for better exploiting the estimated reflection in predicting transmission layer. By incorporating feature maps from the estimated reflection and observation, RAG can be used (i) to mitigate the effect of reflection from the observation, and (ii) to generate mask in partial convolution for mitigating the effect of deviating from linear combination hypothesis. A dedicated mask loss is further presented for reconciling the contributions of encoder and decoder features. Experiments on five commonly used datasets demonstrate the quantitative and qualitative superiority of our RAGNet in comparison to the state-of-the-art SIRR methods. The source code and pre-trained model are available at https://github.com/liyucs/RAGNet. | ['WangMeng Zuo', 'Dongwei Ren', 'Qince Li', 'Yaling Yi', 'Ming Liu', 'Yu Li'] | 2020-12-02 | null | null | null | null | ['reflection-removal'] | ['computer-vision'] | [ 6.91595554e-01 1.34994864e-01 3.82969379e-01 -2.58068144e-01
-7.15162992e-01 3.34731132e-01 4.34584141e-01 -3.32056880e-01
-6.82311878e-02 4.06263351e-01 4.55297828e-01 -7.09319264e-02
2.66561992e-02 -7.98388302e-01 -7.39612818e-01 -1.00390661e+00
3.00191075e-01 -3.12776774e-01 1.51965335e-01 -2.69014210e-01
1.80091724e-01 1.18290819e-01 -1.31059194e+00 4.01611000e-01
1.03187454e+00 9.83127952e-01 6.26842260e-01 3.49836618e-01
1.80115104e-01 7.97348022e-01 -4.24628884e-01 -1.86286107e-01
3.64102185e-01 -5.84782600e-01 -1.04390956e-01 -2.68906187e-02
2.24398360e-01 -7.35326111e-01 -4.34299856e-01 9.12396252e-01
6.19055390e-01 4.95760851e-02 5.10866523e-01 -4.78896976e-01
-5.33797801e-01 6.53249323e-01 -7.89134264e-01 -5.25460541e-02
2.84333497e-01 1.08616054e-01 7.46947229e-01 -8.16806912e-01
3.43773775e-02 1.09803832e+00 5.82650125e-01 4.31112885e-01
-8.02804410e-01 -8.23048115e-01 2.22255483e-01 7.24796653e-02
-1.30734730e+00 -5.48715293e-01 1.04882860e+00 -4.10663225e-02
8.54406297e-01 2.17433691e-01 4.43531096e-01 9.04758692e-01
1.81422010e-01 7.57358432e-01 8.23351145e-01 -4.01111096e-01
-3.72042626e-01 3.15530151e-01 -7.43419752e-02 6.98620915e-01
-4.49009649e-02 3.12499553e-01 -4.05233949e-01 3.29172760e-01
8.02834451e-01 1.53910920e-01 -6.64333105e-01 3.22633028e-01
-7.75575519e-01 4.27015543e-01 8.37959766e-01 1.05833709e-01
-3.86143625e-01 1.29812181e-01 2.41905637e-03 2.93655008e-01
7.72708416e-01 3.01768571e-01 -1.13596961e-01 4.77986664e-01
-8.05245876e-01 -4.86491136e-02 4.90183592e-01 7.45000303e-01
8.08799088e-01 2.13068172e-01 -3.21963072e-01 1.31733811e+00
7.25456715e-01 5.61943412e-01 1.14707381e-01 -7.09695935e-01
5.98734200e-01 5.27704775e-01 4.90709469e-02 -7.43196964e-01
-3.62330824e-01 -8.80803823e-01 -1.15083647e+00 2.64382303e-01
-1.05072670e-01 -7.82091618e-02 -9.61448312e-01 1.42153823e+00
2.43808970e-01 4.64219093e-01 1.55962825e-01 1.17436349e+00
1.16981912e+00 9.39489245e-01 -2.98858702e-01 -1.41609013e-01
1.18254590e+00 -1.20414364e+00 -5.05540490e-01 -2.64318645e-01
3.27949405e-01 -1.05225813e+00 6.52538300e-01 5.29725790e-01
-1.08111382e+00 -6.51567936e-01 -1.17287195e+00 -2.04999983e-01
2.64593929e-01 4.75024730e-01 4.21523809e-01 4.45651889e-01
-8.59157979e-01 4.34317291e-01 -7.61914968e-01 5.86691797e-02
4.55757767e-01 2.06619352e-01 9.16637778e-02 -4.27937478e-01
-1.18688381e+00 5.52128017e-01 -2.11281314e-01 7.16114521e-01
-9.50548172e-01 -9.05971408e-01 -7.47932732e-01 -7.51998574e-02
3.21643472e-01 -6.81432724e-01 9.44305420e-01 -8.60568464e-01
-1.75503445e+00 4.58111435e-01 -1.18463039e-01 -3.73718917e-01
4.96920854e-01 -6.13112986e-01 -3.32490444e-01 2.35982105e-01
-3.86229962e-01 3.02913189e-01 1.11745453e+00 -1.43624270e+00
-3.34625274e-01 -1.04485705e-01 1.00146264e-01 4.80730504e-01
-1.87188104e-01 5.69713907e-03 -6.11642718e-01 -6.24068737e-01
2.07698569e-01 -5.82598746e-01 -1.89333946e-01 2.56613493e-02
-6.74905121e-01 2.27270767e-01 7.53007233e-01 -8.78343523e-01
1.13348651e+00 -2.14745116e+00 -1.98777646e-01 -9.25636943e-03
3.07294279e-01 7.26944566e-01 -3.79829526e-01 5.88302493e-01
-3.92577946e-02 -2.21430942e-01 -3.35564882e-01 -5.54525137e-01
-3.69971305e-01 -3.80371720e-01 -4.42086935e-01 4.78928834e-01
2.64075398e-01 5.61384857e-01 -5.40698528e-01 -9.51419547e-02
5.38551569e-01 1.09505367e+00 -5.83569348e-01 4.82780010e-01
-5.95961250e-02 8.29692960e-01 -5.42638361e-01 5.12235820e-01
1.01639462e+00 -9.90819186e-03 -1.26053646e-01 -6.34688616e-01
-3.51240784e-01 4.65125561e-01 -9.38177228e-01 1.23254764e+00
-9.17594850e-01 5.18331409e-01 1.11340933e-01 -9.00401831e-01
1.17841697e+00 3.33234221e-01 3.48697394e-01 -1.10765731e+00
2.01304317e-01 2.68840164e-01 1.08358353e-01 -4.77030575e-01
3.27078998e-01 -7.20338747e-02 4.62298483e-01 4.00568753e-01
-3.21424842e-01 6.27396582e-03 -2.21145928e-01 -1.36209214e-02
1.03072894e+00 3.08246791e-01 4.11959030e-02 1.94469780e-01
7.76663840e-01 -6.41899049e-01 4.55551147e-01 6.45211339e-01
2.13760838e-01 1.12809491e+00 -4.23903428e-02 -2.92561889e-01
-7.49238372e-01 -9.94526923e-01 -2.72185300e-02 6.47783518e-01
4.57688719e-01 -1.72308862e-01 -5.63890815e-01 -1.35364920e-01
-4.11448509e-01 5.96782565e-01 -2.10031033e-01 -3.79870296e-01
-7.73305297e-01 -8.94300759e-01 1.75037697e-01 2.11992308e-01
1.13686013e+00 -1.02006245e+00 -4.09297913e-01 2.47454450e-01
-3.91926527e-01 -1.12099409e+00 -2.70241231e-01 -1.89361483e-01
-7.07271039e-01 -8.46487582e-01 -8.24839115e-01 -4.10105944e-01
8.07596982e-01 8.24699163e-01 8.37846935e-01 6.05825424e-01
-5.33702612e-01 2.17244744e-01 -4.85555261e-01 -5.72688103e-01
-2.18122691e-01 -1.34431839e-01 -5.00625134e-01 2.14889020e-01
4.26162928e-02 -6.17454410e-01 -1.23608863e+00 4.77139741e-01
-8.35804462e-01 4.88762230e-01 9.32293296e-01 6.12938464e-01
3.50110888e-01 2.56824940e-01 3.56984407e-01 -9.57537830e-01
3.15899283e-01 -3.67448241e-01 -5.75680792e-01 -8.92174840e-02
-3.61693680e-01 -1.87630281e-01 7.62378871e-01 -1.42194271e-01
-1.69331563e+00 -3.57937872e-01 -5.57681024e-01 -1.46829948e-01
1.20911887e-02 2.06330493e-01 -1.73420146e-01 -1.13084726e-01
2.09739774e-01 2.20202506e-01 -1.02918886e-01 -5.69265604e-01
-1.05898276e-01 6.14965498e-01 -2.51276866e-02 -1.87906787e-01
7.74283946e-01 6.29179239e-01 3.79396193e-02 -1.12153995e+00
-1.08410335e+00 -4.76821899e-01 -5.94752803e-02 -3.57952565e-01
6.15132332e-01 -1.24156117e+00 -5.37077487e-01 8.87346745e-01
-9.34674978e-01 -7.03115582e-01 1.16181999e-01 5.85825741e-01
-1.46959394e-01 2.95858949e-01 -5.99554837e-01 -9.75933969e-01
-7.17256308e-01 -1.11159945e+00 1.13355672e+00 3.17025363e-01
3.19640726e-01 -8.30751896e-01 -1.36008337e-01 6.87614918e-01
7.06764162e-01 -2.31855422e-01 6.64172888e-01 9.12157372e-02
-1.02162778e+00 -9.35904309e-02 -6.62701190e-01 7.58045137e-01
2.19629928e-01 -1.19617730e-01 -1.24742746e+00 -2.40638509e-01
2.79654235e-01 -7.32880756e-02 1.37646139e+00 5.65173626e-01
1.15706742e+00 -1.66009843e-01 -2.32622892e-01 8.21934223e-01
1.63647997e+00 2.18740832e-02 1.23215818e+00 1.08612858e-01
8.94407928e-01 6.21118784e-01 7.33114541e-01 4.40080017e-01
2.93179154e-01 6.66339874e-01 6.97575986e-01 -5.25975585e-01
-6.83244884e-01 -6.45531714e-02 4.94491577e-01 8.64625633e-01
-2.69719124e-01 -7.32185900e-01 -3.21919769e-01 2.40932882e-01
-1.55475771e+00 -6.58601522e-01 -3.03479910e-01 2.22603965e+00
5.85697412e-01 -1.13590742e-02 -3.50554407e-01 1.08564921e-01
4.90233928e-01 4.57418472e-01 -3.24772328e-01 -1.73434600e-01
8.57572537e-03 2.62387335e-01 4.08004612e-01 7.23723710e-01
-8.36452782e-01 6.70653462e-01 4.74596024e+00 7.96989083e-01
-1.20779526e+00 -7.93718696e-02 5.08257806e-01 -1.87658425e-02
-4.74578083e-01 -9.87581089e-02 -9.22847986e-01 2.77799368e-01
5.30322373e-01 5.47295868e-01 3.15770924e-01 6.12225309e-02
6.35877609e-01 -2.93309361e-01 -6.75600648e-01 7.48527527e-01
1.64728742e-02 -7.69461453e-01 1.30668074e-01 -1.75497830e-01
5.84983468e-01 1.34092048e-01 2.31848076e-01 1.42732663e-02
3.93054299e-02 -8.57054830e-01 3.82073075e-01 7.46826172e-01
6.92896068e-01 -5.65810680e-01 7.24937022e-01 1.51149586e-01
-1.02690375e+00 -2.97431275e-02 -4.68009472e-01 6.57741576e-02
2.13329613e-01 1.10164237e+00 -6.01025164e-01 8.18904877e-01
6.84482157e-01 9.34677601e-01 -3.79534326e-02 9.90718603e-01
-6.83470845e-01 8.43867540e-01 -2.73107678e-01 3.69242936e-01
1.49420455e-01 -5.47978818e-01 7.90974021e-01 1.07943308e+00
4.14275855e-01 1.75816625e-01 -2.47005492e-01 9.16769862e-01
-5.21798059e-03 -1.37611479e-01 -3.61753464e-01 3.41688931e-01
9.30487067e-02 1.37621558e+00 -3.53317529e-01 1.02850944e-01
-7.06265152e-01 8.85613561e-01 7.06166551e-02 6.70047104e-01
-8.12579989e-01 -3.62323940e-01 6.51031792e-01 4.01003003e-01
3.91157538e-01 -6.49373978e-02 -2.11674906e-03 -1.00684702e+00
1.50797620e-01 -5.92285872e-01 -3.48709151e-02 -9.05073762e-01
-9.61700559e-01 5.59904456e-01 -2.35378101e-01 -1.29502261e+00
4.22015101e-01 -4.64893222e-01 -7.79618680e-01 1.06925225e+00
-2.29525852e+00 -1.15920901e+00 -6.30775034e-01 3.90433580e-01
5.99381149e-01 1.66486129e-01 3.95513088e-01 6.79682434e-01
-7.81633735e-01 4.76988554e-01 -4.95929923e-03 -4.22028564e-02
7.37520754e-01 -6.74043179e-01 1.19959958e-01 9.67999279e-01
-3.39014947e-01 5.79068005e-01 7.31273770e-01 -4.54937756e-01
-1.35111630e+00 -1.31917143e+00 4.02386069e-01 1.79904163e-01
2.14255601e-01 -3.07554692e-01 -1.00747693e+00 4.38918382e-01
2.31247872e-01 -2.61528730e-01 4.80335414e-01 -2.30795264e-01
-4.08476740e-01 -5.68267286e-01 -9.44695652e-01 6.89013243e-01
9.49755728e-01 -2.36698985e-01 1.12901054e-01 1.23531327e-01
6.51827872e-01 -3.93696964e-01 -5.02909303e-01 5.33734024e-01
6.39658093e-01 -1.30983055e+00 1.05298948e+00 4.51767355e-01
9.00351048e-01 -2.95601577e-01 -4.43882309e-02 -1.20578480e+00
-1.99820429e-01 -6.20784283e-01 1.26535855e-02 1.23587704e+00
3.51579636e-01 -9.09444392e-01 5.19478381e-01 2.45487601e-01
-5.75657666e-01 -8.01521182e-01 -3.02959204e-01 -3.80243242e-01
-4.52430457e-01 -4.53908175e-01 3.59850317e-01 3.19386929e-01
-7.58135021e-01 3.21590334e-01 -8.02596807e-01 4.04278338e-01
8.12534571e-01 1.52225807e-01 7.52497137e-01 -7.83330262e-01
-3.46538812e-01 -1.39534056e-01 4.92329011e-03 -1.55567241e+00
-2.05521241e-01 -6.29897952e-01 3.31935436e-01 -1.88507891e+00
1.66143000e-01 -5.99838078e-01 -3.63801390e-01 2.43390307e-01
-3.26111615e-01 5.61854422e-01 4.95489538e-02 1.84723020e-01
-3.98497432e-01 8.23355019e-01 1.58412242e+00 1.25606954e-01
-2.78876156e-01 2.87032425e-01 -1.00538421e+00 6.68794870e-01
8.88584852e-01 -3.82124841e-01 -5.59611261e-01 -8.22075665e-01
3.36823434e-01 -2.36509964e-02 6.48928583e-01 -8.42502356e-01
-1.55156199e-02 3.51955354e-01 1.03883788e-01 -5.94599545e-01
6.00601435e-01 -8.55825424e-01 9.12676379e-02 2.57283747e-01
-2.00233117e-01 -9.40405190e-01 1.61029905e-01 5.57665229e-01
-2.52040982e-01 -1.23493195e-01 9.86388922e-01 -1.10284694e-01
-3.64369035e-01 2.89468586e-01 -1.46559909e-01 -3.63451838e-01
4.56213266e-01 -2.06406891e-01 -4.66408014e-01 -4.63642627e-01
-1.15412854e-01 1.63791940e-01 4.85276617e-02 1.76983774e-01
9.77576971e-01 -7.91297138e-01 -9.81121063e-01 2.16051981e-01
-5.08593619e-02 7.51201585e-02 7.77450383e-01 1.23603451e+00
-4.53032851e-01 1.41333446e-01 2.90106058e-01 -3.55399251e-01
-1.33596897e+00 -2.06946414e-02 4.04519469e-01 -3.87683034e-01
-1.04353809e+00 9.55521524e-01 9.10619199e-01 -2.38204837e-01
8.20179731e-02 -2.24377736e-01 -3.51504773e-01 -3.26972276e-01
7.69576967e-01 4.49453592e-01 1.45124480e-01 -3.70557159e-01
-8.35413579e-03 7.12485611e-01 -3.64979208e-01 3.51427853e-01
1.63413870e+00 -5.46163142e-01 -1.49414480e-01 8.79281089e-02
1.24468577e+00 2.60557681e-02 -1.49805498e+00 -4.63096857e-01
-8.39720905e-01 -4.17654008e-01 3.44367176e-01 -7.84215152e-01
-1.55421114e+00 1.04890168e+00 4.69831824e-01 -1.51508912e-01
1.59816432e+00 -2.96639234e-01 1.01728880e+00 7.59709775e-02
4.02018428e-02 -6.96428239e-01 2.10417986e-01 2.66031712e-01
1.00247037e+00 -1.37196016e+00 2.38146082e-01 -8.39047849e-01
-3.48080635e-01 1.04753590e+00 5.21486938e-01 -2.98791260e-01
7.08085001e-01 3.59178871e-01 2.35533491e-01 -2.76090801e-01
-6.02519989e-01 -4.44688573e-02 3.04718852e-01 4.64633167e-01
6.11995339e-01 -2.24472299e-01 -1.74157396e-01 3.09843510e-01
9.44116414e-02 -1.30711630e-01 4.43450212e-01 5.97701669e-01
-4.02246892e-01 -8.03192616e-01 -1.82831645e-01 5.20587981e-01
-5.91090262e-01 -4.94010806e-01 1.84732564e-02 4.14974183e-01
-1.07840121e-01 1.21335971e+00 -2.49694765e-01 -2.26286635e-01
2.27453455e-01 -8.21004629e-01 4.51130807e-01 -6.66653574e-01
-5.19723833e-01 4.07751143e-01 1.00276008e-01 -6.01784527e-01
-6.84187293e-01 -3.42985779e-01 -1.01353204e+00 6.48386478e-02
-6.16969049e-01 -2.51956016e-01 5.95301926e-01 8.28187406e-01
3.21278483e-01 1.00212467e+00 8.27239156e-01 -8.97861898e-01
-2.13340044e-01 -1.02323842e+00 -6.03607476e-01 6.39602691e-02
6.04483306e-01 -4.95439410e-01 -6.02172613e-01 -1.35639668e-01] | [10.683148384094238, -2.8670310974121094] |
d5bd91c6-6a24-49c4-a60f-e0d61c627db2 | blur-invariants-for-image-recognition | 2301.07581 | null | https://arxiv.org/abs/2301.07581v1 | https://arxiv.org/pdf/2301.07581v1.pdf | Blur Invariants for Image Recognition | Blur is an image degradation that is difficult to remove. Invariants with respect to blur offer an alternative way of a~description and recognition of blurred images without any deblurring. In this paper, we present an original unified theory of blur invariants. Unlike all previous attempts, the new theory does not require any prior knowledge of the blur type. The invariants are constructed in the Fourier domain by means of orthogonal projection operators and moment expansion is used for efficient and stable computation. It is shown that all blur invariants published earlier are just particular cases of this approach. Experimental comparison to concurrent approaches shows the advantages of the proposed theory. | ['Jitka Kostkova', 'Filip Sroubek', 'Matteo Pedone', 'Matej Lebl', 'Jan Flusser'] | 2023-01-18 | null | null | null | null | ['deblurring'] | ['computer-vision'] | [ 1.57848131e-02 -5.56881666e-01 2.24975735e-01 -2.52325654e-01
1.85566620e-04 -4.93648082e-01 5.55712163e-01 -6.01817548e-01
-2.11606219e-01 9.35638309e-01 3.22015524e-01 -8.36613253e-02
-5.11298418e-01 -7.00745434e-02 -7.48864859e-02 -7.22823799e-01
-1.77342385e-01 -1.50993809e-01 3.10712427e-01 -8.72560143e-02
6.32628322e-01 8.87912333e-01 -1.21658492e+00 -2.42845908e-01
9.13163781e-01 7.47093976e-01 3.02771837e-01 1.14996767e+00
2.54056245e-01 9.63978469e-01 -5.74609280e-01 -2.24272743e-01
4.37822104e-01 -3.93280745e-01 -1.10145378e+00 5.44111133e-01
4.23723638e-01 -4.86300200e-01 -9.17258024e-01 1.44309211e+00
3.75863582e-01 1.13911025e-01 7.18536079e-01 -8.34769070e-01
-1.36636460e+00 1.66047201e-01 -5.60502589e-01 8.27359319e-01
7.90304005e-01 -9.23907161e-02 2.88028389e-01 -9.53880787e-01
5.48536241e-01 9.09170449e-01 8.41409028e-01 2.32770994e-01
-1.25620115e+00 1.84544116e-01 -6.54973805e-01 5.16310096e-01
-1.49035430e+00 -4.90545630e-01 8.15260530e-01 -3.46231401e-01
6.45825803e-01 8.72965395e-01 3.31929713e-01 4.11281466e-01
5.19669652e-01 4.90241200e-01 1.76519704e+00 -6.35667205e-01
-1.82945594e-01 2.39107341e-01 6.63121760e-01 7.72148788e-01
5.19918799e-01 5.77902794e-02 -2.27352634e-01 -1.99893117e-01
1.12589979e+00 6.34071454e-02 -9.45532978e-01 -3.25833380e-01
-1.43308771e+00 2.31956124e-01 3.19977641e-01 6.47182941e-01
-3.40735763e-01 1.43648192e-01 1.69825181e-02 4.96766120e-01
3.86464089e-01 7.02943623e-01 -1.71037331e-01 -3.15236062e-01
-1.09550989e+00 -5.96230067e-02 9.22524095e-01 1.13481951e+00
6.50016785e-01 7.56394789e-02 -1.11790098e-01 6.03511393e-01
1.73017439e-02 7.23794878e-01 5.54172873e-01 -1.16139853e+00
-4.37294245e-01 4.22233455e-02 5.38472116e-01 -7.80304611e-01
-9.94120017e-02 -1.39241129e-01 -7.66231239e-01 3.40200603e-01
4.10118133e-01 -1.09751925e-01 -9.60943758e-01 1.01909363e+00
-2.04605937e-01 4.26581979e-01 1.66734219e-01 1.02686048e+00
4.35661376e-01 1.99219868e-01 -7.81783342e-01 -4.87374425e-01
1.30463898e+00 -8.45638812e-01 -1.39389193e+00 3.44995379e-01
-3.60667646e-01 -1.77506018e+00 2.39768922e-01 4.24615085e-01
-1.23800540e+00 -4.40217555e-01 -9.75006938e-01 -1.90284371e-01
-2.81072289e-01 -6.64082319e-02 8.50071073e-01 7.71453202e-01
-1.53204870e+00 6.87199295e-01 -3.84970665e-01 -4.84541535e-01
-1.19150139e-01 4.55493569e-01 -3.21449459e-01 7.38675073e-02
-8.94573569e-01 1.61903310e+00 4.80019189e-02 3.15333158e-01
-3.90526175e-01 -2.12712452e-01 -6.07173443e-01 -1.51890889e-01
-2.51374722e-01 -7.22474217e-01 1.37834728e+00 -7.28885710e-01
-1.53147376e+00 6.83945179e-01 -6.63457632e-01 -3.83339345e-01
6.96476161e-01 -5.17251670e-01 -4.92914468e-01 4.19095129e-01
-1.84442759e-01 -1.60922721e-01 1.36755741e+00 -1.09472227e+00
-1.62513107e-01 1.11436918e-01 1.15524687e-01 2.04663098e-01
3.64141375e-01 2.79099226e-01 -5.75421564e-02 -4.42887962e-01
3.95488799e-01 -7.57151544e-01 -5.20477146e-02 -2.83655047e-01
-9.33564901e-02 2.54208565e-01 1.12612975e+00 -7.31819212e-01
1.23810792e+00 -2.04685855e+00 2.11644605e-01 -8.21894184e-02
3.71989578e-01 3.08188051e-01 5.10271251e-01 5.69419205e-01
-3.86076450e-01 -2.65479147e-01 -8.93092677e-02 6.41553104e-02
-7.07168579e-02 1.22510113e-01 -4.45942551e-01 9.76756036e-01
-7.93729797e-02 5.75114071e-01 -8.17361593e-01 -5.00417292e-01
6.61721408e-01 5.03621578e-01 2.97050122e-02 3.58955085e-01
6.41664922e-01 9.93698463e-02 -1.32669151e-01 5.10763526e-01
1.11760366e+00 -9.59113836e-02 -4.48277086e-01 -6.25751793e-01
-5.77400386e-01 -2.84414500e-01 -1.11383224e+00 1.19701040e+00
-1.21546164e-01 1.23382211e+00 8.36302415e-02 -6.96125448e-01
7.49307454e-01 7.54674375e-01 4.16450679e-01 2.55717337e-01
7.80906305e-02 3.73807788e-01 -5.91346137e-02 -1.01011765e+00
6.65226400e-01 -2.28960231e-01 3.66688401e-01 2.42622048e-01
4.69786935e-02 -5.19461095e-01 2.06724718e-01 6.07400797e-02
1.05733228e+00 -5.55320419e-02 6.37957573e-01 -6.79173172e-01
9.35161471e-01 -1.33250311e-01 4.04092064e-03 8.78852427e-01
-5.61664343e-01 8.49536657e-01 5.31555265e-02 -4.73723024e-01
-1.02948463e+00 -1.03571713e+00 -5.63579798e-01 1.54014915e-01
6.36119545e-01 9.33493674e-02 -9.14157093e-01 -1.79126576e-01
-5.18165715e-02 3.12459558e-01 -3.95767391e-01 1.85321718e-01
-3.77472132e-01 -8.43045354e-01 4.31522042e-01 7.21708536e-02
9.31814313e-01 -5.06979644e-01 -6.70042336e-01 -1.31448582e-01
-8.97404701e-02 -1.09010756e+00 -8.25933754e-01 -2.53145099e-01
-1.03343368e+00 -1.21382630e+00 -1.14077771e+00 -9.13145125e-01
9.05629873e-01 1.02993691e+00 7.01286852e-01 -3.34738009e-02
-5.10790586e-01 8.39544475e-01 -1.65591389e-02 -3.58610451e-02
-2.01378241e-01 -6.67422295e-01 3.08238536e-01 9.15356427e-02
4.69288379e-01 -5.44775605e-01 -6.85520470e-01 2.11698249e-01
-6.75114810e-01 -1.95722491e-01 9.01359439e-01 8.60742867e-01
-2.47435048e-01 1.30241439e-01 -7.85993859e-02 -4.80967402e-01
1.04308844e+00 6.39432296e-02 -5.94132543e-01 2.48282284e-01
-6.73283398e-01 2.79891074e-01 2.41462767e-01 -4.87375677e-01
-1.44492662e+00 -2.39775479e-01 4.53471929e-01 -4.80661839e-01
-2.85135597e-01 2.45409474e-01 5.20620942e-01 -7.26690352e-01
6.56114638e-01 5.95117927e-01 -1.23539870e-03 -7.07352400e-01
3.56326908e-01 8.99999499e-01 9.51907635e-01 -1.68065324e-01
9.80038643e-01 4.76233721e-01 1.51612833e-01 -1.27694130e+00
-4.30501848e-01 -9.38086987e-01 -8.89058173e-01 -3.07453334e-01
6.48938835e-01 -5.31732380e-01 -6.35310113e-01 8.43577206e-01
-1.44939971e+00 4.51275915e-01 -8.87815654e-02 1.01266026e+00
-6.45221472e-01 1.00306129e+00 -9.36219811e-01 -1.03439856e+00
-1.56181097e-01 -1.14468145e+00 4.13948298e-01 5.05054355e-01
-3.85792996e-03 -1.24993551e+00 1.70105640e-02 1.68940142e-01
8.92150164e-01 -6.40612915e-02 1.74053460e-01 -1.89883739e-01
-8.32867384e-01 -2.44162530e-01 -5.74594676e-01 4.94982392e-01
7.02853739e-01 -5.98326232e-03 -1.10878873e+00 -2.00935125e-01
9.38034892e-01 5.43753207e-01 5.71196914e-01 5.26348650e-01
5.58917224e-01 -3.94258261e-01 -1.24192253e-01 7.32149661e-01
1.75256610e+00 2.97562361e-01 7.58775473e-01 2.03922033e-01
3.52358311e-01 -2.65160173e-01 2.72751451e-01 1.43626794e-01
-3.03558528e-01 4.27842617e-01 -4.40177955e-02 -1.01289265e-01
-3.04923564e-01 6.54383600e-01 3.00791651e-01 9.91195560e-01
-8.23048413e-01 4.27241117e-01 -5.25593579e-01 6.00536942e-01
-1.61295617e+00 -1.52865696e+00 -4.58813190e-01 2.04697943e+00
9.73866284e-01 -8.98570046e-02 -5.15983462e-01 -4.10448574e-02
9.77922916e-01 3.20777565e-01 -1.22404573e-02 -4.58939135e-01
8.18721298e-03 2.32541561e-01 8.97494495e-01 1.08270013e+00
-1.15806675e+00 5.01775563e-01 8.36886215e+00 2.36344442e-01
-1.04951990e+00 1.57795221e-01 -1.57330632e-01 5.75592637e-01
1.73673466e-01 2.34052792e-01 -1.97385594e-01 5.29911578e-01
5.37252426e-01 -7.11440682e-01 7.90271938e-01 7.33300745e-01
2.82772273e-01 -6.24392271e-01 -8.27871025e-01 1.21874905e+00
2.50623584e-01 -9.47510064e-01 -3.78423274e-01 -3.31620015e-02
7.65208066e-01 -7.98381418e-02 8.92539546e-02 -4.81614083e-01
3.72310393e-02 -7.17473328e-01 3.38899881e-01 1.33123851e+00
5.87620914e-01 -2.98281819e-01 1.09515238e+00 -1.04513273e-01
-6.94315612e-01 3.73934805e-01 -5.53442180e-01 -4.82223332e-01
2.19500870e-01 7.41185009e-01 -7.92362809e-01 6.25065863e-01
3.30983788e-01 6.33600056e-01 -6.81060970e-01 1.66959774e+00
1.09203011e-01 1.98720515e-01 1.24707818e-01 1.10275134e-01
4.52209152e-02 -5.38246155e-01 9.94305432e-01 1.52508283e+00
5.61404943e-01 2.64234781e-01 -3.68040740e-01 5.84392846e-01
4.10632014e-01 -5.22796273e-01 -8.63776922e-01 2.45723009e-01
4.12466712e-02 1.11001253e+00 -4.42421019e-01 -5.48370957e-01
-6.02314055e-01 1.54802573e+00 -3.65814775e-01 5.29964924e-01
-6.33629262e-01 -8.30243945e-01 7.47221828e-01 -3.33805084e-01
2.27334663e-01 -5.61466873e-01 -1.60085887e-01 -1.74909317e+00
-3.07819575e-01 -5.51001310e-01 -1.51585773e-01 -1.05939853e+00
-1.31854892e+00 6.85262740e-01 3.43291461e-01 -1.30732715e+00
-1.79298624e-01 -9.08405840e-01 -5.73610306e-01 1.49385548e+00
-1.47272718e+00 -7.74606645e-01 -4.20833588e-01 6.49622798e-01
6.31070316e-01 -7.48893842e-02 7.06996739e-01 -2.75528859e-02
-1.11361489e-01 -2.07909688e-01 4.65546548e-01 -1.00350961e-01
9.74669695e-01 -1.86013043e+00 -2.73145847e-02 1.40781236e+00
-2.88508683e-01 1.10550094e+00 1.61865544e+00 -3.83042067e-01
-1.32934332e+00 -3.19451183e-01 1.00999629e+00 -4.95143861e-01
8.38200092e-01 3.74714285e-01 -8.08928609e-01 7.20947206e-01
1.01165366e+00 5.26172295e-02 2.84526706e-01 -2.47059166e-01
-7.67588392e-02 -2.13401437e-01 -1.03485107e+00 3.56598377e-01
6.79137826e-01 -8.62342000e-01 -1.40169227e+00 5.58813512e-01
2.18714952e-01 -6.60097718e-01 -9.23373103e-01 3.21963839e-02
5.67248404e-01 -1.41843069e+00 9.65955019e-01 -2.58855242e-02
-1.75145820e-01 -6.48294210e-01 6.59028590e-02 -1.16547596e+00
-8.44843090e-01 -1.28956544e+00 -1.78983256e-01 7.53020346e-01
-2.10255474e-01 -8.53137374e-01 6.88799098e-02 6.15329504e-01
-2.64806896e-01 -1.47218168e-01 -7.60425031e-01 -8.78891826e-01
-7.82397985e-01 1.12400703e-01 2.78026313e-01 9.18729901e-01
4.61134553e-01 1.56854734e-01 -7.76051462e-01 4.06595260e-01
9.37199473e-01 1.76886708e-01 3.42885613e-01 -9.76435483e-01
-2.03538030e-01 -3.53815526e-01 -7.88403034e-01 -1.14009225e+00
-3.84897321e-01 -2.11886093e-01 -1.81101318e-02 -1.34574199e+00
3.66430879e-01 6.96137026e-02 -3.70629996e-01 -3.48302573e-01
-1.29568785e-01 3.06642026e-01 -1.50072286e-04 6.02364719e-01
-2.78182507e-01 -9.35051516e-02 1.29307568e+00 7.24428101e-03
1.97431192e-01 1.93382710e-01 -3.29200864e-01 8.84264410e-01
6.95824802e-01 -5.92446811e-02 -4.35104668e-01 -3.72799039e-01
-4.94213641e-01 5.64917512e-02 4.42261219e-01 -1.18021810e+00
3.64520818e-01 -2.88427651e-01 3.97963703e-01 -4.70475048e-01
3.87930542e-01 -9.57823932e-01 5.27832389e-01 5.63389659e-01
6.16803057e-02 4.07787651e-01 -1.46431237e-01 4.55818921e-01
-2.65253305e-01 -9.61794019e-01 1.00043082e+00 -4.25109565e-01
-7.47467518e-01 -2.87389278e-01 -4.03485090e-01 -5.78796744e-01
9.18054640e-01 -2.96011686e-01 -5.19886196e-01 -5.38867652e-01
-7.87172318e-01 -7.00638652e-01 8.13103139e-01 -6.08469732e-02
6.18798256e-01 -1.09867048e+00 -2.96069235e-01 1.83040604e-01
-4.87232298e-01 -6.88673794e-01 4.30344865e-02 1.40972102e+00
-1.29334509e+00 7.31274664e-01 -4.48448867e-01 -3.34750980e-01
-1.48076236e+00 9.06751156e-01 5.79109490e-01 1.31983966e-01
-5.57316244e-01 5.60330987e-01 -5.50538264e-02 6.10281885e-01
-3.34392101e-01 -2.60475338e-01 -7.87707791e-02 -4.05321658e-01
6.90029204e-01 7.38397539e-01 -2.09012374e-01 -9.44103003e-01
-3.68111134e-01 6.68653607e-01 -2.52703913e-02 -2.70345181e-01
1.11398935e+00 -6.18956029e-01 -8.76827836e-01 4.76722836e-01
1.03589821e+00 4.44164217e-01 -9.58930552e-01 -1.76469609e-01
1.11348972e-01 -1.10820508e+00 4.80477186e-03 -5.49197137e-01
-5.96314967e-01 5.78790247e-01 7.57077873e-01 7.00884759e-01
1.30228841e+00 -2.27512568e-01 3.85782570e-01 2.26159483e-01
2.31876567e-01 -8.54067981e-01 -1.79685876e-01 4.45081443e-01
1.05843222e+00 -8.76509070e-01 5.00175714e-01 -5.81484079e-01
-2.98908770e-01 1.54946530e+00 2.15320170e-01 -3.53383988e-01
7.35115588e-01 9.18217152e-02 2.05779955e-01 -8.34764466e-02
-3.96066010e-01 -3.77298564e-01 5.58483839e-01 6.28380299e-01
5.95446110e-01 -1.58730745e-01 -8.05799365e-01 -3.03960413e-01
1.85843289e-01 1.87060103e-01 1.12779224e+00 1.10556901e+00
-6.48492813e-01 -7.86477029e-01 -8.77729118e-01 1.42973766e-01
-6.99652970e-01 -1.52119771e-01 -5.23731560e-02 6.22475266e-01
1.32068945e-02 8.14235151e-01 -3.39384973e-01 -1.18437588e-01
9.75113660e-02 1.00114480e-01 8.48895788e-01 -1.40316784e-01
-5.71878478e-02 2.22119674e-01 -1.56012520e-01 -2.77987570e-01
-8.59931529e-01 -8.24532032e-01 -4.72921103e-01 -4.52888787e-01
-6.07453108e-01 5.16014278e-01 4.30521399e-01 7.11607933e-01
1.12763464e-01 1.56118408e-01 6.48677647e-01 -8.88908446e-01
-6.44954145e-01 -1.10257423e+00 -8.54708791e-01 4.85549420e-01
1.04111493e+00 -5.28557718e-01 -7.34708428e-01 6.53772831e-01] | [11.632221221923828, -2.7661683559417725] |
a913ac37-e1d4-45d6-8774-9173074a8357 | effective-features-of-remote-sensing-image | 1401.7743 | null | http://arxiv.org/abs/1401.7743v1 | http://arxiv.org/pdf/1401.7743v1.pdf | Effective Features of Remote Sensing Image Classification Using Interactive Adaptive Thresholding Method | Remote sensing image classification can be performed in many different ways
to extract meaningful features. One common approach is to perform edge
detection. A second approach is to try and detect whole shapes, given the fact
that these shapes usually tend to have distinctive properties such as object
foreground or background. To get optimal results, these two approaches can be
combined. This paper adopts a combinatorial optimization method to adaptively
select threshold based features to improve remote sensing image. Feature
selection is an important combinatorial optimization problem in the remote
sensing image classification. The feature selection method has to achieve three
characteristics: first the performance issues by facilitating data collection
and reducing storage space and classification time, second to perform semantics
analysis helping to understand the problem, and third to improve prediction
accuracy by avoiding the curse of dimensionality. The goal of this thresholding
an image is to classify pixels as either dark or light and evaluation of
classification results. Interactive adaptive thresholding is a form of
thresholding that takes into account spatial variations in illumination of
remote sensing image. We present a technique for remote sensing based adaptive
thresholding using the interactive satellite image of the input. However, our
solution is more robust to illumination changes in the remote sensing image.
Additionally, our method is simple and easy to implement but it is effective
algorithm to classify the image pixels. This technique is suitable for
preprocessing the remote sensing image classification, making it a valuable
tool for interactive remote based applications such as augmented reality of the
classification procedure. | ['T. Balaji', 'Dr. M. Sumathi'] | 2014-01-30 | null | null | null | null | ['remote-sensing-image-classification'] | ['miscellaneous'] | [ 7.44256973e-01 -6.13076210e-01 1.71733052e-01 -4.36954707e-01
-2.03730866e-01 -5.11727929e-01 1.78934664e-01 1.35688201e-01
-4.93848562e-01 4.98676360e-01 -3.96099240e-01 -4.43198472e-01
-4.38083768e-01 -1.31007802e+00 -1.02027744e-01 -1.10631895e+00
-5.09430543e-02 1.53797075e-01 3.52811962e-01 -1.26912102e-01
4.94016945e-01 1.15915298e+00 -2.01729012e+00 1.30629456e-02
9.09240305e-01 9.63941574e-01 6.19266987e-01 7.69129276e-01
-2.25154132e-01 2.32025385e-01 -3.80106300e-01 4.87671643e-01
6.41245902e-01 -5.49917996e-01 -5.11597455e-01 5.42531550e-01
2.50010341e-02 3.76299061e-02 7.32295990e-01 1.25348330e+00
4.83561724e-01 3.90131891e-01 7.78700411e-01 -9.75801468e-01
-2.93345600e-01 2.50251889e-01 -7.15129733e-01 4.41094518e-01
-1.45189213e-02 5.63562755e-03 7.64863789e-01 -8.58769298e-01
2.60845244e-01 1.00385129e+00 4.08352822e-01 -1.72663163e-02
-1.07367778e+00 -3.07744145e-01 -2.01182831e-02 3.73276860e-01
-1.70341039e+00 -9.54054594e-02 7.40222335e-01 -4.25436616e-01
6.48557544e-01 8.82356346e-01 9.60228503e-01 -2.63183802e-01
1.23444796e-01 2.87542552e-01 1.55938601e+00 -7.55381942e-01
3.39524120e-01 2.73696095e-01 4.32737112e-01 5.80233455e-01
2.91509420e-01 -3.69974226e-02 1.60910517e-01 -5.47195002e-02
5.50233960e-01 4.51409459e-01 -3.63956422e-01 5.61757460e-02
-8.55390847e-01 9.02887344e-01 4.49045628e-01 5.70684254e-01
-5.50071716e-01 -1.41422749e-01 1.94332991e-02 3.99707675e-01
4.12400484e-01 4.05386597e-01 -5.58582962e-01 2.76783586e-01
-9.49329853e-01 -1.34026691e-01 3.56326938e-01 6.89465702e-02
1.09813857e+00 8.98199715e-03 2.00149268e-01 8.19919527e-01
5.40638506e-01 7.31025934e-01 4.26715732e-01 -5.73252499e-01
-5.28908335e-02 8.92607272e-01 -3.09240893e-02 -1.28409648e+00
-4.64644372e-01 -1.64669037e-01 -6.85476840e-01 9.85417783e-01
2.52355337e-01 3.26301977e-02 -1.19153476e+00 8.21577907e-01
5.15389442e-01 -2.95079798e-01 -1.21919796e-01 9.08635557e-01
6.44902766e-01 1.09838784e+00 -6.04672246e-02 -3.74436200e-01
1.37844288e+00 -4.10310894e-01 -5.77583849e-01 5.50482459e-02
4.34327394e-01 -9.31803286e-01 1.16638017e+00 5.66380978e-01
-5.40570974e-01 -4.00385946e-01 -9.66533840e-01 3.47009808e-01
-8.16934705e-01 4.81759250e-01 5.98041058e-01 8.73515368e-01
-8.96321952e-01 5.15928030e-01 -7.03967094e-01 -5.41401386e-01
-8.34527835e-02 5.53841472e-01 -9.66056585e-02 3.29608589e-01
-7.12666154e-01 9.06798244e-01 6.45595491e-01 4.15430516e-01
-1.54896860e-03 -1.74198717e-01 -7.01583862e-01 1.21375285e-01
2.40679726e-01 -7.64999241e-02 4.97056395e-01 -1.40629733e+00
-1.38008344e+00 9.08921063e-01 -1.92604914e-01 1.77027471e-02
2.43382573e-01 1.77651614e-01 -3.65122825e-01 1.04161277e-01
-1.15877070e-01 2.02797815e-01 8.14330220e-01 -1.09054744e+00
-7.75087237e-01 -5.45250714e-01 -4.25387293e-01 2.94898182e-01
-1.00047618e-01 1.83978900e-01 4.86621112e-02 -5.80065429e-01
8.05520952e-01 -7.56069303e-01 -4.83328640e-01 1.07761666e-01
-9.47413035e-03 5.71299158e-03 1.21522033e+00 -6.09058380e-01
9.93178666e-01 -2.23080707e+00 -3.46330941e-01 8.57516527e-01
-3.24628651e-01 2.85620630e-01 1.82517737e-01 1.12855189e-01
-1.43052414e-01 3.62088531e-01 -3.00054640e-01 4.92740780e-01
-4.51583654e-01 2.09352851e-01 6.95689544e-02 6.20517313e-01
1.52165830e-01 2.69777328e-01 -4.90252525e-01 -7.31501102e-01
6.19062603e-01 5.05883753e-01 -1.07580490e-01 -1.31939232e-01
8.36295709e-02 3.82188350e-01 -5.43472230e-01 8.08308899e-01
8.50107312e-01 2.04979122e-01 -1.24085531e-01 -2.17554480e-01
-5.00388145e-01 -1.89802095e-01 -1.77062500e+00 6.75014198e-01
-3.21724862e-01 5.79960346e-01 1.61192641e-01 -1.25233614e+00
1.42698562e+00 9.63288173e-02 5.66312551e-01 -4.82027292e-01
2.91194797e-01 3.22525084e-01 -1.56364724e-01 -6.65561795e-01
5.83447456e-01 -1.64019674e-01 4.20239151e-01 2.85525590e-01
-7.09048092e-01 -3.77667695e-01 1.23747729e-01 -4.13491249e-01
3.55734289e-01 9.69741680e-03 7.59169698e-01 -4.22468752e-01
7.08572805e-01 2.96760023e-01 4.96922016e-01 4.64031607e-01
1.96265932e-02 2.59231627e-01 -1.29273176e-01 -7.45197535e-01
-5.91096878e-01 -6.15282774e-01 -4.13630784e-01 1.00414336e+00
2.24050179e-01 4.94659424e-01 -4.68327373e-01 -2.38921285e-01
-1.98587120e-01 3.84570807e-01 -4.67534333e-01 2.86519170e-01
-3.77795845e-01 -1.18168294e+00 -1.60517246e-01 5.22015132e-02
8.87332976e-01 -1.00406384e+00 -1.11388338e+00 1.81155682e-01
-4.23311666e-02 -3.59446675e-01 1.99007783e-02 3.66623223e-01
-1.45065570e+00 -9.51051712e-01 -4.39528883e-01 -6.71221673e-01
9.54869211e-01 8.45126271e-01 8.42124820e-01 6.59478784e-01
-6.36745512e-01 1.86043993e-01 -7.74146318e-01 -5.41960657e-01
-2.61003673e-01 -1.50115252e-01 -3.56786996e-01 2.36128852e-01
4.18576837e-01 -3.53604704e-01 -6.21434689e-01 5.07626891e-01
-1.09076536e+00 -2.53525436e-01 7.85603285e-01 6.46189034e-01
8.70723903e-01 8.25300038e-01 4.88873534e-02 -6.72495365e-01
2.69516528e-01 -8.37758258e-02 -8.49640310e-01 4.58105236e-01
-6.12175286e-01 -1.02190383e-01 4.56917882e-01 -2.00458407e-01
-9.35466051e-01 5.79380393e-01 2.20339857e-02 2.28258967e-01
-3.53968054e-01 6.62179649e-01 -1.93611115e-01 -4.37392771e-01
7.37812400e-01 3.36232543e-01 -8.21604684e-04 -4.85000074e-01
-4.15440239e-02 1.04526472e+00 3.10838558e-02 -8.95602405e-02
6.72948658e-01 5.93432009e-01 3.03633243e-01 -1.45659435e+00
-4.07728046e-01 -9.05579984e-01 -7.22115159e-01 -4.55497533e-01
1.02268577e+00 -2.50944257e-01 -6.06604218e-01 3.96079212e-01
-6.79376245e-01 -4.25719619e-02 2.86548547e-02 6.12366498e-01
-2.68422812e-02 3.49678725e-01 -1.05306804e-02 -1.33966219e+00
-4.65378463e-01 -1.02190959e+00 7.87489116e-01 4.07456994e-01
2.93609887e-01 -9.06053782e-01 -2.37950817e-01 1.63713008e-01
3.69829148e-01 4.13328797e-01 7.47854233e-01 -3.51772249e-01
-5.34756482e-01 -1.31992608e-01 -2.54228592e-01 2.34936059e-01
5.46801269e-01 6.25399053e-01 -6.74714863e-01 -1.50341094e-01
1.29823089e-01 1.81338206e-01 8.89037669e-01 6.28263593e-01
9.85571206e-01 -9.68218297e-02 -2.81684548e-01 4.66497660e-01
1.91074646e+00 5.40737808e-01 8.12906802e-01 6.91446960e-01
3.48367363e-01 7.15873778e-01 1.17666280e+00 3.62397492e-01
-2.24457741e-01 5.33021033e-01 4.74272072e-01 -4.75569069e-01
2.86823273e-01 4.71565038e-01 2.05444530e-01 3.70529562e-01
-4.85560626e-01 3.94668579e-02 -9.42307353e-01 2.96303004e-01
-1.54069602e+00 -1.25567031e+00 -6.37789190e-01 2.34136939e+00
4.15130198e-01 -2.52808481e-01 1.34959416e-02 6.85978830e-01
8.92356157e-01 -1.14979513e-01 -1.23711623e-01 -5.14834583e-01
-1.16502576e-01 1.85100257e-01 7.52235174e-01 5.77863872e-01
-1.28094852e+00 6.22129738e-01 6.04742575e+00 5.72827637e-01
-1.73621833e+00 -1.31492034e-01 5.89151859e-01 3.58782113e-01
-3.57637927e-02 3.99829224e-02 -7.01435924e-01 2.62282193e-01
1.63955986e-01 1.35232285e-01 2.61319429e-01 7.16489971e-01
7.24541903e-01 -8.98973823e-01 -3.44846487e-01 9.16033030e-01
-1.68174639e-01 -9.13507998e-01 1.65276125e-01 1.62284791e-01
5.25808632e-01 -1.23213351e-01 -1.49369150e-01 -3.07168782e-01
1.68934874e-02 -8.71275246e-01 4.46743935e-01 6.06077075e-01
3.33950192e-01 -6.30380332e-01 7.61353374e-01 3.49737018e-01
-1.43645656e+00 -1.40570432e-01 -4.55751657e-01 -3.12335849e-01
-1.13368221e-01 7.47004986e-01 -7.71286666e-01 3.17526370e-01
9.36885536e-01 2.43255243e-01 -5.95115542e-01 1.40911055e+00
1.22539200e-01 3.95718366e-01 -6.67473495e-01 -3.42729956e-01
2.34657213e-01 -8.36106837e-01 4.80783820e-01 1.01570129e+00
5.68554223e-01 4.41521913e-01 4.67584431e-01 5.85741878e-01
6.68174565e-01 6.75381482e-01 -7.33148396e-01 4.41195630e-02
3.11973989e-01 1.33873165e+00 -1.54304349e+00 -2.34787464e-01
-1.86628580e-01 8.03098142e-01 -4.25220639e-01 1.93683073e-01
-5.99966884e-01 -5.81701040e-01 1.71227902e-01 4.99639995e-02
2.91817516e-01 -4.32497889e-01 -5.95142305e-01 -7.25597918e-01
-5.56768440e-02 -6.19476438e-01 5.80305874e-01 -7.71196961e-01
-5.60271025e-01 5.03270626e-01 -8.14081263e-03 -1.39508760e+00
2.94657201e-01 -7.26102412e-01 -5.91913998e-01 7.92509317e-01
-1.62633634e+00 -9.68668759e-01 -6.67314649e-01 5.86124420e-01
5.68479061e-01 1.11661904e-01 6.13871276e-01 1.19479865e-01
-2.37715602e-01 -1.03918023e-01 3.18009347e-01 -4.51865718e-02
2.44892985e-01 -1.19988966e+00 -4.79191959e-01 1.22538841e+00
1.03760421e-01 2.75620461e-01 7.23240197e-01 -7.07742274e-01
-1.13548255e+00 -9.03272688e-01 6.65544093e-01 1.99385539e-01
2.00211406e-01 2.14230880e-01 -8.41930091e-01 2.11350530e-01
-2.07707092e-01 -1.48324639e-01 8.16461682e-01 -2.52202690e-01
3.14220816e-01 -3.05032581e-01 -1.40599036e+00 4.94424641e-01
2.90898502e-01 5.50397672e-02 -5.15336514e-01 3.46874028e-01
-1.14510283e-01 2.89961160e-03 -7.17835903e-01 2.07064182e-01
4.76652563e-01 -1.00234389e+00 8.37928891e-01 1.25205383e-01
-1.62351839e-02 -7.92974114e-01 -1.26232550e-01 -1.03557611e+00
-3.51232022e-01 -2.44792402e-01 1.01349211e+00 9.16680694e-01
6.25900149e-01 -7.81411171e-01 6.47956312e-01 4.93228108e-01
1.97875381e-01 -3.21438819e-01 -6.20336831e-01 -7.64540672e-01
-4.36952919e-01 -1.99275509e-01 3.90850484e-01 9.76128817e-01
-4.51597989e-01 -1.08449282e-02 -1.04088277e-01 4.76158261e-01
5.87842405e-01 4.30179268e-01 6.14415109e-01 -1.68493199e+00
-2.96798423e-02 -5.45002699e-01 -5.39997101e-01 -5.13015985e-01
-4.57715899e-01 -5.47258377e-01 1.06656775e-01 -1.72509491e+00
3.84160429e-02 -8.10219646e-01 7.74465967e-03 6.13776147e-01
-2.42945597e-01 6.89108491e-01 1.35292813e-01 4.06399101e-01
1.30196183e-03 9.63326171e-02 1.13285935e+00 -1.23784013e-01
-8.25019956e-01 2.98588932e-01 -2.24507421e-01 7.19112098e-01
1.02762461e+00 -5.78009188e-01 -2.58031994e-01 -1.62409663e-01
3.43300283e-01 -3.64261031e-01 3.41184735e-01 -8.24999630e-01
9.65274796e-02 -5.77724278e-01 3.82662922e-01 -7.51683652e-01
1.83084220e-01 -1.51746678e+00 4.21135485e-01 7.61389136e-01
2.44538426e-01 -4.88947965e-02 -5.30289002e-02 2.68585265e-01
-2.67931610e-01 -7.20939875e-01 1.08696437e+00 -4.11215723e-01
-1.25462019e+00 -1.07351623e-01 -7.09757090e-01 -8.06097627e-01
1.26876318e+00 -1.09818268e+00 8.51832628e-02 -2.18691394e-01
-9.01504099e-01 -2.67443117e-02 4.90959913e-01 -1.25392497e-01
5.63609540e-01 -9.10243630e-01 -6.43600047e-01 2.54922360e-01
-5.70444986e-02 -2.93237537e-01 1.45873101e-02 1.07127726e+00
-1.18214476e+00 3.07679584e-04 -3.60072225e-01 -7.15897441e-01
-1.92985404e+00 2.66944110e-01 4.68247116e-01 7.25751296e-02
-5.54247022e-01 5.73585987e-01 -3.77678335e-01 -1.50665328e-01
-1.31855518e-01 -5.07133782e-01 -6.91786408e-01 3.41740638e-01
5.42719960e-01 6.62997901e-01 2.31569856e-01 -7.42628753e-01
-4.09560472e-01 1.13172197e+00 4.27671731e-01 -1.84319809e-01
1.41146326e+00 -3.28594685e-01 -5.54445803e-01 4.56745684e-01
9.51981246e-01 7.57483765e-02 -6.88049078e-01 3.33472826e-02
9.53839794e-02 -7.92515457e-01 5.15768468e-01 -7.40063965e-01
-1.09345150e+00 7.34157264e-01 1.05168068e+00 7.17876732e-01
1.70861304e+00 -4.33336079e-01 8.45934376e-02 7.07976103e-01
2.52632529e-01 -1.32070875e+00 -4.64945465e-01 3.45325112e-01
7.64334261e-01 -1.42021132e+00 4.32223320e-01 -7.11773932e-01
-5.63133121e-01 1.50082076e+00 3.96593660e-02 1.09228879e-01
8.55396867e-01 1.02918431e-01 4.91089553e-01 -3.26013416e-01
1.66452639e-02 -7.42844284e-01 3.33421081e-01 4.53849792e-01
3.72629791e-01 6.75476789e-02 -8.76227677e-01 -2.09425345e-01
-2.29211431e-03 -1.99946120e-01 3.84137481e-01 1.05907929e+00
-1.12086630e+00 -9.35419440e-01 -1.10141397e+00 5.45514703e-01
-5.91171563e-01 -4.59400518e-03 -5.32216370e-01 6.73310697e-01
1.66959062e-01 1.04322505e+00 -4.70545851e-02 -1.73144162e-01
1.72516048e-01 2.01125946e-02 1.01679869e-01 -3.76885086e-01
-6.36118948e-01 2.94298410e-01 -2.13631004e-01 -9.88051221e-02
-9.43573952e-01 -6.57128513e-01 -1.30597079e+00 -7.88520202e-02
-8.70051503e-01 3.91025931e-01 1.21574080e+00 8.46294343e-01
-3.32009271e-02 9.78196636e-02 9.53793585e-01 -9.26633716e-01
-1.69247180e-01 -6.98086441e-01 -1.03362000e+00 1.55199423e-01
7.00756013e-02 -6.06269598e-01 -5.55544794e-01 2.23654822e-01] | [9.668492317199707, -1.7505896091461182] |
8483c22a-0795-4f26-9984-b7f8e4fef366 | learning-to-predict-3d-objects-with-an | 1908.01210 | null | https://arxiv.org/abs/1908.01210v2 | https://arxiv.org/pdf/1908.01210v2.pdf | Learning to Predict 3D Objects with an Interpolation-based Differentiable Renderer | Many machine learning models operate on images, but ignore the fact that images are 2D projections formed by 3D geometry interacting with light, in a process called rendering. Enabling ML models to understand image formation might be key for generalization. However, due to an essential rasterization step involving discrete assignment operations, rendering pipelines are non-differentiable and thus largely inaccessible to gradient-based ML techniques. In this paper, we present {\emph DIB-R}, a differentiable rendering framework which allows gradients to be analytically computed for all pixels in an image. Key to our approach is to view foreground rasterization as a weighted interpolation of local properties and background rasterization as a distance-based aggregation of global geometry. Our approach allows for accurate optimization over vertex positions, colors, normals, light directions and texture coordinates through a variety of lighting models. We showcase our approach in two ML applications: single-image 3D object prediction, and 3D textured object generation, both trained using exclusively using 2D supervision. Our project website is: https://nv-tlabs.github.io/DIB-R/ | ['Sanja Fidler', 'Jaakko Lehtinen', 'Alec Jacobson', 'Jun Gao', 'Edward J. Smith', 'Wenzheng Chen', 'Huan Ling'] | 2019-08-03 | learning-to-predict-3d-objects-with-an-1 | http://papers.nips.cc/paper/9156-learning-to-predict-3d-objects-with-an-interpolation-based-differentiable-renderer | http://papers.nips.cc/paper/9156-learning-to-predict-3d-objects-with-an-interpolation-based-differentiable-renderer.pdf | neurips-2019-12 | ['single-view-3d-reconstruction'] | ['computer-vision'] | [ 4.58674192e-01 2.28859987e-02 1.33880526e-01 -3.83897007e-01
-4.46485788e-01 -6.73836589e-01 9.60565329e-01 -8.64829421e-02
-2.78271791e-02 4.38285053e-01 -2.90909290e-01 -5.73266566e-01
5.20675898e-01 -9.52453613e-01 -1.07272005e+00 -6.71542466e-01
1.39821935e-02 5.24331033e-01 2.79732734e-01 6.30270764e-02
4.02290225e-01 1.04612339e+00 -1.80322111e+00 4.99367326e-01
5.87895453e-01 9.72416580e-01 7.64376670e-02 1.15927374e+00
-4.06200975e-01 6.89221919e-01 -3.43032181e-01 -2.43421420e-01
5.40776074e-01 -4.25547749e-01 -5.57383358e-01 3.89794052e-01
1.01011205e+00 -4.93145287e-01 5.61385266e-02 7.90317476e-01
1.99391037e-01 3.24204639e-02 8.73859584e-01 -1.19720411e+00
-5.60662925e-01 -2.60187894e-01 -7.27835536e-01 -1.50788143e-01
3.87348503e-01 3.01354557e-01 8.77777338e-01 -1.01404190e+00
7.86814690e-01 1.33303928e+00 5.48049212e-01 5.15178442e-01
-1.64136267e+00 -1.04885891e-01 2.72209972e-01 -2.35823259e-01
-1.10451806e+00 -3.33286643e-01 8.10677528e-01 -6.48106635e-01
9.53438699e-01 7.23488688e-01 9.63816762e-01 5.42377532e-01
2.47198939e-01 6.94785774e-01 1.46499085e+00 -7.08836198e-01
2.67525524e-01 2.19548613e-01 -1.71996996e-01 1.04756081e+00
-1.82290360e-01 8.22844133e-02 -7.24594593e-01 -1.24144465e-01
1.38853908e+00 -5.27505949e-02 -3.16602647e-01 -5.65203607e-01
-1.19694841e+00 6.45828009e-01 4.06683177e-01 -4.41909075e-01
-4.79487963e-02 2.27769092e-01 -1.57512620e-01 1.92530751e-01
9.09498990e-01 3.16557199e-01 -4.68762845e-01 9.75152627e-02
-7.17975199e-01 4.58951503e-01 8.80851746e-01 8.86670351e-01
1.20323241e+00 -8.79994631e-02 1.66615650e-01 5.02940714e-01
3.46635282e-01 6.57443523e-01 -1.60569087e-01 -1.41602981e+00
3.98381241e-02 5.15804768e-01 1.69922411e-01 -8.99388850e-01
-2.62188673e-01 7.88591206e-02 -5.60292423e-01 1.03484559e+00
5.40393651e-01 1.14269361e-01 -1.03891551e+00 1.15086329e+00
6.43995583e-01 1.62817851e-01 -3.28815103e-01 7.22637713e-01
7.05974162e-01 7.61956751e-01 -3.13640505e-01 -8.88085142e-02
9.08360600e-01 -8.26351047e-01 -1.65475249e-01 1.89748537e-02
5.34601212e-01 -1.06584454e+00 1.48982227e+00 6.53469622e-01
-1.43883777e+00 -3.39918852e-01 -7.58199990e-01 -5.62848151e-01
-3.64828706e-01 -5.68471104e-02 8.40131164e-01 5.81234515e-01
-1.43310678e+00 6.85501814e-01 -8.76222968e-01 -1.77371338e-01
4.78930861e-01 3.62354815e-01 8.41352865e-02 7.64378309e-02
-3.66129637e-01 7.39344656e-01 -1.53418347e-01 5.32331616e-02
-6.17999852e-01 -8.84582996e-01 -6.07787967e-01 -4.10068959e-01
-2.44222060e-02 -8.83858263e-01 1.14046824e+00 -9.30823803e-01
-1.69920778e+00 1.41146684e+00 -4.65348452e-01 -5.30449934e-02
7.53953099e-01 3.12091429e-02 3.39806199e-01 9.66580808e-02
-1.83027774e-01 9.32819963e-01 9.75220680e-01 -1.70656526e+00
-5.44344127e-01 -3.79304737e-01 2.66792685e-01 5.95897853e-01
1.84803292e-01 -1.60255566e-01 -4.61479634e-01 -2.62947112e-01
2.31871188e-01 -7.34038353e-01 -2.73621768e-01 6.36736572e-01
-5.34419417e-01 1.32189253e-02 9.23171818e-01 -5.63360214e-01
5.48763931e-01 -1.85367763e+00 1.03440374e-01 2.03582868e-01
3.42510641e-01 -1.80371210e-01 1.04925349e-01 1.90090820e-01
1.07290477e-01 1.65749192e-01 -3.02422404e-01 -5.86453557e-01
1.82980932e-02 1.13084778e-01 -4.37598497e-01 5.32373250e-01
4.37994093e-01 8.11341882e-01 -7.79297531e-01 -5.02240062e-01
7.48407841e-01 7.68418312e-01 -7.04972863e-01 2.82638103e-01
-8.14770520e-01 7.18357384e-01 -2.75079399e-01 7.03652620e-01
8.70152712e-01 -3.46298784e-01 -6.68048635e-02 -1.66433513e-01
-5.04945397e-01 2.94381499e-01 -1.01322818e+00 1.63094413e+00
-6.00506246e-01 8.48483860e-01 1.36689261e-01 -5.93995452e-01
7.37290919e-01 -1.58825397e-01 4.31390822e-01 -5.10228157e-01
-6.51301742e-02 1.05961472e-01 -4.82981175e-01 -1.04678452e-01
3.58843088e-01 1.31758586e-01 5.12618303e-01 7.07389534e-01
-4.48748082e-01 -8.84529948e-01 -4.65262234e-02 2.74065644e-01
7.92881608e-01 7.40095496e-01 -9.41545814e-02 -3.32141340e-01
1.36818230e-01 2.15247452e-01 6.08325861e-02 6.88202620e-01
4.28578556e-01 9.53139305e-01 6.32780731e-01 -7.55160749e-01
-1.31390572e+00 -1.20272303e+00 -5.14595330e-01 9.58246708e-01
2.41161913e-01 -1.68976471e-01 -7.76363015e-01 -2.68739522e-01
-1.54378349e-02 8.59086633e-01 -6.64202690e-01 3.84455562e-01
-7.29515553e-01 -1.05695498e+00 -4.06762073e-03 1.01145536e-01
1.16440848e-01 -9.47985649e-01 -8.88486862e-01 -7.15908334e-02
1.91653877e-01 -7.99870789e-01 -3.64173502e-01 2.30861485e-01
-9.85499740e-01 -8.67079377e-01 -5.64794421e-01 -4.23472583e-01
9.23740506e-01 3.45470220e-01 1.41492867e+00 3.90905648e-01
-7.83909619e-01 6.85866654e-01 4.50128391e-02 -4.06149983e-01
-3.63427252e-01 -3.42398614e-01 -3.58583689e-01 1.08788200e-01
8.71753916e-02 -6.67731106e-01 -8.02251816e-01 1.81140527e-01
-7.59618640e-01 8.23103607e-01 1.47434086e-01 3.54635268e-01
1.00182462e+00 -4.34208900e-01 -5.19123912e-01 -9.25857663e-01
1.83179379e-01 -3.92707512e-02 -9.39402759e-01 2.71727890e-01
-3.32049906e-01 -1.03598699e-01 2.85301983e-01 -3.25606853e-01
-1.15820444e+00 1.63388401e-01 6.88263625e-02 -3.62884134e-01
-3.63756806e-01 -1.86513141e-01 -7.34732077e-02 -2.75467128e-01
6.46280169e-01 2.63818085e-01 2.08850149e-02 -4.26314235e-01
6.38706565e-01 2.43625522e-01 2.09940016e-01 -8.26632380e-01
7.22624898e-01 1.01937866e+00 3.51611793e-01 -1.20725453e+00
-6.26657963e-01 -4.49869968e-03 -9.67873096e-01 -4.51564729e-01
9.33144808e-01 -5.41637897e-01 -8.23539674e-01 5.75431526e-01
-1.40966713e+00 -1.07091355e+00 -4.18526858e-01 5.29508516e-02
-7.31792390e-01 1.37014270e-01 -5.99031627e-01 -8.67092311e-01
-1.31465029e-03 -1.29962814e+00 1.57771003e+00 1.44952506e-01
2.30657384e-02 -1.16829526e+00 -2.10656449e-01 2.05317810e-01
2.92892665e-01 4.46955293e-01 1.06834054e+00 2.41571113e-01
-1.29794168e+00 1.20207325e-01 -5.47362387e-01 1.58552110e-01
1.16569713e-01 6.36837304e-01 -1.23019028e+00 1.89303726e-01
-1.40920714e-01 -3.17548364e-01 6.10904276e-01 5.72785616e-01
1.39917159e+00 -2.01809436e-01 -2.85133809e-01 1.15060639e+00
1.48762560e+00 -9.24241617e-02 5.65203965e-01 2.34873846e-01
1.15015745e+00 6.74213886e-01 2.51663655e-01 4.25000072e-01
4.03894007e-01 7.07502842e-01 5.24907410e-01 -5.82168579e-01
-5.09373307e-01 -1.23502187e-01 1.45030782e-01 4.98224705e-01
-4.79685992e-01 -1.96230277e-01 -9.13642168e-01 -1.62834913e-01
-1.54100335e+00 -6.96795762e-01 -7.52391219e-01 2.46800804e+00
8.53449941e-01 3.70302461e-02 -3.95570062e-02 -3.56373459e-01
3.24032068e-01 6.77526966e-02 -6.51290238e-01 -4.90432143e-01
-2.72075385e-01 3.13895583e-01 5.99230647e-01 1.06691110e+00
-8.04205060e-01 9.61136043e-01 6.49648380e+00 5.53272605e-01
-1.33395827e+00 -3.20210978e-02 1.05399656e+00 -3.28593999e-01
-8.24393690e-01 1.68220788e-01 -8.59478652e-01 2.49563619e-01
3.81866127e-01 3.03807467e-01 7.96699822e-01 4.47230697e-01
3.47811401e-01 -3.83359790e-01 -1.18640709e+00 8.24388146e-01
9.27595347e-02 -1.48434031e+00 2.21497923e-01 2.96929866e-01
8.34116340e-01 1.34326860e-01 2.58172601e-01 -4.25460696e-01
4.54169661e-01 -9.43082750e-01 9.39277709e-01 5.94007552e-01
1.06271875e+00 -3.47500980e-01 -1.90068364e-01 2.43277401e-01
-8.05007219e-01 5.92490435e-01 -3.56214464e-01 -1.33025125e-01
2.07773954e-01 7.97213435e-01 -8.25345695e-01 2.12833360e-02
6.98880196e-01 5.62501073e-01 -5.39422989e-01 5.42035341e-01
-1.70703381e-01 3.21651667e-01 -6.35132909e-01 7.98907205e-02
8.42987224e-02 -6.78433955e-01 3.46042156e-01 1.14304245e+00
1.57577202e-01 7.18428269e-02 2.00462744e-01 1.11817110e+00
1.07506737e-01 1.42826214e-01 -6.20462835e-01 3.76108706e-01
5.45054860e-02 1.22608519e+00 -1.18343329e+00 -2.69982368e-01
-4.86154079e-01 1.10235202e+00 3.51376891e-01 6.63459480e-01
-7.73330688e-01 4.61641699e-02 5.95997453e-01 5.12947321e-01
7.04034343e-02 -4.21532959e-01 -6.09663069e-01 -1.25529003e+00
3.52696925e-02 -6.38492048e-01 -3.11210841e-01 -1.13573980e+00
-1.05817664e+00 4.01328206e-01 3.60992551e-02 -9.71127868e-01
3.92088592e-02 -1.01807559e+00 -5.65443277e-01 1.01239002e+00
-1.56498945e+00 -1.34974134e+00 -3.84461552e-01 5.68374217e-01
4.55534279e-01 3.47977251e-01 7.00320184e-01 6.01783860e-03
-2.53144503e-01 -6.25548419e-03 1.20901302e-01 -3.03321987e-01
4.57769960e-01 -1.55147958e+00 6.34961665e-01 6.06944859e-01
3.58198673e-01 4.63015705e-01 6.22405052e-01 -5.45416832e-01
-1.55925417e+00 -8.15356672e-01 5.18264890e-01 -8.39761257e-01
4.05516148e-01 -6.76211536e-01 -7.57515609e-01 8.18622291e-01
1.31168947e-01 1.03928395e-01 3.09217483e-01 -9.03080329e-02
-1.96986213e-01 9.11531448e-02 -9.09078717e-01 8.90121341e-01
1.24611676e+00 -4.44085151e-01 1.59468725e-01 7.72200584e-01
3.53655189e-01 -7.84270108e-01 -5.86049139e-01 3.73228751e-02
5.91841042e-01 -1.38906431e+00 1.19754493e+00 -3.39700609e-01
4.76021856e-01 -4.14616883e-01 -2.06797764e-01 -1.02022862e+00
8.45583454e-02 -6.52702332e-01 -2.25214493e-02 7.97387540e-01
3.46526831e-01 -7.30820000e-01 8.19878876e-01 8.83139908e-01
-1.76102251e-01 -7.67312944e-01 -6.36581898e-01 -2.96446294e-01
1.42187485e-02 -5.81340849e-01 4.25274372e-01 8.08812201e-01
-6.78227127e-01 3.77896801e-02 -1.25728250e-01 2.01445669e-01
9.64934886e-01 4.69816178e-01 1.03768945e+00 -1.21224296e+00
-4.24349248e-01 -8.01421881e-01 -7.91927576e-02 -1.45517004e+00
-1.43099889e-01 -9.01555181e-01 -2.54068611e-04 -1.51652122e+00
2.74244212e-02 -9.26907420e-01 4.09392327e-01 2.60828257e-01
4.54145074e-02 4.89895284e-01 6.88371584e-02 3.38762045e-01
-2.00329244e-01 1.37706265e-01 1.49653053e+00 3.79926823e-02
-2.58281261e-01 1.41431009e-02 -1.98517039e-01 9.69186783e-01
8.22607875e-01 -2.15316206e-01 -3.55254650e-01 -8.34310412e-01
3.09364378e-01 -1.99846625e-01 7.34278619e-01 -4.55006570e-01
-5.63803390e-02 -4.39938009e-01 7.13013470e-01 -6.23844743e-01
7.58547902e-01 -4.29366440e-01 1.68920025e-01 -8.89608357e-03
-2.23175287e-01 -1.16716877e-01 8.04985389e-02 1.37483567e-01
3.78862113e-01 -1.74246237e-01 6.41713321e-01 -5.92151880e-01
-3.67593110e-01 4.28419560e-01 -1.39967367e-01 -1.16206452e-01
7.26526022e-01 -4.50385064e-01 -1.24959797e-01 -2.15850756e-01
-6.08396113e-01 -2.53860176e-01 1.18327212e+00 -1.24699838e-01
5.30891120e-01 -1.07371497e+00 -5.52261472e-01 4.57760155e-01
-3.09455663e-01 5.21298587e-01 5.93405701e-02 6.16313159e-01
-1.23014641e+00 7.69009888e-02 8.62506628e-02 -1.00740659e+00
-1.26992249e+00 2.47721583e-01 4.60631132e-01 1.50551528e-01
-8.06169689e-01 1.13147485e+00 6.92665696e-01 -4.77332979e-01
1.15352437e-01 -5.51292002e-01 6.01303399e-01 -2.74293274e-01
4.45271015e-01 2.36106843e-01 6.03248887e-02 -5.01602173e-01
-2.96090655e-02 8.61139953e-01 1.56724900e-01 -3.06993216e-01
1.15563726e+00 -1.55172795e-01 -4.56707090e-01 6.62815034e-01
1.12921596e+00 4.82725911e-02 -1.74684298e+00 1.07231416e-01
-5.02623737e-01 -8.55728269e-01 2.78861463e-01 -6.44356549e-01
-9.71641719e-01 1.24103904e+00 3.90974700e-01 3.54618758e-01
1.01725841e+00 6.30224198e-02 4.74059165e-01 1.46068260e-01
2.99522132e-01 -7.82147944e-01 -1.97993312e-02 1.96792871e-01
8.66001427e-01 -1.21977818e+00 2.20921189e-01 -6.82705104e-01
-2.68558085e-01 1.24716163e+00 5.20442188e-01 -5.35715260e-02
6.84421122e-01 5.26732147e-01 3.37907910e-01 -1.98535323e-01
-6.16279304e-01 -2.06116978e-02 3.63427311e-01 6.19328558e-01
6.86022222e-01 6.73340186e-02 2.36045450e-01 -6.41254008e-01
-1.56142429e-01 -3.02109033e-01 5.00534654e-01 9.31029379e-01
-3.19771886e-01 -1.26931322e+00 -4.36888933e-01 5.50395072e-01
-1.26425818e-01 -2.12755576e-01 -2.67688155e-01 6.46781862e-01
1.81904197e-01 3.59080672e-01 4.89118993e-01 1.63561225e-01
7.89802596e-02 -6.03209585e-02 1.03938639e+00 -7.09542453e-01
-8.20006356e-02 3.34932208e-01 -2.41249576e-01 -6.36263967e-01
-4.58619207e-01 -8.69967639e-01 -1.14761901e+00 -3.50579619e-01
-1.16227634e-01 -3.82694662e-01 1.00614905e+00 5.38127422e-01
2.22280651e-01 1.64882839e-01 5.97418249e-01 -1.64651275e+00
2.98307203e-02 -3.97237062e-01 -5.85861444e-01 2.91694313e-01
4.52749819e-01 -4.63565230e-01 -5.14724076e-01 5.68857193e-01] | [9.28954029083252, -3.146739959716797] |
d6eb185d-b761-4d86-8a00-b6381bf94ecc | 190409409 | 1904.09409 | null | http://arxiv.org/abs/1904.09409v1 | http://arxiv.org/pdf/1904.09409v1.pdf | Funnel Transform for Straight Line Detection | Most of the classical approaches to straight line detection only deal with a
binary edge image and need to use 2D interpolation operation. This paper
proposes a new transform method figuratively named as funnel transform which
can efficiently and rapidly detect straight lines. The funnel transform
consists of three 1D Fourier transforms and one nonlinear variable-metric
transform (NVMT). It only needs to exploit 1D interpolation operation for
achieving its NVMT, and can directly handle grayscale images by using its
high-pass filter property, which significantly improves the performance of the
closely-related approaches. Based on the slope-intercept line equation, the
funnel transform can more uniformly turn the straight lines formed by
ridge-typical and step-typical edges into the local maximum points (peaks). The
parameters of each line can be uniquely extracted from its corresponding peak
coordinates. Additionally, each peak can be theoretically specified by a 2D
delta function, which makes the peaks and lines more easily identified and
detected, respectively. Theoretical analysis and experimental results
demonstrate that the funnel transform has advantages including smaller
computational complexity, lower hardware cost, higher detection probability,
greater location precision, better parallelization properties, stronger
anti-occlusion and noise robustness. | ['Da-Zheng Feng', 'Weixing Zheng', 'QianRu Wei'] | 2019-04-20 | null | null | null | null | ['line-detection'] | ['computer-vision'] | [-5.06316386e-02 -6.55991912e-01 -4.31784302e-01 3.78070399e-02
-2.72183776e-01 -4.56265271e-01 1.08419329e-01 -2.01274175e-02
-2.95084089e-01 4.57030982e-01 -2.25266948e-01 -5.42202890e-01
-6.28923848e-02 -8.71010244e-01 -3.80381495e-01 -6.53347909e-01
-1.52531281e-01 -1.32297873e-01 6.89511716e-01 -4.67765443e-02
5.03624380e-01 7.05412149e-01 -1.05752039e+00 -1.60273314e-01
1.17835855e+00 1.08220029e+00 1.14996498e-02 4.61312413e-01
-3.59879166e-01 1.53951019e-01 -3.28427345e-01 -8.65391046e-02
3.59686702e-01 -3.47517967e-01 -1.37558624e-01 1.58597812e-01
9.97587442e-02 -5.42728424e-01 -3.73069167e-01 1.18122017e+00
3.75009447e-01 -5.62271439e-02 6.32412910e-01 -1.20033014e+00
-6.53068483e-01 -4.22444977e-02 -1.51981509e+00 2.32880220e-01
5.02267301e-01 -3.27590182e-02 7.23409235e-01 -9.83716369e-01
4.43331033e-01 1.08736622e+00 1.15277827e+00 -1.21216871e-01
-1.01886880e+00 -6.51846707e-01 -3.09918851e-01 3.77856135e-01
-1.57504213e+00 6.69086725e-02 8.58564079e-01 -1.09046020e-01
4.80147153e-01 5.44763565e-01 6.52134359e-01 5.71833961e-02
1.03467673e-01 8.89452875e-01 1.06153357e+00 -5.57369590e-01
-2.74262697e-01 -1.11223854e-01 9.98930186e-02 9.72147346e-01
4.21690732e-01 -1.29517447e-03 -6.97798952e-02 3.51539440e-02
1.35578418e+00 6.20176271e-02 -7.51724184e-01 -4.05743271e-01
-1.24578512e+00 5.60554147e-01 6.79313719e-01 4.39629763e-01
-1.98153734e-01 -1.76749289e-01 2.89858639e-01 -4.18276452e-02
-1.54853463e-01 -1.54035494e-01 -5.02302870e-02 3.11548728e-02
-1.05944157e+00 -1.30250081e-01 4.89899725e-01 1.00338447e+00
9.06084895e-01 6.11143894e-02 -1.21104848e-02 5.95060587e-01
2.86523283e-01 7.68638134e-01 6.33497894e-01 -4.56140816e-01
5.20994365e-01 7.96288610e-01 2.21399471e-01 -1.32318187e+00
-7.64551520e-01 -4.36901927e-01 -8.50866914e-01 3.85227352e-01
4.06701326e-01 -1.71174467e-01 -6.67847335e-01 8.81746709e-01
3.98314029e-01 2.25629911e-01 -1.94908455e-01 8.84312272e-01
7.22765982e-01 8.65856349e-01 -5.33637047e-01 -2.77407467e-01
1.60226297e+00 -8.36697638e-01 -7.26418853e-01 -9.83970426e-03
4.51000392e-01 -1.15463221e+00 1.02878642e+00 2.13312805e-01
-8.32783222e-01 -5.52016616e-01 -1.43622577e+00 -2.01186702e-01
-1.89611137e-01 7.18066335e-01 5.44958889e-01 6.58021033e-01
-6.78984165e-01 2.70835668e-01 -6.26738429e-01 -7.38168582e-02
2.57172942e-01 2.39562631e-01 -2.22165868e-01 1.93925083e-01
-8.15577567e-01 4.90936160e-01 3.84930491e-01 2.85501331e-01
4.25312668e-01 -4.78796661e-01 -8.01203907e-01 2.22871631e-01
2.97291189e-01 -4.27750260e-01 8.90148163e-01 -5.67494392e-01
-1.39028132e+00 6.82728112e-01 -4.81337994e-01 -2.58415371e-01
7.49575198e-01 -1.56499390e-02 -6.60021126e-01 4.53922749e-01
7.96994641e-02 6.83664158e-02 8.44431758e-01 -1.01217675e+00
-1.00497663e+00 -3.05382609e-01 -5.41388333e-01 1.61438659e-01
-3.24382484e-01 -7.09729195e-02 -5.29254913e-01 -5.90490937e-01
6.03672981e-01 -4.86490786e-01 -6.90508708e-02 3.73479396e-01
-4.67093468e-01 -1.24979310e-01 1.31645548e+00 -6.62602127e-01
1.63046753e+00 -2.38245130e+00 -5.32581866e-01 6.52095795e-01
1.56434283e-01 5.07667243e-01 3.07899565e-01 4.22971398e-01
4.88934165e-04 -1.05377100e-01 -1.23004347e-01 3.93738240e-01
-3.57219577e-01 -2.44442508e-01 -2.38241389e-01 6.53472006e-01
2.92024249e-03 7.63610303e-01 -6.79246247e-01 -6.43142998e-01
3.80163401e-01 4.07378942e-01 -5.34047708e-02 -3.07800680e-01
5.31476498e-01 -3.78852375e-02 -5.93320787e-01 4.34169263e-01
1.27123475e+00 -1.27047285e-01 -2.39561811e-01 -6.00847840e-01
-7.20066726e-01 -3.06612343e-01 -1.60357296e+00 8.92423511e-01
-2.12184072e-01 7.43623614e-01 -1.79270208e-02 -6.44063413e-01
1.67868364e+00 7.31278732e-02 3.57762814e-01 -8.17860425e-01
1.40919119e-01 5.28155327e-01 -2.73729146e-01 -5.88182509e-01
2.93858737e-01 1.04231350e-01 2.23796681e-01 6.09916933e-02
-5.37862718e-01 -4.67468873e-02 2.18428239e-01 -1.68308049e-01
3.95987451e-01 6.44101650e-02 6.47240281e-01 -3.26318771e-01
9.56220686e-01 3.69998217e-02 7.29897320e-01 3.41644555e-01
-1.35560364e-01 4.44321036e-01 4.59244549e-01 -2.73088425e-01
-7.95466065e-01 -1.03165197e+00 -4.34795469e-01 2.81616330e-01
7.95916080e-01 -1.42244890e-01 -6.64554894e-01 -2.83391178e-01
8.61480609e-02 3.27257007e-01 -2.54031152e-01 -6.71997815e-02
-1.01126301e+00 -4.44549441e-01 3.26305121e-01 7.12476671e-01
1.19858110e+00 -4.54939127e-01 -7.75639653e-01 1.69804394e-01
1.60486773e-02 -9.11450386e-01 -8.59905541e-01 -1.95107311e-01
-9.73270714e-01 -1.20779860e+00 -1.08242822e+00 -1.24331200e+00
8.44619513e-01 7.58390903e-01 3.51876140e-01 1.34004608e-01
-4.74460334e-01 -1.75360948e-01 -1.43986166e-01 -3.66740599e-02
7.96686187e-02 -3.63169104e-01 -3.68926942e-01 1.52765244e-01
3.14377040e-01 -3.32044274e-01 -8.77961040e-01 7.53714263e-01
-5.16238987e-01 3.41210663e-02 6.71294272e-01 9.30520952e-01
5.22779822e-01 4.54484910e-01 3.68938953e-01 -3.91221642e-01
6.56704009e-01 2.35187978e-01 -9.70685303e-01 3.19330901e-01
-4.69753236e-01 6.78228028e-03 8.90119731e-01 -3.48371655e-01
-1.10800338e+00 9.37886313e-02 2.73526311e-02 -1.49936050e-01
2.18578845e-01 3.10861826e-01 1.59831762e-01 -4.86297429e-01
6.45254076e-01 6.30560875e-01 -9.27669462e-03 -3.46717924e-01
3.15334111e-01 7.45181739e-01 1.08246946e+00 -2.33424485e-01
1.10015702e+00 6.74360514e-01 4.52001423e-01 -1.17078650e+00
-1.66335478e-01 -6.91848516e-01 -5.91432989e-01 -3.48446071e-01
5.68922341e-01 -5.51013887e-01 -1.03518927e+00 6.13306046e-01
-9.64056015e-01 2.48187646e-01 1.11260869e-01 5.97231805e-01
-2.89093971e-01 9.90257502e-01 -7.28534698e-01 -8.39226365e-01
-5.05418897e-01 -8.92823100e-01 6.75150096e-01 9.02889371e-01
1.35678530e-01 -1.04526627e+00 -4.36012357e-01 -3.88881087e-01
1.86407432e-01 4.86761808e-01 1.00265110e+00 2.97347046e-02
-5.52790940e-01 -6.61073208e-01 -6.93937361e-01 -4.99553047e-02
2.34671712e-01 3.82707000e-01 -4.29076254e-01 -1.89400256e-01
2.99928505e-02 3.90491426e-01 6.68547451e-01 4.75937515e-01
7.61435807e-01 -8.32757577e-02 -6.95809186e-01 8.49535286e-01
1.62232876e+00 5.41887879e-01 7.50510752e-01 5.57853162e-01
4.60887134e-01 1.24218874e-01 7.91085303e-01 3.13757479e-01
1.41904578e-01 6.10849917e-01 4.03836966e-02 -5.75375497e-01
-1.84210598e-01 -3.42720389e-01 5.29489145e-02 7.42169380e-01
-1.81834713e-01 9.21057090e-02 -6.98826671e-01 2.51415968e-01
-1.67110837e+00 -8.14512014e-01 -1.01122129e+00 2.41170335e+00
5.73284090e-01 2.59857625e-01 2.33758867e-01 6.05560958e-01
1.11179781e+00 -1.02984548e-01 -3.60424459e-01 -3.23558748e-01
-1.60402089e-01 -3.40146087e-02 8.49414408e-01 3.82612467e-01
-1.03570175e+00 5.00064909e-01 6.23399305e+00 1.16223526e+00
-1.38049960e+00 -5.35420060e-01 3.08777750e-01 7.86782384e-01
-1.27653629e-01 1.08692450e-02 -9.39355254e-01 5.29744029e-01
-1.50716484e-01 -4.30249214e-01 -2.36557201e-02 7.34042883e-01
3.33285779e-01 -3.39391321e-01 -4.04570341e-01 1.18640375e+00
-3.45706165e-01 -1.09875596e+00 -1.74248323e-01 -1.76397637e-01
5.41494370e-01 -6.52532458e-01 1.87108293e-01 -1.65296912e-01
-3.79222780e-01 -6.09842062e-01 4.98323709e-01 4.28012371e-01
9.59806800e-01 -8.36534977e-01 5.72155058e-01 3.34848195e-01
-1.74106562e+00 -6.93572611e-02 -3.99437875e-01 9.59888920e-02
2.08522320e-01 6.10915422e-01 -7.78354466e-01 8.76787364e-01
4.10427451e-01 5.59505701e-01 -2.55872548e-01 1.66349065e+00
-2.96262443e-01 2.91271001e-01 -6.63000464e-01 -2.74074435e-01
2.80001849e-01 -7.84610510e-01 4.82933193e-01 1.23374343e+00
6.28782630e-01 -3.09207942e-02 2.33161762e-01 5.11157095e-01
3.61710042e-01 5.98400652e-01 -1.29046008e-01 5.15477657e-01
7.07445264e-01 1.10235894e+00 -1.12363863e+00 -1.71544984e-01
-7.57999539e-01 9.67602491e-01 -1.65346608e-01 3.77091259e-01
-8.99665177e-01 -1.25563920e+00 -2.15782505e-02 1.47200689e-01
5.49732149e-01 -4.65260416e-01 -6.51615739e-01 -8.01487088e-01
2.62381077e-01 -4.56671923e-01 4.36135828e-01 -6.62979066e-01
-7.49166310e-01 3.76147270e-01 -2.41699368e-01 -1.65813243e+00
7.34021738e-02 -6.76065207e-01 -9.90045846e-01 9.72693563e-01
-1.59914565e+00 -9.74135637e-01 -5.82848072e-01 7.57898688e-01
1.81989491e-01 2.96278626e-01 4.35358137e-01 2.63035178e-01
-6.73173666e-01 7.57964849e-01 5.38771808e-01 3.04037124e-01
5.69823086e-01 -1.00073254e+00 3.25094193e-01 1.02902377e+00
-1.86201692e-01 5.03922284e-01 5.47761738e-01 -4.83735800e-01
-1.15917385e+00 -5.07920861e-01 7.02016830e-01 5.49226522e-01
5.74830294e-01 8.18365589e-02 -1.03099084e+00 3.07987124e-01
-2.51731664e-01 -4.43812273e-03 2.02350289e-01 -4.24424291e-01
-1.95942968e-01 -4.04079616e-01 -1.21228051e+00 7.18966186e-01
8.08921278e-01 -1.74567878e-01 -4.68231529e-01 7.87853152e-02
1.54941961e-01 -6.35815501e-01 -6.80687666e-01 4.69945073e-01
5.43475866e-01 -1.23076248e+00 1.16198254e+00 2.61458516e-01
-1.00081302e-02 -8.20296049e-01 4.97695476e-01 -7.65006244e-01
-5.24237573e-01 -7.25783169e-01 1.87342063e-01 1.14975035e+00
3.45076889e-01 -1.07958627e+00 6.03547812e-01 1.29479170e-01
4.36027581e-03 -8.52229655e-01 -9.52125430e-01 -9.76813614e-01
-4.74486738e-01 1.15945034e-01 5.82399845e-01 7.59538710e-01
1.56700611e-01 8.73395056e-02 -2.90228307e-01 2.79741168e-01
5.77232480e-01 5.36864221e-01 7.22843647e-01 -1.08547783e+00
-7.70159140e-02 -7.52206743e-01 -7.08455920e-01 -1.77067256e+00
-6.88108861e-01 -5.31155050e-01 -2.46100456e-01 -1.58558083e+00
-2.99751669e-01 -5.64144254e-01 9.91692320e-02 3.10588688e-01
-2.20591545e-01 3.59629959e-01 -1.29229613e-02 3.92712146e-01
3.20453197e-02 3.21514249e-01 1.54923320e+00 2.44438678e-01
-4.28696692e-01 3.24600786e-01 -2.22726673e-01 9.20343816e-01
6.80691004e-01 5.74810244e-02 -2.14470297e-01 -1.80619285e-01
-1.49778098e-01 2.04894334e-01 8.79754275e-02 -1.08833158e+00
3.88168007e-01 -5.15560992e-02 7.28669465e-01 -8.11159015e-01
9.08140540e-02 -7.82147050e-01 -8.98492411e-02 8.17961991e-01
3.94496202e-01 3.66290025e-02 2.78385311e-01 3.60525310e-01
-2.60537028e-01 -6.91196442e-01 8.22059155e-01 3.85111213e-01
-1.00457823e+00 7.33798891e-02 -1.82159930e-01 -3.99466068e-01
1.42292154e+00 -9.37145948e-01 -2.91589171e-01 -4.08351690e-01
-2.92172492e-01 2.95200080e-01 4.89399999e-01 -3.68007272e-02
7.62745559e-01 -1.45304227e+00 -5.48644006e-01 5.90564191e-01
-2.63822854e-01 -7.85429776e-02 2.55190909e-01 9.56237078e-01
-1.21962392e+00 4.16052461e-01 -1.21169560e-01 -7.09772825e-01
-1.38788390e+00 5.99756956e-01 3.30550134e-01 -2.70383954e-02
-9.15985346e-01 6.10215306e-01 9.59105715e-02 3.11861247e-01
-1.28728926e-01 -3.76869321e-01 -2.53940046e-01 -1.09669827e-01
6.40121400e-01 8.40655923e-01 -3.37953627e-01 -6.61531687e-01
-2.82249749e-01 1.41440654e+00 1.45160690e-01 2.00650170e-01
8.43188167e-01 -1.87332667e-02 -1.34495934e-02 1.08353384e-02
1.32680523e+00 6.21690035e-01 -1.20617545e+00 -1.86884761e-01
-6.33585304e-02 -8.78260911e-01 -2.69577303e-03 -3.37391973e-01
-9.61394250e-01 7.61648595e-01 4.73984718e-01 5.29587567e-01
1.42716765e+00 -4.58026975e-01 1.12260401e+00 -8.36253539e-02
2.48432711e-01 -8.03871751e-01 -1.49510801e-01 6.58445284e-02
6.44726634e-01 -6.70070648e-01 1.95667759e-01 -1.15094340e+00
-1.85919836e-01 1.85919893e+00 4.09988970e-01 -2.38664925e-01
3.53812188e-01 4.28887218e-01 2.13799458e-02 3.27096134e-01
2.81511456e-01 -1.86181515e-01 3.60939622e-01 5.88297904e-01
1.41035497e-01 -8.44796002e-02 -8.16513419e-01 1.98471874e-01
-6.17028177e-02 4.04218072e-03 3.82880300e-01 7.95069277e-01
-8.66626441e-01 -9.86087859e-01 -8.13472331e-01 3.50005776e-01
-1.31018758e-01 1.30122900e-01 1.63631901e-01 1.05622983e+00
-1.44621328e-01 6.96104825e-01 1.79436371e-01 -2.09791392e-01
5.87596536e-01 -3.62715840e-01 3.14440250e-01 2.58602142e-01
-2.60067433e-02 3.95736575e-01 -2.40484893e-01 -1.96697369e-01
-8.68637189e-02 -5.27193189e-01 -1.78900349e+00 -3.97670478e-01
-7.49230444e-01 1.04371682e-01 5.27711749e-01 5.28038561e-01
1.83738619e-01 1.98547229e-01 7.97921419e-01 -3.31605762e-01
-4.38966572e-01 -4.40273851e-01 -5.31909525e-01 1.95060417e-01
2.51868516e-01 -5.28166234e-01 -3.87215704e-01 -1.56305298e-01] | [8.511765480041504, -1.5828051567077637] |
a5973e15-8400-412c-8965-36cba2bbf355 | lstm-pose-machines | 1712.06316 | null | http://arxiv.org/abs/1712.06316v4 | http://arxiv.org/pdf/1712.06316v4.pdf | LSTM Pose Machines | We observed that recent state-of-the-art results on single image human pose
estimation were achieved by multi-stage Convolution Neural Networks (CNN).
Notwithstanding the superior performance on static images, the application of
these models on videos is not only computationally intensive, it also suffers
from performance degeneration and flicking. Such suboptimal results are mainly
attributed to the inability of imposing sequential geometric consistency,
handling severe image quality degradation (e.g. motion blur and occlusion) as
well as the inability of capturing the temporal correlation among video frames.
In this paper, we proposed a novel recurrent network to tackle these problems.
We showed that if we were to impose the weight sharing scheme to the
multi-stage CNN, it could be re-written as a Recurrent Neural Network (RNN).
This property decouples the relationship among multiple network stages and
results in significantly faster speed in invoking the network for videos. It
also enables the adoption of Long Short-Term Memory (LSTM) units between video
frames. We found such memory augmented RNN is very effective in imposing
geometric consistency among frames. It also well handles input quality
degradation in videos while successfully stabilizes the sequential outputs. The
experiments showed that our approach significantly outperformed current
state-of-the-art methods on two large-scale video pose estimation benchmarks.
We also explored the memory cells inside the LSTM and provided insights on why
such mechanism would benefit the prediction for video-based pose estimations. | ['Zhouxia Wang', 'Liang Lin', 'Jimmy Ren', 'Jianbo Liu', 'Jiahao Pang', 'Yue Luo', 'Wenxiu Sun', 'Jinshan Pan'] | 2017-12-18 | lstm-pose-machines-1 | http://openaccess.thecvf.com/content_cvpr_2018/html/Luo_LSTM_Pose_Machines_CVPR_2018_paper.html | http://openaccess.thecvf.com/content_cvpr_2018/papers/Luo_LSTM_Pose_Machines_CVPR_2018_paper.pdf | cvpr-2018-6 | ['2d-human-pose-estimation'] | ['computer-vision'] | [ 1.78394347e-01 -6.06994294e-02 2.98761087e-03 -2.86146533e-02
-4.09320503e-01 -2.19578326e-01 3.32252532e-01 -6.23533487e-01
-6.59306765e-01 5.43620110e-01 2.15785712e-01 -7.80163929e-02
9.30703804e-02 -4.32329357e-01 -1.29256964e+00 -6.53150022e-01
-2.21047938e-01 1.71412408e-01 3.54691982e-01 -1.23067118e-01
4.45526429e-02 5.74240029e-01 -1.67998552e+00 4.03982878e-01
2.66865999e-01 1.04250360e+00 4.42406237e-01 8.29216301e-01
3.45085591e-01 1.09463739e+00 -7.10851729e-01 -3.27191919e-01
2.39162192e-01 -7.90690929e-02 -7.10712731e-01 8.75611827e-02
8.13628495e-01 -8.47193360e-01 -7.73081660e-01 6.33315980e-01
6.45807981e-01 3.66482198e-01 1.24135479e-01 -1.12807572e+00
-5.77982008e-01 3.60466331e-01 -5.60595691e-01 1.95387617e-01
3.62041682e-01 2.33962327e-01 6.59268081e-01 -9.23961520e-01
7.96244442e-01 1.24971211e+00 8.68331969e-01 7.17462480e-01
-8.31142604e-01 -4.63171780e-01 3.09742421e-01 4.07026172e-01
-1.12526977e+00 -4.58067179e-01 4.67503041e-01 -2.30139136e-01
1.39067698e+00 1.82032585e-01 8.62984359e-01 1.42634547e+00
4.77245122e-01 9.81993556e-01 5.60447454e-01 -2.13738188e-01
-1.39240086e-01 -2.48248488e-01 -2.15070695e-01 7.24347651e-01
2.46456880e-02 3.33380461e-01 -6.58405066e-01 2.80725330e-01
1.14750206e+00 1.68256834e-01 -4.30122763e-01 -2.93637931e-01
-1.24783289e+00 4.70894784e-01 7.55818248e-01 3.49865407e-01
-4.03362066e-01 6.69294715e-01 6.97352588e-01 2.39187017e-01
2.72715241e-01 3.23445797e-01 -5.49519718e-01 -2.02605173e-01
-1.12645197e+00 -3.13872402e-03 4.11671281e-01 7.87720799e-01
2.86036670e-01 3.52296203e-01 -1.49609804e-01 6.01925850e-01
1.16327085e-01 3.56013268e-01 4.79029447e-01 -1.19008732e+00
5.83350599e-01 1.85626715e-01 1.39483735e-01 -1.32199907e+00
-6.84767127e-01 -5.98324597e-01 -9.62336302e-01 5.58060333e-02
5.17717421e-01 -1.22765705e-01 -9.22527313e-01 1.82835948e+00
-1.62050620e-01 3.06432337e-01 -1.13966152e-01 1.14854956e+00
7.41268635e-01 6.48925126e-01 2.83010267e-02 -1.02349587e-01
1.18689704e+00 -1.16927528e+00 -7.71767735e-01 -2.30807334e-01
5.59172630e-01 -7.97265649e-01 7.15801120e-01 4.29539114e-01
-1.32526112e+00 -1.01609063e+00 -1.10762763e+00 -1.51258200e-01
-2.48272508e-01 3.64915788e-01 7.29349494e-01 3.08332652e-01
-1.44106925e+00 9.24605012e-01 -1.09213400e+00 -4.38251287e-01
1.52439386e-01 7.19382703e-01 -4.03146654e-01 2.89159358e-01
-1.27272606e+00 1.04060864e+00 1.37632146e-01 7.52972603e-01
-8.33755434e-01 -4.18433011e-01 -9.42621291e-01 5.90157844e-02
5.13608515e-01 -8.81985486e-01 1.21632791e+00 -1.33821762e+00
-1.53288436e+00 5.02304137e-01 -1.80684328e-01 -7.34016657e-01
6.37106299e-01 -8.17682445e-01 -2.63367325e-01 2.25839064e-01
-2.98467517e-01 1.09713900e+00 1.13261592e+00 -1.10911953e+00
-4.02739823e-01 -1.58527344e-01 1.43696338e-01 1.82322949e-01
-4.42404419e-01 4.72871494e-03 -6.86437666e-01 -8.50579619e-01
-3.79675254e-02 -1.28747976e+00 -1.71780109e-01 3.82804386e-02
-2.14128897e-01 1.32491186e-01 1.09925926e+00 -7.00973630e-01
1.27332652e+00 -1.93498456e+00 4.36907411e-01 -1.26491711e-01
1.43309847e-01 5.62942028e-01 -2.91081011e-01 1.58906057e-01
-2.64775008e-01 -9.59267616e-02 3.77294153e-01 -4.95856255e-01
-2.79361904e-01 2.32381105e-01 -4.03954983e-01 6.02902889e-01
2.64438391e-01 1.10084343e+00 -6.75624013e-01 -2.59058267e-01
4.73518759e-01 9.58878458e-01 -6.76314473e-01 2.55891740e-01
-1.49359941e-01 3.45188707e-01 4.13882248e-02 4.77125227e-01
4.67331827e-01 -3.34446192e-01 1.14505284e-01 -6.56302869e-01
1.11328857e-02 -5.91597930e-02 -1.10576212e+00 1.81277096e+00
-5.08975029e-01 9.23835456e-01 -1.42719284e-01 -8.05430770e-01
3.83627236e-01 5.26709199e-01 5.46005905e-01 -8.35017502e-01
3.55984420e-01 1.14728451e-01 5.88000193e-02 -5.12831569e-01
7.87466764e-01 3.36706311e-01 1.52577713e-01 1.72005221e-01
1.35087818e-01 3.92711669e-01 6.80207536e-02 -7.35797435e-02
8.86869788e-01 6.23077631e-01 -8.72151479e-02 -4.81281755e-03
4.54767615e-01 -3.54612440e-01 3.88671994e-01 5.62303245e-01
-2.67243981e-01 8.41665149e-01 1.67907715e-01 -8.96136642e-01
-1.19922781e+00 -8.63806307e-01 2.62058437e-01 1.21637976e+00
7.97610059e-02 -3.75175238e-01 -6.97781563e-01 -3.40193450e-01
-3.40345442e-01 7.91508928e-02 -6.70083225e-01 -2.76609361e-01
-1.03136098e+00 -5.76017737e-01 7.24742413e-01 9.80359972e-01
6.21162653e-01 -1.07427812e+00 -9.67303693e-01 3.68838817e-01
-2.16535330e-01 -1.39820302e+00 -6.32798016e-01 1.53187871e-01
-8.70697975e-01 -1.03310323e+00 -9.67175484e-01 -6.79007530e-01
5.83603144e-01 2.35132694e-01 1.06520259e+00 1.05394170e-01
-1.79648474e-01 1.78050801e-01 -1.11626051e-01 1.37008965e-01
-1.22903464e-02 1.40864462e-01 3.74688953e-01 -1.98663697e-01
-4.51730266e-02 -4.35657203e-01 -7.64163613e-01 5.16808450e-01
-9.41887379e-01 2.15981901e-01 6.19241059e-01 9.69659626e-01
2.34643579e-01 -7.89888948e-02 2.61479288e-01 -4.83973444e-01
2.10479110e-01 9.35176983e-02 -3.58003765e-01 2.14038178e-01
-9.67999920e-02 1.59129769e-01 6.61707342e-01 -6.55724406e-01
-9.82032597e-01 2.82089323e-01 -2.32073307e-01 -7.80413747e-01
1.93613067e-01 1.53023049e-01 1.57906532e-01 -2.67342120e-01
2.34425992e-01 2.34674476e-02 2.13705897e-02 -2.16316611e-01
1.43262863e-01 3.27835411e-01 5.93823850e-01 -3.52374732e-01
5.58917046e-01 3.59935433e-01 6.97945338e-03 -8.63167167e-01
-6.34432554e-01 -1.83483422e-01 -7.07079589e-01 -4.78769720e-01
1.03380120e+00 -1.19526577e+00 -1.00654542e+00 7.55829692e-01
-1.42488480e+00 -3.10645521e-01 1.20618105e-01 4.60793316e-01
-5.99197626e-01 1.97541416e-01 -1.11784744e+00 -6.91164732e-01
-2.65010148e-01 -1.31358147e+00 1.03245473e+00 1.22905038e-01
-4.27469611e-01 -1.05380416e+00 -2.63776690e-01 4.42822278e-01
6.06220603e-01 2.30242774e-01 4.10945743e-01 -1.09787852e-01
-6.51669502e-01 -8.33265707e-02 -2.91012228e-01 2.58037120e-01
-8.99826512e-02 2.46480778e-01 -9.60121572e-01 -5.57965219e-01
-4.38543446e-02 -4.30739701e-01 9.35216725e-01 6.33303642e-01
1.21312499e+00 -3.24392825e-01 -8.67394656e-02 7.27063835e-01
1.21250343e+00 -4.39320467e-02 9.89295125e-01 4.79149789e-01
1.13598025e+00 4.47490722e-01 4.33276534e-01 2.02856064e-01
1.51307702e-01 1.04226005e+00 3.56747121e-01 -5.37479758e-01
-3.02647650e-01 -1.36918217e-01 7.49644637e-01 7.93458700e-01
-4.94294852e-01 -1.93001240e-01 -6.58851564e-01 2.52008140e-01
-2.02909064e+00 -1.06987286e+00 2.50273794e-02 2.01710129e+00
5.28379917e-01 3.53775442e-01 2.18190745e-01 1.57471485e-02
6.44263327e-01 1.77016005e-01 -4.30385023e-01 -4.15018409e-01
-2.44513601e-01 1.72672153e-01 8.42362821e-01 3.50040495e-01
-1.16286385e+00 9.29860115e-01 6.48168516e+00 7.59124994e-01
-1.37594533e+00 2.75393054e-02 7.22698629e-01 -5.30771792e-01
3.77926201e-01 -4.73233610e-01 -7.58335173e-01 2.70055503e-01
9.61603880e-01 5.29569626e-01 2.69547015e-01 7.48412848e-01
4.18718368e-01 -1.08803913e-01 -1.09775126e+00 9.62131381e-01
1.46354169e-01 -1.27732873e+00 2.57732391e-01 1.63263064e-02
6.64890051e-01 7.87292272e-02 2.32105508e-01 1.34753034e-01
-1.35355756e-01 -1.26555634e+00 8.89650762e-01 4.68977839e-01
8.04372907e-01 -7.81549752e-01 8.92942429e-01 -2.33310498e-02
-1.37417281e+00 -2.56430358e-01 -2.45610192e-01 -3.22522759e-01
1.73721582e-01 2.63030380e-01 -4.42011774e-01 3.80147904e-01
8.48903835e-01 7.78604448e-01 -7.03896344e-01 7.48868227e-01
8.36781934e-02 2.05561090e-02 -2.17674911e-01 2.44352043e-01
4.09629107e-01 3.02944660e-01 3.06520849e-01 1.24458551e+00
2.28855148e-01 -3.32366139e-01 8.04332197e-02 3.98853838e-01
2.32418776e-02 -3.72472018e-01 -5.16499817e-01 5.61721846e-02
-1.49208261e-02 1.04676163e+00 -7.22287953e-01 -3.07286799e-01
-3.79498452e-01 1.23989630e+00 3.55162889e-01 4.85055923e-01
-1.21825695e+00 3.88489924e-02 5.78339696e-01 1.89277440e-01
4.84939456e-01 -4.69087213e-01 -9.67883989e-02 -1.20164788e+00
1.90964535e-01 -8.34874570e-01 1.79205373e-01 -8.50674450e-01
-8.20047081e-01 8.33795786e-01 -1.05567276e-01 -1.27102613e+00
-4.42903042e-01 -8.57587337e-01 -3.33667338e-01 3.21063370e-01
-1.32707930e+00 -1.20577347e+00 -1.60021141e-01 6.31432593e-01
6.51647508e-01 1.02513693e-01 7.03509450e-01 6.06059551e-01
-8.52674305e-01 6.45753860e-01 -2.57295489e-01 2.54599750e-01
7.43324280e-01 -9.12703812e-01 4.03369129e-01 9.40630198e-01
-1.28971925e-02 8.36588085e-01 8.20037305e-01 -5.61024725e-01
-1.54217529e+00 -9.97618079e-01 7.11416900e-01 -2.97195047e-01
5.92443347e-01 -3.40634704e-01 -7.26823449e-01 7.10100234e-01
2.87181526e-01 4.43295343e-03 2.63274908e-01 -2.03313351e-01
-1.49340540e-01 3.87666142e-03 -6.25191092e-01 7.27209270e-01
1.12293732e+00 -4.97452050e-01 -2.95935005e-01 9.54353958e-02
6.60187423e-01 -7.53543556e-01 -6.77214921e-01 5.91977119e-01
7.97241092e-01 -1.19844544e+00 1.19163036e+00 -3.88983995e-01
7.37285018e-01 -2.24376425e-01 -3.18577252e-02 -1.00077748e+00
-2.82064915e-01 -5.77495754e-01 -5.41643620e-01 7.23961830e-01
2.78866470e-01 -2.07415655e-01 8.33259702e-01 6.62963033e-01
-3.36472578e-02 -9.86086488e-01 -9.07906115e-01 -7.39468753e-01
-2.12987259e-01 -2.91145861e-01 2.81988904e-02 4.23717648e-01
-3.32864463e-01 1.41257897e-01 -1.00565422e+00 2.78730877e-02
2.90120095e-01 -2.73289472e-01 6.28762662e-01 -6.59839094e-01
-4.56379533e-01 -3.81692946e-01 -4.24588174e-01 -1.28684735e+00
1.78075626e-01 -1.77841499e-01 8.98198411e-02 -1.20907772e+00
9.70657840e-02 6.01603054e-02 -2.06656933e-01 4.73104477e-01
-2.85990797e-02 6.72355235e-01 3.96911830e-01 2.09731340e-01
-8.51057708e-01 3.59799087e-01 1.31386912e+00 4.67123426e-02
-1.95252836e-01 -8.60136524e-02 -1.81580693e-01 7.62908459e-01
6.25325680e-01 -1.09775677e-01 -2.55079925e-01 -7.18832433e-01
3.75105917e-01 1.51499555e-01 6.17911518e-01 -1.42457664e+00
4.97826219e-01 3.11253041e-01 8.71543884e-01 -5.97737432e-01
6.49847090e-01 -9.03407454e-01 3.94530684e-01 8.23421955e-01
-2.18012512e-01 3.84107739e-01 4.15134400e-01 3.38318408e-01
-2.64110893e-01 2.47743592e-01 6.33503616e-01 -1.88540101e-01
-9.64880466e-01 1.99227080e-01 -3.82452875e-01 -3.17800790e-01
8.08148444e-01 -4.49532837e-01 -2.05981597e-01 -4.54818726e-01
-6.97214663e-01 7.39234686e-02 5.05381823e-01 7.33636022e-01
6.98449850e-01 -1.31709528e+00 -3.71574193e-01 2.10605785e-01
-3.56907159e-01 -1.94282189e-01 4.79877234e-01 9.17072952e-01
-6.47001088e-01 7.00226665e-01 -4.42897350e-01 -6.21103525e-01
-1.34408796e+00 5.07233799e-01 4.85040784e-01 -3.96800250e-01
-5.90868056e-01 9.09112573e-01 9.87928063e-02 -2.52876263e-02
7.24393308e-01 -4.13546503e-01 -9.01586935e-02 8.76729265e-02
4.38461840e-01 3.03942025e-01 1.59033284e-01 -8.29450488e-01
-4.44535494e-01 8.54587793e-01 -2.01954544e-01 1.41623631e-01
1.29552042e+00 -2.03956127e-01 1.79947168e-01 3.25576246e-01
1.38509822e+00 -3.71793151e-01 -1.57404208e+00 8.54291394e-02
-2.26867512e-01 -2.94285476e-01 4.50752676e-02 -5.85000396e-01
-1.49786031e+00 7.18418896e-01 5.41262507e-01 -2.29744449e-01
1.12559140e+00 -4.90115672e-01 1.26514637e+00 5.12596309e-01
4.01808709e-01 -1.29263651e+00 4.13061500e-01 6.71984494e-01
8.80804718e-01 -1.16400945e+00 1.04434505e-01 -2.39915535e-01
-5.29623449e-01 1.43209112e+00 7.97117651e-01 -3.69869888e-01
3.50211442e-01 3.82805824e-01 5.00214286e-02 -1.76208541e-01
-8.43211114e-01 5.60697056e-02 5.35952747e-01 3.19308549e-01
6.83171570e-01 -3.10610741e-01 1.24807559e-01 2.70899355e-01
-2.51974732e-01 1.81297511e-01 4.44156200e-01 8.44191968e-01
-1.28366694e-01 -7.92487264e-01 -3.47444683e-01 1.57476097e-01
-5.13238668e-01 -7.00860694e-02 -1.32313371e-01 8.89034212e-01
2.33699530e-01 7.16590822e-01 1.43897250e-01 -6.35027170e-01
2.87719727e-01 -3.02393377e-01 7.45134890e-01 -2.24471226e-01
-8.68750334e-01 1.65949568e-01 5.63885011e-02 -1.08006406e+00
-6.64136827e-01 -4.57272142e-01 -1.03940499e+00 -4.13512290e-01
-3.36130530e-01 -3.33175778e-01 4.35142606e-01 1.00996065e+00
3.81338954e-01 9.69362676e-01 1.93935946e-01 -1.44780576e+00
-3.33636016e-01 -7.90794492e-01 -1.98455796e-01 4.36601043e-01
4.86550570e-01 -8.01547050e-01 -2.48344600e-01 1.16863668e-01] | [7.530975818634033, -0.5758183598518372] |
62ad1cc7-41b7-4aba-a177-fe9bf6f83a46 | automating-cluster-analysis-to-generate | 2006.07197 | null | https://arxiv.org/abs/2006.07197v4 | https://arxiv.org/pdf/2006.07197v4.pdf | Clustering Residential Electricity Consumption Data to Create Archetypes that Capture Household Behaviour in South Africa | Clustering is frequently used in the energy domain to identify dominant electricity consumption patterns of households, which can be used to construct customer archetypes for long term energy planning. Selecting a useful set of clusters however requires extensive experimentation and domain knowledge. While internal clustering validation measures are well established in the electricity domain, they are limited for selecting useful clusters. Based on an application case study in South Africa, we present an approach for formalising implicit expert knowledge as external evaluation measures to create customer archetypes that capture variability in residential electricity consumption behaviour. By combining internal and external validation measures in a structured manner, we were able to evaluate clustering structures based on the utility they present for our application. We validate the selected clusters in a use case where we successfully reconstruct customer archetypes previously developed by experts. Our approach shows promise for transparent and repeatable cluster ranking and selection by data scientists, even if they have limited domain knowledge. | ['Wiebke Toussaint', 'Deshendran Moodley'] | 2020-06-11 | null | null | null | null | ['time-series-clustering'] | ['time-series'] | [-3.45720619e-01 5.25561161e-02 9.18584242e-02 -6.00192666e-01
-6.06498182e-01 -1.06541002e+00 5.22547483e-01 6.92781329e-01
-2.16524854e-01 4.45701301e-01 1.69536069e-01 -4.07528013e-01
-6.08075261e-01 -1.18063354e+00 6.74114451e-02 -8.10580254e-01
-2.08788604e-01 1.11841345e+00 -1.61949337e-01 1.73692890e-02
3.70533824e-01 5.43561280e-01 -1.65669620e+00 3.19834620e-01
1.23923492e+00 5.24958253e-01 3.52994144e-01 9.39031243e-02
-1.35831758e-01 4.15478021e-01 -5.09580135e-01 2.87202168e-02
6.93216100e-02 -4.52994317e-01 -9.35414255e-01 2.26915598e-01
-7.25528598e-01 2.69945025e-01 8.51103604e-01 8.81921887e-01
5.85578740e-01 5.32180592e-02 8.11977267e-01 -1.57576513e+00
-2.36778349e-01 1.17521012e+00 -5.96901514e-02 -1.32834598e-01
5.23297369e-01 -1.01523034e-01 1.02394390e+00 -3.31613541e-01
3.54908228e-01 7.70955026e-01 7.67108917e-01 2.27256026e-02
-1.69596112e+00 -7.38849699e-01 -4.00240794e-02 4.09378946e-01
-1.87516820e+00 -2.79835254e-01 9.32830811e-01 -5.66050172e-01
1.18439841e+00 6.69857442e-01 9.52062011e-01 3.47591013e-01
-7.21744001e-01 4.19246733e-01 1.35181296e+00 -5.78945577e-01
8.54005337e-01 7.09416986e-01 1.43496051e-01 -2.38121286e-01
2.41215438e-01 -1.72303930e-01 2.43697733e-01 -2.40110904e-01
2.41690189e-01 -2.87591100e-01 -4.06721793e-02 -4.25222516e-01
-8.04316521e-01 1.03551638e+00 2.52452940e-01 8.17657053e-01
-4.44448680e-01 -4.14315462e-01 3.31128508e-01 3.32605958e-01
3.68961900e-01 6.40707970e-01 -6.65605009e-01 -1.26497895e-01
-1.22991610e+00 -2.83858240e-01 1.10696173e+00 9.10340190e-01
7.85952270e-01 -1.62384614e-01 2.97753602e-01 7.47801721e-01
3.60779375e-01 2.40684137e-01 3.94972175e-01 -8.76595020e-01
2.06466354e-02 1.01873648e+00 2.00679287e-01 -7.44712234e-01
-6.90471470e-01 -3.86434644e-01 -8.29367161e-01 3.57425421e-01
1.99244972e-02 -2.72755232e-02 -4.86630946e-01 1.21434009e+00
9.16019157e-02 -4.18103933e-01 1.55121267e-01 4.24505234e-01
3.41781199e-01 3.31640840e-01 2.91045189e-01 -5.26788235e-01
1.09447157e+00 -8.15635398e-02 -3.93220574e-01 5.43115258e-01
8.67213726e-01 -4.96816427e-01 6.45203114e-01 6.93347335e-01
-9.99030530e-01 -4.78994489e-01 -6.99070930e-01 7.31974661e-01
-7.86862850e-01 -1.72745347e-01 4.04522717e-01 9.49187219e-01
-1.21381438e+00 8.04737508e-01 -7.41483927e-01 -7.78657079e-01
1.16619855e-01 5.10177672e-01 8.19759965e-02 4.41287547e-01
-1.06855488e+00 1.12356877e+00 1.06798291e+00 -2.04424322e-01
-2.45972365e-01 -5.61555028e-01 -4.39927429e-01 2.63398677e-01
-3.85214388e-02 -2.50129700e-01 7.28265047e-01 -1.04287672e+00
-9.99453068e-01 6.90504193e-01 2.57008225e-01 -4.30413902e-01
5.54080129e-01 5.36993742e-01 -1.00889981e+00 6.72320947e-02
2.68221974e-01 2.85427064e-01 8.46881866e-02 -1.77063847e+00
-1.03688860e+00 -2.54340768e-01 -5.85738480e-01 8.10571015e-02
-2.47095838e-01 -3.16793546e-02 -2.50576645e-01 -2.85312682e-01
1.18356757e-01 -5.90636849e-01 -2.79932499e-01 -9.89617765e-01
-2.05733195e-01 -5.76264679e-01 5.97830534e-01 -5.40076435e-01
1.79795730e+00 -1.78624451e+00 -2.38739341e-01 1.41470814e+00
-1.83625981e-01 1.51914984e-01 5.40673971e-01 7.37428248e-01
-3.34353834e-01 5.57332814e-01 -3.59280080e-01 3.56716096e-01
6.08993471e-01 2.19562978e-01 1.75947100e-01 2.13967562e-01
-2.87182890e-02 6.41700029e-01 -1.14892244e+00 -5.77785075e-01
8.27539206e-01 4.25055474e-01 -1.91758752e-01 -1.18209450e-02
-3.92331406e-02 2.09068999e-01 -3.75214934e-01 4.27242726e-01
5.55998087e-01 -1.70145586e-01 7.26049960e-01 -1.57978505e-01
-2.86646962e-01 7.71354437e-02 -1.49905360e+00 1.09757042e+00
-2.43145481e-01 3.07282627e-01 -1.41858486e-02 -1.16280818e+00
1.01541686e+00 4.69373792e-01 9.02628660e-01 -7.55824506e-01
5.49114123e-02 3.17411184e-01 -5.98140322e-02 -3.96929175e-01
9.89279896e-02 -2.67065972e-01 -1.68774411e-01 6.62158012e-01
-2.68050253e-01 -1.85494870e-01 5.13235867e-01 -1.23738222e-01
9.18615162e-01 -3.36491577e-02 4.14543509e-01 -8.19816113e-01
5.47638357e-01 4.68593121e-01 2.98519969e-01 2.41897494e-01
1.03194728e-01 4.85007465e-01 1.09109923e-01 -1.57779545e-01
-1.30598056e+00 -1.05999243e+00 -4.72316265e-01 7.58549213e-01
-3.92481059e-01 -5.47998369e-01 -7.43364573e-01 -6.04176641e-01
-1.22232869e-01 1.32808232e+00 -4.62556332e-01 3.29179317e-01
-1.89248249e-01 -8.33827853e-01 2.04106465e-01 5.44290304e-01
3.06285530e-01 -1.09455752e+00 -7.63120413e-01 6.73226416e-01
-1.00685798e-01 -4.41516697e-01 1.75381094e-01 4.75927085e-01
-7.04665542e-01 -1.32071388e+00 -3.14841300e-01 -7.50705600e-01
7.95981586e-01 -1.60767078e-01 1.76377189e+00 2.55723894e-01
2.23381706e-02 4.54658002e-01 -6.58051670e-01 -3.85012358e-01
-6.85953677e-01 2.73425967e-01 -1.36406079e-01 -5.71950853e-01
1.15258920e+00 -8.63182366e-01 -6.67393386e-01 7.61053622e-01
-8.44716251e-01 -2.63134599e-01 3.66617411e-01 1.45036295e-01
2.97159314e-01 1.08794439e+00 9.45745170e-01 -1.02228642e+00
8.83847475e-01 -7.68003047e-01 -6.25085890e-01 4.19441849e-01
-1.36769938e+00 4.85397503e-02 5.05633891e-01 1.19150594e-01
-9.04852033e-01 2.53226280e-01 -1.08946197e-01 1.80517316e-01
-7.86636710e-01 7.92196751e-01 -2.48522297e-01 4.56362903e-01
5.05134821e-01 -9.52999666e-02 -2.14649796e-01 -5.34460902e-01
2.89732903e-01 7.41535842e-01 3.25308651e-01 -5.87144136e-01
8.32381964e-01 2.92488784e-01 -3.62693071e-01 -6.08857274e-01
1.98912948e-01 -8.93679261e-01 -1.14622998e+00 -1.78543463e-01
8.29473615e-01 -8.67465973e-01 -8.89725447e-01 -2.61340916e-01
-3.99360120e-01 -3.18823099e-01 -5.80051422e-01 2.69474894e-01
-4.64342117e-01 1.79442644e-01 3.52876246e-01 -9.30172622e-01
-2.04746351e-01 -7.79249430e-01 5.08526385e-01 -2.16519818e-01
-9.86575246e-01 -1.36006570e+00 1.94009274e-01 -5.12380451e-02
3.49430770e-01 6.53508961e-01 1.20262289e+00 -1.00034690e+00
-1.36413306e-01 -7.78684020e-02 1.90473512e-01 2.23762259e-01
3.38085085e-01 2.14971930e-01 -8.96600783e-01 -4.56234694e-01
-2.48462632e-01 2.18556568e-01 3.68137270e-01 1.28296748e-01
9.73520696e-01 -2.66464621e-01 -6.20665073e-01 -1.12504467e-01
1.75728369e+00 6.21033072e-01 7.70199358e-01 5.78960001e-01
3.07075053e-01 1.07875121e+00 3.37057859e-01 5.80410779e-01
5.78458071e-01 4.26018417e-01 -7.35778883e-02 -3.89156044e-01
4.67361629e-01 2.40167692e-01 1.07692204e-01 1.02764237e+00
-2.64171958e-01 -7.96425194e-02 -1.17511618e+00 8.72745156e-01
-1.93255711e+00 -1.25797367e+00 -2.15624243e-01 2.11822653e+00
6.24360025e-01 2.49866262e-01 8.42078149e-01 6.42568052e-01
7.02252626e-01 -6.66327715e-01 -3.14548761e-01 -4.70784545e-01
4.24425602e-02 3.26647282e-01 3.42640936e-01 3.57057184e-01
-6.77615643e-01 2.42150635e-01 6.46458101e+00 6.78564191e-01
-3.04854453e-01 -1.22393280e-01 5.60432374e-01 9.53076258e-02
-7.05015659e-01 6.78704828e-02 -1.81370571e-01 6.14764094e-01
1.21901810e+00 -3.35368961e-01 3.62814486e-01 5.67496359e-01
5.80865681e-01 -2.10246772e-01 -1.33161259e+00 6.14030123e-01
-4.16904002e-01 -1.00457788e+00 -4.77414340e-01 2.70059943e-01
7.57469237e-01 -1.09332010e-01 -5.23768187e-01 2.99108922e-02
1.16649437e+00 -1.27745867e+00 5.53155124e-01 5.16187429e-01
3.70900840e-01 -1.23357856e+00 8.31570685e-01 2.38844603e-01
-1.44904459e+00 -2.18170375e-01 -5.40577210e-02 1.29683435e-01
8.13748911e-02 5.27362525e-01 -9.90790725e-01 8.54827821e-01
1.01892984e+00 4.23471183e-01 -5.99046469e-01 9.66419697e-01
1.08502813e-01 7.35036433e-01 -6.94986880e-01 7.20684379e-02
1.32064357e-01 -7.09686458e-01 -1.93744808e-01 1.49822831e+00
5.96753478e-01 2.36651823e-01 2.60330867e-02 1.06038713e+00
4.14687812e-01 7.02058673e-02 -4.27526414e-01 2.51895547e-01
9.54187810e-01 1.26702416e+00 -1.28589690e+00 -2.64344841e-01
-2.10534796e-01 5.63939095e-01 -2.24410862e-01 4.05652255e-01
-4.66385126e-01 -4.75820333e-01 3.64452124e-01 2.96151400e-01
5.43738842e-01 1.35408849e-01 -3.82726640e-01 -5.54106951e-01
-2.94370562e-01 -6.80942118e-01 5.04071593e-01 -5.57067394e-01
-1.58496451e+00 3.99042130e-01 4.44226235e-01 -1.30809093e+00
-6.50103927e-01 -1.04474023e-01 -9.48190868e-01 8.33103955e-01
-9.89802003e-01 -9.81416106e-01 -2.24026039e-01 6.51819646e-01
2.71694139e-02 -8.17019567e-02 1.18246925e+00 1.26855597e-01
-3.21947157e-01 2.01193169e-02 8.05659294e-01 1.20272785e-01
1.33304015e-01 -1.91785777e+00 8.37295502e-02 3.87486100e-01
-2.18868069e-02 6.40485764e-01 8.61275971e-01 -6.20479107e-01
-6.25639081e-01 -9.84209895e-01 7.64576912e-01 -5.12405217e-01
4.91013229e-01 -3.12229276e-01 -8.04409564e-01 3.55412751e-01
7.21244037e-01 -9.47987139e-01 1.41001129e+00 1.58450559e-01
2.13676170e-01 -1.50030509e-01 -1.58663332e+00 8.24093297e-02
5.36705971e-01 -3.67605805e-01 -6.59069121e-01 9.89003554e-02
-1.50237456e-01 6.73420489e-01 -1.53940856e+00 1.14713825e-01
3.35215360e-01 -1.08752453e+00 9.55853760e-01 -9.08463262e-03
-2.46861026e-01 -6.74455166e-01 -1.01666577e-01 -1.53991997e+00
-7.44471848e-01 -5.00082076e-01 3.18414837e-01 1.99902844e+00
5.57526410e-01 -5.56017160e-01 7.22278655e-01 9.39996779e-01
1.31441534e-01 -1.86888412e-01 -6.21843338e-01 -6.36011660e-01
3.61570686e-01 -3.84883732e-01 1.29756391e+00 1.38527632e+00
4.63188916e-01 2.29984805e-01 4.45802420e-01 1.26030087e-01
8.44231844e-01 2.89247721e-01 4.45750505e-01 -1.79515839e+00
1.77577496e-01 -8.53273869e-01 -2.85610884e-01 1.89812824e-01
1.31830592e-02 -1.13821793e+00 -1.03796378e-01 -2.02492881e+00
5.46158180e-02 -6.85450494e-01 -6.12483323e-01 4.34020877e-01
3.59856725e-01 1.98958367e-01 7.63880014e-02 3.30423445e-01
-2.98762709e-01 -5.35680056e-02 2.10928693e-01 -8.74597207e-02
-5.01544178e-01 4.25770991e-02 -9.65470254e-01 8.24602365e-01
1.16812086e+00 -3.77116442e-01 -7.60701299e-01 1.60488784e-01
5.28725028e-01 -6.42898560e-01 6.44854158e-02 -1.00122142e+00
2.30581567e-01 -2.18020692e-01 7.76804805e-01 -1.00244570e+00
-4.58554357e-01 -1.66743350e+00 1.21644819e+00 3.92531931e-01
-2.28092838e-02 2.38417730e-01 7.05925524e-02 1.05441503e-01
-2.23072730e-02 -3.86692762e-01 3.33351493e-01 -4.83841270e-01
-8.92471910e-01 -2.08499253e-01 -5.93552351e-01 -3.28325897e-01
1.27810121e+00 -6.46664143e-01 2.25750610e-01 -2.28708282e-01
-1.14707232e+00 7.58117199e-01 9.29754972e-01 -1.01836056e-01
1.88509449e-01 -1.49694920e+00 -8.21289301e-01 2.17475165e-02
1.98252618e-01 -1.90053046e-01 -3.26811671e-01 3.71613652e-01
-5.00157595e-01 4.40924555e-01 -2.18996733e-01 -5.62588334e-01
-1.13515532e+00 7.09409475e-01 7.46593922e-02 -1.76705211e-01
-5.01085579e-01 8.01235065e-02 -4.56698358e-01 -4.22959685e-01
2.01584220e-01 -3.80004674e-01 -6.01731241e-01 7.93626606e-01
5.59290610e-02 6.88711584e-01 2.94561774e-01 -7.03432918e-01
-6.55133724e-01 5.11030555e-01 5.93997419e-01 -1.12668939e-01
1.85205460e+00 -6.22309983e-01 -1.32675543e-01 4.33234215e-01
1.05160487e+00 -2.46046528e-01 -8.66290808e-01 2.93338805e-01
7.69735575e-01 -2.09579960e-01 -1.64484680e-01 -1.09245408e+00
-9.12921369e-01 6.42496526e-01 8.88324142e-01 1.29873109e+00
1.57536256e+00 1.66568965e-01 1.27133518e-01 2.57405013e-01
4.25310463e-01 -1.80197394e+00 -5.66662669e-01 -2.74380416e-01
5.69747090e-01 -1.15609407e+00 1.36684731e-01 -1.62482243e-02
-7.93865144e-01 1.06247675e+00 3.50371711e-02 1.62158355e-01
9.50794756e-01 1.67054996e-01 -1.26121426e-03 -4.73022014e-01
-3.19390744e-01 -5.37748039e-01 9.77072641e-02 1.08283818e+00
5.68762481e-01 6.37722552e-01 -3.19447517e-01 5.00676513e-01
-5.12221396e-01 -1.23867013e-01 2.38000721e-01 8.51477981e-01
-4.89749879e-01 -1.31379712e+00 -4.81146723e-01 6.52268589e-01
-2.15077400e-01 4.84273136e-02 -4.77686524e-01 9.69149888e-01
4.82086837e-01 1.40275323e+00 2.98483700e-01 -4.33638334e-01
4.54027504e-01 2.11385548e-01 2.32948393e-01 -5.24527133e-01
-9.46430445e-01 3.81231487e-01 2.99940914e-01 -1.06904842e-01
-9.74286556e-01 -9.53297853e-01 -1.37124336e+00 -5.35451114e-01
-5.61382532e-01 9.59412038e-01 6.19307756e-01 7.21911967e-01
1.93783864e-01 5.31420588e-01 1.08564043e+00 -6.11674249e-01
3.19264531e-02 -1.00111914e+00 -9.42819476e-01 8.42622817e-01
-1.52935252e-01 -3.43199432e-01 -2.77172536e-01 2.71354914e-01] | [7.567553997039795, 4.493282318115234] |
706db914-362f-496d-b46e-d6f6547b6e29 | hybrid-score-and-rank-level-fusion-for-person | 2008.03353 | null | https://arxiv.org/abs/2008.03353v1 | https://arxiv.org/pdf/2008.03353v1.pdf | Hybrid Score- and Rank-level Fusion for Person Identification using Face and ECG Data | Uni-modal identification systems are vulnerable to errors in sensor data collection and are therefore more likely to misidentify subjects. For instance, relying on data solely from an RGB face camera can cause problems in poorly lit environments or if subjects do not face the camera. Other identification methods such as electrocardiograms (ECG) have issues with improper lead connections to the skin. Errors in identification are minimized through the fusion of information gathered from both of these models. This paper proposes a methodology for combining the identification results of face and ECG data using Part A of the BioVid Heat Pain Database containing synchronized RGB-video and ECG data on 87 subjects. Using 10-fold cross-validation, face identification was 98.8% accurate, while the ECG identification was 96.1% accurate. By using a fusion approach the identification accuracy improved to 99.8%. Our proposed methodology allows for identification accuracies to be significantly improved by using disparate face and ECG models that have non-overlapping modalities. | ['Thomas Truong', 'Jonathan Graf', 'Svetlana Yanushkevich'] | 2020-08-07 | null | null | null | null | ['person-identification'] | ['computer-vision'] | [ 3.85588348e-01 -1.25287890e-01 2.34069705e-01 -4.88668382e-01
-6.73041463e-01 -5.37709296e-01 -7.57738948e-02 6.96328357e-02
-4.25984919e-01 7.18279660e-01 -2.73121834e-01 3.02682132e-01
-1.91984728e-01 -4.33190286e-01 -6.48699999e-02 -8.01598728e-01
1.99238151e-01 2.47081280e-01 -5.55800080e-01 4.17758614e-01
1.11514099e-01 5.37352562e-01 -1.62539935e+00 -5.27715497e-02
3.69553477e-01 1.33234465e+00 -5.80940127e-01 5.35398126e-01
2.83203274e-01 2.30698198e-01 -7.02011824e-01 -2.17129126e-01
4.00874794e-01 -3.79047751e-01 -3.91847253e-01 -4.46598604e-02
4.60536271e-01 -4.93524700e-01 2.22222969e-01 8.11828434e-01
1.02402604e+00 -2.22624138e-01 3.46509397e-01 -1.40870416e+00
-6.28261641e-02 2.12396413e-01 -8.26825321e-01 -1.81673035e-01
8.48215342e-01 4.12224280e-03 1.39295822e-02 -4.89271998e-01
3.11248064e-01 8.56047332e-01 1.35545909e+00 7.25578666e-01
-1.30167353e+00 -9.98113871e-01 -8.11287344e-01 8.43832493e-02
-2.09744263e+00 -6.48450613e-01 8.42529774e-01 -1.74624681e-01
7.55124390e-01 5.46577275e-01 7.55830407e-01 8.37232709e-01
2.18327250e-02 -2.57889539e-01 1.63112664e+00 -4.36990499e-01
5.47613055e-02 4.75090921e-01 1.01142280e-01 4.24052745e-01
8.47342193e-01 5.12391478e-02 -6.87211215e-01 -4.53198999e-01
6.74601972e-01 1.06889838e-02 -2.17340868e-02 7.69219771e-02
-7.97649384e-01 3.09804499e-01 -1.31465390e-01 4.90303218e-01
-5.60924649e-01 7.67382011e-02 3.38471919e-01 9.35392752e-02
1.72608998e-02 1.40553817e-01 -8.74579102e-02 -1.83475584e-01
-1.02759433e+00 -1.91402674e-01 8.23285103e-01 6.00421786e-01
6.58930063e-01 -5.47476746e-02 3.59993935e-01 6.16811275e-01
5.20973146e-01 8.04870784e-01 3.93496990e-01 -9.24190521e-01
-1.34574607e-01 6.67345643e-01 7.38963336e-02 -1.17550707e+00
-4.96921569e-01 4.37562093e-02 -7.98674941e-01 3.25677752e-01
3.12199354e-01 -4.08947229e-01 -7.52511263e-01 1.51729667e+00
5.06658435e-01 2.27710858e-01 1.55839995e-01 8.28418314e-01
9.30010736e-01 9.72529799e-02 3.28023136e-01 -5.65481126e-01
1.44415843e+00 2.82811403e-01 -1.00379598e+00 3.75080287e-01
9.66677666e-02 -8.28687310e-01 2.57213205e-01 5.57841480e-01
-8.67364049e-01 -7.47252703e-01 -1.13846052e+00 4.29937065e-01
-2.12258145e-01 2.09076762e-01 3.24541509e-01 1.43714094e+00
-1.05502510e+00 3.86140704e-01 -6.56939328e-01 -7.46138155e-01
3.85011673e-01 1.01712191e+00 -7.68978238e-01 2.42401436e-01
-8.06313157e-01 8.91489089e-01 1.33587852e-01 1.72646329e-01
-3.30247641e-01 -4.37147349e-01 -6.65547848e-01 -6.14710331e-01
-2.07245216e-01 -4.51602966e-01 6.51043713e-01 -1.10271728e+00
-1.28429663e+00 1.09326231e+00 -2.86765605e-01 -1.25293896e-01
5.46438634e-01 -9.65473726e-02 -6.75875962e-01 5.86110055e-01
8.14162381e-03 3.77087355e-01 8.25089574e-01 -1.11858737e+00
-2.52577394e-01 -1.00018036e+00 -6.21397614e-01 1.02090701e-01
-2.82779545e-01 1.53797969e-01 1.23403408e-01 -1.48525268e-01
3.30643773e-01 -9.11100030e-01 3.03821862e-01 -2.17710529e-03
-2.43787035e-01 8.63833129e-02 9.22378778e-01 -9.82189536e-01
1.02673173e+00 -2.06856823e+00 -3.51426721e-01 6.30136967e-01
1.31842896e-01 2.77774245e-01 5.13147593e-01 1.61194652e-01
-1.49712503e-01 3.48320812e-01 -1.50050065e-02 -2.93146431e-01
-4.03496057e-01 1.49963558e-01 4.99529153e-01 9.78807390e-01
-1.75216332e-01 2.99519211e-01 -2.62356430e-01 -7.68380702e-01
5.58674693e-01 9.03028905e-01 2.09470019e-01 2.10383028e-01
9.53382552e-01 7.49486923e-01 -3.00803214e-01 1.09663391e+00
7.94951618e-01 4.02599871e-01 1.99320406e-01 -6.18707538e-01
1.40150085e-01 -4.79672283e-01 -1.57281399e+00 1.46065438e+00
-1.72269657e-01 2.60489434e-01 1.75694406e-01 -6.65983856e-01
1.09829283e+00 1.06044722e+00 1.12796366e+00 -5.01819968e-01
5.15036345e-01 2.13075981e-01 -1.87807426e-01 -7.25794137e-01
6.63715526e-02 -3.80371243e-01 1.17435372e-02 4.49286520e-01
5.15328581e-03 2.42410347e-01 -3.68922591e-01 -2.99882889e-01
6.21961176e-01 -1.80242751e-02 4.45358098e-01 -5.03076911e-02
5.69808304e-01 -2.44526505e-01 5.68899989e-01 6.02865875e-01
-4.40545559e-01 6.75116360e-01 -2.54972517e-01 -3.29833746e-01
-7.03894615e-01 -8.15094292e-01 -5.76437473e-01 -7.67103583e-02
1.29455253e-01 -2.90143013e-01 -8.64132583e-01 -2.77653366e-01
2.57339239e-01 1.38484553e-01 -6.70991004e-01 -2.35649288e-01
-1.93950474e-01 -8.73496115e-01 1.16584551e+00 4.84052092e-01
6.49084806e-01 -5.02925932e-01 -1.12230182e+00 8.74166265e-02
-8.35498124e-02 -9.42929506e-01 3.08053702e-01 -1.14975795e-01
-9.53928113e-01 -1.46365893e+00 -4.53978866e-01 -1.63699791e-01
7.40438879e-01 -1.59949541e-01 7.21702456e-01 4.23240185e-01
-7.05576658e-01 9.85219598e-01 -3.19964081e-01 -7.32186556e-01
-1.43642396e-01 -4.01250690e-01 5.13538599e-01 4.29861248e-01
7.87401557e-01 -4.77870464e-01 -6.68209612e-01 3.01674187e-01
-5.21795332e-01 -5.09303689e-01 8.67566988e-02 3.29792619e-01
6.22629106e-01 -1.39317513e-02 7.20716596e-01 -5.56544602e-01
4.17249292e-01 -4.40023124e-01 -3.04013640e-01 2.81769454e-01
-8.76730740e-01 -4.80605006e-01 8.80433470e-02 -1.73953533e-01
-9.45779860e-01 5.59789956e-01 3.99673805e-02 -3.87176216e-01
-6.41134083e-01 -7.61910900e-02 -1.07508451e-01 -7.88228273e-01
6.06058478e-01 -1.20604455e-01 3.05424750e-01 -4.43812639e-01
-4.82935160e-01 1.09775126e+00 6.74998879e-01 -3.36178243e-01
4.62417632e-01 4.25711662e-01 3.45934838e-01 -1.01386082e+00
2.50096738e-01 -5.43371737e-01 -7.72653401e-01 -7.39343941e-01
9.55987751e-01 -9.31794047e-01 -1.27373958e+00 7.66924083e-01
-7.59315789e-01 6.06798112e-01 2.23157499e-02 6.91406846e-01
-7.26628602e-02 4.17282969e-01 -1.16959244e-01 -1.46435928e+00
-6.40546799e-01 -7.20281482e-01 8.67872357e-01 6.75616264e-01
-7.08056629e-01 -7.71553099e-01 -3.52877490e-02 6.31435156e-01
4.17185038e-01 8.44255805e-01 7.39005283e-02 -5.31016648e-01
2.83731997e-01 -8.12027812e-01 2.35108197e-01 2.32407987e-01
7.11493015e-01 2.13080540e-01 -1.54174185e+00 -2.04143390e-01
3.63979369e-01 -2.68046588e-01 1.96147129e-01 2.18195692e-01
7.49485135e-01 -1.70263182e-02 -3.13530117e-01 6.04600787e-01
1.73271060e+00 5.21929681e-01 6.69593990e-01 -2.22557951e-02
7.15953410e-01 5.92374027e-01 2.68546492e-01 4.99512404e-01
1.87698752e-01 4.72430557e-01 2.97467321e-01 -1.15607500e-01
1.10227332e-01 2.21703395e-01 1.27103820e-01 3.87878835e-01
-6.31342471e-01 2.62630671e-01 -9.19055462e-01 2.75505632e-01
-1.26051235e+00 -9.11809683e-01 -4.82981831e-01 2.39589715e+00
6.41966522e-01 -6.02333128e-01 3.60096425e-01 8.35677385e-01
8.85673583e-01 -6.38051271e-01 -6.03166938e-01 -3.04554969e-01
-1.45720974e-01 5.26471436e-01 6.63358986e-01 2.69077420e-01
-9.68860149e-01 -1.18211294e-02 6.55915880e+00 4.57993196e-03
-1.20397723e+00 1.79254174e-01 6.48661673e-01 -1.91235766e-01
2.32300505e-01 -2.00381279e-01 -5.94418466e-01 5.86615920e-01
1.10353470e+00 3.65009978e-02 1.52626887e-01 3.57699275e-01
1.13814898e-01 -8.38772237e-01 -9.09095407e-01 1.67217338e+00
4.44056600e-01 -6.05345905e-01 -5.36994457e-01 2.05634013e-02
2.66603023e-01 -6.33576572e-01 -1.78965271e-01 -5.63109279e-01
-5.45911014e-01 -1.24053383e+00 3.18539202e-01 8.85115564e-01
1.16255164e+00 -6.59496963e-01 9.64453816e-01 -8.83251727e-02
-1.16296041e+00 1.93094090e-01 1.49869788e-02 -7.28669539e-02
-5.99425472e-02 3.43370318e-01 -8.81791115e-01 6.55782640e-01
9.80409861e-01 4.11347896e-01 -6.27506137e-01 9.84358609e-01
2.38105237e-01 4.89152551e-01 -8.25695038e-01 2.33764261e-01
-6.58567607e-01 -2.08770320e-01 3.18644643e-01 7.61202753e-01
5.18276811e-01 4.30542082e-01 -3.30168873e-01 6.08699560e-01
2.64359683e-01 -8.80113337e-03 -7.63591647e-01 1.80053592e-01
6.02988720e-01 1.33964145e+00 -6.48699880e-01 -2.09363043e-01
-3.02524745e-01 8.35678399e-01 -6.29467726e-01 1.09623484e-02
-7.10323811e-01 -2.80922860e-01 4.20969188e-01 1.65784329e-01
-3.69296581e-01 1.69114128e-01 -5.70288539e-01 -6.69256270e-01
-6.96887076e-03 -6.79935515e-01 5.89741409e-01 -8.17598164e-01
-1.10107064e+00 5.73178649e-01 1.13961756e-01 -1.07533097e+00
-2.73199141e-01 -2.78952688e-01 -2.64946669e-01 1.18840122e+00
-8.18457663e-01 -1.24466836e+00 -6.99255705e-01 8.45906973e-01
-3.22417796e-01 -3.86827514e-02 1.20384288e+00 5.33879161e-01
-5.83146095e-01 8.58883023e-01 -2.69717872e-01 5.88733070e-02
8.76310229e-01 -9.03714895e-01 -6.26287103e-01 4.63941753e-01
-1.55213997e-01 7.11977243e-01 4.16524708e-01 -8.14233363e-01
-1.74396253e+00 -7.44089127e-01 6.09946787e-01 -4.85476196e-01
-3.03349137e-01 8.69952962e-02 -6.27029240e-01 2.77068466e-01
2.48773411e-01 -4.71722595e-02 1.29461396e+00 -2.29161233e-01
6.65681437e-02 -6.55416727e-01 -2.06177163e+00 -2.52257675e-01
4.30035740e-01 -6.81331217e-01 -4.54272509e-01 -8.89330357e-02
-5.39522588e-01 -2.36854240e-01 -1.39753747e+00 3.85038733e-01
1.08968866e+00 -9.90096033e-01 8.96400392e-01 9.84089822e-02
-5.30902982e-01 -5.61457932e-01 -2.07667783e-01 -5.99672556e-01
1.37651369e-01 -5.08351088e-01 2.82911509e-01 1.69328272e+00
-9.54063907e-02 -7.63576150e-01 5.16111553e-01 1.14603639e+00
7.04860806e-01 -1.02194101e-01 -1.18567276e+00 -4.08010364e-01
-6.82520270e-01 -3.56090009e-01 5.46836913e-01 9.15324748e-01
8.06498080e-02 -9.83209163e-02 -5.33826888e-01 2.06859052e-01
1.05454493e+00 -3.25653195e-01 4.81987745e-01 -1.50476003e+00
1.97278365e-01 2.82849878e-01 -8.23815525e-01 5.86501896e-01
-2.79293150e-01 -4.44783509e-01 -2.81034648e-01 -1.09007227e+00
2.56263912e-01 -4.60089684e-01 -3.48746628e-01 8.23589683e-01
8.61635655e-02 1.13167322e+00 -2.55613551e-02 2.16657147e-01
-5.97315542e-02 -2.49583274e-01 4.67578024e-01 2.62520432e-01
-6.37815520e-02 -2.21180990e-01 -5.03829360e-01 6.50973260e-01
9.74096954e-01 -6.07197642e-01 -2.88161457e-01 -8.65726918e-02
1.85848951e-01 2.73021728e-01 5.19685745e-01 -1.52790844e+00
3.14417213e-01 2.54175246e-01 1.11161721e+00 -2.97441542e-01
5.61808825e-01 -1.38648427e+00 1.26455832e+00 6.39029503e-01
1.61997691e-01 2.02102840e-01 3.44378233e-01 2.34562710e-01
-5.39939180e-02 -1.13726474e-01 7.67362952e-01 -2.93381035e-01
-3.82983416e-01 -1.34205505e-01 -3.96820843e-01 -6.63046300e-01
1.43151820e+00 -9.27499235e-01 7.27312341e-02 -3.25473756e-01
-9.93612230e-01 -2.00319171e-01 6.72413230e-01 4.95019518e-02
6.26153708e-01 -1.36278248e+00 -4.63522822e-01 5.45079052e-01
1.45660825e-02 -4.41470474e-01 2.73562908e-01 1.15133941e+00
-5.98704875e-01 1.33016989e-01 -7.65756071e-01 -6.70007706e-01
-1.94838285e+00 2.34238245e-02 5.34349620e-01 6.25295639e-01
-7.01673701e-02 6.77768290e-01 -7.76883304e-01 2.00450439e-02
1.67961597e-01 1.54184848e-01 -2.25321248e-01 4.19133812e-01
5.28854191e-01 7.89550424e-01 2.64274418e-01 -1.18132079e+00
-9.53538895e-01 1.12338161e+00 4.35016036e-01 -2.60492742e-01
1.05976272e+00 -3.81884992e-01 -2.19170347e-01 5.49125731e-01
1.04740047e+00 -2.38440558e-02 -5.57184160e-01 3.21999222e-01
-4.40342426e-01 -5.40286064e-01 -5.02540842e-02 -1.08764184e+00
-1.11405814e+00 7.83616662e-01 1.41132581e+00 1.20699361e-01
1.51504362e+00 -4.34744358e-01 3.90865922e-01 -8.17939639e-02
5.88978827e-01 -1.21817851e+00 -5.36865115e-01 -4.93371338e-01
4.20960039e-01 -1.22324634e+00 3.09504092e-01 -2.77806848e-01
-3.65742356e-01 1.28970182e+00 3.95561576e-01 1.18660405e-01
6.25310004e-01 3.32188249e-01 3.18920106e-01 -1.60860732e-01
-6.56831264e-02 9.10917372e-02 1.60071671e-01 8.64799082e-01
4.74357307e-01 7.88588077e-02 -5.63784420e-01 6.82168424e-01
7.77137354e-02 4.05107707e-01 3.59059900e-01 1.24386096e+00
7.60444030e-02 -1.03841090e+00 -1.10444236e+00 5.46783507e-01
-9.69040871e-01 5.49155116e-01 -8.43145549e-01 6.23939335e-01
6.11095309e-01 1.39663339e+00 8.76286998e-02 -6.10846341e-01
1.05963521e-01 3.99169743e-01 7.60886014e-01 -3.26686241e-02
-9.75623250e-01 1.23229586e-01 1.46106854e-01 -5.70932627e-01
-1.06152725e+00 -9.60628629e-01 -1.20349967e+00 -4.73483711e-01
-2.53270030e-01 8.55424181e-02 1.04899669e+00 6.81446373e-01
4.33066905e-01 2.42442518e-01 5.25332272e-01 -5.55926025e-01
-1.92991734e-01 -8.69972050e-01 -8.49880755e-01 4.69051927e-01
2.96310544e-01 -6.59661710e-01 -2.28510380e-01 4.12547380e-01] | [13.389857292175293, 1.1667373180389404] |
4079e5ce-686f-44b9-afcc-b89fd19be46a | vegfru-a-domain-specific-dataset-for-fine | null | null | http://openaccess.thecvf.com/content_iccv_2017/html/Hou_VegFru_A_Domain-Specific_ICCV_2017_paper.html | http://openaccess.thecvf.com/content_ICCV_2017/papers/Hou_VegFru_A_Domain-Specific_ICCV_2017_paper.pdf | VegFru: A Domain-Specific Dataset for Fine-Grained Visual Categorization | VegFru: A Domain-Specific Dataset for Fine-grained Visual Categorization In this paper, we propose a novel domain-specific dataset named VegFru for fine-grained visual categorization (FGVC). While the existing datasets for FGVC are mainly focused on animal breeds or man-made objects with limited labelled data, VegFru is a larger dataset consisting of vegetables and fruits which are closely associated with the daily life of everyone. Aiming at domestic cooking and food management, VegFru categorizes vegetables and fruits according to their eating characteristics, and each image contains at least one edible part of vegetables or fruits with the same cooking usage. Particularly, all the images are labelled hierarchically. The current version covers vegetables and fruits of 25 upper-level categories and 292 subordinate classes. And it contains more than 160,000 images in total and at least 200 images for each subordinate class. Accompanying the dataset, we also propose an effective framework called HybridNet to exploit the label hierarchy for FGVC. Specifically, multiple granularity features are first extracted by dealing with the hierarchical labels separately. And then they are fused through explicit operation, e.g., Compact Bilinear Pooling, to form a unified representation for the ultimate recognition. The experimental results on the novel VegFru, the public FGVC-Aircraft and CUB-200-2011 indicate that HybridNet achieves one of the top performance on these datasets. The dataset and code are available at https://github.com/hshustc/vegfru.
| ['Yushan Feng', 'Saihui Hou', 'Zilei Wang'] | 2017-10-01 | null | null | null | iccv-2017-10 | ['fine-grained-visual-categorization'] | ['computer-vision'] | [-4.04687107e-01 -4.29559708e-01 -4.35106099e-01 -3.47607136e-01
-2.10209474e-01 -6.32049143e-01 4.25134420e-01 3.72982681e-01
-1.17855616e-01 2.85451680e-01 2.03800157e-01 -7.54776821e-02
2.05936283e-01 -1.19082916e+00 -4.66535151e-01 -7.31487811e-01
6.02328628e-02 -9.84877571e-02 2.50209391e-01 -1.72847658e-01
-8.19675997e-03 2.41438374e-01 -1.72492445e+00 5.35430849e-01
7.92242765e-01 1.65339506e+00 4.93770927e-01 3.15943480e-01
-3.07816982e-01 7.41227865e-01 -2.77513236e-01 -1.68080762e-01
2.12149128e-01 -7.86520615e-02 -6.39616609e-01 3.39794457e-01
4.12392199e-01 -4.67946410e-01 -1.96767032e-01 1.26303327e+00
3.08624893e-01 -1.31831884e-01 9.04588759e-01 -1.38625085e+00
-1.40958035e+00 7.64305472e-01 -7.57597208e-01 1.31831467e-01
-8.44890550e-02 1.28844947e-01 8.20003331e-01 -9.45526600e-01
3.02567631e-01 1.47642136e+00 5.11415839e-01 1.12172328e-01
-1.04038107e+00 -9.62906539e-01 5.52780211e-01 4.40113574e-01
-1.86476994e+00 -8.92085060e-02 4.04100299e-01 -7.16118455e-01
4.47924048e-01 2.72606105e-01 5.79037964e-01 6.40670002e-01
-1.30548954e-01 6.70565486e-01 1.12894583e+00 -1.62628278e-01
-5.07203937e-02 1.06061958e-01 5.89266598e-01 7.77460396e-01
2.40977734e-01 5.53677604e-02 1.22748151e-01 1.81390598e-01
6.49816513e-01 3.41823310e-01 -1.70337930e-01 -4.77525949e-01
-1.41546547e+00 1.07435977e+00 1.14036894e+00 3.89160037e-01
-2.98406690e-01 -1.20010883e-01 7.24235237e-01 2.49410406e-01
4.13055986e-01 -1.76125720e-01 -4.66057450e-01 7.03995228e-01
-5.53985238e-01 1.22509841e-02 3.94036859e-01 1.36410153e+00
1.06230748e+00 -1.30971139e-02 -5.26058733e-01 1.21169889e+00
4.90365505e-01 5.59116900e-01 4.99257237e-01 -8.20141792e-01
4.46355850e-01 7.91678727e-01 1.99269541e-02 -1.39318228e+00
-4.21417743e-01 -1.63787797e-01 -1.41599035e+00 -6.23715185e-02
1.86668739e-01 2.65312284e-01 -1.14426804e+00 1.48486996e+00
3.87659192e-01 -1.91787332e-02 -5.40824160e-02 1.12549925e+00
1.87121856e+00 8.97528052e-01 6.77583575e-01 7.93306082e-02
1.86644137e+00 -1.25452471e+00 -4.24518496e-01 8.93764868e-02
2.79730767e-01 -6.90606236e-01 1.16201937e+00 2.81585693e-01
-4.69128340e-01 -1.11455798e+00 -1.09725261e+00 -2.43269160e-01
-7.98906267e-01 5.80588996e-01 6.90510035e-01 5.11416137e-01
-8.24383438e-01 2.87502915e-01 -1.90340176e-01 -4.16183203e-01
4.26799923e-01 -9.69109833e-02 -3.59942585e-01 -4.10066187e-01
-1.35561442e+00 4.60516214e-01 8.85110378e-01 1.40447557e-01
-9.48437810e-01 -6.13885760e-01 -1.01679099e+00 3.18122476e-01
2.91418999e-01 -2.87177771e-01 9.64963436e-01 -7.44201660e-01
-8.75172198e-01 1.11917794e+00 2.71748692e-01 -1.09468788e-01
2.70103395e-01 3.45901370e-01 -6.36716783e-01 1.67885542e-01
4.26804543e-01 9.68242407e-01 6.84187412e-01 -1.10384583e+00
-9.92174506e-01 -2.43089408e-01 3.99401516e-01 -6.82534575e-02
-5.15098512e-01 8.48478153e-02 -6.24015868e-01 -9.09940124e-01
-1.13838866e-01 -5.93683898e-01 -4.87691537e-02 -2.76325177e-02
-3.90365690e-01 -6.62234604e-01 7.22370565e-01 -7.45852888e-01
1.34619057e+00 -2.29008222e+00 -8.15126225e-02 2.82992944e-02
4.31075364e-01 2.69723743e-01 -3.88985157e-01 2.24010006e-01
6.50063716e-03 2.34413713e-01 6.34007305e-02 3.17029834e-01
3.11304748e-01 -1.07306518e-01 -5.58819473e-02 4.07230139e-01
1.19825944e-01 9.92875516e-01 -8.08830023e-01 -9.28406179e-01
5.14596164e-01 2.30877012e-01 -1.33749008e-01 1.37748569e-01
-1.25905022e-01 7.77478442e-02 -8.57297838e-01 1.03451324e+00
1.25157356e+00 -3.85736912e-01 1.30968317e-01 -7.00631917e-01
-2.95201451e-01 -4.87802029e-01 -1.22271490e+00 1.32149422e+00
-2.62267917e-01 7.71091878e-02 1.71519384e-01 -9.97894168e-01
1.09647012e+00 -2.81960424e-02 2.32068434e-01 -7.06035733e-01
4.36282039e-01 8.60476792e-02 -2.64012873e-01 -2.26050958e-01
3.46457601e-01 2.18218431e-01 -4.27530944e-01 -2.40005646e-02
1.76025733e-01 5.80308959e-02 6.47058070e-01 1.35181565e-02
3.80537331e-01 1.06980957e-01 6.78422749e-01 -5.35707057e-01
7.84793735e-01 2.46037364e-01 7.61334419e-01 4.99649882e-01
-6.05202019e-01 5.37317753e-01 3.40681197e-03 -5.11189222e-01
-8.87780309e-01 -1.10101819e+00 -3.63604158e-01 1.48585629e+00
5.17721355e-01 -3.70344281e-01 -5.95139325e-01 -6.12446129e-01
3.49315286e-01 3.54923248e-01 -9.07199502e-01 -1.14699498e-01
-2.35693470e-01 -5.33059120e-01 4.02735621e-01 6.59141839e-01
1.28832173e+00 -1.38920355e+00 -3.12745780e-01 1.04162127e-01
-4.21508342e-01 -9.32795942e-01 -7.19609499e-01 4.81192116e-03
-2.72716671e-01 -1.23510849e+00 -9.70374107e-01 -1.21361077e+00
3.09626430e-01 8.21386278e-01 1.25404835e+00 1.47509992e-01
-3.82375509e-01 -7.47309476e-02 -7.34697104e-01 -2.97567070e-01
-1.82755932e-01 -6.62063807e-02 -2.16451719e-01 5.05106002e-02
6.62585914e-01 -5.34476936e-02 -7.34810174e-01 6.88421190e-01
-6.08661950e-01 6.66726828e-02 5.15335560e-01 8.21262777e-01
7.83073723e-01 2.84000218e-01 4.95250136e-01 -5.50429046e-01
1.89824730e-01 -7.32725859e-01 -4.89408016e-01 4.72487271e-01
-1.76746190e-01 -4.73141044e-01 9.11709666e-01 -6.49979293e-01
-1.02919710e+00 -5.96940741e-02 -1.75295752e-02 -5.81223845e-01
-5.21406949e-01 2.35767707e-01 -3.63159537e-01 -5.78568317e-02
4.33670044e-01 1.80254430e-01 -5.52400768e-01 -6.68123841e-01
7.35905051e-01 1.07848287e+00 5.79903007e-01 -5.51138639e-01
5.66950917e-01 9.02229398e-02 -3.46402466e-01 -7.54969478e-01
-7.88007438e-01 -6.34123981e-01 -6.57296777e-01 -1.27398565e-01
1.22301197e+00 -1.30070615e+00 -9.05062318e-01 6.61463439e-01
-7.46060073e-01 -1.56497017e-01 1.82871763e-02 3.52491647e-01
-2.90674448e-01 3.50293785e-01 -8.10013592e-01 -2.55552590e-01
-5.25141537e-01 -9.64015126e-01 1.08325279e+00 4.87109900e-01
3.25112551e-01 -5.11114657e-01 -5.46636999e-01 2.59304374e-01
2.24946082e-01 2.72443235e-01 8.97364736e-01 -1.11547679e-01
-3.43183249e-01 3.19690317e-01 -1.11149657e+00 5.22643268e-01
3.91694695e-01 3.12981792e-02 -7.67881870e-01 -4.33276445e-01
-5.20138204e-01 -7.66928494e-01 1.26846397e+00 3.58450770e-01
1.39444995e+00 -5.60881756e-02 -3.40125322e-01 5.83871126e-01
1.59971082e+00 2.57741064e-01 1.91550031e-01 3.42048973e-01
9.98403549e-01 6.19155884e-01 1.06188166e+00 3.90538514e-01
6.82730436e-01 5.46657205e-01 4.76435810e-01 -2.90750146e-01
-3.65899146e-01 -2.35700533e-01 -8.07788819e-02 8.37409317e-01
-8.68103430e-02 -6.04281090e-02 -6.66335881e-01 5.57256401e-01
-1.59601665e+00 -8.82297456e-01 -3.17808211e-01 1.69324887e+00
5.72480619e-01 -3.10648769e-01 1.32054090e-01 5.87115670e-03
9.95023727e-01 2.24005058e-01 -6.75452113e-01 -2.48597786e-01
-1.92189932e-01 -8.33798870e-02 6.69871151e-01 -1.40402585e-01
-1.79768777e+00 1.03330743e+00 5.51227379e+00 1.30803394e+00
-1.06153274e+00 1.88980401e-01 7.25813270e-01 5.28735518e-01
2.39440873e-01 -5.06405592e-01 -8.66807878e-01 6.36769354e-01
3.85876149e-01 -4.34315726e-02 2.49081090e-01 9.74071205e-01
-3.62411737e-02 1.05108283e-01 -5.41366220e-01 1.05430126e+00
-9.83926132e-02 -1.13837004e+00 6.57677427e-02 -2.51378566e-01
6.41179204e-01 -1.78751037e-01 -8.55126679e-02 6.30419910e-01
6.17218852e-01 -9.52620924e-01 1.13213062e+00 1.18770361e-01
1.25645244e+00 -7.30400681e-01 6.29737258e-01 4.01046097e-01
-2.23112106e+00 -4.00463521e-01 -1.07867813e+00 1.45847812e-01
-3.92969579e-01 3.74015838e-01 6.59814253e-02 8.40632558e-01
1.38746691e+00 1.11642826e+00 -9.03844595e-01 8.15869749e-01
-8.36599842e-02 4.45891529e-01 -1.11768626e-01 -9.73189175e-02
4.40126389e-01 -3.31907302e-01 -2.76754081e-01 1.31967521e+00
3.25827092e-01 3.23229373e-01 7.95205712e-01 5.07514656e-01
-2.99586914e-04 4.99594510e-01 -2.92178780e-01 -5.10231126e-03
4.09268320e-01 1.67360985e+00 -9.69671488e-01 -5.52740932e-01
-6.08215451e-01 7.31580675e-01 5.02213717e-01 2.36976445e-01
-9.46835518e-01 -5.00489295e-01 5.95993221e-01 -3.57483298e-01
7.85125852e-01 1.83396488e-01 1.22478262e-01 -1.37591875e+00
-3.74780327e-01 -8.93051386e-01 8.49350214e-01 -7.20007420e-01
-1.56084120e+00 6.56594515e-01 -5.48067270e-03 -1.31867385e+00
3.09392303e-01 -6.86265171e-01 -2.45856375e-01 8.39728236e-01
-1.62709367e+00 -1.61673129e+00 -8.05087805e-01 7.24907994e-01
5.90389967e-01 -9.87432525e-02 8.06732953e-01 4.98595953e-01
-6.17808223e-01 4.48881775e-01 2.14820683e-01 3.23824763e-01
7.11018801e-01 -1.06293178e+00 2.19614476e-01 4.64610040e-01
-2.00444147e-01 4.05809373e-01 2.11899653e-01 -4.83529896e-01
-8.97309840e-01 -1.67788506e+00 5.22416234e-01 3.98910325e-03
5.39710522e-01 -4.50357050e-01 -7.58792520e-01 4.58481878e-01
1.68745935e-01 3.36086065e-01 3.85826647e-01 -1.94686100e-01
-7.01693833e-01 -3.44343096e-01 -1.31288815e+00 1.55487493e-01
1.15103042e+00 -2.39207387e-01 -5.60993612e-01 3.25142145e-01
8.14249396e-01 -1.12045124e-01 -1.04458761e+00 4.73190159e-01
5.15128732e-01 -8.50803018e-01 1.23338985e+00 -3.34906697e-01
4.47978765e-01 -6.99775696e-01 -5.92196107e-01 -1.33178926e+00
-1.16582453e+00 4.87696052e-01 1.21170364e-01 1.50883174e+00
-3.32987845e-01 -4.76712197e-01 3.97339612e-01 -2.82007277e-01
-1.47552565e-01 -4.88353014e-01 -3.12933624e-01 -7.78792202e-01
2.23125800e-01 -8.11156537e-03 1.10114896e+00 9.71529365e-01
-5.79061449e-01 1.64072350e-01 -3.11258495e-01 6.37354627e-02
7.50348806e-01 8.71819496e-01 5.05239844e-01 -1.29761434e+00
1.05666533e-01 -4.73501980e-01 -4.02478844e-01 -9.38472211e-01
1.66916642e-02 -1.12545836e+00 8.00120980e-02 -1.71936893e+00
6.37465358e-01 -5.99254310e-01 -5.60778022e-01 6.53270543e-01
-2.33236998e-01 7.95112729e-01 5.68190157e-01 2.00740620e-01
-5.81611931e-01 3.87964696e-01 1.51948690e+00 -6.91750169e-01
1.42253041e-01 -3.27981234e-01 -1.04759943e+00 6.81463063e-01
8.79746020e-01 6.18926920e-02 -2.17136219e-01 -2.83184588e-01
-4.97277468e-01 -3.66994053e-01 3.78402680e-01 -8.13479066e-01
-1.93337455e-01 -4.39800441e-01 8.24086070e-01 -8.87957931e-01
-5.70935421e-02 -7.96003282e-01 2.82370865e-01 6.71507597e-01
-1.10191271e-01 -3.10693741e-01 1.26646906e-01 3.40113610e-01
-3.99042040e-01 8.76447633e-02 1.14253879e+00 -2.14731425e-01
-1.52572381e+00 6.77164495e-01 -5.57186306e-02 -5.05087152e-02
1.34150374e+00 4.79758680e-02 -5.51531196e-01 1.32667273e-01
-7.70187259e-01 5.58442473e-01 3.18878084e-01 7.24629045e-01
4.09579039e-01 -1.82410502e+00 -1.00424325e+00 7.57379979e-02
7.21816063e-01 -3.15439671e-01 6.98094964e-01 2.28473231e-01
-5.60651720e-01 6.46919668e-01 -7.36054838e-01 -5.27294278e-01
-1.46010244e+00 1.23513770e+00 1.83358029e-01 -2.51436263e-01
-5.36772013e-01 6.89461768e-01 9.76556122e-01 -6.01128519e-01
7.73885921e-02 -5.45575798e-01 -1.02976286e+00 3.76479506e-01
6.79809391e-01 3.13885778e-01 -1.36534497e-01 -1.15328598e+00
-3.94730508e-01 8.85140657e-01 7.33564645e-02 9.51838017e-01
9.59107339e-01 -3.39915723e-01 -1.39235735e-01 2.22092837e-01
1.31003392e+00 -3.21301103e-01 -1.15122402e+00 -3.77299845e-01
-4.18652803e-01 -5.40245354e-01 -3.35261486e-02 -7.50445962e-01
-1.38012516e+00 9.78961289e-01 8.55351448e-01 3.27132940e-01
1.40673828e+00 1.20479561e-01 7.02944756e-01 1.25453740e-01
6.82777822e-01 -7.36189246e-01 -1.92531392e-01 3.18577826e-01
9.15020764e-01 -1.32301474e+00 -2.25559212e-02 -9.23778713e-01
-6.06929243e-01 8.13274443e-01 7.87591398e-01 -1.60519332e-01
7.96459794e-01 -1.38118804e-01 2.54722506e-01 1.49134502e-01
-2.35933185e-01 -4.83085155e-01 4.36844975e-01 8.13576341e-01
2.24185303e-01 6.38828218e-01 -5.16382337e-01 9.46151495e-01
-1.21008940e-02 1.53049052e-01 1.66110843e-02 5.64179480e-01
-6.37342393e-01 -7.61931777e-01 -5.21433175e-01 6.60870194e-01
-1.76549911e-01 -1.50366291e-01 8.90084635e-03 7.53506660e-01
9.13637042e-01 1.38244569e+00 6.82646632e-02 -4.53051150e-01
4.47798997e-01 -4.86874670e-01 3.28872591e-01 -5.25470257e-01
-6.77737832e-01 -1.92485861e-02 -4.23088372e-02 -6.27273023e-01
-4.49120849e-01 -3.24840546e-01 -1.09398866e+00 -6.15153432e-01
-8.04645419e-02 7.20096156e-02 1.77805573e-01 3.56034756e-01
-3.52399275e-02 5.20958900e-01 8.73059273e-01 -9.00229216e-01
-4.15420145e-01 -1.00387180e+00 -1.12896037e+00 5.79783797e-01
6.89777434e-02 -1.06046212e+00 -7.57766664e-02 1.39614165e-01] | [9.679930686950684, 2.072456121444702] |
f5426ea2-3536-4eb7-82b8-596e395d70d5 | a-photometrically-calibrated-benchmark-for | 1607.02555 | null | http://arxiv.org/abs/1607.02555v2 | http://arxiv.org/pdf/1607.02555v2.pdf | A Photometrically Calibrated Benchmark For Monocular Visual Odometry | We present a dataset for evaluating the tracking accuracy of monocular visual
odometry and SLAM methods. It contains 50 real-world sequences comprising more
than 100 minutes of video, recorded across dozens of different environments --
ranging from narrow indoor corridors to wide outdoor scenes. All sequences
contain mostly exploring camera motion, starting and ending at the same
position. This allows to evaluate tracking accuracy via the accumulated drift
from start to end, without requiring ground truth for the full sequence. In
contrast to existing datasets, all sequences are photometrically calibrated. We
provide exposure times for each frame as reported by the sensor, the camera
response function, and dense lens attenuation factors. We also propose a novel,
simple approach to non-parametric vignette calibration, which requires minimal
set-up and is easy to reproduce. Finally, we thoroughly evaluate two existing
methods (ORB-SLAM and DSO) on the dataset, including an analysis of the effect
of image resolution, camera field of view, and the camera motion direction. | ['Jakob Engel', 'Daniel Cremers', 'Vladyslav Usenko'] | 2016-07-09 | null | null | null | null | ['monocular-visual-odometry'] | ['robots'] | [-2.77681518e-02 -5.19667149e-01 1.03642315e-01 -4.12136495e-01
-4.42495197e-01 -9.96612787e-01 7.23758042e-01 -2.95663267e-01
-5.32908797e-01 8.50578487e-01 -4.62908074e-02 7.73516148e-02
1.38131440e-01 -3.11589152e-01 -9.27690148e-01 -6.25023603e-01
-1.65399626e-01 4.60370541e-01 4.65073913e-01 -1.90940220e-02
2.91453242e-01 5.19372702e-01 -1.26426935e+00 -7.84490943e-01
5.85987270e-01 7.74821937e-01 4.32202667e-01 9.44335461e-01
4.96885777e-01 6.64768875e-01 -5.54667175e-01 -3.04982990e-01
6.13560021e-01 -1.75768808e-01 -2.73544043e-01 4.45227146e-01
1.00566089e+00 -5.37372351e-01 -7.10634410e-01 1.24506247e+00
6.42521262e-01 2.00934947e-01 1.11629076e-01 -1.10017943e+00
-2.19423637e-01 -2.78169960e-01 -3.60794097e-01 2.46183753e-01
9.32971299e-01 6.14482999e-01 3.70017678e-01 -5.39114237e-01
1.20963860e+00 9.50718284e-01 8.91231477e-01 1.15933418e-01
-1.15449333e+00 -3.56945753e-01 -1.34914085e-01 9.19431299e-02
-1.38371575e+00 -6.61402106e-01 3.32205594e-01 -7.07443833e-01
5.90928078e-01 -6.06408231e-02 7.85857081e-01 1.02865386e+00
3.43544602e-01 2.03758925e-02 9.42230821e-01 -3.08062971e-01
1.86936229e-01 5.18817231e-02 -4.44450378e-02 5.87468326e-01
7.36283839e-01 5.03441870e-01 -6.97933018e-01 7.40430830e-03
1.02650487e+00 -7.72899985e-02 -6.63862586e-01 -1.11110890e+00
-1.65813994e+00 2.72578120e-01 3.33425671e-01 -1.14670381e-01
-1.22727759e-01 4.26055610e-01 1.10743605e-01 3.53980929e-01
8.50599259e-02 3.29832077e-01 -2.10758880e-01 -3.62793952e-01
-7.09264040e-01 1.90879703e-01 5.92800498e-01 1.61402607e+00
1.09986651e+00 6.17591180e-02 2.69270271e-01 2.03969374e-01
8.62324461e-02 9.15636241e-01 2.61360615e-01 -1.39958072e+00
4.70558465e-01 1.21443048e-01 7.54370570e-01 -9.80973542e-01
-4.82759207e-01 -2.38441661e-01 -2.70641565e-01 2.18682945e-01
5.66660881e-01 -2.88804382e-01 -5.98882139e-01 1.57664013e+00
4.75041032e-01 3.70402098e-01 -1.42644092e-01 1.33370793e+00
4.59896475e-01 2.11327925e-01 -5.97363412e-01 -2.33345255e-01
1.03669369e+00 -7.88268864e-01 -8.38357806e-01 -5.52714050e-01
4.92540061e-01 -1.05045044e+00 8.59622419e-01 3.58601809e-01
-8.67659450e-01 -5.47918379e-01 -1.21870184e+00 -1.58121381e-02
8.64157006e-02 1.89789981e-01 3.52405310e-01 6.51417315e-01
-1.16170025e+00 3.57566923e-01 -7.80868590e-01 -6.72194242e-01
-3.99408042e-01 2.01775759e-01 -5.02784967e-01 -3.00035119e-01
-9.60345745e-01 1.20174587e+00 2.35665828e-01 -1.11472815e-01
-9.06387925e-01 -5.43689191e-01 -1.07583284e+00 -4.67116863e-01
3.95067364e-01 -8.49069834e-01 1.27729142e+00 -6.69727564e-01
-1.67300749e+00 8.90761435e-01 -4.08403963e-01 -5.73702216e-01
7.57637799e-01 -5.81746697e-01 -3.23983729e-01 -2.98909820e-03
8.10206011e-02 4.98564065e-01 5.80873549e-01 -1.19481075e+00
-3.72666240e-01 -3.63811582e-01 4.67036553e-02 5.63476682e-01
5.31570673e-01 -2.71371126e-01 -9.37527657e-01 -1.61128089e-01
5.06950676e-01 -1.35916483e+00 -2.59181947e-01 2.82251295e-02
-1.86392695e-01 8.73486698e-01 6.00431621e-01 -4.60856497e-01
7.63761163e-01 -2.17229843e+00 8.14973637e-02 -1.29740953e-01
-1.22831315e-01 -2.26110920e-01 1.45538241e-01 3.64245236e-01
2.59814978e-01 -6.08681142e-01 1.04038686e-01 -4.35212106e-01
-1.82773679e-01 2.23183990e-01 -2.48793021e-01 1.12928367e+00
-5.54592431e-01 6.02845192e-01 -1.04212022e+00 -1.11480191e-01
8.42185736e-01 2.76975036e-01 -3.18621546e-01 1.90905660e-01
5.62395938e-02 9.29844141e-01 1.76348984e-01 5.73581040e-01
9.69238639e-01 -6.98012933e-02 1.11673750e-01 -1.10977791e-01
-4.48037684e-01 4.98468518e-01 -1.56719565e+00 2.19685912e+00
-2.60372013e-01 1.16804957e+00 -2.90440526e-02 -1.79674640e-01
8.90017986e-01 -5.67103028e-02 4.41660196e-01 -7.91038215e-01
1.61997512e-01 1.54742837e-01 -3.88179958e-01 -3.80538851e-01
1.02949893e+00 1.55202255e-01 -2.81266067e-02 6.19183443e-02
1.22446649e-01 -4.90834475e-01 3.59693229e-01 1.15839675e-01
8.95003974e-01 5.02338111e-01 5.32761753e-01 -1.73975140e-01
3.40708345e-01 3.39570224e-01 5.88762045e-01 7.01817393e-01
-3.07811171e-01 7.93341339e-01 -3.00630871e-02 -5.29551268e-01
-1.22527874e+00 -1.23022473e+00 -5.33347167e-02 3.49212319e-01
1.11193836e+00 -2.86621004e-01 -4.95135725e-01 1.55947521e-01
1.30805090e-01 3.07103336e-01 -2.90881097e-01 2.21039038e-02
-5.77266932e-01 -3.64625961e-01 3.47897649e-01 1.10447399e-01
5.81334829e-01 -3.16091269e-01 -9.35295045e-01 5.29591069e-02
-4.22810018e-01 -1.68537378e+00 -6.39820278e-01 -1.75096527e-01
-8.85736644e-01 -1.25284040e+00 -4.20378119e-01 -3.11628759e-01
5.72495341e-01 8.84014070e-01 1.23035479e+00 -2.73470044e-01
-1.12971686e-01 7.08758056e-01 -4.34019342e-02 -1.25630230e-01
-1.02202669e-01 -3.02457064e-01 4.56448585e-01 -2.52678931e-01
1.77581325e-01 -2.02246472e-01 -5.22385299e-01 7.27541327e-01
-3.93853724e-01 -2.64072046e-02 1.19729474e-01 4.51073825e-01
5.27256072e-01 -5.93660831e-01 -6.14467382e-01 -2.68259138e-01
-1.37859434e-01 -1.67029619e-01 -1.39948857e+00 -1.03986628e-01
-3.18984598e-01 -1.79231569e-01 1.66348174e-01 -3.58178020e-01
-8.57324839e-01 2.64937192e-01 4.99948978e-01 -5.34786403e-01
-6.95937872e-02 -3.64134125e-02 9.21836123e-03 -5.89368880e-01
8.35329711e-01 1.97297022e-01 -2.23236736e-02 -1.12088308e-01
2.97461748e-01 1.76957294e-01 1.25018322e+00 -3.30861717e-01
1.06834888e+00 1.00567174e+00 9.81949642e-02 -1.03529155e+00
-4.67053741e-01 -8.38089049e-01 -8.02201569e-01 -4.72678840e-01
6.85798943e-01 -1.37904549e+00 -8.05037379e-01 5.87029994e-01
-1.11969173e+00 -2.48030275e-01 -4.88552339e-02 1.14881647e+00
-6.91230118e-01 6.55502021e-01 -3.31914127e-01 -5.47753513e-01
1.99465707e-01 -1.27198756e+00 1.13112032e+00 3.03301066e-01
-4.22753952e-02 -9.87139821e-01 5.48814595e-01 6.50704876e-02
1.53437421e-01 4.29002225e-01 -2.49765828e-01 3.14033121e-01
-1.16194689e+00 1.81300920e-02 -9.30458829e-02 -1.07504413e-01
5.10513820e-02 -6.16313294e-02 -8.22206020e-01 -7.32640266e-01
5.55807278e-02 7.72796571e-02 4.25115108e-01 6.88700974e-01
1.24513686e-01 1.04567811e-01 -3.59551787e-01 1.15701365e+00
1.75202763e+00 2.27064267e-01 7.72797465e-01 9.61790740e-01
6.58630848e-01 1.60920724e-01 9.28836763e-01 4.78292912e-01
4.57337707e-01 1.20888603e+00 6.75782144e-01 1.12538852e-01
-8.03525094e-03 -1.68971643e-01 5.61162889e-01 4.21405017e-01
-2.13009253e-01 -1.07400902e-01 -7.22506940e-01 4.39594179e-01
-1.47631681e+00 -9.20632660e-01 -4.43189293e-01 2.80972099e+00
2.99657047e-01 -5.06454036e-02 -6.75656870e-02 -2.79711306e-01
7.13364244e-01 2.88775921e-01 -4.59456563e-01 1.89023077e-01
-2.37846911e-01 -5.78372180e-01 1.34008729e+00 8.68438125e-01
-9.37355399e-01 9.21590865e-01 7.10340118e+00 -2.68068343e-01
-1.25155759e+00 -2.48106159e-02 -4.03554499e-01 -2.83474356e-01
3.01775690e-02 4.50033516e-01 -9.66644585e-01 5.52798808e-01
8.97262931e-01 -1.05009921e-01 6.08573377e-01 8.77149105e-01
1.93811223e-01 -7.03299284e-01 -1.02542019e+00 1.30583465e+00
1.62406847e-01 -1.23167348e+00 -6.10450566e-01 1.61232159e-01
8.46564949e-01 6.27788603e-01 -2.74680823e-01 -2.11669609e-01
5.66039085e-01 -4.01639193e-01 1.01310897e+00 4.32097197e-01
8.41089904e-01 -3.23489904e-01 6.41414404e-01 2.91334689e-01
-1.15680265e+00 1.22283250e-01 -5.74611306e-01 -2.18518302e-01
4.88572627e-01 5.16916811e-01 -8.53907585e-01 8.22092175e-01
7.61624455e-01 8.39718521e-01 -6.44736648e-01 1.35091293e+00
-2.33980924e-01 -2.37108860e-02 -4.18676227e-01 3.20393801e-01
8.97869021e-02 -5.81916571e-01 8.75706375e-01 9.54733849e-01
6.26051903e-01 -1.11876160e-01 1.69463724e-01 2.93925792e-01
2.94426262e-01 -5.27340412e-01 -9.59380567e-01 6.36156261e-01
5.99827290e-01 7.93143094e-01 -1.86876386e-01 -1.94808438e-01
-4.22127604e-01 1.02713287e+00 -7.73783326e-02 4.28394347e-01
-1.00268316e+00 -6.50911182e-02 1.02037382e+00 1.52159303e-01
2.41113022e-01 -9.07427788e-01 -1.28211662e-01 -1.47340798e+00
1.97433099e-01 -6.27170086e-01 7.37581402e-02 -1.26382923e+00
-4.68962193e-01 3.85003030e-01 6.55398965e-02 -1.89604998e+00
-4.33268368e-01 -4.09640580e-01 -9.70758572e-02 6.99061871e-01
-1.49052393e+00 -6.25780344e-01 -1.03412497e+00 5.63176751e-01
3.06054235e-01 8.57093409e-02 3.35107237e-01 5.45852780e-01
-4.69005145e-02 6.68368861e-02 4.26836461e-01 -2.08286308e-02
1.17731130e+00 -1.02523494e+00 7.19271779e-01 1.29329097e+00
1.10626116e-01 6.93759561e-01 1.20277035e+00 -4.82617408e-01
-1.71162844e+00 -7.40867972e-01 6.05530441e-01 -9.67023790e-01
5.58473110e-01 -3.47985357e-01 -4.98033047e-01 1.07541120e+00
8.42534229e-02 1.61084682e-01 -1.31384626e-01 -2.64443487e-01
1.65742934e-02 -1.61557660e-01 -9.01226521e-01 4.95103359e-01
1.19174957e+00 -3.91586453e-01 -3.44910979e-01 5.89838214e-02
6.07974529e-01 -1.32412326e+00 -5.26378334e-01 1.40920922e-01
6.17505133e-01 -1.42685688e+00 1.10987866e+00 1.87250093e-01
-2.66342580e-01 -8.58204663e-01 -3.92405748e-01 -1.29970241e+00
-2.80460209e-01 -9.25573885e-01 3.66069712e-02 7.59511590e-01
-2.25907534e-01 -6.64269269e-01 6.36409998e-01 2.02252969e-01
-8.80239978e-02 2.54739136e-01 -9.68394876e-01 -1.14341736e+00
-8.37443352e-01 -4.09880817e-01 4.48327601e-01 9.10566866e-01
-5.56328297e-01 9.66931507e-02 -8.31613123e-01 6.60413802e-01
1.01295197e+00 3.89465056e-02 1.61163747e+00 -1.05238688e+00
-2.52394557e-01 1.92714617e-01 -9.27455008e-01 -1.50267863e+00
-3.02277029e-01 -2.67179161e-01 8.07165131e-02 -1.20956826e+00
-1.33053884e-01 -9.56180543e-02 4.48636621e-01 -4.47184473e-01
7.13179708e-02 2.32374638e-01 2.13207006e-01 4.38886166e-01
-5.52001178e-01 1.78638577e-01 1.05663395e+00 1.89487457e-01
-1.88469097e-01 -1.01134337e-01 1.06091544e-01 7.40713358e-01
3.51264358e-01 -4.57805037e-01 -3.72961819e-01 -8.70488703e-01
2.32173637e-01 3.50961059e-01 6.48669600e-01 -1.37836623e+00
3.83479327e-01 -2.20581874e-01 3.36829275e-01 -8.20199847e-01
4.98689204e-01 -9.55958366e-01 7.56937981e-01 5.38361371e-01
2.37919375e-01 4.38753009e-01 1.17582254e-01 5.32629371e-01
-1.12570897e-01 -9.18325186e-02 8.57546628e-01 -1.36036992e-01
-1.23507977e+00 1.95488766e-01 -3.19324657e-02 1.24713227e-01
1.03961468e+00 -6.25733256e-01 -6.09625638e-01 -6.26619160e-01
-3.94285023e-01 1.28530920e-01 1.42320013e+00 3.74857217e-01
2.51485705e-01 -1.41357291e+00 -2.60674715e-01 3.29542875e-01
2.94676244e-01 -7.13168979e-02 1.43567696e-01 9.71557319e-01
-1.03512394e+00 5.40085375e-01 -3.48158002e-01 -1.26065481e+00
-1.19210720e+00 5.06543458e-01 5.19262373e-01 1.39220700e-01
-6.37927353e-01 4.13964748e-01 2.98761800e-02 -5.13824582e-01
1.45940423e-01 -5.05353153e-01 3.45803171e-01 -4.14488435e-01
4.83516455e-01 5.26409686e-01 6.30806759e-02 -7.92301476e-01
-5.09871364e-01 1.05537951e+00 5.36023617e-01 -3.42225313e-01
7.32407570e-01 -8.54213178e-01 1.64348304e-01 4.58840728e-01
9.78397846e-01 1.73586532e-01 -1.84439385e+00 -2.32969299e-01
-2.00529266e-02 -1.00930953e+00 -2.47004569e-01 -3.32192481e-01
-5.66416562e-01 4.45582300e-01 7.85784841e-01 -2.73775131e-01
7.83116639e-01 -2.59683281e-01 3.52029979e-01 5.72096527e-01
1.00022399e+00 -9.59486663e-01 -2.15244666e-01 8.22734177e-01
4.63358045e-01 -1.37329435e+00 4.03559953e-01 -4.01913136e-01
-4.97351974e-01 9.34509754e-01 4.59398419e-01 -1.74681306e-01
-2.49365680e-02 2.27176204e-01 3.63394409e-01 4.25847657e-02
-3.40311289e-01 -1.58065796e-01 8.06167908e-03 7.98827589e-01
1.28128737e-01 -1.96869388e-01 4.07465957e-02 -4.94629443e-01
-4.17736381e-01 9.92906615e-02 9.26389515e-01 9.81417477e-01
-4.14995104e-01 -7.35190213e-01 -7.98342705e-01 -3.51078480e-01
9.15405974e-02 2.28578776e-01 -1.55928701e-01 1.10819721e+00
-1.79541156e-01 6.87709153e-01 1.52451813e-01 -3.02143753e-01
4.42346245e-01 -3.72331798e-01 7.67946303e-01 -2.34759241e-01
-1.28081366e-01 -3.70128043e-02 2.17886791e-01 -9.37119067e-01
-6.23953581e-01 -8.72736394e-01 -1.02270854e+00 -6.33385062e-01
-3.35634857e-01 -1.34906530e-01 9.31365371e-01 7.26237357e-01
2.39430636e-01 1.23071894e-01 5.72279572e-01 -1.21452165e+00
-3.01778466e-01 -7.16376722e-01 -4.20335770e-01 3.52905661e-01
7.06245959e-01 -8.23141336e-01 -4.80225593e-01 3.30780119e-01] | [7.638036727905273, -2.16703724861145] |
8d611aba-f886-4858-a2c5-21ca0deb18c7 | assessing-project-level-fine-tuning-of-ml4se | 2206.03333 | null | https://arxiv.org/abs/2206.03333v1 | https://arxiv.org/pdf/2206.03333v1.pdf | Assessing Project-Level Fine-Tuning of ML4SE Models | Machine Learning for Software Engineering (ML4SE) is an actively growing research area that focuses on methods that help programmers in their work. In order to apply the developed methods in practice, they need to achieve reasonable quality in order to help rather than distract developers. While the development of new approaches to code representation and data collection improves the overall quality of the models, it does not take into account the information that we can get from the project at hand. In this work, we investigate how the model's quality can be improved if we target a specific project. We develop a framework to assess quality improvements that models can get after fine-tuning for the method name prediction task on a particular project. We evaluate three models of different complexity and compare their quality in three settings: trained on a large dataset of Java projects, further fine-tuned on the data from a particular project, and trained from scratch on this data. We show that per-project fine-tuning can greatly improve the models' quality as they capture the project's domain and naming conventions. We open-source the tool we used for data collection, as well as the code to run the experiments: https://zenodo.org/record/6040745. | ['Timofey Bryksin', 'Egor Spirin', 'Sergey Zhuravlev', 'Egor Bogomolov'] | 2022-06-07 | null | null | null | null | ['method-name-prediction'] | ['natural-language-processing'] | [-2.17663735e-01 -1.32665187e-01 -1.62456125e-01 -5.19776940e-01
-5.32230437e-01 -4.59030211e-01 2.42403284e-01 3.16612482e-01
-2.43148670e-01 1.17108025e-01 1.88766532e-02 -3.99774551e-01
-1.35262042e-01 -7.54694402e-01 -7.07026899e-01 -6.29667416e-02
4.02317405e-01 2.44063109e-01 3.55386734e-01 -2.62438446e-01
6.17595911e-01 2.85439014e-01 -1.58755386e+00 6.67858958e-01
8.59266579e-01 3.35526228e-01 2.20249236e-01 6.53608739e-01
-2.89423645e-01 7.95197606e-01 -7.03373969e-01 -4.48944509e-01
2.73062289e-01 -2.40876660e-01 -9.38306272e-01 -2.69590169e-01
4.39878434e-01 -8.33687037e-02 3.16990137e-01 9.15386081e-01
3.10586423e-01 -4.11658794e-01 4.48250860e-01 -1.21809530e+00
-4.39618975e-01 8.38343382e-01 -4.34241652e-01 8.83515328e-02
2.60343283e-01 1.70844793e-01 9.71598864e-01 -7.47394681e-01
6.84497833e-01 7.23671973e-01 9.94007468e-01 5.75957298e-01
-1.22985899e+00 -6.23380661e-01 -2.04689577e-01 7.16349483e-02
-1.21178269e+00 -4.55768704e-01 5.89401186e-01 -1.19740677e+00
1.08358634e+00 2.49561742e-01 3.28077585e-01 8.02452981e-01
-4.54667136e-02 2.58704215e-01 8.79282594e-01 -7.80943155e-01
1.18699111e-01 5.60116351e-01 5.79241335e-01 7.14270294e-01
2.65833318e-01 -1.29125327e-01 -3.22520345e-01 -4.49672908e-01
4.84828472e-01 6.97288737e-02 -1.43067881e-01 -6.02716684e-01
-1.16042542e+00 8.57625246e-01 -1.08280480e-02 7.78060734e-01
-2.26215437e-01 2.23138496e-01 3.87148947e-01 7.09773183e-01
3.10751021e-01 9.61140275e-01 -9.25974190e-01 -7.83354521e-01
-9.42937791e-01 3.48301709e-01 1.19633305e+00 1.03875828e+00
1.02037466e+00 -3.48706901e-01 6.02091700e-02 1.02493441e+00
1.67596906e-01 1.30198449e-01 5.13356566e-01 -1.11912501e+00
5.46821952e-01 1.08170962e+00 8.88208225e-02 -7.91153312e-01
-1.55494064e-01 -2.22070172e-01 -3.82323228e-02 5.62406063e-01
7.08721101e-01 -9.37834010e-02 -4.55445349e-01 1.55293918e+00
9.73213278e-03 -4.17548895e-01 -3.04480106e-01 2.82422394e-01
6.40788198e-01 2.27688670e-01 -2.24750653e-01 2.21861154e-01
9.77142334e-01 -1.07297063e+00 -2.64086336e-01 -2.81627744e-01
1.33970070e+00 -1.02609432e+00 1.31881821e+00 5.73833704e-01
-8.91309023e-01 -6.66539609e-01 -9.35106039e-01 4.53733057e-02
-4.44562882e-01 1.48486167e-01 4.54488814e-01 8.87776852e-01
-1.02352405e+00 9.88302350e-01 -9.37023520e-01 -5.01847982e-01
2.44762138e-01 2.16638550e-01 -4.15538013e-01 -3.91353257e-02
-5.27853608e-01 9.98099923e-01 1.55360833e-01 -4.35133189e-01
-6.20910645e-01 -8.43175590e-01 -4.35628742e-01 1.57149121e-01
4.07935232e-01 -5.57264149e-01 1.58816826e+00 -1.02349436e+00
-9.71021712e-01 8.55019867e-01 2.27147654e-01 1.90910339e-01
3.91290635e-01 -3.78948599e-01 2.04514787e-02 -5.91178656e-01
1.58776924e-01 1.80714339e-01 4.75877851e-01 -9.32404101e-01
-6.93098307e-01 -3.53051096e-01 3.05908263e-01 -4.82293963e-01
-6.05424583e-01 5.12005568e-01 -4.83634263e-01 -3.45067263e-01
-4.27600533e-01 -8.93566012e-01 -9.10856500e-02 -1.27112851e-01
-1.39706552e-01 -8.65975171e-02 1.57084554e-01 -6.96216524e-01
1.50950539e+00 -2.12936640e+00 2.01544762e-01 2.39063073e-02
2.12899938e-01 2.39017934e-01 -4.06467110e-01 7.19817162e-01
-2.03310579e-01 3.85553062e-01 -1.70580909e-01 -3.93135697e-02
8.04289803e-02 -6.62486441e-03 1.75130919e-01 1.44767731e-01
1.06950916e-01 4.77411330e-01 -7.68338382e-01 -2.81363130e-01
-9.50716957e-02 3.08501780e-01 -1.09518123e+00 4.18738991e-01
-2.89072335e-01 -3.36872153e-02 -3.93126190e-01 3.47171366e-01
3.52327406e-01 -3.30649018e-01 1.74908668e-01 1.68971881e-01
-1.45711944e-01 3.80424500e-01 -1.28432178e+00 1.69099200e+00
-8.56001794e-01 5.78512251e-01 -1.62328392e-01 -8.14274609e-01
1.06850457e+00 2.68121690e-01 3.03223997e-01 -4.61379707e-01
-3.28630179e-01 4.19186801e-01 2.91601419e-01 -9.14806128e-01
2.27734417e-01 5.21139875e-02 5.18631525e-02 8.98336887e-01
-4.95980531e-02 4.56493497e-02 4.25933778e-01 -2.82878987e-02
1.54119098e+00 2.50569940e-01 5.09081244e-01 -9.54588652e-02
3.43306601e-01 4.34778370e-02 4.44663376e-01 6.83152020e-01
3.04287821e-01 4.16549593e-01 8.20801914e-01 -6.13109767e-01
-1.21891975e+00 -3.96096468e-01 -1.91726405e-02 1.55108261e+00
-5.98866224e-01 -8.23506117e-01 -1.01544178e+00 -1.03138089e+00
7.16891587e-02 7.85278082e-01 -6.13945842e-01 -3.86630557e-02
-5.93231201e-01 -4.56433892e-01 5.36776602e-01 4.20245260e-01
-1.37667894e-01 -1.16124368e+00 -8.87160599e-01 3.09926420e-01
8.26636925e-02 -6.65247083e-01 -3.79088283e-01 7.47516751e-02
-7.71359205e-01 -1.36167395e+00 -4.23428148e-01 -3.42393607e-01
4.99011308e-01 8.09048116e-02 1.61869144e+00 6.09484136e-01
-2.44193047e-01 3.87132555e-01 -5.97491026e-01 -3.51491094e-01
-1.02319896e+00 4.45917904e-01 -4.23948646e-01 -5.48588932e-01
8.00079107e-01 -6.72645152e-01 -7.75557533e-02 2.56779522e-01
-8.13015044e-01 -8.96649361e-02 6.00219727e-01 6.92619860e-01
-1.53078020e-01 -1.15548581e-01 3.69401425e-01 -1.37301970e+00
5.85495293e-01 -6.29275382e-01 -6.72305346e-01 4.38397020e-01
-1.04814637e+00 2.84697562e-01 4.71303225e-01 -2.30855137e-01
-7.76785016e-01 -3.16362716e-02 -2.92989045e-01 1.74187142e-02
-3.48587215e-01 8.03958118e-01 1.31264031e-01 -1.27723679e-01
1.08902013e+00 -1.22538500e-01 -1.15234859e-01 -1.13939989e+00
2.70541877e-01 1.14368606e+00 -8.20518658e-02 -8.40840518e-01
6.05207801e-01 -1.82114154e-01 -5.03751934e-01 -5.28130829e-01
-6.12831831e-01 -4.29306269e-01 -7.61759758e-01 1.51191488e-01
3.41828704e-01 -5.55877328e-01 -3.77325952e-01 2.92687178e-01
-1.43657672e+00 -5.58674157e-01 -1.07982807e-01 2.45437771e-01
-3.59590232e-01 2.78170556e-01 -3.23263824e-01 -5.05446672e-01
-1.55989200e-01 -1.31231034e+00 7.36511171e-01 -1.19824842e-01
-4.70599085e-01 -9.82224286e-01 6.41434669e-01 3.26105535e-01
7.19347894e-01 1.79683268e-01 1.13543224e+00 -8.76926720e-01
-5.30409336e-01 -6.11090846e-02 -9.47590321e-02 7.75007606e-01
2.56148726e-01 4.31342810e-01 -9.31509733e-01 -2.14038610e-01
4.47966764e-03 -2.24430282e-02 4.84534711e-01 -1.11437932e-01
9.35723364e-01 -3.56000036e-01 -1.80518404e-01 4.07458633e-01
1.45207965e+00 -2.34205066e-03 3.66500378e-01 7.66607344e-01
8.34320903e-01 8.18827271e-01 5.65257549e-01 4.29407120e-01
3.12945157e-01 9.35980439e-01 3.22130173e-01 1.50171563e-01
4.44147252e-02 2.03095496e-01 2.87526309e-01 8.91191244e-01
-2.03002870e-01 3.79384696e-01 -1.51217055e+00 7.31024802e-01
-1.87664795e+00 -7.34049797e-01 -3.42572838e-01 2.21491599e+00
1.21458375e+00 1.12380825e-01 3.22240591e-01 1.25032052e-01
3.64352465e-01 -2.86638409e-01 -3.43648251e-03 -7.35190153e-01
6.57727003e-01 2.94791460e-01 7.65872225e-02 3.14418405e-01
-5.75613379e-01 4.95964617e-01 6.04446363e+00 4.45255518e-01
-1.15854621e+00 2.45144397e-01 2.62873434e-02 2.92794500e-02
-2.83437073e-01 2.68567264e-01 -8.08367729e-01 5.49600542e-01
1.34329641e+00 -4.09251332e-01 5.67366540e-01 1.49143434e+00
7.11245611e-02 1.04062758e-01 -1.84286869e+00 6.26111090e-01
-1.91512540e-01 -1.37235594e+00 -4.11038190e-01 1.36597708e-01
5.56339324e-01 2.34201655e-01 -3.73140246e-01 4.85765576e-01
3.50622594e-01 -9.40654874e-01 6.92648888e-01 6.73928618e-01
4.25884068e-01 -4.83466208e-01 8.21425080e-01 7.02762842e-01
-5.68379045e-01 -1.50074750e-01 -3.46661389e-01 -3.82278681e-01
-5.14686823e-01 7.08587885e-01 -1.14085293e+00 3.14858794e-01
8.72635067e-01 6.70365512e-01 -1.20237696e+00 1.05418646e+00
4.23961179e-03 7.50364423e-01 -5.30438609e-02 -6.43254966e-02
-2.31175125e-01 3.55896764e-02 1.36505514e-01 1.57214212e+00
4.01569992e-01 -4.68115866e-01 -3.05684730e-02 1.12589574e+00
6.51151314e-02 3.52691770e-01 -7.65967906e-01 -2.24309742e-01
5.26862562e-01 1.19769120e+00 -2.11429358e-01 -1.94599807e-01
-8.36148322e-01 3.52563143e-01 5.83889186e-01 1.11540399e-01
-5.52136600e-01 -8.53887737e-01 7.42208183e-01 5.32615781e-01
4.62342530e-01 -4.21241820e-02 -2.51405060e-01 -1.03936684e+00
3.57699722e-01 -1.45854533e+00 1.48438498e-01 -6.24324262e-01
-1.09261227e+00 6.28686488e-01 -1.68609768e-02 -8.43576908e-01
-4.35530514e-01 -4.98053193e-01 -5.40411174e-01 9.25676048e-01
-1.27763116e+00 -9.18109834e-01 -4.54201281e-01 -5.79334535e-02
2.45860651e-01 -2.38750920e-01 8.15912604e-01 6.33262873e-01
-3.89229536e-01 6.41854167e-01 1.69490233e-01 2.26177126e-01
9.29746211e-01 -1.34553623e+00 6.41097248e-01 7.33378828e-01
2.68368065e-01 1.24425471e+00 7.36382604e-01 -4.49746758e-01
-1.06137800e+00 -8.29850972e-01 9.71753657e-01 -9.74866807e-01
7.58552790e-01 -3.09241682e-01 -1.30281603e+00 8.47678423e-01
3.70615870e-02 -1.21101320e-01 7.02225447e-01 6.14505410e-01
-7.52871335e-01 -1.44169688e-01 -9.40036297e-01 3.60998958e-02
8.18505824e-01 -4.79910254e-01 -6.57738984e-01 2.64711231e-01
4.09685642e-01 -2.25357562e-01 -1.42259097e+00 1.34909386e-02
5.79514623e-01 -1.43615806e+00 5.74745774e-01 -8.08105469e-01
8.67278814e-01 -1.36063546e-01 -1.11512437e-01 -1.26080775e+00
-3.71375591e-01 -4.35300976e-01 1.79093838e-01 1.55973065e+00
6.48372829e-01 -5.53258479e-01 6.08142853e-01 6.62454963e-01
-3.32138911e-02 -7.28405118e-01 -3.56522381e-01 -9.22831655e-01
3.65933806e-01 -5.53827524e-01 9.94940221e-01 1.09891725e+00
1.52486622e-01 3.07771355e-01 3.22458558e-02 -1.74300581e-01
2.94637829e-01 1.35351062e-01 1.34897447e+00 -1.43738031e+00
-6.51031435e-01 -4.22130048e-01 -3.72627705e-01 -4.51739877e-01
-7.54780322e-02 -9.38403308e-01 -2.46077497e-02 -1.41690075e+00
5.05691588e-01 -6.74254656e-01 -1.71684116e-01 8.79850745e-01
-2.48223051e-01 -1.52668267e-01 1.88941270e-01 4.00465667e-01
-5.20551920e-01 -1.66458070e-01 6.23598635e-01 2.04904765e-01
-1.75847441e-01 1.28725275e-01 -8.72667015e-01 7.48433292e-01
6.97230399e-01 -1.09868157e+00 -4.36865538e-02 -5.90470731e-01
8.08991373e-01 -1.06159419e-01 2.38253921e-01 -1.03324127e+00
3.91686931e-02 -2.14402571e-01 -1.73852816e-01 1.79118402e-02
-2.78039336e-01 -7.07037985e-01 4.77795601e-01 3.67081285e-01
-4.68815565e-01 4.30006161e-02 9.65194553e-02 2.02860753e-03
-7.59309456e-02 -9.32444096e-01 7.69089460e-01 -2.03016594e-01
-4.53698397e-01 -1.81481883e-01 -1.61378488e-01 2.30322182e-01
8.91531289e-01 -1.82314403e-02 -4.41727042e-01 1.88564673e-01
-3.74559671e-01 -1.78086430e-01 1.17415440e+00 5.64178228e-01
7.68797286e-03 -1.14120686e+00 -8.05479467e-01 1.31168470e-01
6.06670439e-01 -4.17226911e-01 -3.00238252e-01 8.37643445e-01
-5.94655752e-01 3.37200612e-01 -3.84144872e-01 -5.21593153e-01
-1.43318808e+00 5.51207185e-01 2.55752563e-01 -3.15014869e-01
-3.84816110e-01 6.05484366e-01 -6.62778169e-02 -8.99091423e-01
9.32275057e-02 -5.65056026e-01 -2.01100677e-01 -1.30960077e-01
4.89248574e-01 5.32454908e-01 4.02623415e-01 -5.91590703e-02
-3.90222549e-01 6.68375134e-01 -2.32830241e-01 5.15097901e-02
1.87458825e+00 3.44630629e-01 -3.76747638e-01 5.68198323e-01
1.07961345e+00 3.46606463e-01 -1.02240384e+00 -2.52082825e-01
5.87387979e-01 -7.47440517e-01 -1.66323215e-01 -9.70191538e-01
-1.08044124e+00 9.69903529e-01 4.39058304e-01 5.57065725e-01
6.65449440e-01 3.12476512e-02 2.35865608e-01 6.43540561e-01
4.18450773e-01 -1.00845826e+00 4.68026958e-02 3.43126982e-01
8.08187902e-01 -1.23612881e+00 1.18639454e-01 -2.40937546e-01
-4.12882149e-01 1.39676011e+00 5.37034392e-01 1.09293900e-01
5.97132385e-01 3.77400875e-01 1.92036435e-01 -3.38907301e-01
-9.95940208e-01 2.26422668e-01 1.76836744e-01 7.22865641e-01
9.68465090e-01 -2.31760204e-01 -1.67793781e-01 7.18368411e-01
-5.57519495e-03 4.51857120e-01 1.03386509e+00 1.22839153e+00
-5.13702273e-01 -1.62450838e+00 -3.40541780e-01 6.63560808e-01
-6.23066723e-01 -1.76606178e-01 -4.89527285e-01 8.87328506e-01
3.48968595e-01 7.24151134e-01 -4.41728085e-01 -3.77670825e-01
6.08523965e-01 2.36781210e-01 6.00582182e-01 -1.36781526e+00
-9.59959745e-01 -3.84203166e-01 3.10097635e-01 -6.85015559e-01
-2.38285214e-01 -8.02328944e-01 -8.16161036e-01 -3.54024917e-01
-4.01059031e-01 3.19379389e-01 1.01095986e+00 7.69741595e-01
5.94189584e-01 4.98423576e-01 5.34277856e-01 -3.64291281e-01
-8.02577436e-01 -1.14230096e+00 -3.60439509e-01 4.35119808e-01
1.09029815e-01 -5.93127012e-01 -5.34266889e-01 4.39420700e-01] | [7.733841419219971, 7.7625813484191895] |
ca746887-1857-4030-82ca-681d31129617 | recurrent-convolutional-fusion-for-rgb-d | 1806.01673 | null | http://arxiv.org/abs/1806.01673v3 | http://arxiv.org/pdf/1806.01673v3.pdf | Recurrent Convolutional Fusion for RGB-D Object Recognition | Providing machines with the ability to recognize objects like humans has
always been one of the primary goals of machine vision. The introduction of
RGB-D cameras has paved the way for a significant leap forward in this
direction thanks to the rich information provided by these sensors. However,
the machine vision community still lacks an effective method to synergically
use the RGB and depth data to improve object recognition. In order to take a
step in this direction, we introduce a novel end-to-end architecture for RGB-D
object recognition called recurrent convolutional fusion (RCFusion). Our method
generates compact and highly discriminative multi-modal features by combining
complementary RGB and depth information representing different levels of
abstraction. Extensive experiments on two popular datasets, RGB-D Object
Dataset and JHUIT-50, show that RCFusion significantly outperforms
state-of-the-art approaches in both the object categorization and instance
recognition tasks. | ['Mirco Planamente', 'Mohammad Reza Loghmani', 'Barbara Caputo', 'Markus Vincze'] | 2018-06-05 | null | null | null | null | ['object-categorization'] | ['computer-vision'] | [ 3.45618487e-03 -3.96361619e-01 -2.20101364e-02 -8.70678663e-01
-8.21134090e-01 -4.24436867e-01 7.04565048e-01 1.55876219e-01
-4.09743965e-01 1.19688205e-01 1.23937130e-01 1.39781490e-01
-9.90891829e-02 -6.96155012e-01 -3.57911050e-01 -7.65282750e-01
3.57296109e-01 2.15796322e-01 2.38087609e-01 -7.27111250e-02
2.43201330e-01 1.01091135e+00 -2.02386904e+00 6.38192952e-01
2.36815602e-01 1.57746375e+00 2.42541030e-01 5.59615970e-01
-4.16109443e-01 6.53124809e-01 -1.09038435e-01 -3.02721232e-01
4.18251753e-01 -2.79704809e-01 -7.84495175e-01 4.04165864e-01
5.61388135e-01 -3.27910274e-01 -4.65746164e-01 6.73326194e-01
6.92612469e-01 8.66999477e-02 3.21398020e-01 -1.29882932e+00
-5.58162391e-01 -1.66003220e-02 -5.09195685e-01 2.29962558e-01
2.58348554e-01 8.77888799e-02 8.15843940e-01 -8.69589269e-01
5.30164123e-01 1.28064013e+00 4.35062349e-01 4.72037137e-01
-1.01874602e+00 -2.84920603e-01 9.91810858e-02 4.21722412e-01
-1.15568042e+00 -2.03995809e-01 9.61625636e-01 -2.39789009e-01
1.28837764e+00 2.96433493e-02 6.96253836e-01 9.07059133e-01
-3.29361260e-01 1.13332736e+00 1.25228107e+00 -4.70874518e-01
1.14086024e-01 -4.03789468e-02 2.91135848e-01 6.32366300e-01
1.98168382e-01 1.96168005e-01 -7.00178564e-01 2.14215338e-01
5.40202081e-01 5.61292112e-01 -7.92894512e-02 -5.91308534e-01
-1.34064639e+00 6.16587102e-01 1.05412567e+00 5.14462531e-01
-5.93477786e-01 1.61181957e-01 3.16272944e-01 3.79758924e-02
2.38264203e-01 1.67206764e-01 -3.13183069e-01 -2.86839962e-01
-5.99575698e-01 1.58318415e-01 3.47035646e-01 6.50612772e-01
6.92683458e-01 -3.33768010e-01 1.20152891e-01 7.49271631e-01
4.78381127e-01 6.77351058e-01 4.07734394e-01 -9.13927615e-01
3.25745314e-01 1.15932870e+00 1.83546711e-02 -7.60056019e-01
-4.85143870e-01 -2.92096227e-01 -8.52992117e-01 5.41959405e-01
4.24871445e-01 4.97293174e-01 -1.13059855e+00 1.25674570e+00
4.91429001e-01 -1.36013389e-01 1.16914846e-01 1.23290074e+00
1.01083338e+00 4.62665915e-01 -3.28837484e-02 5.35661161e-01
1.32685900e+00 -7.92851388e-01 -2.37662852e-01 -1.84300318e-01
3.74901026e-01 -6.06932104e-01 8.59391868e-01 3.54677409e-01
-7.31105328e-01 -7.90588140e-01 -1.00604320e+00 -3.09401870e-01
-7.60467947e-01 1.71902552e-01 9.23811197e-01 6.18800700e-01
-6.60184264e-01 3.71420324e-01 -9.90606308e-01 -2.51766086e-01
7.71410525e-01 3.02176744e-01 -9.31711793e-01 -5.24771512e-01
-6.15958333e-01 8.65204036e-01 2.74391472e-01 2.26566881e-01
-6.01075292e-01 -4.04615551e-01 -6.13110304e-01 -2.96394169e-01
1.72327459e-01 -5.32497942e-01 1.12739754e+00 -6.89841688e-01
-1.33040130e+00 1.21904266e+00 -6.87735006e-02 -2.86561757e-01
3.64301562e-01 -2.62379438e-01 5.78174591e-02 3.83107632e-01
-2.37330168e-01 9.37338114e-01 7.09029377e-01 -1.24572182e+00
-9.49066818e-01 -9.84773159e-01 1.97464414e-02 8.89941454e-02
-2.65745372e-01 -6.97897002e-02 -4.34271067e-01 -2.86063373e-01
4.45110679e-01 -9.17668998e-01 -3.07383779e-02 2.70660847e-01
-2.76175141e-01 -4.78887737e-01 9.74808514e-01 -2.08998829e-01
4.64652807e-01 -2.08710146e+00 4.32716846e-01 -1.21201603e-02
1.46824792e-01 4.67500389e-01 1.55591920e-01 1.24629408e-01
8.26084688e-02 -2.20340461e-01 -1.43190444e-01 -7.87877619e-01
7.22093508e-02 2.95647681e-01 -2.65075564e-01 4.60423291e-01
2.13521540e-01 1.00828969e+00 -7.77223170e-01 -1.83799863e-01
6.18763685e-01 9.48076427e-01 -2.43267462e-01 2.92582989e-01
-4.59271707e-02 4.38556045e-01 -4.84411895e-01 1.06757617e+00
5.03440201e-01 -3.23939741e-01 -3.19131553e-01 -3.34841430e-01
-2.01990068e-01 1.63222715e-01 -1.06811678e+00 2.01153731e+00
-4.14961368e-01 5.32467723e-01 -2.51222402e-01 -9.03948426e-01
9.16103542e-01 8.09899122e-02 5.14481843e-01 -9.37303305e-01
1.96697727e-01 3.02465767e-01 -3.29786807e-01 -4.40989822e-01
2.12544248e-01 -1.21844120e-01 -5.82429543e-02 4.07161981e-01
8.54980387e-03 -2.26002872e-01 -1.31258562e-01 -1.70783341e-01
9.22122180e-01 3.11065584e-01 1.52716950e-01 4.79774386e-01
4.62355316e-01 8.67379084e-02 2.17142746e-01 5.11649549e-01
-3.89220446e-01 7.78436661e-01 2.81420331e-02 -6.84191942e-01
-8.67388904e-01 -9.49818730e-01 -1.40738994e-01 7.29487360e-01
8.99434388e-02 -9.30468515e-02 -2.33719364e-01 -8.08181107e-01
4.85240191e-01 2.29947403e-01 -8.92438829e-01 -1.95868164e-01
-3.78287345e-01 -4.65367258e-01 5.60326755e-01 8.52053881e-01
1.19518256e+00 -9.59959984e-01 -1.37165141e+00 1.30492286e-03
-9.45167542e-02 -1.32647943e+00 2.91075468e-01 4.73256499e-01
-1.05777621e+00 -1.00715196e+00 -7.63659716e-01 -3.94300640e-01
4.75727648e-01 8.38453114e-01 8.23848248e-01 -2.34603792e-01
-7.95191646e-01 6.78876340e-01 -6.96873009e-01 -5.00847220e-01
-2.41887141e-02 7.24913329e-02 -2.32898474e-01 6.07929453e-02
6.86119318e-01 -3.52384090e-01 -7.32880294e-01 2.18926594e-01
-1.03236210e+00 2.14094277e-02 7.25729406e-01 5.07609487e-01
7.08581269e-01 -3.91626507e-01 2.71335244e-01 -1.54298902e-01
4.99963388e-03 -1.84674293e-01 -5.21009803e-01 2.69031525e-01
-3.05478513e-01 9.57617387e-02 7.74442777e-02 -1.39315143e-01
-8.89771342e-01 5.30783832e-01 -2.49027520e-01 -6.40566766e-01
-6.34646535e-01 2.70524204e-01 -2.27794901e-01 -2.23466143e-01
5.15974462e-01 1.61038578e-01 7.03830272e-02 -8.45619142e-01
6.52226806e-01 8.93597066e-01 6.48225069e-01 -1.93553105e-01
4.75076079e-01 8.29317570e-01 2.25214437e-01 -8.35808694e-01
-8.96639049e-01 -7.90107012e-01 -9.04202521e-01 -3.18930030e-01
9.45278883e-01 -8.59437704e-01 -8.60260844e-01 7.98467815e-01
-1.06065071e+00 -4.54861596e-02 -3.20350826e-01 4.01495248e-01
-5.29608011e-01 2.77501568e-02 -2.70369142e-01 -7.94977665e-01
-2.30089813e-01 -1.04377937e+00 1.37260056e+00 3.61163467e-01
2.08198890e-01 -3.59485507e-01 -1.32655771e-02 5.21878183e-01
5.13904631e-01 4.03902590e-01 6.18112862e-01 -3.21312845e-01
-7.63534606e-01 -7.09768236e-01 -6.06157362e-01 5.46962142e-01
2.96389163e-01 -2.00213835e-01 -1.26545608e+00 9.88134071e-02
-1.81547105e-01 -6.89145088e-01 1.24405766e+00 7.40472227e-02
1.04199421e+00 3.01760226e-01 -2.63459504e-01 6.66632891e-01
1.51077998e+00 -1.67850610e-02 6.03702962e-01 3.99089426e-01
8.00957024e-01 4.90432143e-01 4.36172813e-01 5.26128173e-01
5.08451164e-01 8.03311169e-01 6.53563023e-01 -1.26711577e-01
-4.00740474e-01 8.81956983e-03 -1.00444011e-01 2.31960058e-01
-1.52680784e-01 8.77911970e-02 -1.21655512e+00 4.02485788e-01
-1.76639366e+00 -9.23153758e-01 2.94918399e-02 2.00703430e+00
4.70240921e-01 8.65080431e-02 1.91314742e-01 5.95120251e-01
3.32156271e-01 -9.84474272e-02 -4.96639609e-01 -3.03116869e-02
-3.01749557e-01 4.85600196e-02 2.34125048e-01 -2.95120217e-02
-1.17896163e+00 6.99915230e-01 5.09972334e+00 2.75199592e-01
-1.47571445e+00 -1.69088587e-01 4.38506991e-01 -5.35793677e-02
2.53535390e-01 -3.90651733e-01 -9.16838169e-01 1.38534054e-01
7.60300517e-01 3.56580615e-01 5.43289967e-02 1.04876697e+00
-2.62261122e-01 -3.06748509e-01 -1.12980831e+00 1.35247302e+00
3.30274373e-01 -1.33654463e+00 -9.20243338e-02 -1.19022178e-02
4.38486725e-01 3.77039015e-01 1.85834005e-01 3.68612409e-02
2.56077778e-02 -8.62218201e-01 7.08940208e-01 6.54710948e-01
4.36742842e-01 -5.85015357e-01 8.10980022e-01 1.68357223e-01
-1.09098470e+00 -1.92141995e-01 -2.98393846e-01 5.30675240e-02
-1.10888675e-01 5.51426351e-01 -6.76864684e-01 4.99164402e-01
1.13763416e+00 8.22267115e-01 -6.55508995e-01 1.26856256e+00
-1.74054578e-01 3.27832364e-02 -5.64437091e-01 -3.34544778e-02
3.68273288e-01 2.30923295e-01 1.98241472e-01 1.05297148e+00
1.90586865e-01 2.41932333e-01 -4.27731834e-02 6.49883747e-01
-1.61755547e-01 -3.33603650e-01 -6.46112382e-01 -1.84112206e-01
2.02648565e-01 1.28661597e+00 -9.65740800e-01 -2.91408151e-01
-5.70540369e-01 1.12674046e+00 2.91768372e-01 2.61593889e-03
-5.90897918e-01 -3.97657752e-01 7.55796492e-01 -2.47174069e-01
6.93756461e-01 -6.31794214e-01 -4.08692718e-01 -9.40432370e-01
6.25644475e-02 -7.13417828e-01 3.09201151e-01 -8.02201986e-01
-1.19823050e+00 7.70632148e-01 -7.44328871e-02 -1.23142707e+00
-6.88404441e-02 -9.02364314e-01 1.16470471e-01 7.82775223e-01
-1.69588065e+00 -1.43982387e+00 -9.88359571e-01 7.29082286e-01
3.54876399e-01 1.39054256e-02 9.79366601e-01 2.58261085e-01
-2.88951308e-01 1.62998259e-01 -6.08232208e-02 2.08016753e-01
5.30499637e-01 -9.93140876e-01 2.48588189e-01 6.38782501e-01
5.11174679e-01 2.58636057e-01 3.29440027e-01 -1.61017239e-01
-1.82710826e+00 -8.52153122e-01 6.25236630e-01 -5.37975192e-01
1.15691483e-01 -2.30947122e-01 -5.94371915e-01 3.86455983e-01
-2.16560245e-01 4.67960835e-01 7.74377584e-01 -1.01723500e-01
-8.79016519e-01 -4.17043239e-01 -1.07819092e+00 1.22594953e-01
9.38237846e-01 -7.92829156e-01 -7.37231374e-01 9.63142142e-02
4.72630292e-01 -3.40280592e-01 -7.30271339e-01 5.28340399e-01
6.92558348e-01 -1.25831676e+00 1.27003741e+00 -5.30781806e-01
4.63125497e-01 -3.53664726e-01 -7.48517096e-01 -1.06608188e+00
-5.20528369e-02 9.77088735e-02 -1.13964126e-01 1.06113636e+00
-5.62328361e-02 -3.73314410e-01 8.63645256e-01 6.78488791e-01
-1.92090660e-01 -8.24339032e-01 -1.06622183e+00 -5.56000173e-01
-2.03994706e-01 -7.80041039e-01 5.29077947e-01 4.64322090e-01
-2.46778548e-01 7.07997903e-02 6.82870001e-02 1.34151685e-03
7.87189603e-01 8.12887549e-01 8.97882044e-01 -1.31466925e+00
-1.18158795e-01 -6.81095839e-01 -9.74021792e-01 -1.07156932e+00
-2.09588669e-02 -9.37614501e-01 -7.47446865e-02 -1.72386181e+00
6.76273331e-02 -6.40660048e-01 -4.15725321e-01 7.90256977e-01
9.19186100e-02 7.65561104e-01 4.81153250e-01 2.17009649e-01
-7.59266555e-01 6.33748829e-01 1.01161289e+00 -1.86724350e-01
8.10899027e-03 -5.60489111e-02 -5.50110936e-01 5.69573700e-01
6.10219300e-01 -2.13384256e-01 6.89222291e-02 -5.35198510e-01
4.06417018e-03 -1.35589108e-01 6.35546565e-01 -1.20404696e+00
3.59480798e-01 1.56762019e-01 7.65076518e-01 -8.28776658e-01
7.03105807e-01 -1.00065160e+00 -2.18402773e-01 3.79821688e-01
-2.97352493e-01 -1.31547749e-01 3.53421956e-01 4.83526945e-01
-4.73054945e-01 2.88483381e-01 9.32467699e-01 -2.00131670e-01
-1.03571057e+00 2.41200000e-01 1.03006482e-01 -3.40891749e-01
1.09249842e+00 -4.23526883e-01 -3.44909102e-01 7.37101212e-02
-6.00761592e-01 2.47075148e-02 4.02068913e-01 7.13091493e-01
9.62717652e-01 -1.27831066e+00 -4.84839290e-01 4.57533181e-01
4.48955208e-01 -1.31739387e-02 2.32461497e-01 6.99269474e-01
-3.23990971e-01 7.21189201e-01 -6.20042324e-01 -9.01923656e-01
-1.34985089e+00 4.55226749e-01 4.51821297e-01 1.10183775e-01
-7.84169137e-01 1.18319154e+00 -4.21973884e-01 -2.10900709e-01
6.76428080e-01 -5.17030180e-01 1.07992850e-01 2.15454966e-01
7.58218348e-01 3.47863287e-01 4.95554984e-01 -6.96296811e-01
-7.01899827e-01 6.14128470e-01 3.30281556e-02 -4.95488383e-02
1.63700771e+00 -1.40994474e-01 7.52141774e-02 6.69051707e-01
1.40305209e+00 -6.68019056e-01 -1.32394326e+00 -4.26212370e-01
4.34757210e-02 -8.30612421e-01 3.40384960e-01 -9.85472441e-01
-1.08467400e+00 1.26465547e+00 9.46852863e-01 1.56107560e-01
1.14329374e+00 2.11459592e-01 7.19561160e-01 6.30118608e-01
6.90835595e-01 -7.15218246e-01 3.72643203e-01 3.68616343e-01
8.92461598e-01 -1.63614857e+00 -7.95084611e-02 -1.70060888e-01
-5.17579079e-01 1.14224935e+00 3.96587223e-01 -7.25369006e-02
6.13512754e-01 9.42904619e-04 1.46474808e-01 -1.81389421e-01
-5.74980736e-01 -7.33262181e-01 2.98763722e-01 6.03138208e-01
3.47345203e-01 -7.79471174e-02 4.80245173e-01 2.26108313e-01
2.28058934e-01 8.27664509e-02 6.27118722e-03 1.25210011e+00
-5.18140495e-01 -1.16230106e+00 -4.27675992e-01 2.20483199e-01
-3.04103643e-01 4.20829296e-01 -5.80526888e-01 6.93758547e-01
1.44866318e-01 8.55763972e-01 -3.65223549e-03 -4.88716930e-01
6.05371118e-01 2.69963145e-01 8.91407251e-01 -2.81657636e-01
-5.85776210e-01 -2.01248422e-01 -3.31143498e-01 -9.44427967e-01
-6.94182813e-01 -5.92059791e-01 -1.36374521e+00 -1.54128700e-01
-1.72066942e-01 -3.14033478e-01 1.28061426e+00 8.83120775e-01
5.12641788e-01 5.24423182e-01 5.07665992e-01 -1.37910855e+00
-5.31075895e-01 -7.17794836e-01 -3.38973105e-01 5.61019659e-01
4.75124806e-01 -8.88494492e-01 -1.89040840e-01 -6.83770478e-02] | [9.548813819885254, -0.9084527492523193] |
7a982356-0fb4-492f-98e7-432e4155da87 | deepfilternet-perceptually-motivated-real | 2305.08227 | null | https://arxiv.org/abs/2305.08227v1 | https://arxiv.org/pdf/2305.08227v1.pdf | DeepFilterNet: Perceptually Motivated Real-Time Speech Enhancement | Multi-frame algorithms for single-channel speech enhancement are able to take advantage from short-time correlations within the speech signal. Deep Filtering (DF) was proposed to directly estimate a complex filter in frequency domain to take advantage of these correlations. In this work, we present a real-time speech enhancement demo using DeepFilterNet. DeepFilterNet's efficiency is enabled by exploiting domain knowledge of speech production and psychoacoustic perception. Our model is able to match state-of-the-art speech enhancement benchmarks while achieving a real-time-factor of 0.19 on a single threaded notebook CPU. The framework as well as pretrained weights have been published under an open source license. | ['Andreas Maier', 'Alberto N. Escalante-B.', 'Tobias Rosenkranz', 'Hendrik Schröter'] | 2023-05-14 | null | null | null | null | ['speech-enhancement'] | ['speech'] | [ 2.00025678e-01 -9.82857272e-02 4.72134769e-01 -2.62906790e-01
-9.70516205e-01 -2.51059115e-01 5.48173785e-01 2.29696501e-02
-8.14684391e-01 3.94438833e-01 5.09589911e-01 -2.80728012e-01
-6.94474280e-02 -4.71147388e-01 -4.60100502e-01 -5.40404201e-01
-3.16196322e-01 -2.66565174e-01 2.80008674e-01 -5.10308027e-01
-7.20792785e-02 3.78369838e-01 -1.61509860e+00 6.04757428e-01
6.51951671e-01 1.00674856e+00 6.90535247e-01 1.44071758e+00
2.02429056e-01 4.95436221e-01 -8.07340384e-01 -2.97055125e-01
2.25936994e-01 -2.37711236e-01 -2.67966777e-01 -4.84520756e-02
5.27672946e-01 -6.83676839e-01 -7.11686969e-01 1.22745132e+00
1.18432403e+00 3.07579190e-01 6.89181848e-04 -7.46766925e-01
-7.25397766e-02 6.01769745e-01 -5.01831323e-02 5.74129522e-01
2.08250776e-01 2.78869689e-01 8.89865041e-01 -1.03390956e+00
3.42217892e-01 1.20206904e+00 8.64012420e-01 2.36326367e-01
-1.02222681e+00 -6.60148263e-01 -1.63719893e-01 3.05185646e-01
-1.02336371e+00 -1.00605679e+00 6.35625482e-01 2.20790133e-02
1.67834318e+00 1.42510608e-01 6.56906366e-01 1.05376089e+00
2.59577811e-01 6.46049798e-01 1.47995532e+00 -4.95836496e-01
1.40504077e-01 -2.55866885e-01 1.92109179e-02 3.63244921e-01
-3.72647911e-01 8.68852556e-01 -8.71640682e-01 -2.18749158e-02
7.68850088e-01 -6.31339788e-01 -4.50121254e-01 5.82783520e-01
-1.04014838e+00 5.45700490e-01 1.86730817e-01 4.61977780e-01
-4.71707314e-01 3.07445377e-01 6.34540677e-01 6.66706145e-01
6.85656369e-01 2.83051342e-01 -6.83611572e-01 -6.70298934e-01
-1.20712674e+00 1.65994421e-01 7.04979300e-01 6.46248221e-01
4.47752953e-01 5.19517064e-01 -1.68288574e-01 8.09386790e-01
1.58860534e-01 6.17805660e-01 6.69718683e-01 -1.07938051e+00
2.68754482e-01 -6.13868177e-01 -8.74354765e-02 -6.65193617e-01
-5.99822283e-01 -7.59338975e-01 -7.73582101e-01 5.24000168e-01
4.91872519e-01 -4.29911971e-01 -8.02929342e-01 1.55735517e+00
1.77333593e-01 5.92337370e-01 2.03775197e-01 9.63017583e-01
5.89913011e-01 5.97487867e-01 -1.35255754e-01 -3.18908602e-01
1.51021552e+00 -9.74880636e-01 -1.15528429e+00 -2.12269843e-01
1.05276331e-01 -1.39585090e+00 7.91332424e-01 7.67425418e-01
-1.37989235e+00 -9.61048484e-01 -1.24508619e+00 -1.18525580e-01
-1.33507580e-01 -2.04431321e-02 5.46348751e-01 1.14111614e+00
-1.41407847e+00 9.26866114e-01 -1.00810969e+00 1.66937280e-02
1.53396800e-01 4.30606335e-01 -2.12273866e-01 4.10411239e-01
-1.34454095e+00 7.29606152e-01 2.43853673e-01 4.14136127e-02
-8.14200640e-01 -9.78754222e-01 -6.78673863e-01 2.74941027e-01
1.67829394e-01 -4.61695492e-01 1.69369185e+00 -9.70790923e-01
-2.03496695e+00 3.74268949e-01 -1.86485663e-01 -8.48340392e-01
4.70811099e-01 -5.51135480e-01 -9.98452485e-01 5.10280073e-01
-5.50881863e-01 3.78746480e-01 1.28787601e+00 -5.39374411e-01
-6.65521681e-01 -2.16294602e-02 -1.53727094e-02 5.93659803e-02
-4.80952829e-01 2.48338610e-01 -2.19751626e-01 -1.10001004e+00
-1.58408359e-01 -5.58883548e-01 -2.28382185e-01 -1.05583213e-01
7.77504668e-02 4.07833248e-01 7.16650665e-01 -1.18926835e+00
1.28764486e+00 -2.36831975e+00 -3.58152807e-01 -1.48257792e-01
1.95433855e-01 8.31314385e-01 -3.35884780e-01 3.48705649e-01
-8.22544917e-02 -4.17036891e-01 1.06543079e-01 -5.67738771e-01
1.58107474e-01 -9.40643251e-02 -2.57938236e-01 5.44627488e-01
5.62203079e-02 3.08519661e-01 -8.64621758e-01 -6.33458048e-02
4.78083730e-01 9.28969681e-01 -8.20744872e-01 3.27520877e-01
1.19376093e-01 2.13212341e-01 1.29809871e-01 4.18246537e-02
1.09352016e+00 3.37495685e-01 1.71960711e-01 -4.78022426e-01
-4.05717254e-01 6.44872367e-01 -1.48602092e+00 2.01179767e+00
-8.31215143e-01 9.77762163e-01 6.92433834e-01 -5.25814474e-01
7.18335688e-01 7.77996778e-01 2.59450853e-01 -7.25244284e-01
1.62329182e-01 3.96144152e-01 3.26888353e-01 -2.57590979e-01
7.68681288e-01 -1.50231153e-01 5.62046587e-01 4.00562614e-01
5.02758563e-01 -3.33131999e-01 1.43631890e-01 3.70440967e-02
1.10685802e+00 -1.28788844e-01 3.29879910e-01 -3.96636069e-01
5.52546561e-01 -6.64155006e-01 1.80406824e-01 8.42643261e-01
-4.04331267e-01 3.99249107e-01 -1.68928467e-02 -1.07033812e-01
-1.13587248e+00 -1.22792053e+00 -1.78088635e-01 1.26633000e+00
-4.19242740e-01 -6.27422214e-01 -8.91691625e-01 -6.12594821e-02
-4.33720946e-01 5.38720489e-01 -1.54000908e-01 1.02285922e-01
-6.49085402e-01 -3.61417234e-01 7.13895798e-01 5.36827683e-01
6.14580929e-01 -1.01707554e+00 -7.22579718e-01 8.50553393e-01
-2.76307501e-02 -1.40149701e+00 -6.90617681e-01 3.59233081e-01
-6.55000925e-01 -6.26736343e-01 -8.64679992e-01 -4.86141920e-01
-1.00986995e-02 1.44932657e-01 9.83906150e-01 -9.27667692e-02
-2.10926145e-01 3.50839317e-01 -2.81551659e-01 -3.75868738e-01
-6.51115477e-01 -3.94698903e-02 1.95716321e-01 -1.44173875e-01
1.01912290e-01 -9.30728078e-01 -7.11256444e-01 6.66940436e-02
-8.42165530e-01 7.58766942e-03 4.62766320e-01 1.01926434e+00
2.91241318e-01 2.28334650e-01 5.87268949e-01 -3.74627024e-01
8.13644826e-01 5.68579435e-02 -9.18798149e-01 -2.60883778e-01
-5.00221193e-01 3.86076756e-02 5.76576471e-01 -4.77416873e-01
-1.48717499e+00 -4.80742157e-02 -9.89963889e-01 -2.83337265e-01
-1.64266363e-01 3.85942757e-01 3.05080670e-03 -6.31285906e-02
5.46306610e-01 4.99406876e-03 2.49530561e-02 -5.78486443e-01
3.99473250e-01 8.74041498e-01 7.01925457e-01 -5.03663182e-01
4.75650609e-01 3.17723989e-01 -1.20553315e-01 -1.15072477e+00
-4.94257241e-01 -6.48675323e-01 -2.84982264e-01 -1.94137380e-01
5.61711729e-01 -1.23218501e+00 -6.93942428e-01 8.18558395e-01
-1.26295733e+00 -4.40141290e-01 -1.94016203e-01 9.76890385e-01
-5.94701290e-01 3.68601918e-01 -9.81008589e-01 -8.55991185e-01
-6.01656377e-01 -1.06957984e+00 8.60352755e-01 1.06235646e-01
9.68216583e-02 -9.51775491e-01 -4.46760282e-02 -7.90328160e-02
9.86095965e-01 -3.80882144e-01 4.92440276e-02 -5.10157406e-01
-2.57818639e-01 2.94621527e-01 -1.29307747e-01 7.90974319e-01
-5.02689928e-02 -1.48914278e-01 -1.67938316e+00 -5.20722687e-01
4.34456736e-01 9.27279517e-02 9.59144115e-01 6.97479904e-01
8.40306342e-01 -1.11176580e-01 2.56357163e-01 7.94790566e-01
1.20018554e+00 5.88464737e-02 7.58348763e-01 1.10535854e-02
3.76754776e-02 3.61548483e-01 4.89549875e-01 7.22096682e-01
-1.37117982e-01 8.12906623e-01 6.47397712e-02 -3.14738125e-01
-7.27435112e-01 1.48923015e-02 5.15122414e-01 1.23595750e+00
-6.94196448e-02 -1.11144014e-01 -5.72708428e-01 3.88536364e-01
-1.33772230e+00 -1.10562778e+00 -2.23239288e-01 2.01653004e+00
8.73344600e-01 3.55838329e-01 7.74841830e-02 4.25799012e-01
5.67767680e-01 4.24912989e-01 -1.38831958e-01 -6.05720460e-01
-1.92259490e-01 9.82485175e-01 5.69363058e-01 9.71926808e-01
-1.15205944e+00 7.62274384e-01 6.59618521e+00 1.15853238e+00
-1.19962192e+00 4.41920787e-01 1.65700823e-01 -1.54970706e-01
1.22695081e-01 -3.62684458e-01 -6.26521885e-01 1.65026397e-01
1.48529255e+00 -1.81010768e-01 7.57992506e-01 5.19783795e-01
4.75069642e-01 9.02962834e-02 -6.29798889e-01 8.73227835e-01
-2.25327283e-01 -1.16998887e+00 -7.55233109e-01 1.80120051e-01
4.61079359e-01 4.02353078e-01 2.17070639e-01 2.01746568e-01
6.16793744e-02 -6.92604661e-01 7.77996063e-01 1.98513672e-01
9.11107957e-01 -8.22108090e-01 5.86413503e-01 7.77226016e-02
-1.50975966e+00 -3.73164043e-02 -2.89435357e-01 -3.58780980e-01
3.20861548e-01 1.02079010e+00 -9.00763750e-01 4.85738486e-01
6.91158891e-01 4.28373575e-01 -1.29631653e-01 1.02062345e+00
-2.69921690e-01 7.78987706e-01 -3.36127102e-01 2.99502224e-01
1.42468646e-01 2.49606043e-01 7.04285920e-01 1.75842941e+00
5.36049843e-01 2.16386184e-01 -1.63379475e-01 2.76669472e-01
1.75168425e-01 -1.18433103e-01 5.59563935e-02 -5.12003452e-02
2.50767469e-01 1.24218404e+00 -3.33534986e-01 -3.19640607e-01
-4.50556308e-01 9.78531003e-01 -1.14086621e-01 2.89364755e-01
-8.76836896e-01 -6.44488752e-01 1.07241464e+00 -1.18067175e-01
7.00258613e-01 -6.09308124e-01 -1.15691155e-01 -8.89070094e-01
-2.45093986e-01 -1.07121146e+00 -4.77766730e-02 -7.41641521e-01
-9.31623995e-01 9.77127910e-01 -3.91727418e-01 -8.90831709e-01
-2.58050263e-01 -9.39792454e-01 -3.35288942e-01 1.20561528e+00
-1.76043630e+00 -7.57492423e-01 -1.43277168e-01 8.01749825e-01
5.91525435e-01 -2.30121583e-01 1.09816647e+00 6.80505335e-01
-3.14732268e-02 6.18314922e-01 1.35820389e-01 -1.65328428e-01
7.26949513e-01 -1.29784596e+00 8.45825970e-01 1.06924224e+00
1.38799593e-01 4.33144301e-01 1.07013595e+00 -3.62767458e-01
-1.17762578e+00 -7.31307209e-01 9.53733206e-01 2.92564958e-01
9.00700629e-01 -3.86485517e-01 -9.37082589e-01 1.95096686e-01
7.91578591e-01 1.85555890e-01 6.20993972e-01 1.21094376e-01
-4.07628506e-01 -2.60248959e-01 -1.04621029e+00 2.81627625e-01
9.86439645e-01 -9.42368090e-01 -5.12318671e-01 2.05589741e-01
8.09029639e-01 -8.45295429e-01 -8.45022440e-01 3.41763198e-02
5.55745065e-01 -1.30604708e+00 1.12778628e+00 -6.88856989e-02
2.54701108e-01 -1.75492659e-01 -2.95628577e-01 -1.58283973e+00
-7.62521178e-02 -1.30350626e+00 -3.00963104e-01 1.09442401e+00
2.09778324e-01 -7.86067486e-01 2.85282969e-01 -4.71102037e-02
-5.88407815e-01 -2.03315169e-01 -1.13704050e+00 -9.10279393e-01
-2.48625427e-01 -1.02130067e+00 5.76276779e-01 3.27168912e-01
-3.01383696e-02 -4.98874020e-03 -5.08838892e-01 4.42500621e-01
5.77620625e-01 -3.89301717e-01 4.69587743e-01 -8.08703840e-01
-8.97653639e-01 -3.93499702e-01 -2.90288925e-01 -1.13242781e+00
1.56958595e-01 -4.10231650e-01 1.26951247e-01 -9.94655788e-01
-4.74746019e-01 7.05108941e-02 -4.25189078e-01 1.94332097e-02
-1.03733547e-01 1.66213349e-01 4.54076380e-01 -5.00010908e-01
-9.68989879e-02 3.82277846e-01 1.12385917e+00 3.44167575e-02
-4.59408835e-02 6.63314536e-02 -2.47295886e-01 6.80976927e-01
9.34001386e-01 -3.79830956e-01 -3.06823403e-01 -4.47530746e-01
-1.86170772e-01 2.91691810e-01 3.26193839e-01 -1.40514839e+00
4.86184388e-01 3.70737910e-01 1.08211994e-01 -4.24787045e-01
8.29735279e-01 -8.08791161e-01 2.80585289e-02 4.56982583e-01
-3.04841429e-01 5.99656925e-02 7.20231473e-01 3.19983393e-01
-4.31058556e-01 -5.46970852e-02 1.01524305e+00 1.59526616e-01
-7.66272247e-01 -7.05754310e-02 -6.29076421e-01 8.80241469e-02
3.03634346e-01 1.98391929e-01 -3.69626015e-01 -5.04518628e-01
-8.08133185e-01 -5.60164213e-01 -5.49358502e-02 1.60080761e-01
6.04083300e-01 -9.27381277e-01 -8.98123682e-01 4.87713903e-01
-5.55047512e-01 -8.72996747e-01 7.21896231e-01 8.14980328e-01
-4.06569690e-01 4.40178990e-01 -2.58816659e-01 -3.56585324e-01
-1.43450737e+00 4.18599695e-01 6.11769021e-01 -4.14431989e-01
-6.93737686e-01 9.33237255e-01 -1.44290060e-01 2.94332076e-02
2.60337234e-01 -6.02061033e-01 2.45546147e-01 -1.57614350e-01
1.09555078e+00 4.71425503e-01 4.96513844e-01 -4.32327211e-01
-2.90513784e-01 2.34058723e-01 1.60793021e-01 -7.77510762e-01
1.29598033e+00 -9.24460813e-02 3.59884053e-01 1.66522455e-04
1.28532493e+00 4.04956460e-01 -1.55599391e+00 -3.28202426e-01
-3.64642024e-01 -6.48701131e-01 7.89836884e-01 -1.06181288e+00
-1.07297730e+00 1.18095684e+00 1.11608434e+00 1.79442950e-02
1.75116420e+00 -6.01404548e-01 1.05779421e+00 1.85897082e-01
1.58477664e-01 -1.31672919e+00 7.87705183e-02 7.27936983e-01
9.29793477e-01 -1.02386260e+00 -1.28462523e-01 -4.53627408e-01
-3.12652886e-01 1.45340812e+00 6.13648146e-02 8.38388316e-03
8.91942322e-01 1.06031096e+00 3.42268944e-01 2.22513452e-01
-6.26943529e-01 -5.34044683e-01 1.98463514e-01 8.48586082e-01
6.97175980e-01 7.52188563e-02 -8.72167647e-02 4.63289440e-01
-3.95070881e-01 -3.86005268e-02 3.44817013e-01 6.24520838e-01
-4.48802501e-01 -1.12835336e+00 -5.58679700e-01 -8.62771124e-02
-7.68170714e-01 -6.15050673e-01 3.92479002e-01 3.89208645e-01
-3.27868611e-02 1.33966041e+00 -9.51406825e-03 -3.03350180e-01
5.02549112e-01 -2.71135420e-01 5.43661356e-01 -1.12004019e-01
-1.17611372e+00 7.21716821e-01 2.79327273e-01 -6.63301945e-01
-3.72200251e-01 -5.76108754e-01 -1.10311413e+00 -4.17415321e-01
-2.78435409e-01 -2.13901207e-01 1.00894701e+00 7.63698637e-01
3.17310214e-01 1.10763299e+00 4.49236035e-01 -1.11835635e+00
-6.78651392e-01 -1.13795483e+00 -8.12847316e-01 1.24819979e-01
7.27041423e-01 -1.92276910e-01 -5.88019073e-01 2.07380857e-02] | [15.181109428405762, 5.876350402832031] |
1bf59636-3adf-47f7-8bbf-29cc3553e2f8 | artificial-intelligence-security-competition | 2212.03412 | null | https://arxiv.org/abs/2212.03412v1 | https://arxiv.org/pdf/2212.03412v1.pdf | Artificial Intelligence Security Competition (AISC) | The security of artificial intelligence (AI) is an important research area towards safe, reliable, and trustworthy AI systems. To accelerate the research on AI security, the Artificial Intelligence Security Competition (AISC) was organized by the Zhongguancun Laboratory, China Industrial Control Systems Cyber Emergency Response Team, Institute for Artificial Intelligence, Tsinghua University, and RealAI as part of the Zhongguancun International Frontier Technology Innovation Competition (https://www.zgc-aisc.com/en). The competition consists of three tracks, including Deepfake Security Competition, Autonomous Driving Security Competition, and Face Recognition Security Competition. This report will introduce the competition rules of these three tracks and the solutions of top-ranking teams in each track. | ['Mengxin Zhang', 'Kun Hu', 'Huafeng Shi', 'Guoshuai Liu', 'Zhichao Cui', 'Chenhao Lin', 'Chao Shen', 'Yijia Li', 'Junhao Zheng', 'Chen Ma', 'Jitao Sang', 'Haonan Wang', 'Shuang Li', 'YiFei Gao', 'Zhiyu Lin', 'Haoran Lyu', 'Shangbo Wu', 'Yuhang Zhao', 'Yajie Wang', 'Huipeng Zhou', 'Chengqi Duan', 'Yuanzhe Pang', 'Zeyu Liu', 'Kangkang Gao', 'Lin Feng', 'Yu Fu', 'Ye Wang', 'Tianpeng Wu', 'Jian Lin', 'Yu Wu', 'Shouhong Ding', 'Taiping Yao', 'Junyi Cao', 'Zhimin Sun', 'Shen Chen', 'Xuchong Zhang', 'Haoliang Han', 'Changfeng Sun', 'Xuelin Fu', 'Shu Xu', 'Jincai Xu', 'Enhui Xu', 'Yangyi Dong', 'Xinyu Liu', 'Yunteng Tan', 'Jiaxing Li', 'Hanyu Zhao', 'Yi Sun', 'Lianji L', 'Senyou Deng', 'Peng Chen', 'Yinpeng Dong'] | 2022-12-07 | null | null | null | null | ['face-swapping'] | ['computer-vision'] | [-3.82160932e-01 -7.88313523e-02 -1.27622560e-01 1.21874608e-01
-3.34709696e-02 -6.86497152e-01 6.95561469e-01 -2.99972743e-01
-3.43364000e-01 3.98730576e-01 -4.47839707e-01 -6.70301676e-01
-6.64983988e-02 -8.72578502e-01 -4.71374154e-01 -6.10060632e-01
1.49620235e-01 3.72013390e-01 -8.85461122e-02 -4.77594316e-01
2.92713523e-01 8.66393626e-01 -1.27972019e+00 1.47618249e-01
6.88320994e-01 1.45836854e+00 -5.32845557e-01 6.14206076e-01
4.15885568e-01 7.35175610e-01 -7.26995111e-01 -5.19402564e-01
5.53410888e-01 -2.59227008e-01 -9.33848023e-01 -5.81297159e-01
-2.53018349e-01 -5.61823845e-02 -2.09727690e-01 1.35856235e+00
5.34337759e-02 -3.31731647e-01 3.89296979e-01 -2.18283796e+00
-5.05714834e-01 4.51596737e-01 -4.26133007e-01 1.14911228e-01
1.21146783e-01 2.95609117e-01 7.49581635e-01 -6.85735643e-01
5.66364527e-01 9.79959786e-01 1.93360373e-01 9.35038388e-01
-2.66477406e-01 -1.25019860e+00 3.08436900e-01 6.29714310e-01
-1.32975590e+00 -1.05378136e-01 6.59136057e-01 -3.12421352e-01
8.52614343e-01 3.38692367e-01 5.42818964e-01 1.04972637e+00
6.72177494e-01 9.31207716e-01 7.23004997e-01 -3.41833919e-01
5.21318316e-01 1.26395365e-02 4.54732776e-01 6.48314536e-01
7.05271661e-01 8.07617605e-01 -2.66126156e-01 -2.44637668e-01
3.69963646e-01 -6.99143857e-02 1.90998018e-01 -7.78056532e-02
-9.87791955e-01 1.13251758e+00 5.77912569e-01 5.27745068e-01
-4.04992133e-01 7.54890814e-02 6.20266497e-01 4.78577256e-01
1.15487903e-01 4.79416400e-01 -6.02261901e-01 2.34403070e-02
-7.05256313e-02 3.81224990e-01 6.66173816e-01 8.06877434e-01
3.05169880e-01 6.13637030e-01 5.45684934e-01 2.66701430e-01
4.64033842e-01 5.40189683e-01 4.74682413e-02 -1.19845128e+00
9.45643410e-02 4.31124538e-01 -4.41827513e-02 -1.25894916e+00
-3.85048300e-01 -1.54069122e-02 -8.76527190e-01 4.64725435e-01
-1.33658787e-02 -7.65518785e-01 -5.73759913e-01 1.21523964e+00
2.34564394e-01 3.21607262e-01 4.32790816e-01 8.41421247e-01
5.68826675e-01 7.14049995e-01 1.19133443e-01 -2.39575222e-01
1.40218329e+00 -1.02998018e+00 -8.46904039e-01 -1.45086914e-01
4.84331697e-01 -4.09590691e-01 5.34612536e-02 8.98040354e-01
-7.53249824e-01 -4.63893890e-01 -1.39282382e+00 4.58585083e-01
-8.58576596e-01 -1.54280275e-01 8.62947345e-01 9.27333891e-01
-8.17887962e-01 -2.57790118e-01 -6.37746394e-01 -2.63483584e-01
5.07622182e-01 3.94354254e-01 -1.86338320e-01 -1.14850134e-01
-1.71003282e+00 1.22155166e+00 6.06730402e-01 -1.07572917e-02
-1.13213491e+00 -2.82329768e-01 -8.33881021e-01 -2.93716639e-01
4.77820247e-01 -1.80238798e-01 9.29344475e-01 -8.64287555e-01
-1.51235056e+00 6.86385989e-01 4.21607137e-01 -8.34964871e-01
-4.99906614e-02 -2.56620079e-01 -1.00772786e+00 -5.21411821e-02
-1.18453592e-01 7.31010020e-01 7.22059250e-01 -1.08589900e+00
-1.00137830e+00 -6.47476196e-01 -6.17247112e-02 -2.97010958e-01
-2.93930650e-01 5.71607530e-01 -6.52284697e-02 -4.95113134e-01
-5.29861510e-01 -1.17855859e+00 -3.94907951e-01 -2.63885707e-01
-1.47552446e-01 -5.57351768e-01 1.39029956e+00 -3.51412117e-01
9.87789094e-01 -2.23635674e+00 -1.70085788e-01 5.02744675e-01
1.17833681e-01 8.71715784e-01 -1.56755030e-01 2.50636697e-01
-1.84653148e-01 1.09073490e-01 2.02800900e-01 4.57072407e-01
-1.88908905e-01 1.37296826e-01 -5.20072818e-01 2.08940357e-01
2.03365564e-01 8.47492278e-01 -7.37970829e-01 1.41838929e-02
1.78563029e-01 1.55045599e-01 -5.95285632e-02 -8.33968893e-02
6.54704645e-02 2.65352130e-01 -8.88481438e-01 9.10592616e-01
7.79748678e-01 2.31158867e-01 1.66598007e-01 1.68448210e-01
-2.38336533e-01 -1.88275427e-01 -9.61203039e-01 1.11574054e+00
2.09838346e-01 3.11366528e-01 4.43917722e-01 -1.07958174e+00
1.07929921e+00 6.96385264e-01 5.66273034e-01 -5.12582541e-01
5.73786020e-01 4.13733184e-01 1.78081706e-01 -2.86222529e-02
2.75981456e-01 5.27819730e-02 -4.31921422e-01 6.09912336e-01
-2.16387272e-01 -5.04505396e-01 -1.13097377e-01 1.27508864e-01
8.59427512e-01 -3.05058300e-01 3.34313869e-01 -4.66254383e-01
9.06244874e-01 2.06499115e-01 1.03815114e+00 3.95995498e-01
-9.81312811e-01 -6.91076368e-02 1.25548810e-01 -1.20070183e+00
-6.40740514e-01 -6.63945556e-01 3.83635312e-02 6.06496811e-01
4.98964876e-01 -6.86011195e-01 -1.23077345e+00 -1.16119051e+00
-2.08360448e-01 7.13294506e-01 -2.66662687e-01 -7.99114645e-01
-4.16032135e-01 -2.48921290e-01 9.18227434e-01 3.94369900e-01
1.05245042e+00 -1.43885720e+00 -7.50854611e-01 -7.43819028e-02
-1.35925695e-01 -1.37522197e+00 -3.09739172e-01 -1.39659733e-01
-2.04058796e-01 -1.21221745e+00 -9.39108282e-02 -6.06258690e-01
1.70214370e-01 2.87801325e-01 6.84744477e-01 5.48478067e-01
-6.13108456e-01 2.81363904e-01 -2.98992246e-01 -1.33367527e+00
-1.93357497e-01 -7.40602463e-02 4.91400689e-01 1.54330991e-02
8.44117403e-01 3.92611176e-01 -3.30253959e-01 5.38332999e-01
-7.74602532e-01 -2.41180018e-01 1.39437318e-01 7.87920117e-01
-9.53303427e-02 6.92019641e-01 8.15163255e-01 -6.09394491e-01
5.46057403e-01 -3.13755184e-01 -1.08130419e+00 4.22883570e-01
-8.32445443e-01 -6.96300685e-01 5.03910482e-01 -1.23723634e-02
-8.81200194e-01 1.23141535e-01 -1.27753280e-02 -3.16331923e-01
-4.25077081e-01 1.98257938e-01 -6.05775177e-01 -4.69167203e-01
5.16977906e-01 -1.00504488e-01 2.04493627e-01 3.76120716e-01
1.08566046e-01 1.02400649e+00 4.42631751e-01 -5.99926710e-01
1.04196393e+00 2.36363739e-01 7.56854713e-02 -7.34694183e-01
-6.02105498e-01 2.85272319e-02 -2.89000243e-01 -5.24115026e-01
1.30733347e+00 -6.96061790e-01 -1.20542789e+00 9.21503544e-01
-1.28909588e+00 2.00628415e-01 7.80192316e-02 5.52019119e-01
-5.33139855e-02 2.05025256e-01 -9.07664239e-01 -8.37354898e-01
-5.10630608e-01 -1.36775053e+00 6.65210247e-01 5.49779713e-01
-2.36552492e-01 -5.59363306e-01 -1.90208644e-01 7.08438635e-01
4.82703209e-01 2.21299902e-01 5.23430586e-01 -8.56782615e-01
-5.28257608e-01 -4.73685086e-01 9.06947032e-02 7.62405813e-01
-2.46765137e-01 4.73386288e-01 -8.59275997e-01 -1.80492893e-01
2.71749079e-01 -2.78748244e-01 3.15231115e-01 1.61888525e-01
1.07523727e+00 -1.85181081e-01 -5.54657578e-01 1.36375010e-01
9.15814698e-01 1.31917739e+00 9.47925985e-01 7.11295843e-01
3.14691037e-01 8.37515712e-01 8.79599929e-01 2.75311649e-01
2.24287897e-01 4.69780356e-01 5.30443549e-01 1.97084453e-02
7.76343584e-01 1.74331009e-01 5.17285407e-01 5.55668890e-01
-2.67286837e-01 -2.84263100e-02 -1.20513403e+00 -4.38533811e-04
-1.88728070e+00 -7.77559698e-01 -7.10516761e-04 1.77612936e+00
8.02890807e-02 1.89970508e-01 1.23415829e-03 2.90377289e-01
6.36178732e-01 -2.80112505e-01 -3.97678226e-01 -7.60967672e-01
-1.06398113e-01 9.95992422e-02 1.98068619e-01 3.74406993e-01
-1.22225070e+00 1.43104649e+00 6.53023958e+00 1.03677654e+00
-1.16011012e+00 -1.74031649e-02 9.17408347e-01 3.42286557e-01
9.92562771e-02 -1.89993884e-02 -8.72697413e-01 3.13046008e-01
1.26893115e+00 -5.72112679e-01 4.77661520e-01 1.05044150e+00
-4.33807075e-01 2.95914203e-01 -3.01273018e-01 6.74677312e-01
8.49303380e-02 -1.49890053e+00 -1.80454597e-01 4.92647737e-01
6.14269137e-01 -2.47607147e-03 2.38487914e-01 3.56694698e-01
6.69923842e-01 -1.08393407e+00 4.57383603e-01 1.25312597e-01
3.23898107e-01 -1.48911631e+00 9.11023796e-01 3.73091161e-01
-9.98964369e-01 -7.75453806e-01 -2.81603068e-01 -1.87792376e-01
-9.60179120e-02 2.65046448e-01 -9.40950662e-02 9.31626856e-01
1.04162753e+00 4.11872119e-01 -2.00746268e-01 5.78112245e-01
-3.67183447e-01 3.73243123e-01 4.91205379e-02 -1.42884240e-01
2.97135234e-01 -4.56543118e-01 4.88856614e-01 4.45782751e-01
-1.84398461e-02 7.03579903e-01 3.79046589e-01 4.46445972e-01
1.94156691e-01 -7.08025038e-01 -1.00845373e+00 -2.54605025e-01
5.58249474e-01 1.35534155e+00 -6.66401267e-01 -2.37751827e-01
-1.75613433e-01 4.63367820e-01 -3.59973907e-01 1.75896838e-01
-9.43164587e-01 -4.23461676e-01 1.05656922e+00 -5.12636662e-01
-3.62283915e-01 -3.89435142e-01 -7.45474517e-01 -8.99481833e-01
-6.44954503e-01 -1.38458002e+00 7.30541348e-01 -5.86215258e-01
-1.15395606e+00 1.20824230e+00 4.46508788e-02 -1.07835805e+00
-2.72288769e-01 -9.92651224e-01 -6.69460833e-01 6.26672626e-01
-1.12098157e+00 -1.27517450e+00 -5.09476513e-02 7.46858895e-01
3.12977314e-01 -1.03900063e+00 1.15872145e+00 2.56724544e-02
-9.61877108e-01 6.37364089e-01 -4.03934389e-01 2.61781305e-01
1.52721271e-01 -5.16743004e-01 5.21524489e-01 1.07697225e+00
-1.07423149e-01 5.31261325e-01 2.88773686e-01 -5.92765093e-01
-1.62513828e+00 -9.13713157e-01 7.72300243e-01 -6.10723376e-01
6.78056896e-01 -3.40324908e-01 -5.31372786e-01 7.15005338e-01
7.24449039e-01 -1.42407596e-01 6.16236687e-01 -4.28319067e-01
-4.89116907e-01 -2.68024147e-01 -1.37185287e+00 7.21699476e-01
4.59625244e-01 -4.01798189e-01 -2.02103436e-01 2.78003693e-01
9.62340415e-01 -8.47062543e-02 -5.18713295e-01 6.83608055e-01
2.89344430e-01 -4.95715737e-01 8.78789306e-01 -6.17218018e-01
7.63195455e-02 -5.35390496e-01 2.44360212e-02 -8.81066561e-01
-3.64791214e-01 -7.26671278e-01 6.06355928e-02 7.94940591e-01
2.06945226e-01 -9.29280043e-01 9.35138464e-01 9.60179090e-01
-8.86992887e-02 -4.36640948e-01 -9.90035415e-01 -9.54120398e-01
3.30185682e-01 -5.86712599e-01 9.05979037e-01 9.33239937e-01
2.38819763e-01 2.64564276e-01 -4.04916197e-01 2.02913970e-01
6.22231126e-01 -4.28577423e-01 6.46822274e-01 -1.35800517e+00
2.41371751e-01 -6.23402953e-01 -7.28433907e-01 -2.29994804e-01
4.97818500e-01 -4.19213146e-01 -1.51124552e-01 -9.35991168e-01
-4.65960175e-01 -1.46044269e-01 -4.56428528e-01 8.11695039e-01
3.93051803e-02 1.61615148e-01 5.01662195e-01 1.00462750e-01
-3.66030604e-01 2.95200169e-01 1.00147998e+00 -4.59619403e-01
4.39000428e-01 3.20248157e-01 -8.62308264e-01 6.19971156e-01
1.15089977e+00 -3.19952965e-02 -3.10161382e-01 -2.36219287e-01
-3.40696096e-01 -7.64779523e-02 2.25067735e-01 -1.02240181e+00
6.67872369e-01 -6.31371439e-01 2.06793725e-01 -4.51429188e-01
3.84296685e-01 -1.30981696e+00 2.85001725e-01 1.16636646e+00
1.10100862e-02 2.93319583e-01 2.70458132e-01 -1.47591159e-01
-4.05804992e-01 5.85131757e-02 9.01506722e-01 7.87760541e-02
-7.17450738e-01 4.23390388e-01 -9.66008186e-01 -6.28235161e-01
2.01413679e+00 -1.85947612e-01 -9.07026410e-01 -1.12254158e-01
-2.55534649e-01 6.66870236e-01 -2.34065838e-02 8.98800313e-01
9.19089556e-01 -1.35971653e+00 -7.52458453e-01 5.24768054e-01
-2.47636624e-02 -3.68352920e-01 2.22486913e-01 4.67223227e-01
-6.29240036e-01 7.51616120e-01 -5.54086626e-01 -7.04528317e-02
-1.40022659e+00 9.04859483e-01 2.04805061e-01 -1.07627101e-01
-2.72488534e-01 8.15984130e-01 2.29935631e-01 -6.26652718e-01
4.71060961e-01 6.84721529e-01 -4.63193387e-01 -5.90516925e-01
9.57821906e-01 3.34689736e-01 -1.14966566e-02 -6.60463631e-01
-6.41367257e-01 3.74947995e-01 -3.52738261e-01 -1.04528703e-01
8.83717418e-01 3.43979269e-01 -4.98681366e-01 -1.88472718e-01
8.64824891e-01 -6.73067391e-01 -5.17978013e-01 4.72079575e-01
5.53599000e-01 -2.19569638e-01 4.02877629e-02 -9.80458021e-01
-1.49702072e+00 8.14484537e-01 5.80998838e-01 2.21747294e-01
1.32040036e+00 -5.63216567e-01 1.05655468e+00 3.95709932e-01
9.57870662e-01 -1.08097005e+00 -2.76603047e-02 1.09376907e+00
9.48900998e-01 -9.40344155e-01 -4.18078929e-01 -5.23378074e-01
-1.18447852e+00 1.14050496e+00 1.13252592e+00 -2.50804484e-01
1.32555521e+00 6.36068761e-01 5.25243461e-01 -3.03291380e-01
-9.56075191e-01 1.80883244e-01 1.38816729e-01 1.20306540e+00
-1.77387279e-02 1.87079832e-01 -4.50219303e-01 7.08849907e-01
-4.08840291e-02 1.82695702e-01 3.22218210e-01 1.15995681e+00
-1.17189355e-01 -1.32691395e+00 -5.30494213e-01 -7.63783827e-02
-4.34523851e-01 2.16616541e-01 -9.75796103e-01 6.24648273e-01
3.10374707e-01 1.29078734e+00 -2.27621153e-01 -1.22450793e+00
3.26411575e-01 -4.30058479e-01 -2.90432096e-01 -5.23441173e-02
-1.10940385e+00 -2.00472504e-01 1.61404774e-01 -6.90490603e-01
1.50513485e-01 -3.46165746e-01 -1.49025154e+00 -6.18380427e-01
-3.12768996e-01 1.01334453e+00 1.04107380e+00 6.83655918e-01
5.78060389e-01 2.93449789e-01 9.79662120e-01 -5.24170339e-01
-5.99044710e-02 -2.96277940e-01 -5.52612185e-01 -4.10728455e-01
-3.73669267e-01 -5.63523650e-01 -1.67096421e-01 -3.56972188e-01] | [5.515892028808594, 7.406710624694824] |
cc447856-ef0c-49bb-9294-05dc74fb1ad8 | learning-scene-dynamics-from-point-cloud | 2111.08755 | null | https://arxiv.org/abs/2111.08755v1 | https://arxiv.org/pdf/2111.08755v1.pdf | Learning Scene Dynamics from Point Cloud Sequences | Understanding 3D scenes is a critical prerequisite for autonomous agents. Recently, LiDAR and other sensors have made large amounts of data available in the form of temporal sequences of point cloud frames. In this work, we propose a novel problem -- sequential scene flow estimation (SSFE) -- that aims to predict 3D scene flow for all pairs of point clouds in a given sequence. This is unlike the previously studied problem of scene flow estimation which focuses on two frames. We introduce the SPCM-Net architecture, which solves this problem by computing multi-scale spatiotemporal correlations between neighboring point clouds and then aggregating the correlation across time with an order-invariant recurrent unit. Our experimental evaluation confirms that recurrent processing of point cloud sequences results in significantly better SSFE compared to using only two frames. Additionally, we demonstrate that this approach can be effectively modified for sequential point cloud forecasting (SPF), a related problem that demands forecasting future point cloud frames. Our experimental results are evaluated using a new benchmark for both SSFE and SPF consisting of synthetic and real datasets. Previously, datasets for scene flow estimation have been limited to two frames. We provide non-trivial extensions to these datasets for multi-frame estimation and prediction. Due to the difficulty of obtaining ground truth motion for real-world datasets, we use self-supervised training and evaluation metrics. We believe that this benchmark will be pivotal to future research in this area. All code for benchmark and models will be made accessible. | ['Anand Rangarajan', 'Sanjay Ranka', 'Patrick Emami', 'Pan He'] | 2021-11-16 | null | null | null | null | ['scene-flow-estimation'] | ['computer-vision'] | [ 2.55892128e-01 -5.32267094e-01 -1.82561984e-03 -1.83329895e-01
-4.01323706e-01 -5.63010573e-01 8.60344946e-01 4.16666158e-02
-4.29720819e-01 6.06716573e-01 -1.18447185e-01 -3.36218417e-01
1.03375174e-01 -8.05399060e-01 -6.54853463e-01 -5.08477390e-01
-4.78874922e-01 3.59082550e-01 8.17463756e-01 -1.24067143e-01
4.29599285e-01 9.72591579e-01 -1.91606307e+00 2.70945877e-01
5.86817026e-01 9.59434509e-01 5.58736205e-01 1.19889188e+00
-1.79915383e-01 1.00741243e+00 -4.60383952e-01 -8.57934728e-02
5.54073334e-01 -2.14410543e-01 -7.76586354e-01 1.14424601e-01
7.17618823e-01 -4.69714135e-01 -4.33973104e-01 7.20600724e-01
1.43096283e-01 3.46496910e-01 3.45799506e-01 -1.47030675e+00
2.34834000e-01 -2.02998966e-01 -5.25172472e-01 6.37409210e-01
4.21195358e-01 3.71073544e-01 6.77155435e-01 -7.65830040e-01
8.27396810e-01 1.27182114e+00 6.61283553e-01 3.22782457e-01
-9.85637844e-01 -4.60149050e-01 2.30582416e-01 5.13679504e-01
-1.19287395e+00 -4.07485276e-01 8.35177541e-01 -7.12969244e-01
1.29274023e+00 2.50933945e-01 9.36668694e-01 6.99282825e-01
1.73933938e-01 6.79050744e-01 6.79647744e-01 -1.35877684e-01
2.43917897e-01 -3.47832441e-01 -7.05706477e-02 5.27332008e-01
1.13884717e-01 4.06807780e-01 -5.70837736e-01 4.27708998e-02
9.79610145e-01 4.97099161e-02 -1.99450806e-01 -3.75211447e-01
-1.68909872e+00 6.84342027e-01 3.15350920e-01 2.10424766e-01
-5.21461666e-01 4.43851769e-01 3.19187015e-01 2.45695144e-01
7.26313710e-01 6.84435144e-02 -3.30962747e-01 -4.62603241e-01
-1.09865177e+00 4.39815164e-01 9.09270644e-01 1.12215924e+00
9.56622660e-01 1.66930452e-01 2.43808702e-01 2.58561045e-01
1.26565889e-01 6.06427550e-01 1.45039305e-01 -1.58913088e+00
3.98775399e-01 2.61436403e-01 2.99655288e-01 -1.32420754e+00
-3.76270860e-01 -9.74712148e-02 -7.95407891e-01 3.71206194e-01
5.22836745e-01 -5.84740192e-02 -4.83718365e-01 1.35005450e+00
4.60054457e-01 1.16380882e+00 1.24678060e-01 9.68270600e-01
5.46996653e-01 1.05752981e+00 -1.48933172e-01 -4.47032630e-01
9.20667589e-01 -9.70311046e-01 -4.35718030e-01 -3.36214080e-02
5.60587883e-01 -8.47183764e-01 6.93926156e-01 8.14282894e-02
-1.03545344e+00 -6.69542134e-01 -8.41879189e-01 -1.05759716e-02
-1.58561707e-01 -3.19461524e-01 7.31008708e-01 1.83026969e-01
-1.21527791e+00 7.36197829e-01 -1.38759267e+00 -5.92894256e-01
3.39872569e-01 2.40608007e-01 -2.11752176e-01 4.86299619e-02
-7.24054575e-01 8.32656085e-01 1.40573248e-01 1.11862995e-01
-7.18775392e-01 -9.24643934e-01 -8.31160188e-01 -1.77616313e-01
8.76457840e-02 -9.46759939e-01 1.29223788e+00 -7.78787971e-01
-1.47888947e+00 5.67794442e-01 -7.63708353e-01 -8.09894621e-01
6.22767091e-01 -1.46193936e-01 -1.69604287e-01 2.91944712e-01
1.95672587e-01 8.79291415e-01 7.41151512e-01 -1.23716259e+00
-1.09961534e+00 -1.48636311e-01 1.36139348e-01 2.35780045e-01
1.48917511e-01 -5.04415296e-02 -3.85400862e-01 -4.26002085e-01
2.38464735e-02 -1.07477498e+00 -5.06041229e-01 1.00092366e-01
-2.44726241e-02 -9.67276245e-02 1.17458940e+00 -2.12160736e-01
7.94214666e-01 -1.83740497e+00 -1.33493096e-02 -1.05823472e-01
5.96388355e-02 2.47450262e-01 -9.66700763e-02 3.62481356e-01
7.70737678e-02 -1.59084707e-01 -2.99405783e-01 -6.40814781e-01
-4.31156933e-01 5.10055006e-01 -7.23267078e-01 6.37380064e-01
2.64537275e-01 7.41247296e-01 -1.13332033e+00 -5.26623905e-01
9.17002082e-01 5.57706892e-01 -6.49878323e-01 1.85621588e-03
-3.73055458e-01 7.86157310e-01 -3.66488010e-01 3.24808031e-01
8.54167402e-01 -4.06503767e-01 -2.56682813e-01 -7.70678669e-02
-6.21351898e-01 1.39789090e-01 -1.26796973e+00 1.80210006e+00
-5.38929045e-01 1.03316605e+00 -3.96251947e-01 -8.66592407e-01
7.83785641e-01 8.38476419e-02 1.01801693e+00 -4.16402549e-01
-8.06169286e-02 9.02584717e-02 -3.50139320e-01 -4.94296402e-01
9.12429988e-01 8.81458372e-02 2.78435767e-01 1.55178502e-01
-2.53267199e-01 -5.59940040e-01 4.83719170e-01 2.00511724e-01
1.14801669e+00 2.91795433e-01 1.35327369e-01 -3.64663079e-02
7.65428066e-01 5.23423374e-01 5.72234809e-01 6.55917704e-01
-4.06966597e-01 6.91703081e-01 1.82422325e-01 -8.04066956e-01
-1.21744311e+00 -9.07173991e-01 6.22303747e-02 4.22785461e-01
6.18225932e-01 -5.80843568e-01 -2.48110101e-01 -2.75626779e-01
-1.46804778e-02 6.36140287e-01 -2.56235242e-01 3.99907410e-01
-1.05066705e+00 -5.02350509e-01 1.15491711e-01 3.89369816e-01
4.20397460e-01 -8.79393101e-01 -1.22586155e+00 3.52057755e-01
-3.54277581e-01 -1.67408717e+00 -2.39535302e-01 -2.97269940e-01
-1.13070333e+00 -1.09003198e+00 -4.81517285e-01 -5.40462613e-01
3.27094942e-01 9.20731187e-01 1.18541431e+00 1.57952771e-01
-5.71379401e-02 5.77408850e-01 -2.98362255e-01 -3.69815677e-01
-4.09529567e-01 -1.22643895e-01 -4.40694131e-02 -4.67483234e-03
3.09865624e-01 -8.13719690e-01 -7.06294835e-01 3.01934004e-01
-7.79704809e-01 3.25620979e-01 -4.16699946e-02 3.05347174e-01
7.03182042e-01 -1.87483639e-01 2.89749317e-02 -4.17290717e-01
8.24140236e-02 -3.58752012e-01 -1.04843402e+00 -1.63641661e-01
-1.66558940e-02 -1.84212908e-01 5.31134903e-01 -2.09891081e-01
-9.05716896e-01 3.82420301e-01 -4.55057435e-02 -8.98450196e-01
-3.46956193e-01 1.84610799e-01 5.60857415e-01 -1.53305203e-01
3.97226036e-01 2.25426599e-01 2.98401527e-02 -3.27282622e-02
2.34000072e-01 2.65300095e-01 7.87445009e-01 -3.33531320e-01
8.88476610e-01 1.13923371e+00 4.65931028e-01 -1.11332023e+00
-6.54741883e-01 -9.36913371e-01 -8.26246440e-01 -6.14191413e-01
9.53007281e-01 -1.05630839e+00 -1.02398241e+00 4.27244902e-01
-1.75029838e+00 -3.67630064e-01 -3.39349657e-01 7.31093228e-01
-9.12122667e-01 5.72876155e-01 -5.47839582e-01 -8.23390007e-01
2.89105624e-02 -1.18948293e+00 1.32073581e+00 9.03392560e-04
-9.28154811e-02 -1.15992832e+00 4.22834873e-01 8.79477188e-02
2.91685909e-01 5.22588789e-01 1.84459046e-01 -8.01337212e-02
-1.24324429e+00 1.47078589e-01 -1.01942331e-01 1.40142664e-02
1.13434918e-01 3.63983810e-01 -8.94929230e-01 -2.44521528e-01
1.98772982e-01 1.39149904e-01 9.13734615e-01 6.56866491e-01
9.91792083e-01 5.71227595e-02 -2.98352331e-01 7.68358767e-01
1.58598447e+00 2.93054044e-01 5.16363561e-01 4.79547799e-01
7.77332783e-01 5.56548655e-01 7.74508774e-01 5.79444349e-01
6.94087684e-01 7.70171940e-01 6.21891260e-01 1.01630993e-01
-2.09080279e-01 -7.37657174e-02 2.23572195e-01 8.59316826e-01
-4.93439376e-01 -3.54844898e-01 -1.09606421e+00 7.09046721e-01
-2.06637406e+00 -1.45339942e+00 -5.55689871e-01 1.98527336e+00
8.51464048e-02 -1.00248314e-01 6.38510734e-02 9.00479555e-02
6.38002813e-01 3.69993418e-01 -5.38609087e-01 8.10599700e-02
-1.33243233e-01 -1.11555876e-02 6.47570431e-01 8.41902256e-01
-1.15046835e+00 1.12142646e+00 5.94836855e+00 3.68071526e-01
-1.32636762e+00 5.27944900e-02 3.07635307e-01 -1.64566085e-01
-2.78876796e-02 2.10244894e-01 -8.36367130e-01 5.45259297e-01
9.98365581e-01 -2.13433966e-01 3.11823487e-01 5.57195902e-01
6.58994257e-01 -3.24081779e-01 -1.01411188e+00 1.21586823e+00
-1.25693902e-01 -1.80599439e+00 -5.24687907e-03 -3.70152071e-02
7.32241869e-01 5.66021919e-01 -2.55403519e-01 -1.56885847e-01
3.50339293e-01 -4.63296980e-01 7.25349963e-01 5.57252526e-01
3.84842575e-01 -5.10298848e-01 4.26036060e-01 4.37400520e-01
-1.55155981e+00 9.27736536e-02 -4.45207745e-01 -2.97184080e-01
7.53021121e-01 5.78359604e-01 -7.34872580e-01 7.86056936e-01
7.45094955e-01 1.51013660e+00 -3.09912384e-01 1.28705370e+00
2.12051436e-01 3.02610010e-01 -7.10796833e-01 2.24978969e-01
3.32907587e-01 -3.34319800e-01 9.09950078e-01 1.17422140e+00
6.56845212e-01 1.90163448e-01 2.53689021e-01 6.13035262e-01
4.42864895e-01 -2.60940999e-01 -8.48940492e-01 4.23460156e-01
3.95908356e-01 1.00046313e+00 -8.75602722e-01 -5.27589142e-01
-6.47077739e-01 7.75007248e-01 -2.18176516e-03 3.33492160e-01
-7.81547248e-01 5.78902848e-02 1.03271067e+00 8.73114448e-03
4.13469344e-01 -8.74386966e-01 -2.18234524e-01 -1.48519874e+00
-1.39950007e-01 -2.29085043e-01 2.13361546e-01 -8.27987075e-01
-1.04772151e+00 7.33807683e-01 2.43499056e-01 -1.88385975e+00
-5.91064274e-01 -3.12753171e-01 -5.40613294e-01 6.50160015e-01
-2.06637645e+00 -9.76859212e-01 -6.53209090e-01 7.49731302e-01
9.01411951e-01 1.35836959e-01 3.68996799e-01 2.42527053e-01
-2.39711910e-01 -3.95876795e-01 -1.02024630e-01 -2.83620358e-01
3.32435131e-01 -8.88593018e-01 8.61033976e-01 1.10716987e+00
3.16558331e-01 2.35564653e-02 9.23165917e-01 -6.47963643e-01
-1.33249807e+00 -1.27542377e+00 1.02545249e+00 -6.36329412e-01
7.95581937e-01 -6.98565170e-02 -8.19443762e-01 7.47598171e-01
8.34810659e-02 3.98842365e-01 1.79072455e-01 -6.09925449e-01
2.40734622e-01 -8.65722522e-02 -7.86554039e-01 3.73697132e-01
1.33833623e+00 -3.25590342e-01 -2.46055678e-01 4.05410022e-01
9.67923522e-01 -5.87995470e-01 -8.18576992e-01 5.78200161e-01
1.77130967e-01 -1.19848073e+00 1.14081001e+00 -2.18785465e-01
5.07369518e-01 -6.49165630e-01 -2.97936141e-01 -1.12786770e+00
8.51269520e-04 -5.90309560e-01 -2.37331137e-01 6.56613767e-01
-1.43715709e-01 -4.54714030e-01 1.18310881e+00 3.76973838e-01
-2.51559794e-01 -2.62231380e-01 -9.13786292e-01 -8.69907498e-01
-2.15390161e-01 -8.74123573e-01 5.54233551e-01 9.35497999e-01
-5.34859836e-01 7.97748044e-02 -2.77112305e-01 5.72935879e-01
7.53992498e-01 4.91364151e-01 1.17374730e+00 -1.12451911e+00
2.20879186e-02 -3.18430990e-01 -7.85424113e-01 -1.43123293e+00
3.95682722e-01 -6.25728011e-01 -1.01469398e-01 -1.46436298e+00
-3.83075565e-01 -5.11042416e-01 3.05889040e-01 -9.22355652e-02
9.09671560e-02 2.05810010e-01 7.21851468e-01 5.16686440e-01
-6.14552796e-01 6.08057976e-01 1.29012716e+00 -1.76245645e-02
-3.36056411e-01 2.25611478e-01 3.43964666e-01 7.78898299e-01
6.26458943e-01 -3.04142356e-01 -5.89056671e-01 -6.64437771e-01
-7.76871890e-02 4.54741925e-01 7.01315165e-01 -1.43319309e+00
6.58990383e-01 -4.16991264e-01 8.26453716e-02 -1.22614539e+00
7.65798092e-01 -1.00953245e+00 3.73925388e-01 4.60106760e-01
3.26684155e-02 5.28530300e-01 3.53974789e-01 6.38062119e-01
-3.91272068e-01 7.60057718e-02 6.28983378e-01 -2.81423271e-01
-1.22093534e+00 6.11439705e-01 -3.87799144e-01 -1.58879459e-01
1.25818598e+00 -4.78533447e-01 -3.43980461e-01 -5.07579684e-01
-4.99618113e-01 3.00458699e-01 5.74780107e-01 4.10655171e-01
7.80216217e-01 -1.09256732e+00 -7.31764913e-01 1.57109782e-01
-1.04822852e-01 4.48075622e-01 2.68616945e-01 7.05111325e-01
-9.63089824e-01 5.96566498e-01 -1.83380842e-01 -1.30505776e+00
-1.32380509e+00 4.91230786e-01 2.52629489e-01 -1.08137324e-01
-7.92759597e-01 4.23533469e-01 1.18574739e-01 -2.85913199e-01
-9.60587189e-02 -5.90515316e-01 -1.96323723e-01 -1.33446738e-01
4.69605744e-01 5.23150265e-01 2.09992826e-02 -1.06588197e+00
-2.32151598e-01 8.26017857e-01 3.07050288e-01 -8.31675082e-02
1.50010598e+00 -2.91702569e-01 -9.97222438e-02 5.85800886e-01
1.26342177e+00 -3.23452294e-01 -1.63321114e+00 -1.34730011e-01
-8.77106935e-02 -8.39602828e-01 3.06925108e-03 7.77628794e-02
-1.12972617e+00 9.02611375e-01 5.09612739e-01 2.69845873e-01
1.00577271e+00 -1.31901696e-01 8.47307444e-01 3.63619983e-01
6.76597834e-01 -4.49596494e-01 -7.02566877e-02 8.48688185e-01
5.78109562e-01 -1.18869221e+00 -7.69836502e-03 -6.40240669e-01
-3.04272532e-01 1.16320634e+00 5.05107164e-01 -2.49482691e-01
7.03165472e-01 3.17672580e-01 -1.43007129e-01 -1.01625152e-01
-1.08629966e+00 -3.12112093e-01 -9.11364332e-02 5.36225975e-01
4.87808250e-02 -2.46856481e-01 2.11266384e-01 -5.65443933e-01
-3.14810991e-01 3.58644485e-01 7.42820680e-01 1.12034297e+00
-2.86210209e-01 -9.96897817e-01 -3.83126348e-01 2.04530180e-01
-1.75068993e-03 9.15939510e-02 1.06307223e-01 7.57466018e-01
-9.14257243e-02 9.86618876e-01 7.02773213e-01 -2.59230524e-01
3.71756017e-01 -4.43510771e-01 4.72543418e-01 -3.05884808e-01
-2.99693733e-01 -1.18439570e-01 -1.34148881e-01 -9.55192983e-01
-1.20640481e+00 -9.96070862e-01 -1.15993643e+00 -7.23547935e-01
4.45642881e-02 -3.52185145e-02 6.67315960e-01 7.36901879e-01
4.97537911e-01 3.40350121e-01 6.50511861e-01 -1.49700737e+00
7.78419003e-02 -4.71599728e-01 -1.48508266e-01 3.65393609e-01
8.10935557e-01 -5.96141279e-01 -3.78923059e-01 4.09268081e-01] | [8.505866050720215, -1.9682425260543823] |
7f9bb06a-14f6-49b3-bd85-9cb0e30b341a | learning-to-execute | 1410.4615 | null | http://arxiv.org/abs/1410.4615v3 | http://arxiv.org/pdf/1410.4615v3.pdf | Learning to Execute | Recurrent Neural Networks (RNNs) with Long Short-Term Memory units (LSTM) are
widely used because they are expressive and are easy to train. Our interest
lies in empirically evaluating the expressiveness and the learnability of LSTMs
in the sequence-to-sequence regime by training them to evaluate short computer
programs, a domain that has traditionally been seen as too complex for neural
networks. We consider a simple class of programs that can be evaluated with a
single left-to-right pass using constant memory. Our main result is that LSTMs
can learn to map the character-level representations of such programs to their
correct outputs. Notably, it was necessary to use curriculum learning, and
while conventional curriculum learning proved ineffective, we developed a new
variant of curriculum learning that improved our networks' performance in all
experimental conditions. The improved curriculum had a dramatic impact on an
addition problem, making it possible to train an LSTM to add two 9-digit
numbers with 99% accuracy. | ['Ilya Sutskever', 'Wojciech Zaremba'] | 2014-10-17 | null | null | null | null | ['learning-to-execute'] | ['computer-code'] | [ 4.77658540e-01 -1.22217610e-01 -8.57981592e-02 -3.06431562e-01
-6.65885091e-01 -7.63139427e-01 4.90209162e-01 1.54726237e-01
-7.78512955e-01 7.65476227e-01 -1.30868614e-01 -9.73593116e-01
1.69667035e-01 -1.03542626e+00 -1.12091970e+00 -4.23863024e-01
-3.11234325e-01 2.01986015e-01 3.46316606e-01 -4.59543228e-01
4.28760827e-01 4.29789752e-01 -1.53284490e+00 5.44074833e-01
8.07136834e-01 6.58365428e-01 3.27563763e-01 1.32119608e+00
-2.49600455e-01 1.41497552e+00 -7.78068244e-01 -4.23055977e-01
-1.30242053e-02 -1.85028464e-01 -9.81310964e-01 -5.09792268e-01
3.87650162e-01 -5.31283259e-01 -3.83501023e-01 9.80601728e-01
3.62105131e-01 1.81512594e-01 3.25467050e-01 -8.08465838e-01
-6.39414966e-01 7.89311290e-01 -9.50841084e-02 2.15665892e-01
4.66306746e-01 1.56044349e-01 1.11315489e+00 -6.66223884e-01
5.76944530e-01 1.10314977e+00 9.10504818e-01 5.77005148e-01
-1.31480861e+00 -4.84773874e-01 -1.95875853e-01 2.78079323e-02
-9.37956750e-01 -2.51134306e-01 3.01160187e-01 -4.15951908e-01
1.61285949e+00 1.43834680e-01 4.60627407e-01 1.01663911e+00
2.22036660e-01 8.36965442e-01 9.74524021e-01 -6.28091753e-01
-1.14613511e-01 1.01280391e-01 2.17916369e-01 8.70844722e-01
-9.95590761e-02 3.57586056e-01 3.23521085e-02 -6.58099502e-02
1.00027454e+00 -6.19033128e-02 -1.31089553e-01 -2.09213480e-01
-1.34779024e+00 9.48137283e-01 5.42225003e-01 6.18885934e-01
3.36673334e-02 6.28283620e-01 7.72664309e-01 9.58707035e-01
1.06276032e-02 8.06808829e-01 -4.80586976e-01 -3.71534348e-01
-8.26698780e-01 2.63652503e-01 9.47737813e-01 7.96074510e-01
5.54403961e-01 5.35308123e-01 -8.79682600e-02 7.14279413e-01
-4.76920038e-01 3.74493092e-01 6.02764010e-01 -7.46986568e-01
5.23380280e-01 4.08963025e-01 -1.86176628e-01 -7.51031101e-01
-3.19685191e-01 -5.21647632e-01 -7.45474219e-01 2.97904462e-01
5.80221653e-01 -4.36358035e-01 -7.99057543e-01 1.74864495e+00
-4.42103326e-01 7.99857527e-02 1.30808532e-01 5.26889145e-01
3.62001956e-01 1.07291305e+00 1.25240371e-01 8.57986975e-03
7.81979978e-01 -7.44988501e-01 -2.83019990e-01 -2.16151461e-01
1.39964533e+00 -6.24112129e-01 1.40452051e+00 3.15569729e-01
-1.23301554e+00 -7.90596366e-01 -1.28700352e+00 -7.32504800e-02
-5.88998020e-01 -7.60351047e-02 6.96214378e-01 5.91427505e-01
-1.34120142e+00 1.17698908e+00 -4.63632166e-01 -8.39529466e-03
1.55993655e-01 5.62747002e-01 -2.85155714e-01 8.54426920e-02
-1.33648348e+00 1.24563420e+00 7.77539074e-01 -1.06921978e-01
-8.97062898e-01 -5.52476883e-01 -9.52727497e-01 3.66605371e-01
2.31750607e-01 -4.04317260e-01 1.64736497e+00 -1.53100491e+00
-1.48627996e+00 7.56727338e-01 1.49298012e-01 -6.90896630e-01
3.70913327e-01 1.04975007e-01 -4.15713042e-01 4.10874411e-02
-4.11189139e-01 5.47030628e-01 6.38017654e-01 -7.36941516e-01
-4.87304419e-01 9.10120159e-02 3.05419892e-01 -2.84035951e-01
-5.81012726e-01 3.14961135e-01 2.49168590e-01 -6.69172227e-01
-3.10573667e-01 -9.34529305e-01 -2.60569364e-01 -3.34841996e-01
-8.83195549e-02 -3.19158912e-01 5.72074592e-01 -5.17750084e-01
1.22734582e+00 -1.97329283e+00 1.18668884e-01 2.65055031e-01
6.92708641e-02 6.87881768e-01 -3.32751930e-01 2.81052858e-01
-3.83765817e-01 3.81331891e-01 -2.20051870e-01 2.09941342e-01
2.54391819e-01 2.90377468e-01 -5.72191834e-01 2.81706631e-01
4.66602117e-01 1.28366995e+00 -9.02471185e-01 -3.88156742e-01
4.20008693e-03 2.77647704e-01 -5.95283806e-01 3.61756563e-01
-5.68055868e-01 -1.08543165e-01 -6.12108447e-02 2.12145478e-01
5.26601672e-02 -3.43710124e-01 1.43478215e-01 5.67764759e-01
-1.42149597e-01 4.49950933e-01 -7.05696166e-01 1.52657235e+00
-6.82068765e-01 1.09243393e+00 -4.92058456e-01 -1.20344269e+00
8.32608163e-01 5.17383099e-01 1.13738924e-02 -9.97744858e-01
-5.98178152e-03 4.93444473e-01 2.76717722e-01 -5.00515103e-01
6.43183112e-01 -3.35780084e-01 -3.01872939e-01 6.29709959e-01
5.38630560e-02 -1.29779771e-01 3.74708444e-01 -8.83529894e-03
1.04673111e+00 1.01096831e-01 3.45882066e-02 -2.08123207e-01
4.98408020e-01 -6.77720457e-02 1.77049682e-01 8.83646905e-01
2.67312437e-01 2.22907573e-01 7.32700646e-01 -7.07956612e-01
-1.51096916e+00 -7.77564585e-01 2.27165759e-01 1.62964606e+00
-7.56727517e-01 -2.16624424e-01 -6.79287314e-01 -3.22293222e-01
-1.96440086e-01 8.21011484e-01 -5.40736377e-01 -2.83445030e-01
-1.18965089e+00 -3.14970315e-01 1.06031168e+00 9.07206655e-01
3.13241452e-01 -1.32318401e+00 -1.05128407e+00 4.96795952e-01
1.84941232e-01 -7.84223258e-01 -2.24351093e-01 7.20096290e-01
-1.11160815e+00 -9.15485620e-01 -1.14461851e+00 -1.08781707e+00
5.93554020e-01 -3.54471475e-01 1.26735544e+00 3.67947221e-01
-7.55161569e-02 -1.49970930e-02 3.05055995e-02 -2.08046019e-01
-8.93089116e-01 3.75348628e-01 -1.87032193e-01 -6.65818512e-01
1.50241002e-01 -5.76918244e-01 -1.04111833e-02 4.13725004e-02
-1.05042100e+00 -1.23434499e-01 6.70753777e-01 1.05142117e+00
4.32866141e-02 -1.01754382e-01 6.45633698e-01 -1.07366669e+00
9.17736053e-01 -3.56883481e-02 -9.08174276e-01 3.02457869e-01
-3.71757776e-01 5.26821554e-01 1.12856591e+00 -7.12763190e-01
-6.66341960e-01 -1.86099932e-02 -4.76767987e-01 -2.27155328e-01
1.31188139e-01 8.43533635e-01 3.49354953e-01 -4.37526733e-01
8.37305129e-01 3.59624892e-01 -1.90686524e-01 -7.10910037e-02
1.84136420e-01 3.57356608e-01 6.28234744e-01 -8.89689744e-01
5.34038961e-01 -4.48541850e-01 -1.17884144e-01 -6.10852957e-01
-3.02817076e-01 1.86986849e-02 -4.86213177e-01 -2.97230855e-02
4.00107533e-01 -5.94119012e-01 -1.06859958e+00 3.84604067e-01
-1.29328775e+00 -9.35367286e-01 -1.66651547e-01 9.94377881e-02
-6.24380171e-01 1.82134658e-01 -1.00418210e+00 -5.42192101e-01
-2.43834108e-01 -1.24016309e+00 4.61161017e-01 -9.60429013e-02
-4.14469361e-01 -1.24144685e+00 -3.18110641e-03 -5.45851707e-01
7.99323440e-01 3.09133023e-01 1.60381043e+00 -6.39564335e-01
-4.15822864e-01 -3.17303002e-01 -1.57203898e-01 5.81774652e-01
-4.15024102e-01 2.42985517e-01 -9.90606129e-01 -2.18763500e-01
-1.04085319e-01 -7.18557835e-01 9.34303164e-01 1.58228800e-01
1.24356222e+00 -2.96271771e-01 1.01896487e-01 4.81202960e-01
1.43176758e+00 3.01390350e-01 8.71809304e-01 4.69594568e-01
4.68065739e-01 3.23073119e-01 6.27117828e-02 6.28155103e-05
-1.78734228e-01 3.64371359e-01 1.35077806e-02 1.00642880e-02
1.98091060e-01 -3.42087418e-01 5.88854313e-01 1.10589933e+00
-1.45309987e-02 3.71799134e-02 -1.17647660e+00 5.65172791e-01
-1.54305887e+00 -1.02012014e+00 2.84745153e-02 2.24412060e+00
9.63131547e-01 6.76438391e-01 2.91471574e-02 3.72431725e-01
5.72642386e-01 1.10278204e-01 -2.68677473e-01 -1.07909417e+00
-5.60016856e-02 8.07896078e-01 5.54351568e-01 3.84802848e-01
-9.79964197e-01 8.57222199e-01 7.47769976e+00 7.68770695e-01
-1.23088264e+00 -1.13250710e-01 6.25936687e-01 7.19839036e-02
-4.17431623e-01 -2.33007148e-01 -4.73341107e-01 1.92413464e-01
1.68656611e+00 -1.74111158e-01 2.77921259e-01 9.31194961e-01
-2.43312359e-01 1.96283888e-02 -1.37414706e+00 6.76461518e-01
-3.09602201e-01 -1.51274073e+00 -8.57511722e-03 -2.97286421e-01
8.14906776e-01 -5.44837043e-02 2.43841469e-01 9.66569424e-01
5.10117292e-01 -1.54890192e+00 5.04900277e-01 2.46697903e-01
1.04321229e+00 -1.16045606e+00 6.25570893e-01 4.93011057e-01
-1.02006865e+00 -3.54672074e-01 -5.81776857e-01 -4.05676216e-01
-3.39210063e-01 1.34862483e-01 -9.04440701e-01 -8.71977434e-02
1.62719443e-01 4.78535414e-01 -5.96644521e-01 9.27536905e-01
-3.65672767e-01 4.88472790e-01 -3.17138769e-02 -6.13559067e-01
7.09174991e-01 1.55034408e-01 1.42307967e-01 1.57937515e+00
4.25103456e-01 -1.74023155e-02 -9.05224383e-02 9.40545082e-01
-3.82653087e-01 -2.02347800e-01 -9.85168397e-01 -4.43825871e-01
4.42386419e-02 7.35796750e-01 -5.52371204e-01 -5.36875069e-01
-2.11681619e-01 7.08853304e-01 5.53323567e-01 3.86891395e-01
-7.64150262e-01 -9.74046052e-01 1.00700200e-01 -2.01225504e-01
4.11021709e-01 -2.91790903e-01 -2.85663098e-01 -9.65719879e-01
-2.54450785e-03 -1.21883678e+00 2.66689330e-01 -7.10189879e-01
-6.67212248e-01 7.56900549e-01 -2.06627280e-01 -8.32926393e-01
-8.38817358e-01 -7.88766921e-01 -6.41321242e-01 8.91641915e-01
-1.47994077e+00 -7.28359461e-01 1.96897611e-01 6.24948800e-01
3.42107356e-01 -1.40602767e-01 1.20973623e+00 3.85981828e-01
-2.77324438e-01 9.02918398e-01 2.13687152e-01 4.50403422e-01
2.73393899e-01 -1.36036396e+00 6.11172259e-01 7.72962272e-01
1.93788912e-02 9.67558384e-01 7.37266183e-01 -2.41680682e-01
-1.56948006e+00 -9.86170173e-01 1.08721566e+00 -1.78676248e-01
8.52020085e-01 -3.84016216e-01 -1.00122917e+00 1.01987088e+00
1.64143771e-01 -6.94388226e-02 3.94246250e-01 9.68262181e-02
-6.43244982e-01 1.79115564e-01 -6.71718121e-01 6.79207981e-01
6.48540080e-01 -1.01638162e+00 -7.16447055e-01 2.05443755e-01
7.98158288e-01 -5.58274925e-01 -8.27845871e-01 2.74530768e-01
6.37971520e-01 -8.15074980e-01 8.60033989e-01 -8.48661959e-01
1.04267490e+00 2.25806117e-01 -1.17130540e-01 -1.26624167e+00
-6.40832782e-02 -5.62006652e-01 -3.25952210e-02 7.23540246e-01
6.11227155e-01 -6.95970714e-01 5.64631224e-01 5.56610405e-01
-2.36249909e-01 -6.82149470e-01 -5.40238142e-01 -9.13975060e-01
6.49806440e-01 -3.88940334e-01 4.89439845e-01 7.56245852e-01
3.19315821e-01 2.45680228e-01 -3.35570484e-01 -3.07005852e-01
-2.08796281e-02 3.52768868e-01 5.20954370e-01 -1.02851558e+00
-5.98606229e-01 -8.36687684e-01 -2.86172539e-01 -1.11801350e+00
5.28522134e-01 -1.02351367e+00 6.26062080e-02 -1.00687778e+00
4.35862765e-02 -3.21710855e-01 -3.83662164e-01 4.13059413e-01
-5.20889536e-02 -1.24598257e-01 1.62587807e-01 -1.75630957e-01
-3.66528779e-01 -1.85798763e-05 1.03292084e+00 -2.88657993e-01
9.96302143e-02 2.83880383e-01 -4.03535277e-01 5.00379145e-01
7.46537626e-01 -5.30459762e-01 -3.06079447e-01 -5.35987198e-01
7.09412515e-01 3.66775900e-01 2.55570024e-01 -1.02379441e+00
4.19080496e-01 1.12702779e-01 4.24397290e-01 -4.58956182e-01
1.67619795e-01 -5.78066647e-01 -1.29476160e-01 8.15371633e-01
-9.88262832e-01 4.08034235e-01 6.15340114e-01 7.43259564e-02
-2.07832798e-01 -7.95675278e-01 7.49059021e-01 -4.41602558e-01
-9.27396774e-01 -5.61768487e-02 -6.04945958e-01 -2.18639970e-02
6.68019354e-01 -8.91873538e-02 -2.27788940e-01 -4.08764064e-01
-3.99825126e-01 -9.97712240e-02 3.98222268e-01 -2.38555297e-02
6.42121851e-01 -1.16224110e+00 -6.24759018e-01 3.02428752e-01
-4.75806296e-02 -2.60767311e-01 -1.76665321e-01 5.10029078e-01
-8.49047482e-01 8.38668466e-01 -4.75813925e-01 -4.19593245e-01
-1.37100554e+00 7.11948693e-01 4.44856584e-01 -4.92798001e-01
-5.59135437e-01 9.53670323e-01 -1.78784817e-01 -7.64499247e-01
5.24576664e-01 -6.90727174e-01 8.45038425e-03 -2.42316246e-01
6.50745809e-01 3.27891335e-02 6.86046556e-02 -2.05687415e-02
3.65761667e-02 4.47001845e-01 -6.73913881e-02 4.16794745e-03
1.44395518e+00 7.07351089e-01 -2.09420279e-01 6.82903349e-01
1.54156232e+00 -4.09889042e-01 -8.82828236e-01 -2.21939713e-01
5.46147883e-01 -1.08284034e-01 -1.35127634e-01 -5.81053019e-01
-8.14498723e-01 1.41794181e+00 3.10714275e-01 2.70439714e-01
9.72192824e-01 -7.17891812e-01 1.00243318e+00 1.06635535e+00
3.31147015e-01 -9.95530009e-01 3.00575227e-01 1.20777774e+00
4.74450409e-01 -1.01479769e+00 -3.06233674e-01 2.61791259e-01
-3.16721648e-01 1.76505423e+00 3.77843469e-01 -3.17054242e-01
3.96756083e-03 5.32528698e-01 -3.16292524e-01 6.89092577e-02
-9.75853264e-01 8.53978992e-02 7.35731199e-02 3.17334235e-01
6.33974671e-01 -3.20135802e-02 3.54418725e-01 1.16233595e-01
-2.83905715e-01 3.57130975e-01 7.31102705e-01 1.14139915e+00
-6.13695860e-01 -9.58490729e-01 -1.89292714e-01 4.28182095e-01
-5.43229699e-01 -2.91053057e-01 -1.09385572e-01 8.46370220e-01
-4.24953550e-01 5.19703209e-01 3.68410200e-02 -4.59909409e-01
1.23822600e-01 3.47957432e-01 6.30788624e-01 -5.85200548e-01
-1.00618982e+00 -3.55598450e-01 2.59981036e-01 -3.32978547e-01
1.11691281e-03 -1.86914608e-01 -1.24406362e+00 -7.73744226e-01
-1.82332367e-01 1.63694043e-02 5.45504093e-01 8.76718163e-01
-1.13888435e-01 8.30462754e-01 4.68458623e-01 -5.44075549e-01
-1.08151305e+00 -7.90676236e-01 -3.53892446e-01 1.68809101e-01
6.66256070e-01 -9.26403031e-02 -1.53794497e-01 -1.02686891e-02] | [8.67476749420166, 7.216052055358887] |
ea813b8e-edaa-455b-97b9-8feebdeaecd1 | lifelong-learning-of-spatiotemporal | 1805.10966 | null | http://arxiv.org/abs/1805.10966v4 | http://arxiv.org/pdf/1805.10966v4.pdf | Lifelong Learning of Spatiotemporal Representations with Dual-Memory Recurrent Self-Organization | Artificial autonomous agents and robots interacting in complex environments
are required to continually acquire and fine-tune knowledge over sustained
periods of time. The ability to learn from continuous streams of information is
referred to as lifelong learning and represents a long-standing challenge for
neural network models due to catastrophic forgetting. Computational models of
lifelong learning typically alleviate catastrophic forgetting in experimental
scenarios with given datasets of static images and limited complexity, thereby
differing significantly from the conditions artificial agents are exposed to.
In more natural settings, sequential information may become progressively
available over time and access to previous experience may be restricted. In
this paper, we propose a dual-memory self-organizing architecture for lifelong
learning scenarios. The architecture comprises two growing recurrent networks
with the complementary tasks of learning object instances (episodic memory) and
categories (semantic memory). Both growing networks can expand in response to
novel sensory experience: the episodic memory learns fine-grained
spatiotemporal representations of object instances in an unsupervised fashion
while the semantic memory uses task-relevant signals to regulate structural
plasticity levels and develop more compact representations from episodic
experience. For the consolidation of knowledge in the absence of external
sensory input, the episodic memory periodically replays trajectories of neural
reactivations. We evaluate the proposed model on the CORe50 benchmark dataset
for continuous object recognition, showing that we significantly outperform
current methods of lifelong learning in three different incremental learning
scenarios | ['Cornelius Weber', 'Stefan Wermter', 'Jun Tani', 'German I. Parisi'] | 2018-05-28 | null | null | null | null | ['continuous-object-recognition'] | ['computer-vision'] | [ 4.51004744e-01 -1.28617771e-02 -1.43793285e-01 -1.55062079e-01
7.23107904e-02 -4.12372887e-01 9.15924847e-01 3.00280839e-01
-6.64570153e-01 1.17318368e+00 9.60578844e-02 3.09450567e-01
-5.12233496e-01 -9.69033241e-01 -1.10503078e+00 -9.97197390e-01
-5.14839351e-01 5.72597384e-01 5.72459519e-01 -5.98242953e-02
2.85489231e-01 5.56392372e-01 -2.26111913e+00 2.68320858e-01
5.82859814e-01 8.39481056e-01 6.33890450e-01 5.56031346e-01
-1.78916991e-01 8.34967554e-01 -3.82387608e-01 4.93243933e-01
-1.91619955e-02 -4.26191986e-01 -8.13036263e-01 9.02141817e-03
1.23542786e-01 1.13066301e-01 -5.35427928e-01 4.79931295e-01
3.35618734e-01 6.75526440e-01 5.42146087e-01 -8.30805004e-01
-8.84078503e-01 5.11715949e-01 -2.42050476e-02 5.81764400e-01
1.11700937e-01 3.12794656e-01 3.72382432e-01 -1.32058787e+00
7.97100782e-01 8.45671654e-01 6.87730610e-01 6.66431963e-01
-1.10158622e+00 -2.45071605e-01 4.18479711e-01 5.63231409e-01
-1.17860365e+00 -5.38566589e-01 4.75896120e-01 -1.80341229e-01
1.60939276e+00 -3.34618181e-01 1.06736505e+00 1.37508440e+00
7.45353460e-01 6.12199128e-01 1.07976699e+00 -2.00614825e-01
6.57280922e-01 -6.47712126e-02 2.57208765e-01 4.06785965e-01
2.53309697e-01 6.01877451e-01 -1.14019620e+00 1.06623113e-01
5.18738151e-01 5.54634094e-01 -1.72321096e-01 -3.82808894e-01
-1.10420513e+00 4.03072715e-01 5.71778238e-01 5.64975262e-01
-7.01298833e-01 1.19418621e-01 2.07508788e-01 7.26269364e-01
1.51448667e-01 5.28549612e-01 -5.46771944e-01 -2.08172619e-01
-7.48730838e-01 4.97362688e-02 4.96459633e-01 7.45584071e-01
9.62237298e-01 3.35719436e-01 -3.04014590e-02 9.68964875e-01
-1.96633577e-01 3.97984415e-01 1.36598933e+00 -4.92970765e-01
-2.47572601e-01 5.23134470e-01 -2.97596604e-02 -5.41516542e-01
-4.35959369e-01 -9.49568689e-01 -9.19166744e-01 2.08410755e-01
1.96059830e-02 2.48248249e-01 -1.25770080e+00 2.00028634e+00
-1.46964327e-01 4.57645953e-01 4.01545405e-01 4.96663839e-01
4.93176311e-01 9.33804393e-01 2.53651023e-01 -7.36876488e-01
8.61448884e-01 -8.96586180e-01 -5.43927133e-01 -5.75111210e-01
2.31633157e-01 1.02404319e-02 9.99914229e-01 2.35659912e-01
-1.02446771e+00 -6.52256906e-01 -1.28969932e+00 2.82632291e-01
-8.64588499e-01 -7.87506521e-01 3.11692059e-01 -5.43567464e-02
-1.19039714e+00 8.43440175e-01 -7.57134378e-01 -7.70919859e-01
5.07519305e-01 3.02710354e-01 -2.36833304e-01 -1.09534092e-01
-1.32726622e+00 9.67995763e-01 7.58415699e-01 -3.89391324e-03
-1.55291998e+00 -6.34691119e-01 -5.53845882e-01 2.04679564e-01
5.88130541e-02 -7.61577308e-01 1.02848542e+00 -8.99254322e-01
-1.48718536e+00 8.28547478e-01 -3.38979885e-02 -8.96127164e-01
-8.53293017e-02 -3.73229951e-01 -3.15821856e-01 -1.37921661e-01
-1.40507475e-01 6.80650294e-01 1.20059109e+00 -1.06564105e+00
-5.14056265e-01 -4.65225637e-01 -3.88920695e-01 3.57286841e-01
-5.23463964e-01 -7.38056719e-01 7.34932572e-02 -8.12826335e-01
8.58110562e-02 -1.08556259e+00 -1.16586521e-01 -3.45396966e-01
5.40329993e-01 -9.92238373e-02 1.13356721e+00 4.55800742e-02
9.75190341e-01 -2.16085005e+00 5.50564289e-01 -1.08461678e-01
-5.51830418e-02 3.48437726e-01 -3.14253300e-01 4.58326578e-01
-8.95606801e-02 -3.61878455e-01 -5.47236204e-01 3.54930796e-02
-4.27922934e-01 6.83337927e-01 -5.90997517e-01 7.93264657e-02
1.36480480e-01 1.14795184e+00 -1.06401420e+00 2.30166372e-02
-2.02016816e-01 3.91342252e-01 -1.59890518e-01 3.27551126e-01
-4.69328344e-01 3.34498227e-01 1.44028693e-01 4.79595602e-01
3.00784826e-01 -4.21023965e-01 6.83963373e-02 4.43846107e-01
-1.80104747e-01 -4.21909057e-02 -6.52544260e-01 1.95885026e+00
-6.12052739e-01 5.19709766e-01 -4.86644596e-01 -1.15111077e+00
9.58121002e-01 2.69515783e-01 2.23951593e-01 -1.17119777e+00
-1.48514390e-01 1.24985255e-01 -8.07492957e-02 -2.23520517e-01
3.57344627e-01 -2.45276570e-01 -4.02198173e-02 8.84447813e-01
6.34582877e-01 1.98952004e-01 1.82097912e-01 3.51027064e-02
1.39906979e+00 -5.98831177e-02 2.50092685e-01 -1.87521204e-01
1.76198155e-01 -1.23732187e-01 5.61918020e-01 1.16847086e+00
-1.59410790e-01 1.69583976e-01 -2.16823488e-01 -9.09981847e-01
-9.01505113e-01 -1.64711833e+00 -5.60095608e-02 1.57385504e+00
2.71214638e-02 7.18520954e-02 -1.40576154e-01 -2.83152044e-01
-7.11360574e-02 7.94312179e-01 -8.88873577e-01 -1.08019221e+00
-7.36938119e-01 -9.08209503e-01 2.58409113e-01 5.47710478e-01
5.73887944e-01 -1.98289704e+00 -1.33286417e+00 5.97360313e-01
3.77361447e-01 -5.57484567e-01 -1.37657121e-01 8.87826383e-01
-1.24849463e+00 -7.88594663e-01 -7.84920096e-01 -1.15367651e+00
6.94841146e-01 3.45530540e-01 1.09358740e+00 5.48697542e-04
-4.71970379e-01 6.42725885e-01 -3.70141536e-01 -1.82604343e-01
-1.10139266e-01 3.22903156e-01 5.48507929e-01 -8.27043802e-02
1.52868092e-01 -1.31142199e+00 -5.26381195e-01 1.79798469e-01
-1.23343635e+00 -1.00413896e-01 9.84003425e-01 1.36819816e+00
9.26323354e-01 1.89320803e-01 1.43535030e+00 -9.05548990e-01
5.52993894e-01 -9.13082957e-01 -2.08645090e-01 2.74099886e-01
-8.98449540e-01 4.05783057e-01 6.19042099e-01 -8.85431528e-01
-1.26478672e+00 3.07717118e-02 2.37846985e-01 -3.64888996e-01
-1.63302213e-01 7.33036697e-01 1.25593841e-01 6.02279697e-03
8.03823352e-01 1.07322812e+00 -1.32803708e-01 -2.12160394e-01
4.35971648e-01 2.17613820e-02 7.09869385e-01 -5.88417351e-01
4.42118138e-01 2.70204216e-01 -2.28102431e-01 -9.42710400e-01
-8.68747771e-01 -8.11231136e-02 -7.84559846e-01 -4.24519211e-01
3.32984507e-01 -4.99556601e-01 -2.20027179e-01 8.25498462e-01
-7.81578779e-01 -6.72878206e-01 -1.11368716e+00 2.13879883e-01
-9.20539379e-01 -4.14584875e-01 -7.06580997e-01 -4.91965622e-01
-2.47643515e-01 -4.08805907e-01 3.51680070e-01 5.92269480e-01
-2.97456458e-02 -8.52425873e-01 5.14982462e-01 -5.88144779e-01
7.95235753e-01 1.04516707e-01 1.09928989e+00 -3.33246976e-01
-6.62903488e-01 1.35991022e-01 2.52139151e-01 -5.35395257e-02
3.33951175e-01 -8.23687613e-01 -9.91611660e-01 -6.36865079e-01
4.70484160e-02 -6.82895660e-01 1.51450074e+00 9.26289260e-02
8.23992610e-01 -5.58214903e-01 -3.52274537e-01 3.75145137e-01
1.22679377e+00 4.93304074e-01 5.29496312e-01 3.37188303e-01
1.15498915e-01 4.77818072e-01 3.77088845e-01 4.14921165e-01
-4.05248366e-02 1.84480324e-01 3.23656529e-01 5.95413983e-01
-2.27166548e-01 -4.43766683e-01 3.38296741e-01 1.05140722e+00
7.80718997e-02 1.54985696e-01 -9.45956349e-01 9.87741709e-01
-2.00578046e+00 -1.21962559e+00 8.93213332e-01 2.18460417e+00
1.10617769e+00 2.81731308e-01 -2.20939636e-01 -1.79535579e-02
4.10893083e-01 1.61334455e-01 -1.31684995e+00 -2.28617430e-01
-5.42044401e-01 2.97224462e-01 5.85920252e-02 2.69028723e-01
-7.70432830e-01 1.01932323e+00 6.32130623e+00 3.83130997e-01
-1.14734817e+00 2.72102714e-01 4.30118263e-01 -4.98664796e-01
3.33748907e-02 -1.63698211e-01 -6.44488335e-01 3.64018768e-01
1.48929262e+00 -4.17967826e-01 5.24788499e-01 6.47483885e-01
-3.31480443e-01 -1.70635119e-01 -9.83231485e-01 8.75147104e-01
5.24403006e-02 -1.55150330e+00 3.01311702e-01 -3.33290815e-01
9.73897219e-01 2.25825563e-01 6.24168873e-01 6.47097588e-01
2.77496815e-01 -1.16064715e+00 4.31662798e-01 1.30399275e+00
6.00321412e-01 -6.35701835e-01 4.37681973e-01 6.36352003e-01
-1.04123974e+00 -5.62159300e-01 -3.52252871e-01 -1.52989075e-01
2.61495747e-02 3.16481054e-01 -5.98266959e-01 -2.09697425e-01
9.56831753e-01 9.44635570e-01 -7.57376730e-01 1.13571727e+00
2.71539446e-02 5.54924190e-01 -2.31053621e-01 -3.21888328e-02
8.57827514e-02 3.38020504e-01 6.66451752e-01 1.00155091e+00
4.86925989e-01 3.03745806e-01 -2.31226400e-01 5.52802742e-01
-2.17411369e-01 -3.42248172e-01 -1.02592874e+00 -2.70857699e-02
5.92192352e-01 6.89350605e-01 -8.39739144e-01 -3.54433388e-01
9.92641747e-02 1.08819103e+00 7.98737764e-01 5.96901000e-01
-2.47311503e-01 -1.19285934e-01 1.58600986e-01 1.32385060e-01
4.87746418e-01 -4.21614945e-01 -4.00103703e-02 -8.64519358e-01
-9.90051478e-02 -4.00268704e-01 6.25611186e-01 -6.75512791e-01
-1.16624618e+00 8.99518192e-01 -2.61737376e-01 -8.53208423e-01
-5.20131648e-01 -1.36667237e-01 -5.91105282e-01 2.33397961e-01
-1.33956170e+00 -9.21054721e-01 -3.84781927e-01 8.21491659e-01
1.03015554e+00 -4.46801901e-01 1.21971893e+00 -1.29754434e-03
-1.17353551e-01 4.27891403e-01 4.20494646e-01 -6.07765853e-01
5.69187164e-01 -9.26438987e-01 1.60667039e-02 5.73105097e-01
3.39667171e-01 8.10931265e-01 7.35572159e-01 -8.12773347e-01
-1.38599098e+00 -1.48774326e+00 6.27486527e-01 -2.76180595e-01
3.53390604e-01 -4.22464043e-01 -1.55069840e+00 6.46779954e-01
1.33648053e-01 1.57996625e-01 5.78455806e-01 4.81179282e-02
-4.40295994e-01 -2.34341368e-01 -7.66924083e-01 2.41216287e-01
1.33553493e+00 -4.74326879e-01 -1.01986837e+00 -9.32959374e-03
7.82172441e-01 1.92588940e-01 -4.63521034e-01 3.78415644e-01
6.74913585e-01 -8.38213205e-01 6.42331421e-01 -7.46169031e-01
-1.95187986e-01 -3.34241718e-01 2.02928483e-02 -1.65317965e+00
-4.93988603e-01 -3.43821943e-01 -6.92230344e-01 7.66524732e-01
3.82191926e-01 -6.49135351e-01 6.76380038e-01 -9.15343985e-02
-2.28958607e-01 -8.72374713e-01 -1.12464881e+00 -1.05062878e+00
3.82718705e-02 4.47367411e-03 3.42649043e-01 6.36070073e-01
1.64253283e-02 4.48965490e-01 -4.16243047e-01 -1.09498389e-01
5.01083374e-01 2.65794843e-01 1.39214396e-01 -1.34102273e+00
-1.29370674e-01 -2.98116863e-01 -3.26342523e-01 -7.69677401e-01
3.11281949e-01 -9.30166304e-01 2.46711060e-01 -1.17461491e+00
3.24761063e-01 -3.04614484e-01 -9.51479852e-01 5.29549003e-01
1.54638648e-01 7.38631040e-02 8.39878619e-03 5.87071002e-01
-1.11917770e+00 1.12794781e+00 7.29355872e-01 -3.19152325e-01
-4.81704712e-01 -1.00180231e-01 -3.75936925e-01 6.11937046e-01
7.75715709e-01 -5.09823740e-01 -7.97544599e-01 -2.49160901e-01
4.35122311e-01 -1.01957619e-01 3.48016202e-01 -1.41221666e+00
7.52103806e-01 -7.74731347e-03 5.23386478e-01 -4.80045170e-01
2.94799298e-01 -6.96883321e-01 2.33496547e-01 7.32797265e-01
-5.27203500e-01 5.91616258e-02 3.18989098e-01 1.05281854e+00
-1.88944876e-01 -7.22574741e-02 9.68628824e-01 -3.08967054e-01
-1.13745534e+00 4.47746664e-01 -9.41827476e-01 -1.38550997e-01
1.10111046e+00 -3.92914981e-01 -3.60945165e-01 -2.39028126e-01
-1.39616334e+00 7.11750910e-02 3.17387491e-01 9.22118545e-01
1.22899497e+00 -1.42760468e+00 -3.96744311e-01 4.90808517e-01
2.07146853e-01 -3.29467148e-01 4.54380184e-01 3.99279684e-01
1.42486140e-01 4.63266879e-01 -8.74008358e-01 -4.77395713e-01
-5.55620492e-01 8.28336596e-01 3.32011998e-01 -2.14353412e-01
-6.43713534e-01 8.46198916e-01 2.14580566e-01 -1.18516535e-01
3.35426152e-01 6.45168051e-02 -4.28539485e-01 2.22994894e-01
7.09632695e-01 2.93397337e-01 4.25887853e-02 -4.84940290e-01
-1.39374614e-01 3.40705365e-01 -6.03983164e-01 1.40405698e-02
1.75977719e+00 -2.55888164e-01 -1.34714499e-01 1.29292285e+00
8.11608434e-01 -9.12805378e-01 -1.40631664e+00 -6.56090200e-01
2.87395149e-01 2.44182013e-02 -2.83092022e-01 -8.95516813e-01
-7.53352165e-01 7.58496344e-01 1.09118259e+00 -1.14645749e-01
1.27440453e+00 2.22310111e-01 7.82978714e-01 8.83845627e-01
7.70703316e-01 -1.41296315e+00 8.57986450e-01 1.11817789e+00
1.21596742e+00 -7.31066942e-01 -3.74526054e-01 5.53850293e-01
-1.66258037e-01 9.82221365e-01 7.71932900e-01 -3.65249515e-01
8.75038505e-01 7.70440325e-02 -3.80359054e-01 -2.25915700e-01
-1.39032543e+00 -9.88280121e-03 -2.04271395e-02 7.69188225e-01
-1.06935248e-01 -2.32580140e-01 1.62720650e-01 6.44809365e-01
-6.45008609e-02 -1.01637924e-02 3.75190169e-01 1.25599658e+00
-1.00907242e+00 -6.70914888e-01 -4.08323444e-02 6.72704637e-01
1.39617443e-01 -1.01928944e-02 -1.93230897e-01 3.34186643e-01
2.11061835e-01 4.08388764e-01 1.72765687e-01 -1.59155235e-01
1.21609107e-01 4.87380266e-01 5.53976893e-01 -7.60006785e-01
-1.76771373e-01 -4.14745748e-01 -4.56935674e-01 -4.25974488e-01
-3.80649805e-01 -9.05341923e-01 -1.41745758e+00 2.27300674e-01
1.21724956e-01 1.69063866e-01 1.63835868e-01 7.76400685e-01
6.63399577e-01 6.71163619e-01 3.01771402e-01 -9.18116927e-01
-2.04329535e-01 -9.25955355e-01 -5.71868896e-01 3.36648732e-01
4.86707687e-01 -1.03438616e+00 -2.37168133e-01 3.29375267e-01] | [9.822473526000977, 3.390072822570801] |
9b2e0570-8c65-4ba7-8603-541907611231 | on-the-performance-of-differential-evolution | 1904.06960 | null | http://arxiv.org/abs/1904.06960v1 | http://arxiv.org/pdf/1904.06960v1.pdf | On the Performance of Differential Evolution for Hyperparameter Tuning | Automated hyperparameter tuning aspires to facilitate the application of
machine learning for non-experts. In the literature, different optimization
approaches are applied for that purpose. This paper investigates the
performance of Differential Evolution for tuning hyperparameters of supervised
learning algorithms for classification tasks. This empirical study involves a
range of different machine learning algorithms and datasets with various
characteristics to compare the performance of Differential Evolution with
Sequential Model-based Algorithm Configuration (SMAC), a reference Bayesian
Optimization approach. The results indicate that Differential Evolution
outperforms SMAC for most datasets when tuning a given machine learning
algorithm - particularly when breaking ties in a first-to-report fashion. Only
for the tightest of computational budgets SMAC performs better. On small
datasets, Differential Evolution outperforms SMAC by 19% (37% after
tie-breaking). In a second experiment across a range of representative datasets
taken from the literature, Differential Evolution scores 15% (23% after
tie-breaking) more wins than SMAC. | ['Anett Schülke', 'Mischa Schmidt', 'Shahd Safarani', 'Tobias Jacobs', 'Sebastien Nicolas', 'Julia Gastinger'] | 2019-04-15 | null | null | null | null | ['smac-1', 'smac'] | ['playing-games', 'playing-games'] | [ 1.02721982e-01 -7.57882893e-02 -3.58845234e-01 -2.40520492e-01
-8.43392074e-01 -5.06376624e-01 6.83502972e-01 4.54700410e-01
-8.69231343e-01 8.59731972e-01 -1.54644504e-01 -2.74698317e-01
-8.46614897e-01 -3.99791002e-01 -3.26582789e-01 -1.08171093e+00
-4.78737522e-03 9.86690044e-01 1.99133202e-01 1.18916221e-02
6.35994434e-01 3.39410990e-01 -1.51367140e+00 -1.68375876e-02
5.22870600e-01 6.84032083e-01 -1.45685047e-01 6.44888043e-01
1.62614524e-01 -1.67627037e-02 -7.87781596e-01 -6.30610943e-01
3.94113541e-01 -2.64922559e-01 -6.28836811e-01 -2.92873532e-01
8.94784462e-03 5.40803254e-01 1.96811423e-01 6.19860232e-01
9.07891929e-01 1.51726976e-01 1.00615001e+00 -1.20188046e+00
1.79214954e-01 1.01542270e+00 -6.52679741e-01 2.98370481e-01
7.51861855e-02 5.15765727e-01 1.08174860e+00 -2.86497831e-01
3.79332632e-01 1.11062336e+00 7.94270217e-01 2.49842301e-01
-1.49407995e+00 -6.55501008e-01 -2.15256244e-01 4.76954244e-02
-1.45265114e+00 -1.93234563e-01 4.62365597e-01 -4.97467667e-01
9.97089565e-01 3.40550512e-01 4.65627193e-01 1.02121258e+00
2.47450143e-01 3.93653840e-01 1.05800843e+00 -6.22706831e-01
7.40188777e-01 4.23990726e-01 4.29651946e-01 3.04717034e-01
7.53551960e-01 2.88958371e-01 -4.84079361e-01 -5.23904443e-01
2.68639140e-02 -7.62615204e-01 -1.30560249e-01 -3.61838341e-01
-1.01837802e+00 9.98226047e-01 -3.84306945e-02 2.50631392e-01
-3.78253341e-01 9.07915533e-02 4.41255063e-01 3.16222161e-01
3.13966930e-01 1.25268745e+00 -7.57432997e-01 -6.16919816e-01
-9.14506972e-01 3.85555714e-01 1.18121612e+00 5.36916077e-01
2.00661287e-01 -5.93367107e-02 -2.87930757e-01 9.11070824e-01
9.31006968e-02 1.57169208e-01 7.43393183e-01 -8.39444339e-01
3.29988867e-01 4.58087385e-01 1.76004574e-01 -9.05410707e-01
-8.84942412e-01 -9.81270373e-01 -6.55107856e-01 1.81019887e-01
5.38539886e-01 -3.57097358e-01 -4.72060531e-01 1.47563004e+00
3.81980002e-01 -1.82268560e-01 -4.49872240e-02 5.71015000e-01
2.64312476e-01 5.14746010e-01 2.73769855e-01 -5.51578760e-01
1.07918799e+00 -5.50714552e-01 -2.24634692e-01 5.21640144e-02
7.29099631e-01 -7.02701986e-01 1.05777180e+00 8.35583806e-01
-8.38086009e-01 -2.99786240e-01 -1.21903682e+00 7.51738250e-01
-2.75494576e-01 -1.64552957e-01 3.79692733e-01 1.20706499e+00
-6.74914181e-01 6.97675049e-01 -7.01237679e-01 -2.16788858e-01
2.14237258e-01 3.87211114e-01 1.45648584e-01 2.51865476e-01
-9.98588920e-01 1.13994563e+00 8.39257598e-01 -4.88464266e-01
-3.85488510e-01 -1.08310568e+00 -1.70840219e-01 6.42288998e-02
5.11358023e-01 -7.03322053e-01 1.20854807e+00 -6.32997990e-01
-1.67544854e+00 6.82303727e-01 4.23078954e-01 -8.61259401e-01
9.30639505e-01 -3.04915339e-01 -1.14346571e-01 -3.92362714e-01
-4.61643815e-01 3.95866156e-01 7.57680118e-01 -1.18878067e+00
-6.64411366e-01 -1.78315416e-01 -2.32253477e-01 8.79310593e-02
-2.67921597e-01 -3.86892669e-02 -2.71806091e-01 -6.80993915e-01
-2.04116896e-01 -1.20099962e+00 -1.98225752e-01 -6.97666407e-01
-2.96189457e-01 -3.67541909e-01 5.45444310e-01 -5.45010231e-02
1.67733598e+00 -1.75678778e+00 3.89365166e-01 4.35020536e-01
-9.79779884e-02 2.94622928e-01 1.40293807e-01 3.65039796e-01
-1.17618069e-01 2.53420860e-01 -4.12190199e-01 -2.50710160e-01
1.07464030e-01 9.09221768e-02 1.15971733e-02 4.22825783e-01
-1.32133335e-01 4.63761151e-01 -5.20380795e-01 -5.10395288e-01
6.54791147e-02 2.06645548e-01 -5.98430276e-01 -2.72999585e-01
-3.43012184e-01 5.29896170e-02 -3.62353355e-01 4.12045836e-01
1.90384015e-01 -2.84589589e-01 1.88137636e-01 3.32933292e-02
2.22184043e-03 -2.15848446e-01 -1.37239110e+00 1.24064267e+00
-5.38214922e-01 7.20493436e-01 -4.29394662e-01 -9.68656957e-01
9.96993601e-01 1.07375734e-01 4.19919670e-01 -3.53754222e-01
4.19337809e-01 1.77643329e-01 5.40247023e-01 -2.56762505e-01
3.46494108e-01 2.40088552e-01 -1.69688419e-01 8.81701589e-01
-3.80287282e-02 -4.03836757e-01 3.75290334e-01 -3.50782946e-02
1.08915150e+00 -1.63240612e-01 4.39937502e-01 -5.80125511e-01
3.92735451e-01 1.29806340e-01 4.57427025e-01 1.03353274e+00
-3.79658975e-02 4.60368156e-01 6.26406968e-01 -2.97870010e-01
-9.62056935e-01 -5.48122406e-01 -4.97791588e-01 1.30099916e+00
-3.42753798e-01 -2.86155045e-01 -9.22004402e-01 -4.97547805e-01
2.51153827e-01 1.28785014e+00 -6.59381986e-01 -5.32371819e-01
-2.21882820e-01 -1.62215447e+00 4.94920552e-01 1.57484170e-02
3.65666091e-01 -9.01865065e-01 -1.14590812e+00 1.57728404e-01
3.64789754e-01 -7.31362998e-01 -1.78236980e-02 3.71110022e-01
-8.98641288e-01 -1.09296107e+00 -5.62370837e-01 1.11189354e-02
8.94772559e-02 -4.30461109e-01 1.21222305e+00 6.88106567e-02
-5.40757477e-01 3.83635104e-01 -3.63991439e-01 -6.71564639e-01
-5.92115819e-01 6.98164165e-01 1.42370194e-01 -8.32011402e-02
3.70119810e-01 -3.67866784e-01 -2.27447316e-01 6.03678584e-01
-5.91571331e-01 -3.16073686e-01 6.63032115e-01 8.51513386e-01
2.96511531e-01 3.89616340e-01 4.45810944e-01 -8.77330303e-01
8.92118812e-01 -5.56103230e-01 -1.01143527e+00 4.16500390e-01
-1.34765339e+00 4.13250118e-01 1.97059438e-01 -7.91104853e-01
-9.55855250e-01 -1.41668633e-01 5.42360425e-01 -2.07493633e-01
1.39892474e-01 5.48140168e-01 1.24488391e-01 -7.85405934e-02
1.15738678e+00 -1.73554093e-01 -3.40140723e-02 -5.09986222e-01
-7.95703083e-02 7.01040089e-01 3.02949518e-01 -8.17286253e-01
5.66216946e-01 -1.36414066e-01 1.98090777e-01 -5.48889101e-01
-8.67659330e-01 -2.10527316e-01 -4.87086356e-01 -2.64651835e-01
4.69837368e-01 -3.83320302e-01 -7.24347055e-01 5.05176425e-01
-6.37740314e-01 -4.95371252e-01 -1.65460914e-01 7.34340787e-01
-5.18415809e-01 -2.18794271e-01 8.39243382e-02 -7.76745737e-01
-3.77367526e-01 -1.15417647e+00 5.52506804e-01 4.16156292e-01
-7.52705216e-01 -1.00919604e+00 3.52243602e-01 3.73784721e-01
4.10176307e-01 3.07428718e-01 1.04893053e+00 -1.11959183e+00
4.59942557e-02 -3.89520019e-01 2.84095556e-01 5.41152470e-02
-1.16707243e-01 5.07009268e-01 -7.42851853e-01 -3.85656029e-01
-2.79160798e-01 -1.28056958e-01 7.43548334e-01 6.14988446e-01
1.16776013e+00 4.10979204e-02 -4.82310295e-01 6.24911547e-01
1.33096159e+00 4.11500812e-01 3.71383429e-01 1.00938046e+00
3.53358164e-02 6.46447420e-01 7.12189734e-01 7.72296548e-01
-1.88553646e-01 7.29268014e-01 1.89573869e-01 4.39032376e-01
2.34630167e-01 2.52146959e-01 -4.73774271e-03 2.69104630e-01
-3.96531560e-02 -3.86567116e-01 -1.39072359e+00 1.72599003e-01
-1.60228705e+00 -6.09200597e-01 1.60330988e-03 2.44804025e+00
1.02681136e+00 7.21377075e-01 3.63859296e-01 3.21276188e-01
8.35085511e-01 -2.32179686e-01 -8.58812869e-01 -5.54423809e-01
-1.05891787e-01 2.22991705e-01 5.98409414e-01 3.06626588e-01
-9.68446374e-01 5.54038703e-01 6.64906454e+00 1.04834151e+00
-8.57224226e-01 -2.45009705e-01 7.16009974e-01 -5.49649894e-01
1.17860667e-01 -2.30365656e-02 -9.40010369e-01 6.99521601e-01
1.28499365e+00 -4.16205853e-01 2.68648773e-01 7.03262746e-01
5.56821488e-02 -5.29748201e-01 -1.03990531e+00 9.99021471e-01
-3.88605222e-02 -1.20155501e+00 -2.05354527e-01 -8.28090310e-02
1.04394698e+00 -1.53464690e-01 1.95144013e-01 3.35996300e-01
6.35637164e-01 -1.07807648e+00 5.13806224e-01 5.47271967e-01
3.79091024e-01 -1.03260612e+00 8.85625780e-01 2.86588311e-01
-3.45427930e-01 -3.20454448e-01 -1.88601494e-01 1.65969342e-01
-1.06656127e-01 8.12547326e-01 -9.57210302e-01 5.09008825e-01
8.84126186e-01 1.47885039e-01 -9.53112662e-01 1.41918457e+00
1.90249756e-01 9.36751008e-01 -5.22426367e-01 -4.80466485e-01
1.57401100e-01 -6.59496486e-02 7.76428521e-01 1.28709817e+00
1.17393993e-01 -6.63573146e-02 -3.21306497e-01 4.46593076e-01
2.12378636e-01 4.40830737e-02 1.96699891e-02 -2.81758048e-02
1.01801550e+00 9.52111959e-01 -7.22121358e-01 -1.13413766e-01
2.61557430e-01 3.36042851e-01 -2.84484867e-02 1.83112785e-01
-9.02841747e-01 -4.70063418e-01 3.62534910e-01 -9.74801257e-02
3.05302944e-02 5.04836254e-02 -7.15594351e-01 -3.90995502e-01
-4.42445815e-01 -1.01403654e+00 9.06874955e-01 -5.59255362e-01
-1.12123775e+00 4.44023997e-01 6.78955078e-01 -9.50533032e-01
-4.66265857e-01 -4.64487940e-01 -5.90648532e-01 6.14393353e-01
-8.30146909e-01 -3.01498443e-01 -3.03018868e-01 4.32754382e-02
4.48448449e-01 -7.35169649e-01 5.76894641e-01 -2.81998873e-01
-9.09537673e-01 8.17580640e-01 5.91538310e-01 -4.04854238e-01
8.98579836e-01 -1.07409549e+00 -3.90605330e-02 3.72700542e-01
-6.92840442e-02 4.23105001e-01 1.15564454e+00 -4.37529653e-01
-9.59208429e-01 -7.55710721e-01 3.44484538e-01 -2.73107022e-01
5.13120294e-01 2.07524613e-01 -9.23209727e-01 1.30258322e-01
1.57196999e-01 -6.12257600e-01 7.59835482e-01 3.61799300e-01
-2.50408769e-01 -8.77139941e-02 -1.19331920e+00 4.89859939e-01
6.50090635e-01 9.51964408e-04 -5.26968837e-01 1.55768767e-01
3.54665905e-01 -2.57358462e-01 -1.06083894e+00 5.23231447e-01
6.87417984e-01 -9.72402394e-01 8.26785982e-01 -6.11316383e-01
2.14028209e-01 -2.09014434e-02 -1.30820647e-01 -1.50170302e+00
-1.01738505e-01 -9.00474727e-01 -5.11786807e-03 1.25964236e+00
6.61099494e-01 -7.88145304e-01 7.69557595e-01 6.48516715e-01
3.44738364e-01 -8.90558004e-01 -8.35093796e-01 -9.22624648e-01
2.83793986e-01 -4.74815488e-01 5.29955447e-01 8.31949949e-01
-4.27462995e-01 3.16536158e-01 1.09195888e-01 -1.70319855e-01
7.45363235e-01 -1.11175254e-01 7.59316027e-01 -1.54462671e+00
-5.36771238e-01 -1.13360596e+00 -3.11038494e-01 4.62544076e-02
4.47252505e-02 -8.04949820e-01 -2.80813634e-01 -7.54059494e-01
3.04835737e-01 -6.05893970e-01 -2.32291669e-01 3.25927585e-01
-2.70758897e-01 -1.59824133e-01 1.01209223e-01 1.29481196e-01
-2.19348148e-01 1.72299579e-01 5.56842864e-01 -5.35693206e-02
-5.00366867e-01 3.55056554e-01 -5.07028461e-01 5.50004900e-01
1.05054092e+00 -8.30706775e-01 -3.76327366e-01 -4.99782823e-02
2.95651942e-01 -1.90610901e-01 3.42777446e-02 -1.08216596e+00
2.40803719e-01 -1.57740593e-01 1.99960187e-01 -2.07166031e-01
-2.94440147e-02 -4.53441292e-01 5.65618575e-01 8.54416847e-01
-7.51582325e-01 2.02966541e-01 4.20424640e-01 5.08246779e-01
2.44363874e-01 -7.19994783e-01 1.05372548e+00 3.52375299e-01
-3.15600723e-01 -2.78582573e-01 -3.25449497e-01 4.50307965e-01
1.19939196e+00 -1.83077633e-01 -7.71944672e-02 -3.22350442e-01
-5.28550923e-01 2.74961799e-01 2.72613496e-01 3.25800478e-01
-2.73668598e-02 -6.95930183e-01 -8.97982240e-01 -7.03666955e-02
3.30698639e-02 -2.30579078e-01 -1.80149719e-01 7.77833045e-01
-4.08218414e-01 4.89204496e-01 2.01860778e-02 -7.84499705e-01
-1.43004739e+00 1.16034657e-01 5.23851812e-01 -5.08638442e-01
-6.19468875e-02 1.02381074e+00 -5.73121428e-01 -2.31549561e-01
6.09919965e-01 5.16066179e-02 5.17352000e-02 4.62134212e-01
1.30552024e-01 7.81107247e-01 4.01716232e-01 -3.20341475e-02
-2.51886725e-01 6.27182543e-01 -7.16804191e-02 -3.98630440e-01
1.24942684e+00 2.61994183e-01 2.85896659e-01 5.59584081e-01
9.62251782e-01 -4.83679265e-01 -1.12894416e+00 -3.51588607e-01
4.71014708e-01 -3.19381237e-01 4.19764191e-01 -1.19840443e+00
-7.82705784e-01 3.90631169e-01 6.32816315e-01 8.65248963e-02
9.65838373e-01 -3.16005677e-01 -1.36727676e-01 6.44552231e-01
4.82012510e-01 -1.29365706e+00 -4.86095324e-02 2.39512011e-01
9.38919306e-01 -1.03868055e+00 5.38200200e-01 2.91737407e-01
-8.30621421e-01 1.14490736e+00 4.25255537e-01 2.27845341e-01
7.47498631e-01 2.31167525e-01 -2.37588435e-01 -5.31521253e-02
-1.10610378e+00 1.45780697e-01 3.54792833e-01 3.88935298e-01
5.23367375e-02 -4.43439297e-02 -5.32709420e-01 3.96428734e-01
-6.09950662e-01 -3.29240859e-01 2.77771711e-01 8.04336309e-01
-3.82613748e-01 -1.03889775e+00 -4.79707271e-01 6.43958509e-01
-2.71964043e-01 8.28419998e-02 -5.44966877e-01 1.31781101e+00
-1.57835737e-01 8.46328497e-01 2.70540208e-01 -3.37943316e-01
1.72956735e-01 1.89420357e-01 5.13491213e-01 -2.46864244e-01
-9.01010156e-01 -5.96029237e-02 1.72835872e-01 -2.47415900e-01
-2.72515386e-01 -1.03911245e+00 -9.69385087e-01 -3.38363558e-01
-5.56674421e-01 3.99517119e-01 8.34873617e-01 7.06159592e-01
3.28786999e-01 5.15890062e-01 5.25901854e-01 -5.52028179e-01
-1.09546685e+00 -1.03559363e+00 -3.91181856e-01 6.03529029e-02
-8.17402303e-02 -1.09684062e+00 -7.79949784e-01 -3.47597480e-01] | [6.694573402404785, 3.9825186729431152] |
cc2ce379-f68f-40a6-b7b2-e37531ede16d | segmentation-of-multiple-myeloma-plasma-cells | 2111.05125 | null | https://arxiv.org/abs/2111.05125v1 | https://arxiv.org/pdf/2111.05125v1.pdf | Segmentation of Multiple Myeloma Plasma Cells in Microscopy Images with Noisy Labels | A key component towards an improved and fast cancer diagnosis is the development of computer-assisted tools. In this article, we present the solution that won the SegPC-2021 competition for the segmentation of multiple myeloma plasma cells in microscopy images. The labels used in the competition dataset were generated semi-automatically and presented noise. To deal with it, a heavy image augmentation procedure was carried out and predictions from several models were combined using a custom ensemble strategy. State-of-the-art feature extractors and instance segmentation architectures were used, resulting in a mean Intersection-over-Union of 0.9389 on the SegPC-2021 final test set. | ['Danijel Skočaj', 'Tomaž Martinčič', 'Dejan Štepec', 'Álvaro García Faura'] | 2021-11-08 | null | null | null | null | ['image-augmentation'] | ['computer-vision'] | [ 3.75123620e-01 4.02925521e-01 4.03396845e-01 -4.59914982e-01
-1.01217699e+00 -3.02108735e-01 6.70532703e-01 5.89105308e-01
-8.47363174e-01 8.70496631e-01 -5.69037735e-01 -8.88340510e-05
-1.20474249e-01 -4.52498823e-01 -3.86243343e-01 -9.22974646e-01
2.80098081e-01 1.34045756e+00 4.28677469e-01 -3.50760780e-02
2.71112084e-01 7.99279690e-01 -1.06027544e+00 7.53648818e-01
8.54884923e-01 8.26312840e-01 2.71162629e-01 1.07467461e+00
-3.74418169e-01 5.36866724e-01 -6.38321102e-01 -4.23965394e-01
-8.33172500e-02 -3.17022830e-01 -9.26719069e-01 3.03827137e-01
1.63476691e-01 3.85818303e-01 8.98021460e-02 7.72585273e-01
4.35569316e-01 -3.21524560e-01 1.04366446e+00 -1.02107167e+00
1.02578208e-01 5.08952379e-01 -4.50759083e-01 4.15918320e-01
1.25569418e-01 2.49136910e-01 4.25513029e-01 -7.58526921e-01
1.08272290e+00 8.01019311e-01 6.26314878e-01 5.25371313e-01
-1.60461378e+00 -2.80135900e-01 -3.96001846e-01 1.65231705e-01
-1.39558280e+00 -2.03230828e-01 1.72050074e-01 -8.20486069e-01
1.14145803e+00 3.90493155e-01 7.84542382e-01 7.76757061e-01
4.09129947e-01 7.83119142e-01 1.42188895e+00 -7.29575634e-01
2.96336979e-01 4.10483211e-01 3.82089615e-01 7.79166937e-01
1.11172915e-01 -9.93812084e-02 -1.92128852e-01 -5.76869287e-02
5.99456191e-01 -2.76717067e-01 -8.53548050e-02 -2.66460299e-01
-1.06047392e+00 8.26328695e-01 3.35586369e-01 7.03482568e-01
-2.99600869e-01 -2.23604247e-01 2.75028825e-01 2.29859315e-02
2.52437979e-01 5.90919137e-01 -3.86347711e-01 2.93451667e-01
-1.14655232e+00 1.93185121e-01 6.57621980e-01 5.80124795e-01
6.02993965e-01 -5.84929168e-01 -2.75591373e-01 6.54457808e-01
2.12910682e-01 9.08296257e-02 4.62528944e-01 -7.30120361e-01
-1.34655550e-01 1.21308374e+00 -1.99902147e-01 -3.71000767e-01
-1.20142531e+00 -6.42737567e-01 -7.28296220e-01 3.04220468e-01
8.55652094e-01 -2.14371998e-02 -1.50241733e+00 1.00762665e+00
2.56210029e-01 -4.05771695e-02 -8.71349871e-02 7.94244766e-01
7.11327910e-01 9.89943668e-02 9.38746408e-02 -1.22704364e-01
1.39673638e+00 -8.95657837e-01 -5.20243764e-01 2.57807761e-01
9.73496079e-01 -8.39873731e-01 5.01837671e-01 6.64152563e-01
-8.67940903e-01 -2.13591531e-01 -9.09250498e-01 3.30753267e-01
-6.24363482e-01 1.62747785e-01 5.82614243e-01 6.80492699e-01
-1.13134897e+00 5.16316950e-01 -9.72885251e-01 -5.60187042e-01
7.11657524e-01 9.18821216e-01 -6.37795508e-01 5.09643219e-02
-5.92557013e-01 1.04194891e+00 6.17813110e-01 -1.49357721e-01
-5.18838584e-01 -7.74272263e-01 -3.38479310e-01 -4.28765655e-01
1.18633188e-01 -8.12063217e-01 9.29132819e-01 -8.76176536e-01
-1.38588274e+00 1.61300063e+00 1.54017240e-01 -7.69809902e-01
9.44873631e-01 1.92548558e-01 -2.52040505e-01 3.26599419e-01
-7.99398571e-02 8.94338131e-01 4.11236584e-01 -1.37105429e+00
-9.65330422e-01 -4.84711945e-01 -5.39926350e-01 -1.68805137e-01
2.90225536e-01 -4.05165069e-02 -6.92038715e-01 -3.06091964e-01
3.13779861e-02 -1.04472053e+00 -6.07859612e-01 -4.35102046e-01
-5.48614502e-01 -2.26106495e-02 5.61873615e-01 -5.48164845e-01
6.97481513e-01 -1.70194471e+00 4.61938530e-01 6.53228760e-01
3.42352182e-01 2.24915594e-01 2.25357205e-01 7.83275962e-02
5.61142415e-02 1.31740179e-02 -3.90019774e-01 -6.45491481e-01
-1.33128300e-01 2.17524037e-01 2.80558735e-01 4.60908294e-01
2.13539436e-01 7.51554906e-01 -5.37223339e-01 -7.58173347e-01
5.71093976e-01 5.33109009e-01 -3.47145230e-01 2.74179876e-01
-3.10032934e-01 8.29669416e-01 -1.62078947e-01 8.87543619e-01
6.13204956e-01 -4.76048917e-01 5.61178923e-02 5.62859215e-02
2.92585120e-02 -5.82350552e-01 -7.71117449e-01 1.92472088e+00
2.35893503e-02 2.53550291e-01 1.29668236e-01 -7.76384592e-01
8.63630235e-01 2.37560809e-01 9.90317941e-01 -4.86317426e-01
5.96493840e-01 4.44574952e-01 8.69522542e-02 -2.64536917e-01
7.52915069e-02 -2.72847354e-01 2.11054698e-01 8.75571929e-03
6.25318289e-01 -3.57777625e-01 6.41179085e-01 8.75330791e-02
1.12123764e+00 -1.00626118e-01 1.44669950e-01 -5.80197036e-01
1.09108067e+00 4.73422438e-01 3.99043471e-01 6.62666202e-01
-3.85050476e-01 9.98263001e-01 4.94527608e-01 -5.61032116e-01
-8.39668274e-01 -7.96605706e-01 -5.72131455e-01 5.03016770e-01
-3.65169764e-01 -1.55816659e-01 -1.21492875e+00 -1.08100891e+00
-5.22398390e-02 6.00355029e-01 -8.19347978e-01 2.80505449e-01
-4.11279470e-01 -1.25015867e+00 5.48383474e-01 5.98135479e-02
-4.63966951e-02 -1.16456854e+00 -5.32851100e-01 4.27578479e-01
1.20049402e-01 -1.25040758e+00 2.87838995e-01 6.76052928e-01
-7.79765666e-01 -1.50973058e+00 -9.49018776e-01 -7.22348332e-01
8.90159249e-01 -5.75026512e-01 1.17460752e+00 2.68165231e-01
-8.65327299e-01 6.73034042e-02 -1.52360037e-01 -6.69842422e-01
-5.19599557e-01 2.02052191e-01 -4.29870486e-01 -1.56258538e-01
6.68443978e-01 -8.97317752e-02 -3.66150677e-01 1.84078440e-01
-8.90800416e-01 2.72878438e-01 7.32001960e-01 1.02507925e+00
1.09688628e+00 -1.24049045e-01 3.60856920e-01 -1.33256650e+00
1.45240247e-01 -7.43045509e-02 -6.57338440e-01 2.28947505e-01
-4.78976250e-01 -1.11929461e-01 4.40386176e-01 -7.57406577e-02
-6.58094466e-01 7.09827006e-01 -4.98537153e-01 -2.45596588e-01
-6.76124096e-01 2.01306030e-01 6.97885081e-02 -5.34931719e-01
6.25817597e-01 8.52230191e-02 2.10463136e-01 -3.48110020e-01
8.80354792e-02 4.42234576e-01 5.35196066e-01 -3.01396906e-01
2.60704726e-01 5.33578038e-01 3.97967696e-01 -7.30280101e-01
-6.92572892e-01 -6.07332051e-01 -1.04507983e+00 -3.06875318e-01
9.52376246e-01 -4.63195443e-01 -4.23611313e-01 9.05449688e-01
-8.84061813e-01 -4.53377843e-01 -3.45638901e-01 2.83331692e-01
-7.57103682e-01 5.93665019e-02 -6.51567161e-01 -5.18669963e-01
-5.60436547e-01 -1.47354448e+00 9.87533331e-01 5.81044614e-01
-2.99812019e-01 -9.57041025e-01 3.40229392e-01 6.52874768e-01
4.42466170e-01 4.30353016e-01 7.93355465e-01 -1.30117774e+00
-3.04821312e-01 -4.02348965e-01 -2.37498492e-01 1.97589044e-02
-8.82927626e-02 2.19200641e-01 -1.07680345e+00 -2.32317030e-01
-1.94053367e-01 -1.94600597e-01 1.15430343e+00 4.40983891e-01
1.06744874e+00 5.35324097e-01 -6.88801885e-01 6.77427649e-01
1.55802464e+00 2.28569329e-01 6.14287257e-01 5.77178955e-01
3.98498803e-01 7.10849047e-01 3.92531842e-01 2.31872544e-01
-6.69520050e-02 6.18986905e-01 3.46244574e-01 -2.44826302e-01
-1.37549162e-01 4.31047171e-01 -4.80196953e-01 1.91389829e-01
9.60179605e-03 -2.01765522e-01 -1.33102548e+00 4.20784414e-01
-1.59762204e+00 -1.44045100e-01 -5.82387924e-01 1.71661544e+00
5.68472445e-01 5.49103916e-01 3.93316858e-02 1.77737504e-01
5.11823177e-01 -7.10209072e-01 -1.70145944e-01 -1.42437637e-01
-4.49556828e-01 4.30809349e-01 4.46065545e-01 5.12034893e-01
-1.13624310e+00 8.64415050e-01 7.08634853e+00 9.41831589e-01
-1.03710830e+00 -2.44574510e-02 8.87760997e-01 -2.36932766e-02
3.19576561e-01 -1.83020979e-01 -9.43065882e-01 4.15551513e-01
1.14733088e+00 3.52110952e-01 -3.51256989e-02 3.92472804e-01
-2.28939995e-01 -5.11525869e-01 -9.56859767e-01 8.44359338e-01
4.14664336e-02 -1.52050400e+00 -6.59312168e-03 2.40023881e-01
7.18739092e-01 8.53929669e-02 -1.67744964e-01 1.46018729e-01
1.52454227e-01 -1.20153034e+00 3.04435730e-01 9.76674318e-01
6.12642825e-01 -9.20160711e-01 1.45765054e+00 2.15106934e-01
-6.14043295e-01 2.33497411e-01 -1.93058550e-01 4.67130810e-01
2.22015753e-01 8.91352654e-01 -1.18764579e+00 7.05520034e-01
3.27612281e-01 1.88631684e-01 -9.45500016e-01 1.28406262e+00
2.10690290e-01 3.07657123e-01 -4.91381168e-01 1.83800846e-01
1.47962570e-01 -1.35454938e-01 3.94848377e-01 1.68746340e+00
7.37756863e-02 -1.29353702e-01 6.60477430e-02 6.16178513e-01
1.04077153e-01 2.64081120e-01 2.59974189e-02 -8.21364950e-03
-1.15108803e-01 1.72384071e+00 -1.40459466e+00 -3.19839776e-01
-1.85282797e-01 7.91971624e-01 3.89500201e-01 -1.25944108e-01
-7.55111456e-01 -3.68722826e-02 1.08497590e-01 1.85262024e-01
1.79655418e-01 3.02265733e-01 -7.11801767e-01 -7.74127781e-01
-6.33726299e-01 -6.45280361e-01 6.65792108e-01 -4.39916253e-01
-1.35545301e+00 9.02751267e-01 -4.61034536e-01 -4.84568179e-01
-1.48088917e-01 -1.03585494e+00 -5.66388845e-01 7.78752685e-01
-1.22747672e+00 -1.35964394e+00 -3.08971167e-01 3.86019439e-01
2.31018454e-01 -5.87331295e-01 1.27604604e+00 4.61968295e-02
-6.29171789e-01 4.07881647e-01 8.00991431e-03 -9.31792632e-02
5.28504252e-01 -1.68440425e+00 -3.00386138e-02 4.00893658e-01
-8.46035313e-03 1.23607323e-01 7.63632119e-01 -4.47502166e-01
-8.89135003e-01 -9.24510658e-01 7.92233467e-01 -5.82314909e-01
7.12749302e-01 -5.58199100e-02 -8.84433270e-01 4.51348364e-01
3.27353209e-01 8.16019252e-02 9.89258945e-01 -3.06355149e-01
3.33384037e-01 2.97037959e-01 -1.63029027e+00 2.18100995e-01
3.38585109e-01 -1.41637206e-01 -4.41677272e-01 4.37377334e-01
1.35050535e-01 -6.79158390e-01 -1.08314753e+00 6.16782069e-01
1.29734516e-01 -1.13098896e+00 7.50373721e-01 -4.70128119e-01
7.42776394e-02 -2.23861247e-01 7.31149390e-02 -1.03124952e+00
-1.80930451e-01 -4.38668787e-01 -1.39265284e-01 9.81806755e-01
4.96066093e-01 -2.43834883e-01 1.37536037e+00 5.39309502e-01
-2.49130726e-01 -1.16833556e+00 -1.14505816e+00 -2.97907144e-01
2.68838733e-01 -1.29037291e-01 2.15650275e-01 5.38840234e-01
-5.31815784e-03 1.14089146e-01 3.96879464e-01 -2.08212689e-01
6.87203050e-01 -2.74119645e-01 6.84387386e-01 -1.46403635e+00
-5.04205115e-02 -7.89759040e-01 -7.32856691e-01 -1.94049016e-01
6.87029883e-02 -1.00534785e+00 -2.64674455e-01 -1.59500527e+00
3.91981572e-01 -3.03289741e-01 -6.59420311e-01 3.57893944e-01
-6.23021498e-02 4.84137684e-01 -9.79442615e-03 1.97309092e-01
-5.98729730e-01 -7.20687658e-02 1.01139331e+00 -9.19335186e-02
-8.00845847e-02 1.19197287e-01 -4.23920900e-01 8.30540061e-01
9.01026428e-01 -5.48838019e-01 2.52804816e-01 2.58117974e-01
-2.72296339e-01 7.80701265e-03 2.69928664e-01 -1.30427277e+00
2.79395968e-01 7.47103766e-02 7.22976029e-01 -8.93900633e-01
3.95447671e-01 -8.78632247e-01 4.71298665e-01 7.11701572e-01
-8.79932344e-02 -3.75083596e-01 2.62789726e-01 2.86366880e-01
-1.48630053e-01 -4.64221209e-01 1.24547911e+00 -2.68505186e-01
-4.32763904e-01 4.03264314e-02 -5.16224623e-01 -2.62048185e-01
1.53867137e+00 -2.07801238e-01 -3.51309299e-01 4.11227167e-01
-1.25522435e+00 3.07404071e-01 8.58299136e-01 -8.84115472e-02
2.42413536e-01 -7.20189214e-01 -8.52893770e-01 3.36794436e-01
1.28224358e-01 1.43054962e-01 4.00557965e-01 1.27369940e+00
-9.59854364e-01 7.61379421e-01 -5.26021481e-01 -1.00017703e+00
-1.41464758e+00 2.61942983e-01 8.08398545e-01 -1.05369890e+00
-5.86038351e-01 1.12172282e+00 -3.50694284e-02 -5.35243571e-01
-3.93342860e-02 -1.44218970e-02 -5.09702861e-01 8.54678228e-02
4.70529079e-01 2.50375658e-01 5.87117136e-01 -7.70271897e-01
-4.41343009e-01 2.87763685e-01 -4.58720148e-01 -3.52208652e-02
1.40063143e+00 1.94488153e-01 -2.66751915e-01 1.98691472e-01
7.70246506e-01 -3.31107765e-01 -1.10019672e+00 -4.73170355e-02
5.05309463e-01 -9.74224731e-02 5.06472364e-02 -1.26884794e+00
-1.13708270e+00 5.47292709e-01 9.18342948e-01 1.53870419e-01
1.08094656e+00 -4.95800488e-02 3.80920798e-01 1.06350981e-01
4.49111253e-01 -1.11549556e+00 -3.71662438e-01 4.75670546e-01
6.04055643e-01 -1.34327292e+00 -1.37094921e-02 -5.69941163e-01
-6.61465526e-01 1.32462227e+00 4.89607900e-01 -1.51404053e-01
4.94707078e-01 5.47662914e-01 2.80087173e-01 -4.05294091e-01
-6.88860059e-01 -3.06503892e-01 3.72736841e-01 7.98079848e-01
5.90721607e-01 1.30733550e-01 -4.31051612e-01 9.54203904e-01
-1.21052288e-01 2.84269661e-01 4.04223442e-01 9.91906941e-01
-5.57787895e-01 -1.26904428e+00 -4.37913120e-01 7.11610794e-01
-7.45996714e-01 2.98751503e-01 -7.02813089e-01 9.06367242e-01
3.27674180e-01 8.15586925e-01 -9.25802812e-02 -2.17366263e-01
3.57641548e-01 3.06033224e-01 6.50326490e-01 -4.29443151e-01
-9.68383551e-01 2.81291425e-01 -7.74006918e-02 -3.77322644e-01
-3.25385183e-01 -6.52025223e-01 -1.75388277e+00 -7.08746240e-02
-3.49759579e-01 1.39794558e-01 7.97536254e-01 1.08551300e+00
5.76624162e-02 9.89835918e-01 1.38283461e-01 -9.22057867e-01
1.71440037e-03 -1.09999835e+00 -7.25825429e-01 4.29924875e-01
-1.94363862e-01 -5.67073524e-01 -3.43491018e-01 1.52502105e-01] | [15.033942222595215, -3.0063188076019287] |
1a6d9091-05c0-442e-bc04-ace6955be07f | disentangled-representation-for-age-invariant | null | null | http://openaccess.thecvf.com//content/ICCV2021/html/Hou_Disentangled_Representation_for_Age-Invariant_Face_Recognition_A_Mutual_Information_Minimization_ICCV_2021_paper.html | http://openaccess.thecvf.com//content/ICCV2021/papers/Hou_Disentangled_Representation_for_Age-Invariant_Face_Recognition_A_Mutual_Information_Minimization_ICCV_2021_paper.pdf | Disentangled Representation for Age-Invariant Face Recognition: A Mutual Information Minimization Perspective | General face recognition has seen remarkable progress in recent years. However, large age gap still remains a big challenge due to significant alterations in facial appearance and bone structure. Disentanglement plays a key role in partitioning face representations into identity-dependent and age-dependent components for age-invariant face recognition (AIFR). In this paper we propose a multi-task learning framework based on mutual information minimization (MT-MIM), which casts the disentangled representation learning as an objective of information constraints. The method trains a disentanglement network to minimize mutual information between the identity component and age component of the face image from the same person, and reduce the effect of age variations during the identification process. For quantitative measure of the degree of disentanglement, we verify that mutual information can represent as metric. The resulting identity-dependent representations are used for age-invariant face recognition. We evaluate MT-MIM on popular public-domain face aging datasets (FG-NET, MORPH Album 2, CACD and AgeDB) and obtained significant improvements over previous state-of-the-art methods. Specifically, our method exceeds the baseline models by over 0.4% on MORPH Album 2, and over 0.7% on CACD subsets, which are impressive improvements at the high accuracy levels of above 99% and an average of 94%. | ['Shengjin Wang', 'YaLi Li', 'Xuege Hou'] | 2021-01-01 | null | null | null | iccv-2021-1 | ['age-invariant-face-recognition'] | ['computer-vision'] | [ 2.03126654e-01 -8.72362852e-02 -2.36960754e-01 -6.17839992e-01
-5.76814711e-01 -9.25877914e-02 4.40658659e-01 -2.92451590e-01
-3.58952850e-01 7.51014054e-01 1.57930925e-01 4.02358115e-01
-2.18128368e-01 -5.19134998e-01 -2.99108565e-01 -9.30663049e-01
-2.26490915e-01 5.32481849e-01 -5.40856302e-01 -6.44683540e-02
1.27882175e-02 4.75070328e-01 -1.94667423e+00 2.94360053e-02
8.56436491e-01 1.23713017e+00 -5.04347682e-01 2.49357864e-01
3.38247329e-01 2.70434976e-01 -4.07456994e-01 -7.50488520e-01
2.64835835e-01 -1.65496379e-01 -5.40139079e-01 -3.65931317e-02
1.00896287e+00 -3.41917813e-01 -4.74398166e-01 9.69131947e-01
9.61056471e-01 -2.43695423e-01 1.02243710e+00 -1.63502431e+00
-7.44522035e-01 3.42108935e-01 -1.10806239e+00 8.20799768e-02
5.87513112e-02 -1.75727174e-01 7.89356887e-01 -1.03727746e+00
2.64629930e-01 1.63141954e+00 6.16470456e-01 1.19714642e+00
-1.28312409e+00 -1.26172292e+00 8.20695534e-02 4.30442989e-01
-1.59351623e+00 -8.51350367e-01 6.43987119e-01 -5.27819276e-01
4.67426330e-01 3.04621667e-01 2.32099891e-01 1.16346383e+00
1.04378052e-01 5.65054893e-01 1.34678066e+00 -2.65686095e-01
-2.00541601e-01 -2.54403859e-01 7.28293508e-02 1.02918410e+00
4.65420008e-01 1.26916394e-01 -8.98598492e-01 -6.94561601e-02
6.63250506e-01 -1.27026945e-01 -1.08653985e-01 -1.98720425e-01
-8.68784308e-01 5.54449558e-01 1.14745654e-01 -1.29261568e-01
-8.11228156e-02 -1.47796765e-01 4.16636705e-01 3.89536500e-01
7.86251843e-01 -1.46912098e-01 -3.77982825e-01 2.16244787e-01
-8.66016269e-01 1.92141742e-01 3.40362102e-01 4.24376488e-01
5.87203145e-01 9.90402475e-02 -2.20865101e-01 1.22275722e+00
5.64034700e-01 7.38383234e-01 4.04431254e-01 -7.83220947e-01
1.62768558e-01 5.89731276e-01 -4.97171462e-01 -7.48411894e-01
-2.96303213e-01 -2.42745385e-01 -1.12784004e+00 6.67370200e-01
6.18911684e-01 -1.65969916e-02 -1.01366079e+00 2.33826113e+00
3.59304279e-01 3.33735287e-01 -2.50598401e-01 6.13235474e-01
8.24013174e-01 -1.88177824e-02 3.16300720e-01 -5.82264066e-01
1.72333121e+00 -5.89945257e-01 -5.95619857e-01 -3.86061579e-01
1.16487771e-01 -6.03239954e-01 4.42136794e-01 2.80271679e-01
-1.02848577e+00 -5.76788664e-01 -1.28300464e+00 1.22297004e-01
3.97823984e-03 3.25566709e-01 7.37981200e-01 9.87339258e-01
-1.02051258e+00 7.41252482e-01 -5.56271911e-01 -9.07114893e-02
8.77920687e-01 7.90190816e-01 -1.03789949e+00 -2.63848901e-01
-1.04067492e+00 7.87091970e-01 -2.05993026e-01 1.26868382e-01
-9.20825303e-01 -1.11323500e+00 -8.72409165e-01 -4.08319652e-01
8.36896002e-02 -9.42281663e-01 7.04363763e-01 -8.04273248e-01
-1.17742717e+00 1.58311248e+00 -1.96653590e-01 6.81763664e-02
5.00302076e-01 -3.86805892e-01 -6.98436081e-01 -4.12579030e-02
3.93166468e-02 5.54063022e-01 1.17477691e+00 -1.04763651e+00
-2.26530090e-01 -1.36875892e+00 -3.52378219e-01 1.33103982e-01
-5.32815218e-01 4.17982101e-01 -2.98829168e-01 -7.00502336e-01
6.69304505e-02 -9.71112847e-01 3.38430315e-01 7.19324648e-01
-1.65784240e-01 -3.23173434e-01 5.78967690e-01 -1.01920974e+00
1.09417367e+00 -2.05616760e+00 5.89306891e-01 -2.32348666e-01
5.76002896e-01 9.04443637e-02 -2.57230848e-01 -1.80116013e-01
-5.57873130e-01 6.69137901e-03 -2.19838887e-01 -8.18479836e-01
-2.00396404e-01 -6.21673279e-02 3.36973876e-01 7.58022666e-01
2.41344854e-01 4.10357207e-01 -4.63293970e-01 -6.07890427e-01
-3.75058055e-01 7.05650032e-01 -3.16057950e-01 2.74931878e-01
2.51005769e-01 5.18110812e-01 -1.71910912e-01 9.09043729e-01
9.68676865e-01 2.52757311e-01 2.20266938e-01 -6.67928576e-01
2.53835708e-01 -4.15933996e-01 -9.60420787e-01 1.68687356e+00
-2.54323512e-01 2.75503099e-01 1.63014904e-01 -8.57015073e-01
1.02178466e+00 2.72551835e-01 7.25686491e-01 -6.16416633e-01
3.38179499e-01 1.35207519e-01 1.64649516e-01 -1.83026120e-01
-2.87243456e-01 -3.31424743e-01 1.70737222e-01 5.06593764e-01
3.36543888e-01 5.38039625e-01 2.15891823e-02 8.49515013e-03
7.96555996e-01 -7.14149922e-02 1.69130728e-01 -4.45830196e-01
7.89870322e-01 -1.19342530e+00 9.66587961e-01 3.57954577e-02
-6.84910715e-01 7.49748468e-01 4.79198128e-01 -3.37079912e-01
-8.24882030e-01 -1.31215072e+00 -3.36100012e-01 1.11071682e+00
-2.73881406e-01 -2.22798437e-01 -7.92795122e-01 -8.43418896e-01
3.74683499e-01 5.11022992e-02 -1.09108150e+00 -5.72561622e-01
-4.31828350e-01 -1.13529491e+00 7.32651114e-01 5.22743762e-01
6.13105297e-01 -6.01127803e-01 3.55021417e-01 -5.49853921e-01
2.80437153e-02 -9.94409204e-01 -6.88605249e-01 -4.58478719e-01
-7.90474355e-01 -1.22297621e+00 -8.78825724e-01 -6.52263582e-01
8.71314645e-01 -5.61306328e-02 1.07525706e+00 -4.24580276e-02
-7.20409214e-01 2.20290810e-01 1.72638044e-01 -2.63251275e-01
-4.66606468e-02 -1.57057703e-01 6.49465024e-01 5.73041499e-01
3.37097496e-01 -9.66960371e-01 -9.51547444e-01 4.61406380e-01
-3.92078370e-01 -1.12911627e-01 6.19356751e-01 7.22784221e-01
3.48666787e-01 -3.52153748e-01 1.01750195e+00 -6.36187494e-01
2.07267910e-01 -3.03784460e-01 -2.96043474e-02 4.31624204e-01
-9.82392371e-01 2.20012948e-01 -8.00761655e-02 -4.00822580e-01
-1.10235357e+00 -6.19089827e-02 -6.59302399e-02 -3.86590838e-01
1.01858638e-01 -8.53561684e-02 -8.09420824e-01 -2.39087328e-01
4.76096541e-01 -6.69034943e-02 6.55306756e-01 -4.90406454e-01
1.71913728e-01 6.91147327e-01 4.47149664e-01 -8.04386854e-01
8.24027479e-01 4.13596034e-01 2.03485131e-01 -6.31979346e-01
-9.63104129e-01 -5.31882569e-02 -7.57306039e-01 -3.92705172e-01
7.33248174e-01 -1.17317653e+00 -9.24664557e-01 1.00575316e+00
-7.76613474e-01 1.68368980e-01 2.53100663e-01 2.33881921e-01
-3.08809310e-01 4.74115461e-01 -5.24399638e-01 -8.26533020e-01
-7.64359593e-01 -9.36880410e-01 1.08496463e+00 3.32083762e-01
-8.98717195e-02 -6.92444921e-01 4.40478809e-02 8.38442802e-01
1.33817956e-01 6.44354761e-01 1.04111087e+00 -5.21460593e-01
-5.34754880e-02 -1.50427938e-01 -4.04772907e-01 4.27436739e-01
4.90618646e-01 -2.55961604e-02 -1.23233068e+00 -5.56349099e-01
-2.33262017e-01 -5.56115091e-01 1.01517880e+00 1.83828890e-01
1.21959698e+00 -1.45307824e-01 -2.66838104e-01 5.91105759e-01
1.13970900e+00 -9.25545543e-02 7.83613861e-01 -1.92713633e-01
7.76068151e-01 7.93297708e-01 2.48828873e-01 4.96827483e-01
3.74726713e-01 7.30174661e-01 4.28067356e-01 1.88999251e-01
-5.04122794e-01 9.28967595e-02 3.45245600e-01 8.65136623e-01
-4.42057967e-01 3.37305158e-01 -5.44330060e-01 2.73163110e-01
-1.53735888e+00 -8.05693388e-01 2.17990145e-01 2.43695688e+00
1.02292347e+00 -1.15514278e-01 3.85539263e-01 2.60254502e-01
8.94908607e-01 3.08772504e-01 -8.54810476e-01 -4.61098999e-02
-3.16847473e-01 3.91796708e-01 1.52213806e-02 2.62374043e-01
-1.06710517e+00 4.66009319e-01 5.78675222e+00 7.42627740e-01
-8.47128093e-01 8.35863501e-02 1.00582933e+00 -3.38963062e-01
1.46733865e-01 -5.56848288e-01 -9.24553812e-01 4.73488867e-01
7.15519547e-01 -3.24551731e-01 4.20429379e-01 6.25535131e-01
-1.26779810e-01 2.53289729e-01 -1.39796996e+00 1.42034960e+00
5.74518263e-01 -6.83914363e-01 5.06845377e-02 2.43457451e-01
5.06502271e-01 -5.05740821e-01 5.99351108e-01 2.19692990e-01
-1.27109677e-01 -1.38545811e+00 3.43108654e-01 6.35734558e-01
1.39605892e+00 -9.30382490e-01 3.89571100e-01 -1.99526250e-01
-1.29642236e+00 -4.57875021e-02 -1.39819786e-01 3.91400866e-02
-2.03651726e-01 6.70527935e-01 -1.62900820e-01 5.92348337e-01
5.94950497e-01 7.36189842e-01 -7.70037830e-01 7.00306952e-01
5.14618196e-02 2.43582681e-01 2.72792336e-02 6.06995821e-01
-7.37817407e-01 -1.95916310e-01 2.56753415e-01 5.23257136e-01
1.48931131e-01 1.51671041e-02 -1.76785141e-01 4.93752003e-01
-4.57038969e-01 -9.89308860e-03 -2.44578868e-01 -1.90835372e-02
4.44462568e-01 1.33313882e+00 -2.42635295e-01 2.45630052e-02
-3.47405761e-01 1.12100565e+00 5.50462008e-01 1.64492801e-02
-7.24620581e-01 -9.89565998e-02 1.39872742e+00 4.21867594e-02
-1.30168542e-01 -3.77031825e-02 -1.22618251e-01 -1.12676525e+00
1.30138248e-01 -1.11114860e+00 5.86613715e-01 -2.10998222e-01
-1.64468980e+00 5.79866588e-01 -1.63209870e-01 -6.89911366e-01
3.43873054e-02 -7.30834544e-01 -4.23115909e-01 9.68987763e-01
-1.20301092e+00 -1.55693913e+00 -3.90315861e-01 4.68270212e-01
4.30295110e-01 -5.19287109e-01 1.05157101e+00 8.03555787e-01
-1.21769035e+00 1.27738035e+00 -1.08720839e-01 2.09452331e-01
1.18822491e+00 -1.07976079e+00 8.39861482e-02 5.99484205e-01
-1.32479236e-01 6.67351604e-01 4.56737399e-01 -5.27976930e-01
-1.34026289e+00 -9.10834014e-01 7.12708712e-01 -5.61202645e-01
3.40999633e-01 -5.17135501e-01 -7.93626964e-01 3.62783968e-01
-5.09729013e-02 2.58113325e-01 1.06053591e+00 3.66011828e-01
-1.14486766e+00 -5.92046916e-01 -1.30327570e+00 5.42885363e-01
1.57942498e+00 -6.25103176e-01 -3.41585338e-01 1.26234561e-01
4.82669115e-01 9.14720520e-02 -1.21886528e+00 8.39958310e-01
1.19708741e+00 -1.02163446e+00 1.24553931e+00 -7.27184832e-01
3.64960998e-01 -3.07065677e-02 -1.24558985e-01 -1.01297677e+00
-3.87659013e-01 -4.69395727e-01 -4.48942542e-01 1.73542738e+00
1.46440417e-01 -4.74780798e-01 9.25753593e-01 6.17840827e-01
3.48333478e-01 -9.78574216e-01 -1.07499123e+00 -6.91531181e-01
2.77323246e-01 7.07582161e-02 4.93222445e-01 8.58657777e-01
-4.16106611e-01 4.06157136e-01 -4.64500904e-01 1.66613579e-01
1.24270487e+00 -2.34159395e-01 3.16510439e-01 -1.68715882e+00
5.89812249e-02 -4.92838621e-01 -7.51282811e-01 -2.53841966e-01
6.68449342e-01 -8.64890397e-01 -3.62142414e-01 -1.05275178e+00
7.09193468e-01 -2.86144733e-01 -5.62424064e-01 5.05291760e-01
-3.68913144e-01 5.92345417e-01 -9.42136198e-02 5.84967295e-03
-2.67230451e-01 8.20967376e-01 1.09078109e+00 -3.41885686e-01
3.29601020e-01 4.70503457e-02 -9.31037664e-01 6.14714921e-01
6.87588573e-01 -1.78508118e-01 -4.10167485e-01 -2.55684316e-01
-8.71630535e-02 -2.28703082e-01 8.90378430e-02 -1.02742195e+00
-6.68675974e-02 9.74505991e-02 7.39199758e-01 -1.06455840e-01
6.39767230e-01 -3.82660508e-01 1.59587830e-01 5.32253444e-01
-9.48221534e-02 -7.12741353e-03 1.54177949e-01 4.72016275e-01
3.81108932e-02 1.39893904e-01 1.11575925e+00 2.44948566e-01
-4.51097667e-01 1.06653905e+00 1.81532994e-01 7.61329606e-02
9.25922275e-01 -1.48333892e-01 -3.28666598e-01 -1.60478558e-02
-8.17836344e-01 4.45581879e-03 1.71380967e-01 7.06066966e-01
6.50535405e-01 -1.68640113e+00 -1.13196337e+00 3.45120221e-01
2.77878255e-01 -4.84978616e-01 6.42328262e-01 7.46684134e-01
3.38072726e-03 -1.27566651e-01 -6.76743925e-01 -3.19388449e-01
-1.94412613e+00 2.88033605e-01 4.39093739e-01 -6.61457926e-02
4.96072732e-02 1.10282719e+00 4.49182898e-01 -2.83122420e-01
2.21179947e-01 4.84775156e-01 -3.81941706e-01 4.11967903e-01
8.10317934e-01 6.35839641e-01 2.49543861e-02 -9.51958597e-01
-6.01420820e-01 9.42490697e-01 -4.84790683e-01 1.58053726e-01
1.30521965e+00 -9.81942564e-02 -3.98814440e-01 2.08307281e-01
1.37550485e+00 -1.64783299e-01 -1.21845675e+00 -1.29739702e-01
-2.13821083e-01 -4.92908031e-01 -7.68952519e-02 -8.15168858e-01
-1.48863852e+00 7.74387240e-01 1.22569585e+00 -5.43137431e-01
1.14305496e+00 7.28774518e-02 4.88343596e-01 -2.81040341e-01
3.57601941e-01 -7.91905820e-01 4.80233133e-01 3.98879386e-02
1.11599422e+00 -1.39196932e+00 2.81271696e-01 -6.50610626e-01
-2.63648301e-01 8.11563313e-01 1.02945566e+00 1.13026313e-01
8.74293923e-01 -1.65751986e-02 -4.92524505e-02 -1.15151189e-01
-5.24399102e-01 -4.77771759e-02 6.36155367e-01 8.55856001e-01
5.79787135e-01 1.92353167e-02 -2.68942773e-01 7.92164505e-01
-8.82548839e-02 -3.23615134e-01 -2.60084242e-01 4.55004930e-01
-2.20175102e-01 -1.37585068e+00 -2.08599657e-01 5.44528008e-01
-5.69867730e-01 1.97537988e-01 -4.74619389e-01 5.84250867e-01
3.09350610e-01 6.77297115e-01 2.62657390e-03 -4.98427838e-01
1.35076925e-01 2.90728748e-01 1.09650338e+00 -4.19805437e-01
-1.58061549e-01 -3.71426195e-01 1.81391329e-01 -5.44700623e-01
-3.12722057e-01 -8.70980740e-01 -7.54217267e-01 -3.77558827e-01
-2.20630854e-01 -3.28076780e-01 5.51173985e-01 7.88239419e-01
4.63763744e-01 3.38887006e-01 8.88058782e-01 -6.98958993e-01
-6.18873239e-01 -1.01521099e+00 -7.63598025e-01 4.85226572e-01
1.43914178e-01 -1.14912808e+00 -3.93427074e-01 -6.08665720e-02] | [13.35424518585205, 0.7041165828704834] |
ac559824-9877-48fc-9ff5-97fc27bf408b | detecting-mitoses-with-a-convolutional-neural | 2208.12437 | null | https://arxiv.org/abs/2208.12437v2 | https://arxiv.org/pdf/2208.12437v2.pdf | Detecting Mitoses with a Convolutional Neural Network for MIDOG 2022 Challenge | This work presents a mitosis detection method with only one vanilla Convolutional Neural Network (CNN). Our method consists of two steps: given an image, we first apply a CNN using a sliding window technique to extract patches that have mitoses; we then calculate each extracted patch's class activation map to obtain the mitosis's precise location. To increase the model performance on high-domain-variance pathology images, we train the CNN with a data augmentation pipeline, a noise-tolerant loss that copes with unlabeled images, and a multi-rounded active learning strategy. In the MIDOG 2022 challenge, our approach, with an EfficientNet-b3 CNN model, achieved an overall F1 score of 0.7323 in the preliminary test phase, and 0.6847 in the final test phase (task 1). Our approach sheds light on the broader applicability of class activation maps for object detections in pathology images. | ["Xiang 'Anthony' Chen", 'Shino Magaki', 'Neda Zarrin-Khameh', 'Christopher Kazu Williams', 'Shuo Ni', 'Mohammad Haeri', 'Hongyan Gu'] | 2022-08-26 | null | null | null | null | ['mitosis-detection'] | ['medical'] | [ 5.76396525e-01 4.62457299e-01 -2.66983479e-01 -1.07433490e-01
-1.33263886e+00 -4.01884764e-01 5.15877426e-01 5.33554256e-01
-1.00138807e+00 8.18981588e-01 -2.09244788e-01 -3.56522143e-01
2.99007773e-01 -6.95430577e-01 -6.16609395e-01 -1.06335294e+00
-5.95106557e-02 4.56353217e-01 6.59176946e-01 6.39217123e-02
1.23282582e-01 5.51850975e-01 -1.10304558e+00 4.75435853e-01
6.75117910e-01 1.09161067e+00 4.01506759e-03 1.09372973e+00
1.59616902e-01 8.66118729e-01 -6.27011657e-01 -3.04224312e-01
5.90980574e-02 -8.14595968e-02 -9.26521599e-01 1.06038116e-01
3.09607446e-01 -2.27956623e-01 -1.37787044e-01 8.59653056e-01
5.81447542e-01 -2.16606602e-01 6.73872709e-01 -7.09999323e-01
-3.60792652e-02 4.11669135e-01 -6.35707319e-01 7.06931412e-01
-2.38053694e-01 6.83842361e-01 7.90454924e-01 -8.77877235e-01
9.02582705e-01 5.39507449e-01 8.81827176e-01 6.31122649e-01
-1.36438668e+00 -4.05468792e-01 -2.87553668e-01 -7.17071742e-02
-1.29252601e+00 -4.76985455e-01 3.19924802e-01 -4.15900856e-01
1.02664042e+00 2.49825373e-01 8.03784132e-01 1.01278365e+00
3.17943931e-01 1.03777444e+00 9.20374930e-01 -3.54810268e-01
3.73882443e-01 1.93793210e-03 1.93991080e-01 5.43209970e-01
1.87910244e-01 -3.44028734e-02 -2.19574869e-01 -6.52903458e-03
6.87519372e-01 -2.01044977e-01 -8.85581374e-02 -1.15693547e-01
-1.32611835e+00 6.86297655e-01 6.84828341e-01 4.24551457e-01
-2.57519811e-01 1.19900517e-01 4.95508939e-01 3.20824385e-02
7.08120883e-01 5.09667814e-01 -3.20757031e-01 2.06730828e-01
-1.20177674e+00 2.10237503e-01 4.89039332e-01 2.64194787e-01
5.56541145e-01 -4.91602540e-01 -4.34275717e-01 5.22783518e-01
1.63155541e-01 -3.69828939e-03 3.58304858e-01 -5.51857650e-01
6.01052940e-02 1.02367842e+00 -3.10511529e-01 -4.12086487e-01
-6.07252717e-01 -7.05139756e-01 -8.79623413e-01 5.78178465e-01
8.32984090e-01 -1.10206522e-01 -1.38616669e+00 1.30874670e+00
3.43957365e-01 1.83355957e-01 -5.22909779e-03 7.15078056e-01
8.71892631e-01 1.38966054e-01 2.97481835e-01 -5.88193648e-02
1.38107979e+00 -9.30326641e-01 -4.34307843e-01 -3.37862730e-01
9.49429274e-01 -6.07192397e-01 8.08247507e-01 3.70108575e-01
-1.14265656e+00 -3.47644001e-01 -1.31152141e+00 -3.66686344e-01
-6.07832074e-01 4.11849648e-01 6.46200538e-01 4.75567281e-01
-1.32884848e+00 4.18750733e-01 -1.12287605e+00 -4.49841857e-01
1.00986075e+00 7.43677318e-01 -4.31913018e-01 1.12093233e-01
-6.74853861e-01 7.79421449e-01 6.18005991e-01 -2.64648553e-02
-1.08231354e+00 -9.15901601e-01 -6.23117805e-01 2.03076571e-01
2.37504750e-01 -7.57949531e-01 1.14861941e+00 -1.05209613e+00
-1.05493963e+00 1.32441247e+00 1.71238169e-01 -7.07195520e-01
5.46016634e-01 4.93398130e-01 2.35562958e-02 3.19224626e-01
-2.53955908e-02 1.14026332e+00 3.65671843e-01 -7.53301024e-01
-8.17586362e-01 -4.26470727e-01 -4.71085384e-02 -5.36544435e-02
-2.85648376e-01 -1.92292139e-01 -6.80536568e-01 -6.02846682e-01
-1.42608300e-01 -8.55142653e-01 -5.14014304e-01 3.42937708e-01
-4.81342643e-01 -5.67787066e-02 5.84233403e-01 -6.61025047e-01
7.47946143e-01 -2.30855417e+00 -1.12085417e-02 2.65716463e-01
5.84952235e-01 2.49689564e-01 -4.92952392e-02 -2.18961164e-01
-1.92909032e-01 7.28663504e-02 -3.86976600e-01 -6.71562552e-01
-4.00563836e-01 -1.55758649e-01 2.65845239e-01 5.84291995e-01
8.49810719e-01 1.13461995e+00 -8.32411885e-01 -6.22738242e-01
-7.31758103e-02 4.94130552e-01 -6.26117229e-01 9.06105340e-02
-1.50822058e-01 3.25005651e-01 8.51719305e-02 1.06798756e+00
5.47809601e-01 -5.82799971e-01 1.19050734e-01 5.81814088e-02
5.61513454e-02 -8.94177929e-02 -6.96441472e-01 1.73587132e+00
2.31164619e-02 9.03998137e-01 2.45341703e-01 -9.58563268e-01
6.32007897e-01 3.20366889e-01 6.69075310e-01 -5.21787941e-01
2.09712908e-01 3.61278445e-01 1.25536919e-01 -3.03545475e-01
3.15805584e-01 2.07334965e-01 4.93507572e-02 3.41754630e-02
3.89755994e-01 1.97564065e-01 4.92862582e-01 2.46735066e-01
1.79068303e+00 -1.95064664e-01 3.51091921e-01 -3.12124729e-01
4.59695131e-01 2.00344473e-01 5.79980731e-01 5.02243578e-01
-5.50281107e-01 8.86529624e-01 9.03997719e-01 -5.61461508e-01
-1.11309195e+00 -9.51390266e-01 -1.95806101e-01 6.40398800e-01
-2.81463712e-01 -3.18941846e-02 -5.83221614e-01 -1.14434445e+00
-1.55725598e-01 5.17150983e-02 -1.16998589e+00 -2.07884774e-01
-4.65861380e-01 -1.23223531e+00 6.77913845e-01 5.98859787e-01
5.27343333e-01 -1.08552146e+00 -5.81021488e-01 1.86720341e-01
-1.16158202e-02 -8.70747030e-01 -1.99840918e-01 6.49844646e-01
-8.92774999e-01 -1.38470829e+00 -9.41633582e-01 -9.62266982e-01
9.91855025e-01 -2.28703380e-01 1.14050817e+00 3.48568052e-01
-7.89006472e-01 -2.46239945e-01 -1.42405540e-01 -5.03648221e-01
-3.96015942e-01 3.26673150e-01 -6.82510555e-01 -2.47313887e-01
5.17419517e-01 -1.40880421e-01 -8.70697856e-01 -1.09320588e-01
-9.35311317e-01 7.59699382e-03 9.77070510e-01 1.15126920e+00
8.51192415e-01 -2.38254622e-01 3.70698154e-01 -1.04302287e+00
4.69771363e-02 -3.68376672e-01 -5.60076773e-01 -7.24215480e-03
-3.53945106e-01 -4.77538496e-01 2.11745545e-01 -4.26693052e-01
-7.38126040e-01 7.79221416e-01 -2.80541152e-01 -4.73944023e-02
-1.10132229e-02 3.83451790e-01 8.19540322e-02 -3.09182435e-01
9.60072517e-01 5.85729890e-02 2.17443615e-01 -7.45968074e-02
-1.39419481e-01 3.59436691e-01 7.37970948e-01 1.49700463e-01
4.92734551e-01 6.25550270e-01 5.67701012e-02 -5.24333954e-01
-7.50032365e-01 -7.29467332e-01 -6.18734300e-01 -1.94298759e-01
9.51045156e-01 -9.44272697e-01 -6.00315273e-01 6.82406306e-01
-9.30672824e-01 -6.21754527e-01 -4.09885496e-01 4.62517202e-01
-3.59264225e-01 6.72231242e-02 -8.68306577e-01 -3.77716392e-01
-5.21933854e-01 -1.13834906e+00 1.18069625e+00 3.97933304e-01
-2.46728435e-01 -9.02053654e-01 3.10053527e-01 6.04105413e-01
3.66851479e-01 4.31639880e-01 8.47405612e-01 -9.61685658e-01
-7.17159510e-01 -4.61021423e-01 -2.75339663e-01 1.98984101e-01
-1.81956857e-01 2.25596488e-01 -1.32633829e+00 -4.78503436e-01
-5.61847091e-01 -4.82658386e-01 1.37394583e+00 5.22401571e-01
1.28207409e+00 1.39955506e-01 -6.85155690e-01 7.49198973e-01
1.43773866e+00 4.65259962e-02 9.23362553e-01 4.39269096e-01
1.84608012e-01 3.02638322e-01 4.03814852e-01 -1.16269983e-01
-1.33377798e-02 3.48994076e-01 6.41271412e-01 -8.59367549e-01
-5.18763185e-01 3.50147367e-01 -1.64222699e-02 2.42735192e-01
6.11186735e-02 -2.39285201e-01 -1.19843340e+00 9.16307330e-01
-1.60979784e+00 -6.07509673e-01 1.89480446e-02 1.99465525e+00
9.42545295e-01 5.95305681e-01 4.65585887e-02 3.63829464e-01
6.13219619e-01 -2.54929811e-01 -6.06500685e-01 -8.15493986e-02
-1.75014988e-01 4.48907822e-01 5.77336192e-01 1.96715474e-01
-1.50087655e+00 6.24118984e-01 6.60307980e+00 9.68023002e-01
-1.16804767e+00 1.28287852e-01 1.35311544e+00 -1.11261621e-01
2.67217845e-01 -1.15249939e-01 -7.05430567e-01 4.13370937e-01
8.15314770e-01 2.63975382e-01 -2.25742906e-01 6.47475183e-01
-9.51993242e-02 -4.92174268e-01 -9.75630403e-01 5.10989070e-01
-5.70199341e-02 -1.69839835e+00 -3.61430377e-01 4.43060130e-01
6.31373286e-01 1.08787924e-01 6.96635991e-02 4.01971847e-01
1.64191082e-01 -1.19604361e+00 1.69929624e-01 4.99205649e-01
8.05788994e-01 -7.97763109e-01 1.28091753e+00 2.63804197e-01
-8.61794472e-01 -5.25678173e-02 -2.80360788e-01 1.61077961e-01
-2.71822780e-01 7.61434972e-01 -1.59701955e+00 1.41855299e-01
4.99958336e-01 3.46157074e-01 -9.67411995e-01 1.59735906e+00
1.73794284e-01 6.55975163e-01 -3.79969418e-01 -1.16494790e-01
1.11754164e-01 4.69046801e-01 4.35334146e-01 1.40747428e+00
1.09498017e-01 -1.26077071e-01 6.43798113e-02 7.93187559e-01
-2.63027549e-01 3.26022059e-02 -7.99353123e-02 -1.39328718e-01
1.84098393e-01 1.78956437e+00 -1.46213353e+00 -4.00334865e-01
-5.51430210e-02 7.01434672e-01 3.98205370e-01 5.64000234e-02
-6.00378871e-01 -5.98825812e-01 1.76516086e-01 1.38338149e-01
2.23113939e-01 2.02862337e-01 -6.77693367e-01 -8.36134374e-01
-2.83482581e-01 -6.68604672e-01 5.17732918e-01 -5.16392469e-01
-1.04246891e+00 5.06663680e-01 -6.08681321e-01 -1.09044659e+00
3.22243050e-02 -6.75439477e-01 -8.35078359e-01 6.92418337e-01
-1.30051899e+00 -1.25947452e+00 -3.62656981e-01 2.23784909e-01
4.53587234e-01 -1.36082172e-01 8.08374286e-01 4.77091938e-01
-6.32806361e-01 7.34566629e-01 -1.77775607e-01 4.56520975e-01
6.79708064e-01 -1.61061943e+00 4.18520093e-01 8.69677603e-01
-2.23100901e-01 4.01044756e-01 3.87876540e-01 -5.61623335e-01
-8.13085020e-01 -1.14222801e+00 9.21912193e-01 -4.11649585e-01
4.77174073e-01 -4.46794719e-01 -7.92691469e-01 5.90948164e-01
1.40765905e-01 3.74999881e-01 7.47527599e-01 -1.28745422e-01
9.68044177e-02 7.10431337e-02 -1.46054065e+00 3.48328859e-01
5.17220974e-01 -2.34159678e-01 -1.30817622e-01 4.68434542e-01
4.37418431e-01 -6.44374073e-01 -8.29294741e-01 6.62923694e-01
2.67779112e-01 -7.81476796e-01 7.99673200e-01 -3.05146635e-01
4.27497327e-01 -1.16882585e-01 3.06772798e-01 -1.06529582e+00
-4.21191692e-01 -1.99170411e-01 -2.79840529e-02 7.73378074e-01
8.79279792e-01 -2.99561650e-01 1.37378430e+00 1.63071062e-02
-3.99482638e-01 -1.21774685e+00 -1.16464460e+00 -1.57191291e-01
1.75151214e-01 8.78871605e-03 8.84060934e-02 6.34871244e-01
-5.94783537e-02 1.58987463e-01 3.06480974e-01 -6.54513091e-02
4.48723316e-01 -5.76869369e-01 5.62640190e-01 -1.24202287e+00
-2.91495830e-01 -4.58610296e-01 -7.40153074e-01 -4.14304614e-01
-1.78151518e-01 -8.21345687e-01 1.74815327e-01 -1.22888052e+00
6.13176405e-01 -2.45120108e-01 -4.94599760e-01 7.79468536e-01
-3.56318742e-01 9.83670652e-01 -1.83720604e-01 6.86441511e-02
-7.02692866e-01 -1.55155852e-01 1.09911704e+00 -4.00207758e-01
6.66911975e-02 4.96365968e-03 -5.30144811e-01 5.82067192e-01
6.60383642e-01 -4.28199440e-01 -1.36864722e-01 7.40302131e-02
1.81983545e-01 -2.24805027e-01 6.49137557e-01 -1.28842258e+00
4.53009516e-01 3.00954103e-01 1.13921249e+00 -8.62901866e-01
4.25149560e-01 -5.04394531e-01 -5.33604249e-03 8.06229949e-01
-4.26910549e-01 -2.89004236e-01 4.23197716e-01 3.73399824e-01
-1.03788348e-02 1.19919172e-02 1.03566480e+00 -1.19322814e-01
-2.89578080e-01 3.74845654e-01 -6.30762041e-01 -2.42314219e-01
1.12445164e+00 -4.78711486e-01 -5.38158059e-01 2.89483041e-01
-1.00770378e+00 2.69359112e-01 4.93959457e-01 -4.35655154e-02
3.20136487e-01 -1.04515934e+00 -6.83140516e-01 2.64767021e-01
2.11980239e-01 3.16298187e-01 2.82310277e-01 1.14251256e+00
-8.60919714e-01 2.55114973e-01 -2.60060549e-01 -9.66409743e-01
-1.35869181e+00 2.87562162e-01 6.55283868e-01 -8.94498944e-01
-4.12098169e-01 1.28454196e+00 1.20637663e-01 -2.87986547e-01
3.00963938e-01 -2.80846089e-01 -3.77892315e-01 1.05944462e-01
5.46232820e-01 2.01564834e-01 6.86375856e-01 -2.87654728e-01
-4.76688623e-01 1.65415034e-02 -4.81070727e-01 4.70449403e-02
1.21674311e+00 4.20100421e-01 -8.81132334e-02 1.80562407e-01
1.29861462e+00 -3.25448096e-01 -1.32445776e+00 -9.70731899e-02
1.85291439e-01 -4.93521094e-02 4.41507906e-01 -1.16540337e+00
-1.21081591e+00 5.83855152e-01 1.03285348e+00 3.99749242e-02
1.19179797e+00 -4.18448960e-03 5.55977941e-01 2.41628274e-01
-1.10752955e-01 -9.95658875e-01 1.70054391e-01 4.09896046e-01
3.60603988e-01 -1.32049155e+00 -2.71531362e-02 -3.96981597e-01
-1.83272764e-01 1.22487199e+00 7.65424609e-01 -8.38251635e-02
3.31071496e-01 6.38654530e-01 1.89761713e-01 -4.17242289e-01
-1.08684886e+00 -2.50925601e-01 2.19365135e-01 4.98970449e-01
5.16281545e-01 -1.00034915e-01 -1.90700859e-01 5.11210740e-01
1.24339268e-01 3.38538647e-01 5.57320595e-01 9.42807615e-01
-5.06424427e-01 -7.73396254e-01 -6.24720342e-02 4.99029517e-01
-7.39928246e-01 7.36939684e-02 -7.44496107e-01 9.26145017e-01
2.94303298e-01 4.83081490e-01 5.48078299e-01 -1.53961167e-01
8.40021074e-02 -1.85729176e-01 3.57419193e-01 -7.40818918e-01
-9.88656640e-01 2.20685005e-01 -8.71051475e-03 -3.91294390e-01
-2.17304394e-01 -4.58219856e-01 -1.28022575e+00 -1.80843726e-01
-4.54024911e-01 -1.94536939e-01 4.90591168e-01 7.50253737e-01
1.37092367e-01 8.32137406e-01 3.41713935e-01 -6.50720775e-01
-2.65284687e-01 -1.03744066e+00 -4.35515583e-01 1.22052833e-01
4.39958870e-01 -3.37604940e-01 -3.87485027e-01 1.81847230e-01] | [15.096780776977539, -3.0804970264434814] |
773780c4-dbba-419e-866a-b9d00c427bcf | exploiting-visual-semantic-reasoning-for | 2006.08889 | null | https://arxiv.org/abs/2006.08889v1 | https://arxiv.org/pdf/2006.08889v1.pdf | Exploiting Visual Semantic Reasoning for Video-Text Retrieval | Video retrieval is a challenging research topic bridging the vision and language areas and has attracted broad attention in recent years. Previous works have been devoted to representing videos by directly encoding from frame-level features. In fact, videos consist of various and abundant semantic relations to which existing methods pay less attention. To address this issue, we propose a Visual Semantic Enhanced Reasoning Network (ViSERN) to exploit reasoning between frame regions. Specifically, we consider frame regions as vertices and construct a fully-connected semantic correlation graph. Then, we perform reasoning by novel random walk rule-based graph convolutional networks to generate region features involved with semantic relations. With the benefit of reasoning, semantic interactions between regions are considered, while the impact of redundancy is suppressed. Finally, the region features are aggregated to form frame-level features for further encoding to measure video-text similarity. Extensive experiments on two public benchmark datasets validate the effectiveness of our method by achieving state-of-the-art performance due to the powerful semantic reasoning. | ['Caili Guo', 'Zheng Li', 'Zhimin Zeng', 'Zerun Feng'] | 2020-06-16 | null | null | null | null | ['video-text-retrieval'] | ['computer-vision'] | [ 2.20183939e-01 -1.64092943e-01 -3.79017949e-01 -3.76504213e-01
-2.53907025e-01 -2.78151900e-01 7.22142339e-01 2.56619960e-01
-7.96853155e-02 4.08539414e-01 4.35866177e-01 2.43011233e-03
-2.20222771e-01 -1.00276554e+00 -6.64463341e-01 -4.51025009e-01
6.10049106e-02 -2.77700782e-01 5.16705871e-01 -3.32453698e-01
2.73024559e-01 2.10750595e-01 -1.39108932e+00 4.40021873e-01
7.56998003e-01 1.14158392e+00 -3.97840030e-02 4.48702425e-02
-3.74701768e-01 1.18990040e+00 -3.82589340e-01 -4.98083413e-01
8.41167644e-02 -6.67995453e-01 -7.28475869e-01 2.72006720e-01
2.45825008e-01 -2.29460850e-01 -8.23423743e-01 1.23099554e+00
1.87940374e-01 4.25082624e-01 2.35555366e-01 -1.37493205e+00
-7.48481452e-01 5.83484709e-01 -6.47674561e-01 3.38987648e-01
7.98917830e-01 8.38851184e-02 1.11386049e+00 -7.79921651e-01
7.00485826e-01 1.41987681e+00 2.95641631e-01 1.19169295e-01
-7.75190532e-01 -5.62947989e-01 3.99997026e-01 6.95917189e-01
-1.41735768e+00 -2.24887252e-01 1.08825803e+00 -3.21482301e-01
7.75777876e-01 7.08103925e-02 8.39153528e-01 8.66594791e-01
-1.06876165e-01 8.86430204e-01 6.58328652e-01 -9.15189162e-02
-4.62287245e-03 -2.53152937e-01 1.27098903e-01 1.10741639e+00
1.21556565e-01 -4.63003576e-01 -6.63034141e-01 1.49270013e-01
6.87689900e-01 4.37611133e-01 -5.47956228e-01 -5.97505391e-01
-1.36096573e+00 6.91489279e-01 8.71479928e-01 3.43136370e-01
-4.12014246e-01 1.86204284e-01 6.91810310e-01 1.45716861e-01
5.01011431e-01 2.58092815e-03 1.39066890e-01 9.56570059e-02
-7.41097033e-01 1.58365127e-02 4.66542274e-01 9.63576555e-01
7.14018583e-01 -1.29061759e-01 -6.29421175e-01 9.01426256e-01
2.37263829e-01 2.62114972e-01 3.27542603e-01 -9.05079484e-01
6.41243458e-01 1.17670047e+00 -4.22347635e-01 -1.79098904e+00
-6.64921105e-02 -2.68253446e-01 -9.55893397e-01 -3.58639061e-01
-7.27339238e-02 2.95260072e-01 -5.32094896e-01 1.42500615e+00
2.37064928e-01 6.07196569e-01 7.08847716e-02 1.21584237e+00
1.11388063e+00 6.74941540e-01 1.72767848e-01 -1.54687956e-01
1.48906422e+00 -1.17112029e+00 -7.81612396e-01 7.67933726e-02
4.45959181e-01 -4.98385936e-01 7.60289907e-01 -6.43009320e-02
-1.09733033e+00 -6.37554407e-01 -9.45479751e-01 -1.43551707e-01
-3.83594871e-01 -1.10451132e-01 5.88685572e-01 3.15723047e-02
-8.91967952e-01 4.65461880e-01 -6.36867642e-01 -4.11862552e-01
7.73041129e-01 2.05595139e-02 -4.04957920e-01 -3.87789160e-01
-1.40567541e+00 3.78071666e-01 5.95545650e-01 6.41222596e-02
-5.25189996e-01 -4.37095642e-01 -1.01102781e+00 2.45089307e-01
8.02347124e-01 -8.24937582e-01 6.44597292e-01 -1.00794911e+00
-1.16549134e+00 5.62040687e-01 -1.90965533e-01 -3.72327983e-01
4.14875329e-01 -9.57832560e-02 -5.48355162e-01 7.85242915e-01
2.01129213e-01 5.61455965e-01 7.48671055e-01 -1.03360033e+00
-5.08562088e-01 -2.93077886e-01 6.56988859e-01 3.40512097e-01
-6.43725872e-01 -6.49990607e-03 -1.12430668e+00 -8.26860070e-01
1.67072415e-01 -6.16153777e-01 1.82161443e-02 1.55157283e-01
-3.33010495e-01 -4.26180482e-01 9.98997629e-01 -7.07705855e-01
1.28197718e+00 -2.18784738e+00 3.33895415e-01 2.60139823e-01
4.64277685e-01 2.67908543e-01 -1.94340929e-01 3.37600142e-01
1.95079332e-03 7.64828548e-02 -6.98633417e-02 7.01048151e-02
-1.68875292e-01 8.05333033e-02 -2.17524260e-01 3.21886510e-01
3.51061463e-01 1.11811173e+00 -1.16289699e+00 -8.15748453e-01
3.09306741e-01 6.09648049e-01 -4.80747372e-01 1.77832037e-01
-3.15100104e-01 2.27447927e-01 -1.00131404e+00 6.68984294e-01
5.39820254e-01 -4.79085475e-01 2.52630055e-01 -5.92770219e-01
2.04428852e-01 -1.32642359e-01 -7.74343491e-01 2.01075006e+00
-2.50950634e-01 6.61383212e-01 -3.19477797e-01 -1.34634960e+00
8.30118895e-01 1.28093027e-02 6.62080228e-01 -9.03326511e-01
2.14184910e-01 -2.16019645e-01 -3.10543329e-01 -7.26264715e-01
4.16372001e-01 2.82670766e-01 8.84580538e-02 3.03543564e-02
-7.14783743e-02 1.92867488e-01 3.27401131e-01 6.92304909e-01
1.10580266e+00 4.16582793e-01 1.95438579e-01 -7.60996118e-02
1.03504848e+00 -5.73179089e-02 5.52983046e-01 3.26470107e-01
-1.33130491e-01 5.10051370e-01 6.06741428e-01 -3.61722022e-01
-6.08179033e-01 -8.71460259e-01 3.38623464e-01 8.66364121e-01
7.85037935e-01 -8.24169874e-01 -8.00548375e-01 -7.26361871e-01
-7.55189359e-02 2.45118216e-01 -6.73513174e-01 -3.45869988e-01
-5.29925108e-01 -1.87971443e-01 3.33896041e-01 5.58176100e-01
1.07923925e+00 -9.50618923e-01 -4.07094270e-01 1.58141814e-02
-5.51496863e-01 -1.41645753e+00 -4.59249496e-01 -6.76664174e-01
-5.93498766e-01 -1.33575833e+00 -8.40147853e-01 -7.85163403e-01
6.42428458e-01 8.14545214e-01 9.92584050e-01 5.19747496e-01
-2.86375135e-01 5.46073616e-01 -7.48550713e-01 1.94792315e-01
4.41358946e-02 -2.19353899e-01 -4.21678424e-01 3.31851929e-01
3.08996588e-01 -2.56670594e-01 -9.17027652e-01 2.16649100e-01
-1.04177105e+00 2.57586032e-01 3.94614041e-01 6.44136071e-01
5.78238249e-01 1.46599799e-01 3.79505038e-01 -7.14514136e-01
5.40214896e-01 -6.94293141e-01 -4.19371068e-01 5.14613330e-01
-5.85821345e-02 5.54489121e-02 6.47893965e-01 -4.27114591e-02
-1.14448440e+00 -2.52882242e-01 3.31770808e-01 -8.10117781e-01
4.27591987e-02 6.00483835e-01 -2.56267518e-01 -2.40798499e-02
8.26407373e-02 2.53957927e-01 9.51674022e-03 -3.81648093e-02
4.88228023e-01 3.90354604e-01 4.20196623e-01 -4.92897123e-01
7.66719997e-01 6.86024904e-01 2.37681314e-01 -5.85738063e-01
-9.73569095e-01 -5.37547708e-01 -3.42500716e-01 -4.68612254e-01
1.18131089e+00 -1.03944468e+00 -7.33667016e-01 1.12032458e-01
-1.12266827e+00 2.52722412e-01 -9.22381878e-04 4.22652066e-01
-4.55823660e-01 8.36200416e-01 -5.29073834e-01 -3.89657438e-01
-2.95729131e-01 -1.26325226e+00 1.08156097e+00 2.86020160e-01
1.86968580e-01 -9.34294224e-01 -2.81920344e-01 6.27646863e-01
1.93478778e-01 3.88418496e-01 8.89847517e-01 -2.66277105e-01
-8.67455244e-01 -8.98947194e-03 -8.20713222e-01 1.35101974e-01
1.27545580e-01 -1.63426585e-02 -5.44284344e-01 -1.14265956e-01
-3.43307495e-01 -1.18460201e-01 1.13405299e+00 6.46760762e-02
1.51524913e+00 -1.07902028e-01 -5.27119339e-01 5.17131269e-01
1.35616064e+00 2.98522282e-02 5.64650476e-01 3.45673054e-01
9.75622237e-01 5.54290831e-01 7.41408408e-01 4.98858958e-01
5.26328385e-01 7.04437256e-01 4.54851449e-01 -1.99802201e-02
-3.22604060e-01 -3.46715689e-01 1.77414551e-01 9.24397290e-01
-3.03869486e-01 -2.78558701e-01 -7.43211031e-01 4.25892800e-01
-2.27767801e+00 -1.12110281e+00 -1.05436608e-01 1.81108701e+00
4.02394265e-01 1.10103153e-02 -1.02313794e-01 -6.61909878e-02
9.72422719e-01 4.87217665e-01 -3.84674788e-01 2.11730227e-01
-1.86914444e-01 -1.47594765e-01 1.90693319e-01 -3.07756439e-02
-1.03349972e+00 1.09157979e+00 4.43618441e+00 9.03116465e-01
-7.66292274e-01 -2.58176234e-02 4.93944079e-01 3.40657681e-02
-2.58983284e-01 1.11141913e-01 -2.15948761e-01 4.43900198e-01
3.48714441e-01 -1.98833346e-01 4.68766570e-01 5.64552188e-01
1.48817614e-01 1.25565454e-02 -7.76590168e-01 1.22578275e+00
4.68141049e-01 -1.47888565e+00 5.70488751e-01 -3.10830235e-01
7.18538344e-01 -4.30027038e-01 -1.63108826e-01 2.10922807e-01
2.80151609e-02 -6.63145959e-01 6.26996398e-01 8.15862298e-01
6.16222858e-01 -8.55644822e-01 7.44212270e-01 -1.52432740e-01
-1.77540445e+00 1.45432595e-02 -4.94261175e-01 1.68946758e-01
1.56594232e-01 5.70284486e-01 -3.03185552e-01 1.13400209e+00
7.53786087e-01 1.44561982e+00 -7.17896402e-01 9.03970957e-01
-2.36241609e-01 3.29264790e-01 8.50848779e-02 -9.00082104e-03
3.33552539e-01 -2.94990748e-01 3.53336066e-01 1.14552927e+00
2.14293525e-01 3.20266634e-01 3.13673735e-01 8.19444835e-01
-3.45153213e-01 2.73193687e-01 -6.46142125e-01 -9.08782855e-02
2.77194589e-01 1.23710310e+00 -9.50623751e-01 -5.40367603e-01
-8.76613140e-01 1.18820703e+00 3.96798939e-01 5.55357754e-01
-1.08475471e+00 -3.64362955e-01 5.40685296e-01 -1.01977199e-01
2.48428136e-01 -5.55748269e-02 2.64307261e-01 -1.52742159e+00
1.57112315e-01 -5.90998709e-01 4.59259987e-01 -8.75665724e-01
-1.19866264e+00 4.91108924e-01 1.31603241e-01 -1.35000408e+00
6.34010509e-02 -2.93465227e-01 -3.70400429e-01 3.86387408e-01
-1.68568420e+00 -1.15864420e+00 -8.79142284e-01 9.03712451e-01
8.00807595e-01 -2.04695627e-01 3.52167130e-01 3.34825605e-01
-5.92946708e-01 3.44893336e-01 -1.79303870e-01 4.94504571e-01
4.06085938e-01 -7.35133410e-01 1.35230929e-01 7.88279712e-01
1.78469673e-01 6.36921346e-01 1.83606416e-01 -6.44217968e-01
-1.59158981e+00 -1.36876667e+00 4.95467693e-01 1.40587529e-02
7.89597332e-01 -1.52156442e-01 -9.70543206e-01 3.48258138e-01
3.23554993e-01 4.56692517e-01 4.75208521e-01 -3.45349818e-01
-5.47461033e-01 1.17036272e-02 -7.58115709e-01 8.33091438e-01
1.59930301e+00 -7.09054351e-01 -6.70629501e-01 3.31642836e-01
9.12813187e-01 -1.03855669e-01 -7.70708799e-01 4.41571593e-01
3.90193433e-01 -9.97745335e-01 1.08844960e+00 -4.85619277e-01
7.14499056e-01 -4.41724867e-01 -1.06654227e-01 -1.02782393e+00
-2.87957758e-01 -2.47324333e-01 -6.54876083e-02 1.31393933e+00
-2.18768343e-01 -3.34908903e-01 5.40137768e-01 1.73240870e-01
6.42237514e-02 -6.78476870e-01 -4.16381806e-01 -6.30965590e-01
-4.55650419e-01 -2.60719657e-01 6.20297670e-01 1.04830945e+00
8.23246315e-02 3.73185992e-01 -1.33168548e-01 -4.77444977e-02
5.73547602e-01 4.10030752e-01 5.92848539e-01 -1.10638368e+00
-5.71476184e-02 -5.84867418e-01 -9.24490690e-01 -1.11308789e+00
6.27855897e-01 -1.01047754e+00 -1.64927498e-01 -1.76932216e+00
5.13116658e-01 -7.66879097e-02 -5.62793255e-01 4.15582508e-01
-4.24305230e-01 3.78659487e-01 4.66766506e-01 2.27782935e-01
-1.30740535e+00 7.40906835e-01 1.36918628e+00 -3.68355244e-01
2.36805588e-01 -5.93631983e-01 -4.35134798e-01 7.24117219e-01
6.53368711e-01 -2.94201914e-02 -7.70345628e-01 -4.40353096e-01
2.21017569e-01 2.34532818e-01 6.83660567e-01 -1.11622000e+00
2.37867132e-01 -1.54682085e-01 2.84348756e-01 -4.01646733e-01
2.41357312e-01 -9.18232262e-01 -4.82016429e-03 1.93682343e-01
-4.33949292e-01 2.72183539e-03 -1.53204009e-01 9.85692441e-01
-6.74729168e-01 1.30096689e-01 4.30288404e-01 -1.39743656e-01
-1.14571595e+00 5.17251372e-01 6.62089735e-02 1.26731277e-01
1.23870885e+00 -2.18242690e-01 -3.35835725e-01 -3.96628767e-01
-4.12960738e-01 3.15856427e-01 4.10262138e-01 7.44353056e-01
9.66053069e-01 -1.46772802e+00 -5.06973863e-01 7.18313977e-02
4.45276350e-01 -1.28269285e-01 4.53733742e-01 7.29973912e-01
-5.08469820e-01 4.57723916e-01 -1.78348422e-01 -5.50965309e-01
-1.32919276e+00 8.00594687e-01 3.48824896e-02 7.07631698e-03
-8.55455935e-01 5.91008067e-01 3.69140863e-01 2.62023449e-01
1.06529310e-01 -2.70437956e-01 -5.25592446e-01 1.13436565e-01
4.12536532e-01 2.63683617e-01 -2.68401504e-01 -9.45186615e-01
-4.49355155e-01 6.85128450e-01 8.20297971e-02 3.99274051e-01
1.10319889e+00 -2.26832375e-01 -1.89486459e-01 1.11182377e-01
1.41835093e+00 -2.06617713e-01 -1.08551693e+00 -5.13424695e-01
-1.73857838e-01 -7.18404830e-01 4.79115061e-02 -2.15406969e-01
-1.55032349e+00 8.78830969e-01 2.71008879e-01 1.76965445e-01
1.18105233e+00 7.15440810e-02 8.74192655e-01 2.78322399e-01
3.03457171e-01 -8.34842503e-01 4.88769501e-01 2.70302325e-01
7.05664158e-01 -1.15685380e+00 6.78227693e-02 -8.11502159e-01
-5.64208925e-01 1.15647471e+00 6.28531933e-01 -3.08446318e-01
4.94898111e-01 -2.76033103e-01 -3.76476556e-01 -3.21422756e-01
-4.78411943e-01 -4.31970328e-01 5.36161363e-01 3.59972596e-01
3.51166487e-01 -9.85811055e-02 -4.35525656e-01 4.16539431e-01
3.63409549e-01 1.60907939e-01 3.43282931e-02 8.42531025e-01
-3.21348637e-01 -7.62566090e-01 -2.41039656e-02 4.09390867e-01
-3.21060151e-01 -2.35193938e-01 -2.27266237e-01 6.08056843e-01
-1.87358968e-02 1.00511718e+00 5.42917065e-02 -3.75964046e-01
2.13779852e-01 -1.82149589e-01 2.89116621e-01 -3.15249562e-01
-4.19616699e-01 -1.18061170e-01 -1.00638434e-01 -9.31597054e-01
-9.16859448e-01 -4.32015985e-01 -1.59263837e+00 -8.73107314e-02
-1.08871594e-01 1.52373597e-01 1.61680490e-01 9.77427959e-01
5.01630604e-01 7.77346492e-01 4.94825274e-01 -5.21781087e-01
-2.07194760e-02 -4.86492515e-01 -3.38855326e-01 9.15893018e-01
8.38153064e-02 -8.84615660e-01 -7.93552995e-02 1.22081138e-01] | [10.1245756149292, 0.9215598106384277] |
e4eb5396-7b43-4e6d-b6d6-67129f328552 | contextual-guided-segmentation-framework-for | 2106.03330 | null | https://arxiv.org/abs/2106.03330v2 | https://arxiv.org/pdf/2106.03330v2.pdf | Contextual Guided Segmentation Framework for Semi-supervised Video Instance Segmentation | In this paper, we propose Contextual Guided Segmentation (CGS) framework for video instance segmentation in three passes. In the first pass, i.e., preview segmentation, we propose Instance Re-Identification Flow to estimate main properties of each instance (i.e., human/non-human, rigid/deformable, known/unknown category) by propagating its preview mask to other frames. In the second pass, i.e., contextual segmentation, we introduce multiple contextual segmentation schemes. For human instance, we develop skeleton-guided segmentation in a frame along with object flow to correct and refine the result across frames. For non-human instance, if the instance has a wide variation in appearance and belongs to known categories (which can be inferred from the initial mask), we adopt instance segmentation. If the non-human instance is nearly rigid, we train FCNs on synthesized images from the first frame of a video sequence. In the final pass, i.e., guided segmentation, we develop a novel fined-grained segmentation method on non-rectangular regions of interest (ROIs). The natural-shaped ROI is generated by applying guided attention from the neighbor frames of the current one to reduce the ambiguity in the segmentation of different overlapping instances. Forward mask propagation is followed by backward mask propagation to further restore missing instance fragments due to re-appeared instances, fast motion, occlusion, or heavy deformation. Finally, instances in each frame are merged based on their depth values, together with human and non-human object interaction and rare instance priority. Experiments conducted on the DAVIS Test-Challenge dataset demonstrate the effectiveness of our proposed framework. We achieved the 3rd consistently in the DAVIS Challenges 2017-2019 with 75.4%, 72.4%, and 78.4% in terms of global score, region similarity, and contour accuracy, respectively. | ['Minh-Triet Tran', 'Tam V. Nguyen', 'Trung-Nghia Le'] | 2021-06-07 | null | null | null | null | ['video-instance-segmentation'] | ['computer-vision'] | [ 5.14078021e-01 9.39134955e-02 -4.03703935e-02 -2.72267938e-01
-6.94450855e-01 -5.62015831e-01 2.64282763e-01 -1.60550550e-01
-3.23178947e-01 7.09164262e-01 -4.82562967e-02 1.16373911e-01
4.93982658e-02 -5.44897437e-01 -8.71992052e-01 -7.15749800e-01
3.41251135e-01 5.38298190e-01 9.01188910e-01 2.01699585e-01
2.31122524e-01 5.56529284e-01 -1.38445270e+00 4.07445908e-01
9.18532968e-01 9.59967554e-01 2.75857866e-01 5.12891710e-01
-5.20801432e-02 4.23639476e-01 -6.31473660e-01 -2.55666375e-01
3.78872365e-01 -4.33635354e-01 -1.00532413e+00 7.09857523e-01
5.48667133e-01 -5.89349568e-01 4.09694165e-02 1.04643321e+00
7.46159479e-02 3.42740595e-01 5.99167049e-01 -1.19463265e+00
-9.41648781e-02 3.41207415e-01 -1.10316718e+00 1.46476582e-01
2.61529177e-01 2.68352002e-01 4.65241045e-01 -9.34035659e-01
9.50718880e-01 1.18716598e+00 3.14828604e-01 5.21809220e-01
-9.80785191e-01 -6.08925521e-01 6.92989469e-01 4.13922183e-02
-1.28212821e+00 -2.30383202e-01 7.58606672e-01 -6.25570595e-01
3.10673773e-01 2.05036655e-01 6.25310898e-01 6.12332702e-01
-3.27521265e-02 8.96897316e-01 8.44064355e-01 -1.63755044e-01
1.57985359e-01 -7.52466843e-02 1.48007333e-01 7.73708463e-01
6.83763996e-02 -3.49144568e-03 3.32858115e-02 2.98919559e-01
8.10632229e-01 -1.67197138e-02 -3.91128182e-01 -3.00028652e-01
-1.32056057e+00 2.78097779e-01 2.25781485e-01 1.85830787e-01
-4.55540150e-01 -3.74527544e-01 2.54434943e-01 -2.85716712e-01
3.63433182e-01 -9.89856794e-02 -5.73003590e-01 2.22711131e-01
-1.27822685e+00 2.46369600e-01 4.03434098e-01 1.16010821e+00
9.14316297e-01 -1.02374822e-01 -4.82050836e-01 7.93639004e-01
1.03070199e-01 1.94285125e-01 3.02637815e-01 -1.15198135e+00
3.69961590e-01 6.99368834e-01 1.85545057e-01 -9.88748908e-01
-2.89043427e-01 -3.99680227e-01 -9.33556855e-01 7.06668049e-02
4.79310334e-01 -1.00648209e-01 -1.45584345e+00 1.45460820e+00
1.00177753e+00 7.11207032e-01 -5.50394095e-02 1.25353789e+00
1.01369572e+00 6.08513772e-01 1.79731116e-01 -5.16042769e-01
1.46735823e+00 -1.29832673e+00 -4.89653647e-01 -1.27796531e-01
1.31587312e-01 -7.19169199e-01 7.21922398e-01 3.36212307e-01
-1.19718802e+00 -8.42608392e-01 -6.82938814e-01 6.16861619e-02
9.70097333e-02 2.04219326e-01 5.88828772e-02 3.00674319e-01
-6.04206145e-01 3.87361318e-01 -8.62899125e-01 -1.35411337e-01
5.23808241e-01 1.77932292e-01 -3.48230541e-01 -2.39094958e-01
-8.68490398e-01 2.48429105e-01 5.50711691e-01 1.73006445e-01
-1.13919401e+00 -8.45393121e-01 -6.94469333e-01 -1.82042852e-01
7.74710774e-01 -6.37237251e-01 7.81488121e-01 -1.17325723e+00
-1.12524903e+00 8.54713798e-01 -3.37452292e-01 -2.83684939e-01
7.02648640e-01 -2.21833289e-01 -1.57699376e-01 3.29603672e-01
2.68804014e-01 8.53228152e-01 1.01123095e+00 -1.53058743e+00
-1.03079069e+00 -3.38355124e-01 1.08206980e-01 2.55057365e-01
3.75394195e-01 3.67293917e-02 -1.12487578e+00 -6.08597636e-01
3.29731047e-01 -1.04530776e+00 -1.92727268e-01 -6.86992481e-02
-7.01147854e-01 -1.32056713e-01 1.10392785e+00 -1.10667551e+00
1.23705494e+00 -2.27128673e+00 3.35407823e-01 2.63656259e-01
7.23280534e-02 3.04300249e-01 -3.10731996e-02 -4.29068118e-01
-2.49222480e-02 2.46882409e-01 -7.65524924e-01 -2.52086669e-01
-4.11160767e-01 3.04316282e-01 -3.15035731e-02 2.76886880e-01
2.85989702e-01 5.75047553e-01 -7.64235139e-01 -8.65779579e-01
4.14191335e-01 3.83798808e-01 -5.62939465e-01 3.40271354e-01
-2.71083117e-01 9.39141750e-01 -4.83423769e-01 7.79796422e-01
9.42147672e-01 1.09582944e-02 -1.48834273e-01 -6.44101739e-01
-1.09456711e-01 -3.66434544e-01 -1.73884070e+00 1.61401200e+00
-4.62244526e-02 6.82248846e-02 2.29426965e-01 -1.02063227e+00
5.79858303e-01 2.03076914e-01 6.33495510e-01 -2.79574275e-01
3.38008329e-02 1.79851633e-02 -1.65790781e-01 -6.35948896e-01
2.88861036e-01 2.67285913e-01 2.74440318e-01 1.71065807e-01
3.57958749e-02 -7.24653080e-02 5.51447749e-01 1.30294800e-01
5.49064159e-01 6.33382976e-01 1.35644570e-01 -7.14808330e-02
9.24487114e-01 -7.63450041e-02 1.00730908e+00 5.41118443e-01
-3.28010798e-01 1.25391614e+00 3.70583177e-01 -4.01024938e-01
-7.25796103e-01 -8.97254646e-01 -4.98712398e-02 8.33226264e-01
6.16836369e-01 -1.20991811e-01 -1.10594654e+00 -8.66616011e-01
-3.83769095e-01 4.68497843e-01 -4.49252725e-01 1.10114165e-01
-8.32025945e-01 -5.79869986e-01 6.43592477e-02 4.11540091e-01
8.30321193e-01 -1.17433095e+00 -5.87938428e-01 3.47026944e-01
-4.99702722e-01 -1.42988789e+00 -8.27176809e-01 -2.89667368e-01
-8.66104662e-01 -1.29611146e+00 -9.59138751e-01 -8.12178373e-01
9.05504227e-01 9.67273787e-02 8.53026032e-01 2.67170072e-01
-2.55851239e-01 1.74560755e-01 -1.81638375e-01 1.74627885e-01
-2.71322727e-01 -1.78051651e-01 -2.40862414e-01 4.86402839e-01
-2.53944725e-01 -2.07340613e-01 -9.57119346e-01 7.02234030e-01
-1.03641260e+00 2.95919418e-01 3.55551600e-01 5.67221045e-01
1.16553557e+00 2.97483802e-01 2.77313083e-01 -8.98392081e-01
5.54307736e-02 -3.77869368e-01 -4.97090697e-01 3.49280775e-01
-5.92667684e-02 -3.77894938e-01 3.07048023e-01 -5.38352251e-01
-1.28524625e+00 4.13445383e-01 -3.19127692e-03 -6.85424089e-01
-7.00920343e-01 9.28808227e-02 -4.07371312e-01 2.41398022e-01
1.82926044e-01 2.07080856e-01 -3.03360164e-01 -2.23589122e-01
1.86066315e-01 3.35372865e-01 8.01448464e-01 -8.18664253e-01
6.64260209e-01 3.66071880e-01 -2.95316607e-01 -6.31940305e-01
-8.61221492e-01 -3.74873132e-01 -8.49410474e-01 -4.36998963e-01
1.51061153e+00 -8.15152884e-01 -2.70265222e-01 6.01141810e-01
-1.13767481e+00 -3.91427249e-01 -1.08810879e-01 4.35005844e-01
-3.44330996e-01 4.35697705e-01 -5.42935669e-01 -6.13727689e-01
-2.40061864e-01 -1.55784988e+00 1.26954710e+00 5.55450499e-01
-1.05858907e-01 -7.38232315e-01 -5.08776426e-01 5.27014554e-01
-1.27366126e-01 5.64032495e-01 6.22620940e-01 -4.64569122e-01
-7.62901127e-01 7.65580982e-02 -2.67544478e-01 4.71361548e-01
2.23313689e-01 3.81611973e-01 -7.21912324e-01 -8.71193334e-02
-1.67256311e-01 1.26408413e-01 6.54733300e-01 6.58492029e-01
1.31294000e+00 -1.92263052e-01 -4.02475595e-01 6.53352916e-01
1.07955432e+00 5.55594265e-01 5.87075174e-01 3.90342809e-02
9.67340708e-01 7.26351380e-01 9.47519004e-01 1.96718648e-01
1.62457570e-01 6.57906234e-01 2.78494149e-01 -1.74949750e-01
-3.07940364e-01 -3.72362919e-02 8.54062587e-02 3.48842412e-01
-3.72696579e-01 -1.30048230e-01 -6.09119892e-01 5.14160454e-01
-1.66581595e+00 -8.90522361e-01 -2.83324569e-01 2.10634971e+00
6.33514404e-01 2.73973852e-01 3.03172737e-01 -8.63962099e-02
1.23177099e+00 -3.18495855e-02 -6.55757725e-01 5.21544702e-02
-7.01018944e-02 6.47416264e-02 1.00653365e-01 5.75073421e-01
-1.18492639e+00 1.02618504e+00 3.93551397e+00 9.93209004e-01
-8.83113623e-01 2.79356465e-02 1.14645648e+00 1.74563408e-01
-6.79435208e-02 3.62357572e-02 -8.40930700e-01 6.73635781e-01
1.30281165e-01 2.26488829e-01 2.88938969e-01 4.96713370e-01
2.78180599e-01 -3.68822277e-01 -8.97324860e-01 6.41707540e-01
-2.36504688e-03 -1.04276645e+00 1.25622705e-01 -2.19744906e-01
1.00993299e+00 -5.39324403e-01 -2.62730747e-01 2.91405678e-01
-1.70613647e-01 -6.40724659e-01 8.89463723e-01 5.93293130e-01
7.03726888e-01 -7.56353557e-01 6.97357595e-01 5.06955564e-01
-1.51876795e+00 2.14213073e-01 -5.90731353e-02 5.51641524e-01
3.97025973e-01 6.04379237e-01 -2.00874224e-01 7.94990301e-01
7.17952251e-01 6.62295401e-01 -3.54220420e-01 1.02020717e+00
-1.60882488e-01 5.65413892e-01 -3.70903224e-01 7.65557349e-01
1.35132819e-01 -3.84212404e-01 6.93428278e-01 1.05434620e+00
1.25761166e-01 5.83081901e-01 5.47331691e-01 9.21342373e-01
1.50651112e-01 -4.64300439e-02 2.58560991e-03 5.97151697e-01
2.67056048e-01 1.42443907e+00 -1.18157268e+00 -6.72653317e-01
-2.95509100e-01 1.22884214e+00 -2.02038456e-02 6.48652017e-01
-1.10998583e+00 -3.30226459e-02 1.25640154e-01 2.16408759e-01
4.51814085e-01 1.47098422e-01 -6.32764623e-02 -1.12191045e+00
1.56177208e-01 -8.24002981e-01 5.67338824e-01 -7.33535647e-01
-9.36619878e-01 7.23716736e-01 2.60967612e-01 -1.16458666e+00
-1.16448559e-01 -9.02570635e-02 -6.59081995e-01 7.88419366e-01
-1.15016174e+00 -1.10887492e+00 -5.07840991e-01 7.32771873e-01
1.00701690e+00 2.64052510e-01 1.82081312e-01 3.70917976e-01
-7.67357349e-01 4.93953586e-01 -5.67982614e-01 2.27034420e-01
5.75359762e-01 -8.84765148e-01 1.22337453e-01 1.06136155e+00
-1.11031443e-01 4.65924025e-01 3.77429098e-01 -9.71288383e-01
-6.70608163e-01 -1.51898420e+00 4.01965946e-01 -5.42740524e-02
1.36483848e-01 -8.16406533e-02 -1.13787341e+00 5.54721057e-01
-1.54227570e-01 3.18441540e-01 1.87898614e-02 -5.55311620e-01
1.93119213e-01 -1.87233314e-02 -1.38920784e+00 5.51904798e-01
1.05639267e+00 -5.95213138e-02 -4.09032166e-01 1.71383888e-01
8.38723779e-01 -9.29163396e-01 -8.55193377e-01 8.60835075e-01
4.29058433e-01 -9.09216881e-01 1.05673468e+00 -2.98756570e-01
6.52569115e-01 -8.11819375e-01 4.35272139e-03 -8.39425862e-01
-1.68877006e-01 -4.98555124e-01 -3.93845551e-02 1.53661430e+00
1.53611794e-01 -8.84858966e-02 7.84707129e-01 7.15474367e-01
-3.93651754e-01 -8.95729423e-01 -7.34143257e-01 -3.66361588e-01
-1.32258520e-01 -4.51688617e-01 5.35083473e-01 8.74457359e-01
-7.23388612e-01 -3.95557769e-02 -2.57573158e-01 5.00535607e-01
7.33247995e-01 2.34463856e-01 7.83335090e-01 -9.45389867e-01
-2.06884637e-01 -2.52608538e-01 -1.83156073e-01 -1.13452816e+00
1.77092478e-01 -6.33962035e-01 2.53601633e-02 -1.50481939e+00
3.27019006e-01 -3.78790110e-01 -2.50134081e-01 4.28770751e-01
-6.35467708e-01 4.92849290e-01 2.76767403e-01 2.12675750e-01
-6.23190165e-01 1.16547413e-01 1.61885548e+00 -1.54240191e-01
-4.50650454e-01 1.87627077e-01 -2.76072145e-01 9.41683412e-01
6.32272184e-01 -2.97040761e-01 -2.84760594e-01 -2.17506409e-01
-5.10200679e-01 5.35187900e-01 6.00805223e-01 -1.02184570e+00
1.10742986e-01 -2.65118033e-01 4.92630035e-01 -8.49675894e-01
1.32540837e-01 -8.55295122e-01 3.51504922e-01 3.43934327e-01
-1.70872554e-01 -2.29402006e-01 1.44643128e-01 6.00077748e-01
-2.38319516e-01 -2.49234930e-01 9.03188348e-01 -2.05001667e-01
-8.66914988e-01 5.66781878e-01 -7.85717145e-02 3.61183345e-01
1.26632822e+00 -6.31637514e-01 9.47738811e-02 -4.83425567e-03
-1.19653964e+00 3.58764648e-01 3.59632224e-01 3.97651285e-01
7.46078789e-01 -9.95533645e-01 -8.33588719e-01 3.19720596e-01
-6.37876689e-02 6.16354167e-01 7.84652531e-01 1.03831553e+00
-3.04932773e-01 -2.12178573e-01 3.13150212e-02 -9.45460975e-01
-1.34696281e+00 4.83676553e-01 5.44228852e-01 -1.28481343e-01
-7.70712554e-01 6.88320935e-01 8.70375276e-01 -1.78656951e-01
1.54725134e-01 -4.46040750e-01 -2.65499860e-01 1.48893259e-02
3.83912265e-01 3.10183436e-01 -8.42486024e-02 -9.80962694e-01
-5.11460125e-01 1.01367939e+00 -1.39237553e-01 2.56814007e-02
8.85154843e-01 -2.19305590e-01 -1.29531967e-02 1.57240316e-01
1.03368735e+00 -3.57672088e-02 -1.59702849e+00 -4.47907969e-02
-3.40794235e-01 -4.48847115e-01 -2.73671210e-01 -8.17498505e-01
-1.62806106e+00 5.12697399e-01 5.84187388e-01 -2.33105212e-01
1.18768609e+00 7.90688470e-02 8.26918781e-01 -4.51037496e-01
2.21480563e-01 -9.56801534e-01 -2.35762000e-01 3.20565820e-01
8.00613105e-01 -1.10975206e+00 1.56208258e-02 -6.47355080e-01
-7.49518096e-01 1.01147103e+00 9.61211920e-01 -2.56807934e-02
4.34726238e-01 2.07852289e-01 -2.10755363e-01 2.26176772e-02
-2.68759251e-01 -2.82555483e-02 5.80523789e-01 4.95948017e-01
5.52951731e-02 -3.21608922e-03 -3.51542771e-01 6.16608143e-01
1.15175813e-01 -4.07493860e-02 4.42289740e-01 6.61391675e-01
-3.71902198e-01 -4.63761032e-01 -5.67301333e-01 3.06768119e-01
-5.41759133e-01 1.63662061e-01 -1.05191626e-01 8.53505611e-01
6.64087117e-01 9.72423375e-01 2.17791781e-01 -1.43569767e-01
4.52114493e-01 -1.58295751e-01 2.71704406e-01 -4.78884667e-01
-5.54481149e-01 5.61682522e-01 -5.23659959e-02 -6.70439720e-01
-6.25194013e-01 -7.83404589e-01 -1.69326532e+00 1.14113823e-01
-3.12212795e-01 3.15740146e-02 2.81756431e-01 1.02472782e+00
1.01190269e-01 6.67306483e-01 4.15764272e-01 -9.98605669e-01
1.22149527e-01 -6.56974971e-01 -3.27301353e-01 7.40703821e-01
2.34476462e-01 -6.48439407e-01 -4.30617809e-01 6.07538164e-01] | [9.201553344726562, -0.20248228311538696] |
825aaff8-07c9-4176-ad2f-8f54618789d1 | selectively-hard-negative-mining-for | 2303.00181 | null | https://arxiv.org/abs/2303.00181v1 | https://arxiv.org/pdf/2303.00181v1.pdf | Selectively Hard Negative Mining for Alleviating Gradient Vanishing in Image-Text Matching | Recently, a series of Image-Text Matching (ITM) methods achieve impressive performance. However, we observe that most existing ITM models suffer from gradients vanishing at the beginning of training, which makes these models prone to falling into local minima. Most ITM models adopt triplet loss with Hard Negative mining (HN) as the optimization objective. We find that optimizing an ITM model using only the hard negative samples can easily lead to gradient vanishing. In this paper, we derive the condition under which the gradient vanishes during training. When the difference between the positive pair similarity and the negative pair similarity is close to 0, the gradients on both the image and text encoders will approach 0. To alleviate the gradient vanishing problem, we propose a Selectively Hard Negative Mining (SelHN) strategy, which chooses whether to mine hard negative samples according to the gradient vanishing condition. SelHN can be plug-and-play applied to existing ITM models to give them better training behavior. To further ensure the back-propagation of gradients, we construct a Residual Visual Semantic Embedding model with SelHN, denoted as RVSE++. Extensive experiments on two ITM benchmarks demonstrate the strength of RVSE++, achieving state-of-the-art performance. | ['Zhongtian Du', 'Zerun Feng', 'Xin Wang', 'Caili Guo', 'Zheng Li'] | 2023-03-01 | null | null | null | null | ['text-matching'] | ['natural-language-processing'] | [ 2.38130823e-01 -3.77782690e-03 -4.61007357e-01 -5.61791539e-01
-4.29159850e-01 -4.62317979e-03 5.02542317e-01 -1.23669868e-02
-4.08252001e-01 1.92764342e-01 -1.38043523e-01 -2.96876818e-01
6.83345348e-02 -7.57632852e-01 -8.18589687e-01 -6.75413072e-01
1.90813377e-01 1.58204675e-01 2.85397828e-01 -8.79876167e-02
4.16429490e-01 -1.01860560e-01 -1.36845446e+00 4.03101027e-01
1.01770020e+00 1.34247947e+00 5.47316611e-01 1.55316532e-01
-4.35913622e-01 7.85054684e-01 -3.35088938e-01 -5.72385132e-01
6.66746199e-01 -3.68983835e-01 -5.37990272e-01 1.70897990e-02
5.51992238e-01 -2.05578163e-01 -2.84161270e-01 1.39015079e+00
3.71429503e-01 2.69980244e-02 4.99205530e-01 -1.60236716e+00
-7.58693039e-01 2.78837651e-01 -8.77018869e-01 6.78666905e-02
6.51734918e-02 9.24288705e-02 1.11222792e+00 -1.31276870e+00
6.13435149e-01 1.21082437e+00 7.28372574e-01 5.19379854e-01
-1.10885394e+00 -8.00924480e-01 3.56759071e-01 3.87710154e-01
-1.40424681e+00 -2.67321765e-01 9.50167358e-01 -1.92197755e-01
8.56933594e-01 1.61907122e-01 5.84907055e-01 5.23952782e-01
1.30976975e-01 1.18417394e+00 8.70831490e-01 -4.60800469e-01
8.91904458e-02 4.86293167e-01 1.24371322e-02 1.02365339e+00
-2.72347014e-02 -1.40655398e-01 -7.51341462e-01 -8.80046748e-03
5.27856588e-01 1.49222568e-01 -3.44431669e-01 -5.60191154e-01
-8.18496168e-01 8.74401450e-01 7.85124481e-01 1.94661766e-01
-2.58047998e-01 -1.29185738e-02 2.76693046e-01 6.91863954e-01
4.27373111e-01 2.22200140e-01 -1.40867010e-01 1.03354178e-01
-8.77476990e-01 -6.77931011e-02 2.68077046e-01 8.62150252e-01
1.11747348e+00 -3.11986625e-01 -2.01238558e-01 1.21798337e+00
3.45436662e-01 2.14317068e-01 6.92341328e-01 -7.34611630e-01
8.89482617e-01 1.04844820e+00 -1.00659728e-01 -1.18838573e+00
5.43509014e-02 -3.92213762e-01 -8.19441557e-01 1.42713264e-01
1.81856051e-01 2.39901066e-01 -9.47692633e-01 1.73630905e+00
3.01856071e-01 1.57752380e-01 -1.72117099e-01 1.16702986e+00
6.37111783e-01 7.45870769e-01 -4.45802808e-02 -7.63392374e-02
8.40219080e-01 -1.11371601e+00 -5.66015244e-01 -4.59694326e-01
7.85918653e-01 -7.20302045e-01 1.32025731e+00 1.16105929e-01
-1.07639241e+00 -5.01183093e-01 -1.11845338e+00 -4.51742969e-02
-1.99853390e-01 5.22539280e-02 4.00651962e-01 2.14840963e-01
-9.79751647e-01 7.84651577e-01 -6.55743301e-01 -7.85601735e-02
4.80459571e-01 3.81797850e-01 -2.04644248e-01 -2.05000430e-01
-1.08012199e+00 7.40876257e-01 1.48705646e-01 2.17970923e-01
-5.31037688e-01 -6.93396986e-01 -7.45207250e-01 -4.40307967e-02
3.17043096e-01 -5.22316992e-01 8.94948721e-01 -1.32064211e+00
-1.17052102e+00 8.78886819e-01 -3.57488334e-01 -3.28156322e-01
7.46538520e-01 -8.98070037e-02 -1.05686225e-01 1.59979053e-03
1.97452962e-01 9.79377687e-01 1.23115110e+00 -1.27409053e+00
-5.96732438e-01 -3.52467984e-01 -1.04066707e-01 2.72719383e-01
-1.00628293e+00 -1.70108452e-01 -7.60888815e-01 -4.57878679e-01
3.23915809e-01 -7.21477509e-01 -2.58447766e-01 5.19034684e-01
-2.91238964e-01 -3.97165865e-01 8.19876313e-01 -4.24426436e-01
1.44103134e+00 -2.30361319e+00 3.51209119e-02 4.39134955e-01
1.34946808e-01 4.58671242e-01 -4.26792979e-01 1.62998423e-01
-1.98811945e-02 -4.41060439e-02 -2.55461425e-01 -5.77162802e-01
-3.76539938e-02 2.68956214e-01 -1.64814562e-01 3.90428692e-01
4.06207919e-01 9.09102738e-01 -8.63579273e-01 -8.66555333e-01
2.84034342e-01 2.43021518e-01 -7.11016178e-01 2.45506898e-01
-7.52413943e-02 -1.18298396e-01 -4.30020690e-01 5.51712871e-01
8.67695212e-01 -6.13688052e-01 2.15805411e-01 -7.26978853e-02
-1.42465502e-01 1.38375968e-01 -1.14738142e+00 1.32206464e+00
-4.68535870e-01 5.97624660e-01 5.92543483e-02 -1.21729088e+00
9.57553864e-01 -1.11036701e-02 4.00947571e-01 -1.09404230e+00
1.19326033e-01 3.45133752e-01 -1.32537231e-01 -2.89453238e-01
4.06364411e-01 -4.20784801e-02 5.30586004e-01 1.02803797e-01
-1.91040784e-01 2.73483336e-01 4.49930318e-02 2.33307853e-01
7.57347882e-01 -5.08107767e-02 -2.48370953e-02 -1.93672031e-01
7.42279291e-01 -2.53996015e-01 7.78313160e-01 7.29572177e-01
-3.27685177e-01 5.95309556e-01 4.64063197e-01 -2.73474902e-01
-8.73922646e-01 -9.51197088e-01 -6.99138716e-02 1.02074373e+00
5.77486813e-01 -5.42104661e-01 -6.40070319e-01 -8.60006630e-01
3.04563083e-02 2.04360157e-01 -6.27990127e-01 -5.30212045e-01
-5.82387149e-01 -8.04742396e-01 -1.08573847e-01 5.11219442e-01
7.01907456e-01 -1.01483643e+00 -3.88003320e-01 1.85568601e-01
-2.04850093e-01 -8.29515517e-01 -6.33096337e-01 8.02771822e-02
-8.95657897e-01 -8.88671279e-01 -8.79703701e-01 -1.13724899e+00
1.00511289e+00 7.15707719e-01 8.92707407e-01 4.51196015e-01
-1.23327903e-01 -1.03386506e-01 -2.89114147e-01 -1.41600505e-01
-2.45094895e-01 4.27691406e-03 -3.85316372e-01 2.57637471e-01
3.09148967e-01 -3.90946031e-01 -8.26191962e-01 4.77870703e-01
-8.42242181e-01 1.87617078e-01 6.01924837e-01 1.06303549e+00
7.00485289e-01 -1.23050231e-02 4.95448202e-01 -5.78242064e-01
4.57865745e-01 -3.25190634e-01 -5.05120277e-01 3.65124375e-01
-1.15076828e+00 1.25439063e-01 8.15669298e-01 -5.70370376e-01
-7.25308061e-01 -5.20975888e-02 -4.23827693e-02 -7.04732955e-01
5.37856281e-01 3.38360876e-01 -7.31576681e-02 -2.42840305e-01
1.61135793e-01 4.51816380e-01 8.20917916e-03 -3.78278941e-01
1.01027295e-01 7.23394930e-01 2.26836815e-01 -2.14037493e-01
8.44832659e-01 6.31699264e-01 -1.72789037e-01 -6.21271253e-01
-8.64369154e-01 -7.68260539e-01 -8.83782133e-02 -3.65936369e-01
5.07758498e-01 -7.56620586e-01 -5.15068233e-01 5.28683424e-01
-8.20167184e-01 -3.35425645e-01 4.45311032e-02 2.82927662e-01
-3.15488279e-01 6.52218997e-01 -3.76018047e-01 -7.84697652e-01
-5.62035441e-01 -1.02729011e+00 1.05044925e+00 1.75119042e-01
-6.67666420e-02 -9.52004433e-01 -2.08420128e-01 1.86258510e-01
4.04292792e-01 -3.85248184e-01 1.01430154e+00 -2.23148674e-01
-4.89111751e-01 -1.00438304e-01 -4.83682215e-01 4.45935339e-01
-1.61203608e-01 -2.47553542e-01 -6.63383424e-01 -4.15975720e-01
-1.77206248e-02 -2.57699132e-01 1.24379361e+00 2.11719945e-01
1.28965580e+00 -3.35792750e-01 -3.78831238e-01 7.92061269e-01
1.55550528e+00 4.72008660e-02 7.77949929e-01 7.38853693e-01
6.69525564e-01 4.86721575e-01 7.89178491e-01 4.83063668e-01
3.06112885e-01 6.98187590e-01 5.69110930e-01 -2.66198277e-01
-5.68106286e-02 -5.60767531e-01 5.08193254e-01 8.49080324e-01
3.71378511e-01 -1.03907160e-01 -4.89459246e-01 5.56687176e-01
-1.96804512e+00 -6.88165724e-01 -8.22401196e-02 2.44713593e+00
1.00327122e+00 4.26536858e-01 -2.18056455e-01 3.18966329e-01
7.60166883e-01 4.15310204e-01 -5.76072514e-01 -2.90249228e-01
-2.01145768e-01 2.22947933e-02 4.35796231e-01 4.75998312e-01
-9.67212796e-01 8.49984050e-01 4.92216682e+00 9.51672137e-01
-1.37751675e+00 5.62450923e-02 5.65392673e-01 -1.24759367e-02
-5.17183244e-01 2.03177750e-01 -7.41210699e-01 7.10328341e-01
2.52348125e-01 2.21204534e-01 4.14442331e-01 8.15062642e-01
1.09647766e-01 9.91551578e-03 -8.29024434e-01 1.08006799e+00
3.36500779e-02 -1.10529447e+00 1.75273240e-01 -1.33350268e-01
7.30828822e-01 1.15568954e-02 1.32731289e-01 2.82654285e-01
-6.34101406e-02 -6.59567475e-01 7.58853972e-01 2.20318839e-01
5.52058339e-01 -6.34547770e-01 6.76136613e-01 4.86191124e-01
-1.19154561e+00 -2.52084941e-01 -5.98895133e-01 1.96965367e-01
7.76480064e-02 8.12518358e-01 -7.67547429e-01 3.84337455e-01
7.53050745e-01 1.00107682e+00 -5.81830502e-01 9.71867919e-01
-6.35294467e-02 3.86463284e-01 -3.82702768e-01 -1.35454610e-01
4.03981984e-01 -4.84568328e-01 5.12183964e-01 1.03556430e+00
2.01362371e-01 -3.86737853e-01 2.29720011e-01 8.31338406e-01
-2.45405823e-01 4.42201287e-01 -3.60178232e-01 9.87593979e-02
2.59498119e-01 9.93431449e-01 -4.20156896e-01 -4.08836901e-01
-6.20216012e-01 1.21803486e+00 6.55234039e-01 2.87138879e-01
-7.38653839e-01 -4.40520525e-01 4.19877857e-01 1.78182259e-01
4.23828930e-01 2.07498357e-01 -5.59844732e-01 -1.16618800e+00
5.54282665e-01 -5.78368664e-01 3.89868438e-01 -5.37755251e-01
-1.30677176e+00 2.82781154e-01 -4.89533901e-01 -1.34358025e+00
1.25073314e-01 -4.39936578e-01 -6.87687278e-01 5.96298873e-01
-1.85264409e+00 -8.40512931e-01 -3.03658694e-01 5.19033015e-01
4.59984660e-01 7.96131715e-02 2.33768716e-01 5.71292341e-01
-6.91752672e-01 1.01473022e+00 4.59601171e-02 8.89762342e-02
6.19882941e-01 -9.93388534e-01 1.78380162e-01 6.48003519e-01
5.27009852e-02 4.77387190e-01 5.29008389e-01 -4.44277525e-01
-1.26465690e+00 -1.21582484e+00 1.10329366e+00 6.83572516e-02
4.05857056e-01 -3.08110893e-01 -1.17638862e+00 3.18854451e-01
-1.36153638e-01 -8.40522647e-02 3.10089111e-01 -1.61409006e-01
-4.42856997e-01 -3.42224777e-01 -1.01439834e+00 7.32978642e-01
1.05201125e+00 -4.00781453e-01 -4.00820106e-01 3.96945328e-01
4.82700855e-01 6.26374185e-02 -6.00870669e-01 4.94768679e-01
4.46462393e-01 -1.07177818e+00 8.74426603e-01 -4.23845917e-01
6.59914315e-01 -2.04517737e-01 -1.46307856e-01 -1.05609548e+00
-1.98362261e-01 -5.72405696e-01 -1.77202061e-01 1.19646490e+00
4.63391095e-01 -8.25075746e-01 8.75226796e-01 2.81727463e-01
-1.17188878e-01 -1.21227372e+00 -9.81640875e-01 -9.25415576e-01
2.63708998e-02 -2.89268404e-01 3.86483908e-01 9.68116224e-01
6.55317232e-02 1.79384902e-01 -4.86683190e-01 -6.92243129e-02
5.60287297e-01 2.68539965e-01 5.59536278e-01 -1.01469076e+00
-2.01936439e-01 -6.34535849e-01 -3.95142704e-01 -1.58186328e+00
1.52128696e-01 -9.58290339e-01 1.05102368e-01 -1.49206734e+00
5.35078287e-01 -5.79545021e-01 -5.55641592e-01 5.11604488e-01
-4.80951428e-01 3.67618054e-01 3.00948143e-01 4.09757286e-01
-7.00780272e-01 7.87865579e-01 1.34607172e+00 -2.02785715e-01
-2.13476270e-01 -2.64116198e-01 -5.20274758e-01 6.60406470e-01
7.00145841e-01 -5.46650529e-01 -3.12342852e-01 -4.61033076e-01
3.16374689e-01 -1.99239790e-01 3.58870119e-01 -6.80970609e-01
2.42720798e-01 -1.11775875e-01 3.22120398e-01 -5.58088183e-01
3.68657827e-01 -7.00241923e-01 -3.59495699e-01 5.84683537e-01
-4.24316496e-01 -1.25988321e-02 -2.01665670e-01 4.22182292e-01
-4.32221800e-01 -4.01831210e-01 8.88744593e-01 5.80499806e-02
-8.31259310e-01 4.44618523e-01 1.06072128e-01 1.99999154e-01
7.42047668e-01 -4.18187410e-01 -2.00022236e-01 -3.28392923e-01
-1.29031911e-01 6.82663500e-01 6.15170181e-01 5.77861428e-01
8.60412717e-01 -1.50846982e+00 -4.31320637e-01 3.88418674e-01
1.88819051e-01 -1.05054170e-01 1.88324787e-02 1.06193602e+00
-1.06928654e-01 2.51500398e-01 2.04452053e-01 -6.65579021e-01
-1.26226234e+00 5.59953690e-01 4.29743677e-01 -2.47898430e-01
-6.91434205e-01 7.92906761e-01 3.34061325e-01 -3.28428805e-01
4.51598018e-01 -5.79433795e-03 8.59024972e-02 -1.92551389e-01
4.82410133e-01 2.34344602e-01 9.17024631e-03 -3.26936215e-01
-4.02450979e-01 6.25317037e-01 -2.05012068e-01 -1.14336759e-02
1.16297090e+00 -2.27319419e-01 -8.46220329e-02 2.64634460e-01
1.75659168e+00 -3.22183818e-01 -1.31581271e+00 -3.81225020e-01
1.88613087e-02 -7.74779797e-01 1.47564858e-01 -3.51123542e-01
-1.46655619e+00 1.03256404e+00 6.18912697e-01 2.12543383e-02
1.33729875e+00 -4.36117239e-02 1.28637695e+00 3.69028091e-01
2.08554082e-02 -1.46507478e+00 3.07103008e-01 2.51296550e-01
5.65665185e-01 -1.51982105e+00 -1.71574771e-01 -3.69614452e-01
-6.18224084e-01 7.92274773e-01 7.57626951e-01 -1.31072372e-01
6.44297481e-01 2.60466128e-04 8.06745291e-02 1.67864468e-02
-7.92168021e-01 -7.17618177e-03 2.45075092e-01 2.81811118e-01
1.91087380e-01 -1.52100638e-01 -4.77082253e-01 1.47965088e-01
-1.58397015e-02 6.74661249e-02 5.96128069e-02 1.01238978e+00
-6.27867579e-01 -1.13850904e+00 -1.43311650e-01 5.63761711e-01
-3.36148471e-01 -2.57956952e-01 -3.57539058e-01 5.57547510e-01
-6.04417315e-03 8.22449565e-01 1.96020812e-01 -4.30984795e-01
3.48568201e-01 -1.77050065e-02 3.78627121e-01 -1.80667430e-01
-4.02511775e-01 -2.08672971e-01 -1.94597334e-01 -5.45600951e-01
-2.45136201e-01 -4.28135544e-01 -1.48636067e+00 -2.18430847e-01
-6.29968166e-01 6.03919327e-02 4.90626216e-01 8.49055886e-01
2.50394523e-01 9.83342379e-02 1.05245495e+00 -4.26202148e-01
-8.42271388e-01 -7.05054283e-01 -3.61262470e-01 7.86580861e-01
3.21739525e-01 -5.91077089e-01 -5.23943901e-01 -2.00859040e-01] | [9.446127891540527, 3.1464428901672363] |
a2e6861e-ba70-45d8-a45d-8ef878447d1d | hearing-lips-improving-lip-reading-by | 1911.11502 | null | https://arxiv.org/abs/1911.11502v1 | https://arxiv.org/pdf/1911.11502v1.pdf | Hearing Lips: Improving Lip Reading by Distilling Speech Recognizers | Lip reading has witnessed unparalleled development in recent years thanks to deep learning and the availability of large-scale datasets. Despite the encouraging results achieved, the performance of lip reading, unfortunately, remains inferior to the one of its counterpart speech recognition, due to the ambiguous nature of its actuations that makes it challenging to extract discriminant features from the lip movement videos. In this paper, we propose a new method, termed as Lip by Speech (LIBS), of which the goal is to strengthen lip reading by learning from speech recognizers. The rationale behind our approach is that the features extracted from speech recognizers may provide complementary and discriminant clues, which are formidable to be obtained from the subtle movements of the lips, and consequently facilitate the training of lip readers. This is achieved, specifically, by distilling multi-granularity knowledge from speech recognizers to lip readers. To conduct this cross-modal knowledge distillation, we utilize an efficacious alignment scheme to handle the inconsistent lengths of the audios and videos, as well as an innovative filtering strategy to refine the speech recognizer's prediction. The proposed method achieves the new state-of-the-art performance on the CMLR and LRS2 datasets, outperforming the baseline by a margin of 7.66% and 2.75% in character error rate, respectively. | ['Rui Xu', 'Mingli Song', 'Haihong Tang', 'Ya Zhao', 'Xinchao Wang', 'Peng Hou'] | 2019-11-26 | null | null | null | null | ['lipreading'] | ['computer-vision'] | [ 4.12402153e-01 6.17864057e-02 -5.81824481e-01 -4.25234810e-02
-1.23728144e+00 -3.81569415e-01 4.12343085e-01 -2.77713835e-01
-3.02646637e-01 5.97544074e-01 5.53002656e-01 -2.00728133e-01
-7.24080205e-02 -9.33060199e-02 -5.00045121e-01 -9.11079943e-01
5.38311124e-01 -1.05249183e-02 1.59155071e-01 8.09929371e-02
3.86633754e-01 3.97214442e-01 -2.01994419e+00 1.89841956e-01
1.11467957e+00 1.30481398e+00 4.36378509e-01 5.66479266e-01
-2.87185878e-01 3.57821286e-01 -4.19574976e-01 -4.18612957e-01
-6.48723766e-02 -3.29607517e-01 -5.83158970e-01 1.64975435e-01
4.87632930e-01 -3.58953685e-01 -3.42116743e-01 9.80042756e-01
1.02096748e+00 3.51162255e-02 5.27061820e-01 -1.08029604e+00
-6.14658713e-01 4.61786807e-01 -5.93877316e-01 -5.82269020e-02
4.88196313e-01 2.10918203e-01 1.05783021e+00 -1.07582939e+00
2.88493872e-01 1.15183163e+00 5.71142256e-01 7.88417399e-01
-9.79729176e-01 -6.95130408e-01 5.30256331e-02 5.44893324e-01
-1.41643167e+00 -1.28187895e+00 9.25086319e-01 -2.66399741e-01
6.31569922e-01 1.77152097e-01 4.09024924e-01 1.22725070e+00
-4.29793298e-01 1.30918849e+00 1.03363323e+00 -5.60807943e-01
2.51959395e-02 -2.11625770e-02 -2.16079265e-01 5.15565336e-01
-4.40667234e-02 1.39134526e-01 -1.06494653e+00 3.15323412e-01
3.19830269e-01 -3.13610554e-01 -5.07279992e-01 -1.21284626e-01
-1.11537540e+00 4.38790619e-01 -8.69449973e-02 1.41860873e-01
-3.11918497e-01 -2.98772573e-01 3.34844470e-01 -1.16555870e-01
3.23528051e-01 5.59706986e-02 -5.74150980e-01 -5.05698025e-01
-1.07512569e+00 -2.90844053e-01 5.14122009e-01 7.18180597e-01
3.62156630e-01 9.97150876e-03 -3.53682116e-02 1.05548298e+00
6.00038171e-01 6.72268808e-01 8.36921215e-01 -6.22672677e-01
7.23748863e-01 3.77935350e-01 -1.34826139e-01 -7.19982266e-01
-3.06154430e-01 -1.55322507e-01 -8.69574070e-01 1.13186575e-02
5.61628580e-01 -5.41759469e-02 -9.06516552e-01 1.79414117e+00
3.90065223e-01 3.64439845e-01 3.55241716e-01 6.69865549e-01
9.22031045e-01 4.29542750e-01 -4.40990031e-02 -4.23166335e-01
1.24729991e+00 -8.63807380e-01 -8.91533434e-01 -7.80420601e-02
3.34045529e-01 -9.60600019e-01 1.17832923e+00 3.42926741e-01
-1.07994330e+00 -7.94189990e-01 -8.94660175e-01 -3.81026827e-02
-2.12993562e-01 5.63823402e-01 2.75308937e-01 6.36637092e-01
-1.04470825e+00 1.43815562e-01 -6.47459626e-01 -2.15661898e-01
5.29891968e-01 4.69338298e-01 -3.71475995e-01 1.18950196e-01
-1.07449365e+00 6.99502826e-01 2.34138981e-01 1.85402229e-01
-4.81632411e-01 -6.45190537e-01 -8.34709585e-01 7.79866800e-03
5.28740287e-01 -2.59953022e-01 1.14805186e+00 -7.62947619e-01
-1.93702567e+00 8.54661405e-01 -5.25741935e-01 -2.61659741e-01
5.23062229e-01 -2.48898387e-01 -4.61878002e-01 3.05222064e-01
-1.74205184e-01 7.33164072e-01 1.19610643e+00 -1.01419377e+00
-8.38616610e-01 -4.16964710e-01 -3.32491636e-01 3.06636363e-01
-3.57191771e-01 -1.77808404e-02 -5.97886682e-01 -5.86431623e-01
2.64772177e-01 -7.28155434e-01 4.39253956e-01 -8.42364598e-03
-4.60509777e-01 -6.00331604e-01 9.15011168e-01 -1.00359094e+00
1.13951862e+00 -2.33373523e+00 1.32744446e-01 -9.97066200e-02
1.99015781e-01 8.15712273e-01 -2.56057858e-01 1.82947040e-01
5.63917048e-02 5.50154150e-02 8.41503367e-02 -4.98973608e-01
1.38100758e-02 -4.42666300e-02 -3.33605766e-01 4.73499924e-01
2.89286375e-01 9.32146549e-01 -7.35817254e-01 -6.51275516e-01
2.68431544e-01 5.74621677e-01 -3.22116852e-01 3.82558435e-01
-5.73184565e-02 4.78207409e-01 -2.94625729e-01 1.06737649e+00
7.33405292e-01 -7.76747242e-02 1.88826397e-02 -3.12584698e-01
-1.52920067e-01 6.00135505e-01 -1.03826046e+00 1.74289763e+00
-3.05334479e-01 7.97864616e-01 1.78952277e-01 -1.13747704e+00
1.01248598e+00 5.26646554e-01 5.24911702e-01 -7.02393949e-01
7.04091089e-03 2.81520277e-01 -3.77605669e-02 -8.88867199e-01
2.11140990e-01 -4.96990904e-02 2.57510841e-01 1.94195524e-01
-1.00302540e-01 5.55938557e-02 -1.22657772e-02 -2.18084559e-01
5.18952429e-01 1.06602766e-01 2.50579149e-01 1.42330125e-01
1.10355818e+00 -7.23053098e-01 4.97831911e-01 3.52450609e-01
-3.73731375e-01 4.43500578e-01 2.98897564e-01 1.82157040e-01
-7.13463545e-01 -1.16146278e+00 -3.39670181e-01 1.11846209e+00
2.38450691e-01 -2.76908785e-01 -7.53228903e-01 -5.92722058e-01
-3.48501615e-02 1.68838441e-01 -2.29254246e-01 -1.89865753e-01
-4.23942953e-01 -3.02521139e-01 8.77702832e-01 6.65644348e-01
7.72235096e-01 -9.91326511e-01 -2.63524979e-01 4.78015728e-02
-5.07396817e-01 -1.20900357e+00 -5.48162341e-01 -1.41816020e-01
-4.28692311e-01 -1.03088188e+00 -9.82572317e-01 -8.91199410e-01
2.75469273e-01 2.62547582e-01 5.62207997e-01 -6.41295910e-02
-6.34610355e-02 3.39364558e-01 -4.05190289e-01 -2.57065713e-01
-4.36922163e-01 3.29006195e-01 4.40420091e-01 3.18312407e-01
4.03118014e-01 -4.13477063e-01 -5.71031153e-01 3.60620618e-01
-6.73381925e-01 1.67679980e-01 1.11527383e+00 1.10733438e+00
5.99651456e-01 -1.86954066e-01 1.13808215e+00 -9.80370026e-03
4.98317450e-01 -2.45344222e-01 -4.93121326e-01 4.46618140e-01
-7.29165912e-01 -3.61406896e-03 4.97578114e-01 -7.31680334e-01
-1.12183654e+00 4.46676463e-02 -5.54462850e-01 -3.56327116e-01
-3.84532481e-01 3.47923249e-01 -7.09480405e-01 -4.94666994e-02
5.92006044e-03 6.42500162e-01 3.66743147e-01 -7.57536054e-01
3.29615921e-01 1.26779473e+00 8.31078708e-01 -4.21724051e-01
5.47318757e-01 8.32341015e-02 -1.84253812e-01 -1.19784653e+00
-6.43635392e-01 -6.20377243e-01 -5.99963784e-01 -2.29026705e-01
8.89686882e-01 -7.83008814e-01 -1.12037230e+00 1.01822007e+00
-1.09730816e+00 -7.41700530e-02 -1.07558571e-01 5.54785132e-01
-6.98535681e-01 5.44143140e-01 -3.47363740e-01 -9.51583862e-01
-3.02823067e-01 -1.35671616e+00 1.21460724e+00 4.65418488e-01
-6.44352660e-03 -6.15046978e-01 -1.98425770e-01 9.44906354e-01
3.87676060e-01 -4.07462686e-01 7.82385111e-01 -8.64428222e-01
-6.16486311e-01 -6.40125498e-02 -4.10228431e-01 6.11452639e-01
3.66594195e-01 -1.89298630e-01 -1.22327316e+00 -1.85511544e-01
-2.71111429e-01 -5.37716568e-01 7.86467373e-01 3.75940561e-01
1.13858461e+00 -2.81493634e-01 -3.50477159e-01 5.53201973e-01
7.84750462e-01 2.34589323e-01 7.08180964e-01 6.28349707e-02
6.18194520e-01 6.56047821e-01 5.81187546e-01 3.62392604e-01
3.68744344e-01 9.94845569e-01 1.88121215e-01 -5.04817814e-02
-6.48351789e-01 -4.97731268e-01 6.22951746e-01 1.11644614e+00
3.21197473e-02 -1.59666568e-01 -7.72531569e-01 4.00147051e-01
-1.66428006e+00 -8.97670567e-01 3.81804764e-01 2.09965706e+00
1.06321299e+00 -1.42674074e-01 1.22456200e-01 4.86232877e-01
9.08057511e-01 3.45782548e-01 -7.94980228e-01 7.27168694e-02
-2.04190180e-01 5.36080413e-02 8.11265931e-02 4.39925879e-01
-9.94502664e-01 1.07745945e+00 5.87722921e+00 1.07311726e+00
-1.47327328e+00 -1.89286590e-01 4.86266524e-01 6.98503926e-02
1.06412187e-01 -3.98234397e-01 -1.23120618e+00 6.48133337e-01
7.60337830e-01 1.63382641e-03 3.87286931e-01 5.44467390e-01
3.27638417e-01 -2.50424296e-02 -9.69363928e-01 1.24311173e+00
3.78223240e-01 -1.24545956e+00 -1.09295703e-01 -7.31828995e-03
4.43937570e-01 -7.95373619e-02 3.55548978e-01 1.67393237e-01
-3.40552896e-01 -9.59815323e-01 6.01437688e-01 5.67465544e-01
1.17256749e+00 -6.12813115e-01 6.17566049e-01 4.48328644e-01
-1.35816729e+00 -1.41091406e-01 1.03311762e-01 3.01140219e-01
1.14278093e-01 1.36399731e-01 -1.09369791e+00 3.81020278e-01
4.80950594e-01 7.21325815e-01 -3.32322985e-01 9.56668615e-01
-3.01258415e-01 6.37623072e-01 -2.76845932e-01 -2.37229038e-02
-1.41002998e-01 1.06850810e-01 5.41162968e-01 9.42795575e-01
2.58176386e-01 -8.61609355e-02 7.50441253e-02 5.58577418e-01
-2.46583238e-01 2.14260533e-01 -3.55380833e-01 -2.35287815e-01
6.67950034e-01 9.46943164e-01 -1.99754581e-01 -1.58616230e-01
-5.43445349e-01 5.02176404e-01 1.45660847e-01 3.79903406e-01
-6.71021581e-01 -4.06935394e-01 8.53454351e-01 -6.51805894e-03
4.54064906e-01 1.59993749e-02 -1.24513432e-01 -1.22811985e+00
2.83808172e-01 -1.20632780e+00 4.15101834e-03 -6.17065668e-01
-1.08820808e+00 4.08161938e-01 -3.97232294e-01 -1.18018103e+00
-4.87082869e-01 -5.40626764e-01 -3.67602021e-01 8.94073069e-01
-2.01671100e+00 -1.32052517e+00 -1.35069147e-01 7.39072919e-01
8.22626829e-01 -3.28329831e-01 6.55863345e-01 2.68443614e-01
-5.57411432e-01 1.13031983e+00 1.35829791e-01 2.17196241e-01
8.84611785e-01 -7.68498838e-01 -7.65166208e-02 7.93550491e-01
1.46099746e-01 3.89491588e-01 3.98635000e-01 -4.67592627e-01
-1.53137314e+00 -7.04740107e-01 9.46748495e-01 -4.19319533e-02
7.29511738e-01 -2.18134388e-01 -9.21847641e-01 1.63371846e-01
-2.91610602e-02 -1.03774130e-01 7.84789681e-01 6.41031191e-02
-4.37235117e-01 -4.35855657e-01 -8.63920569e-01 5.39100409e-01
1.07289779e+00 -9.02754486e-01 -8.63416791e-01 -2.55095921e-02
6.49183214e-01 -3.60990703e-01 -6.88354254e-01 6.16923809e-01
8.81787896e-01 -7.95904279e-01 1.05165172e+00 -4.76426125e-01
1.41727105e-01 -1.83514714e-01 -2.71762848e-01 -9.86140728e-01
2.61709452e-01 -8.35292161e-01 -3.54523212e-01 1.73732352e+00
3.49910736e-01 -6.74260259e-01 9.03650165e-01 2.91855991e-01
-1.50537491e-01 -8.93225312e-01 -1.19976771e+00 -7.73105025e-01
-1.42681032e-01 -4.36708272e-01 5.47764599e-01 6.18335485e-01
2.71131486e-01 3.59112054e-01 -4.69412863e-01 2.09306121e-01
3.34222853e-01 1.02323487e-01 7.76584029e-01 -1.26844788e+00
-3.09624732e-01 -6.35320604e-01 -4.44285959e-01 -1.69392490e+00
6.42233908e-01 -7.40983307e-01 3.05062354e-01 -1.23258841e+00
2.93374341e-02 -4.16324615e-01 -3.07291865e-01 6.00054562e-01
-3.34120542e-01 4.11724299e-02 3.31400096e-01 2.27340698e-01
-5.76302707e-01 7.58376181e-01 1.18718612e+00 -2.73263097e-01
-3.65018457e-01 3.38207930e-01 -6.37918949e-01 7.56106436e-01
6.45674527e-01 1.24790557e-01 -3.77342284e-01 -2.45515093e-01
-1.85606599e-01 3.04899067e-01 1.98248312e-01 -8.98405433e-01
5.29269814e-01 1.12900674e-01 1.71122029e-01 -9.22924519e-01
6.28397703e-01 -5.52318096e-01 -4.11178380e-01 1.43899724e-01
-5.03264725e-01 -3.45040381e-01 2.69974321e-01 5.56536257e-01
-4.47299063e-01 -3.13628539e-02 7.11098194e-01 4.76658642e-01
-7.57510543e-01 2.38210335e-01 -2.28593424e-01 2.27716535e-01
7.40707517e-01 -2.41448820e-01 -3.68561357e-01 -4.03177381e-01
-5.37254989e-01 -3.72059047e-02 2.11178750e-01 5.72984338e-01
7.33869970e-01 -1.20000792e+00 -5.80853939e-01 5.36557674e-01
3.31788473e-02 -4.33751978e-02 2.93920577e-01 1.13007450e+00
2.49495730e-01 5.00663757e-01 3.33823338e-02 -7.63407648e-01
-1.57308269e+00 4.57213342e-01 1.15109719e-01 2.03485768e-02
-5.98567665e-01 7.95617044e-01 5.55256605e-02 -2.20038053e-02
7.52038062e-01 -1.83064431e-01 -3.39586318e-01 3.72104257e-01
7.82640815e-01 3.92325222e-01 7.76447654e-02 -8.50837946e-01
-3.78500491e-01 7.68863618e-01 -1.43628597e-01 9.92407128e-02
1.10646355e+00 -4.98927951e-01 3.11196178e-01 2.77763188e-01
1.12440228e+00 4.26560581e-01 -1.44517577e+00 -3.02315563e-01
7.63923824e-02 -4.14526910e-01 -5.41611761e-02 -8.60629559e-01
-9.70617175e-01 1.12243211e+00 5.60600042e-01 2.06890143e-02
1.27141023e+00 2.85871774e-01 1.03149056e+00 2.00072482e-01
5.56785380e-04 -1.14816189e+00 1.53841719e-01 3.34482431e-01
7.32857764e-01 -1.51553822e+00 -4.65761721e-01 -3.40769619e-01
-5.22582293e-01 1.06424665e+00 3.07217866e-01 6.45906627e-01
3.40887308e-01 2.29124010e-01 1.89477980e-01 4.53470290e-01
-5.47824025e-01 -4.94929999e-01 5.84470093e-01 7.92114198e-01
2.27182865e-01 -8.96289945e-02 1.05722584e-02 6.91095889e-01
-9.12851691e-02 1.16223797e-01 -7.56644160e-02 3.74825597e-01
-5.22571802e-01 -1.14512074e+00 -4.19906378e-01 2.06072628e-01
-4.45644468e-01 -1.98518127e-01 -2.53058583e-01 6.48324907e-01
1.36981547e-01 1.13744175e+00 -2.08239585e-01 -4.89755958e-01
1.52702659e-01 3.63656282e-01 1.92453310e-01 -1.30817756e-01
1.11621082e-01 3.04953963e-01 -6.99671805e-02 -4.51956242e-01
-4.35718775e-01 -8.63940239e-01 -1.14785337e+00 -1.08137459e-01
-5.79402447e-01 -6.75534233e-02 7.70098984e-01 1.18852234e+00
5.27263045e-01 2.29979306e-01 7.10355282e-01 -9.40257013e-01
-9.14023697e-01 -9.18474853e-01 -4.07979697e-01 1.81528792e-01
6.31302655e-01 -8.39292049e-01 -5.36560178e-01 1.10674366e-01] | [14.326064109802246, 5.000001907348633] |
92fea935-7193-4df6-b758-53a239915cb4 | on-the-curious-case-of-ell-2-norm-of-sense | 2210.14815 | null | https://arxiv.org/abs/2210.14815v1 | https://arxiv.org/pdf/2210.14815v1.pdf | On the Curious Case of $\ell_2$ norm of Sense Embeddings | We show that the $\ell_2$ norm of a static sense embedding encodes information related to the frequency of that sense in the training corpus used to learn the sense embeddings. This finding can be seen as an extension of a previously known relationship for word embeddings to sense embeddings. Our experimental results show that, in spite of its simplicity, the $\ell_2$ norm of sense embeddings is a surprisingly effective feature for several word sense related tasks such as (a) most frequent sense prediction, (b) Word-in-Context (WiC), and (c) Word Sense Disambiguation (WSD). In particular, by simply including the $\ell_2$ norm of a sense embedding as a feature in a classifier, we show that we can improve WiC and WSD methods that use static sense embeddings. | ['Danushka Bollegala', 'Yi Zhou'] | 2022-10-26 | null | null | null | null | ['word-sense-disambiguation'] | ['natural-language-processing'] | [ 1.74398586e-01 -6.00205697e-02 -2.07692593e-01 -3.77095133e-01
-4.68073845e-01 -7.95475304e-01 6.88529313e-01 8.73943746e-01
-1.04017866e+00 6.33844316e-01 3.86099041e-01 -5.07554531e-01
-1.60572648e-01 -9.65341330e-01 -1.02827482e-01 -7.45558202e-01
-1.35011554e-01 1.71798930e-01 2.90265799e-01 -7.64951229e-01
4.06404048e-01 2.45178729e-01 -1.66247475e+00 -1.67195946e-01
6.54618740e-01 8.88207436e-01 3.67998928e-01 2.36344874e-01
-5.14937401e-01 -4.95149828e-02 -7.11710989e-01 -2.04779327e-01
8.64911266e-03 -3.46484572e-01 -9.77273166e-01 -6.10651195e-01
1.32806793e-01 6.88673556e-01 1.12306647e-01 1.26300085e+00
4.13132340e-01 5.82962751e-01 5.79922915e-01 -8.64750028e-01
-8.51213396e-01 4.43804651e-01 -2.32347727e-01 4.11894947e-01
8.89440417e-01 -1.03282079e-01 1.54463315e+00 -7.74015903e-01
8.88548672e-01 1.12404346e+00 7.00927198e-01 3.89698774e-01
-1.15062344e+00 -4.03054476e-01 3.14047933e-01 1.45203620e-01
-1.35119092e+00 1.68890193e-01 6.70205355e-01 -3.72557342e-01
1.39811969e+00 2.55325884e-01 9.51354623e-01 9.72784400e-01
1.84346825e-01 6.45057082e-01 1.04666877e+00 -7.30589747e-01
5.08471489e-01 7.16467053e-02 6.02809370e-01 3.49250227e-01
4.98480856e-01 -2.97433347e-03 -3.68288100e-01 -4.34401929e-01
2.72733241e-01 -1.07000694e-01 -3.69924814e-01 -4.60797697e-02
-1.08633649e+00 9.96804953e-01 3.88030022e-01 9.86386597e-01
-1.60694212e-01 2.28420019e-01 3.90119493e-01 4.38990623e-01
5.23574769e-01 1.09807789e+00 -6.74282074e-01 -3.97990972e-01
-6.07102454e-01 3.72237325e-01 6.93844199e-01 5.85744202e-01
1.08891857e+00 -1.59072012e-01 4.49687056e-02 9.14622247e-01
1.85007185e-01 5.90388358e-01 9.21015501e-01 -3.09100419e-01
-1.25068948e-01 9.15842712e-01 -9.78296716e-03 -9.97292757e-01
-3.35312665e-01 -2.19390556e-01 -2.55758464e-01 -1.07821368e-01
1.63720459e-01 8.61147419e-02 -6.07649803e-01 2.09726548e+00
2.77325451e-01 2.50763118e-01 3.70893739e-02 6.38980150e-01
4.10456777e-01 3.82303417e-01 2.73324221e-01 -2.18986228e-01
1.68025112e+00 -1.13132007e-01 -5.28435946e-01 -6.55963600e-01
8.03268492e-01 -9.38628435e-01 1.48926258e+00 -1.52775729e-02
-4.24093157e-01 -2.25404486e-01 -1.28860283e+00 -9.25112590e-02
-1.07382393e+00 -5.28650582e-01 9.51942146e-01 7.27010429e-01
-8.85641158e-01 4.61149246e-01 -2.91002691e-01 -6.81273818e-01
-5.23882732e-02 9.12048668e-03 -5.78569531e-01 -1.55639052e-01
-1.58980775e+00 1.04742134e+00 6.22178793e-01 -6.09301746e-01
7.54317194e-02 -8.10978293e-01 -1.23369300e+00 -4.21041958e-02
2.96615690e-01 -3.75547320e-01 6.86786056e-01 -5.08640051e-01
-6.69971287e-01 1.05322301e+00 -4.98183042e-01 -2.52282798e-01
-3.88666600e-01 -1.73089698e-01 -8.32917154e-01 -2.22335488e-01
3.98809075e-01 2.31917873e-01 3.35208029e-01 -1.05440485e+00
-6.85263038e-01 -3.78664613e-01 3.85132998e-01 5.80993183e-02
-3.24891061e-01 -2.97268510e-01 1.07381262e-01 -9.08223033e-01
1.30501822e-01 -8.01035047e-01 -3.16481441e-01 -1.39203206e-01
-1.29997924e-01 -8.27568650e-01 5.96326888e-01 -1.27652963e-03
1.63165677e+00 -2.27260160e+00 -5.29444665e-02 3.04786950e-01
1.49523601e-01 4.55475390e-01 -4.09032375e-01 6.85523748e-01
-2.75565177e-01 4.61428463e-01 -3.23564440e-01 -8.85342136e-02
1.40299976e-01 5.73099613e-01 -2.09223658e-01 7.74654672e-02
1.78707376e-01 7.11230457e-01 -1.51302016e+00 -1.12067722e-01
1.73791498e-01 3.39169323e-01 -4.60171878e-01 -9.45241600e-02
-1.43447936e-01 -3.96131188e-01 -5.99491119e-01 1.14009835e-01
4.02083099e-01 -1.54084638e-01 5.85959136e-01 -4.86702099e-02
8.98096059e-03 6.53430223e-01 -1.23947370e+00 1.72071457e+00
-7.28396416e-01 6.06636465e-01 -3.86894107e-01 -9.52413380e-01
9.03729618e-01 1.64141893e-01 4.64303792e-01 -6.66077971e-01
-1.28491700e-01 2.41084203e-01 -5.85313775e-02 -2.98212290e-01
8.08264136e-01 -6.04855537e-01 -5.52436352e-01 7.19155669e-01
2.01953381e-01 -1.79318354e-01 5.02224147e-01 1.24080010e-01
1.25446022e+00 -3.82844090e-01 8.23583245e-01 -8.45282853e-01
3.40811521e-01 1.66484602e-02 6.46429777e-01 7.20464230e-01
-1.52849928e-01 3.45517218e-01 5.57758808e-01 -4.42685932e-01
-6.76221251e-01 -1.22235656e+00 -2.37073645e-01 1.20665956e+00
3.26189667e-01 -8.14464390e-01 -5.09324253e-01 -5.13372302e-01
2.15209812e-01 9.63918030e-01 -8.74779224e-01 -2.74047613e-01
-4.86803144e-01 -7.69054830e-01 5.96411347e-01 6.05687201e-01
-5.79373725e-03 -8.97134185e-01 -6.11886621e-01 4.09005910e-01
6.99210241e-02 -7.71053910e-01 -7.00610578e-01 6.09906077e-01
-4.91911292e-01 -1.26707375e+00 -1.24230817e-01 -6.77973866e-01
2.22676083e-01 2.86312044e-01 1.28666985e+00 2.37409130e-01
-3.96737069e-01 3.50251615e-01 -9.10746098e-01 -4.63158369e-01
-5.40265217e-02 6.33319914e-02 4.62997556e-01 -4.82323885e-01
1.00495398e+00 -7.22085893e-01 -4.61620003e-01 -5.10786027e-02
-1.42424762e+00 -7.73111820e-01 7.49696344e-02 1.02868152e+00
7.02481329e-01 -2.51552433e-01 4.83790338e-01 -9.56772625e-01
1.13585460e+00 -3.83196205e-01 -3.40487063e-01 3.95662189e-01
-1.09919381e+00 5.04623353e-01 3.25536817e-01 -5.10313451e-01
-4.84975934e-01 -3.71835470e-01 -5.46627939e-01 2.16348663e-01
-1.01355165e-01 7.91278481e-01 5.96445762e-02 3.07448387e-01
7.55813062e-01 3.28989685e-01 -2.96187520e-01 -5.04194319e-01
5.75179994e-01 4.96509314e-01 1.93784069e-02 -6.06328666e-01
6.23350918e-01 1.94677904e-01 -1.72059998e-01 -1.08977461e+00
-1.06582522e+00 -7.96552718e-01 -5.52978277e-01 4.42815065e-01
1.06597114e+00 -4.33961213e-01 -5.33777535e-01 -2.81489134e-01
-9.61451828e-01 -9.66977235e-03 -4.06397730e-01 5.06535232e-01
-2.18827456e-01 4.68855917e-01 -4.48983721e-02 -8.60524237e-01
-2.19907194e-01 -9.16460335e-01 9.78118122e-01 1.94507375e-01
-8.74902189e-01 -1.48256814e+00 3.07407707e-01 -1.96333036e-01
4.61717844e-01 3.58540863e-01 1.47560430e+00 -1.07197821e+00
2.16956317e-01 -2.82215178e-01 3.87766585e-02 2.71689594e-01
3.96012694e-01 -4.48718697e-01 -8.78403664e-01 -1.45716771e-01
-1.44766554e-01 -6.09714985e-02 9.25989509e-01 2.18984317e-02
6.32046282e-01 -2.38693416e-01 -3.60960752e-01 2.61449009e-01
1.80851305e+00 1.75750270e-01 6.45036459e-01 2.93810040e-01
2.49569654e-01 1.63964137e-01 5.30124605e-01 3.67958248e-01
1.53235346e-01 5.85436881e-01 -1.85659025e-02 2.19364360e-01
1.15711089e-04 -3.57393801e-01 1.82120457e-01 6.58065736e-01
1.69844583e-01 -8.48897696e-02 -9.68965948e-01 9.56088066e-01
-1.46023440e+00 -1.00063550e+00 -3.84415314e-02 2.20458889e+00
1.05036235e+00 2.00283781e-01 -1.75941780e-01 3.25673163e-01
3.32675427e-01 6.92212641e-01 -1.05358303e-01 -7.30138838e-01
-3.10239315e-01 1.01635003e+00 3.11527491e-01 6.49440706e-01
-6.78441167e-01 1.14939725e+00 6.59879494e+00 1.02576518e+00
-1.03707552e+00 2.08779469e-01 -1.66784391e-01 1.89311206e-02
-9.45170820e-01 1.18215017e-01 -3.50562394e-01 4.58753079e-01
5.21715999e-01 -4.22435462e-01 3.41168642e-01 8.30997229e-01
-2.58616179e-01 -2.25639209e-01 -1.16756201e+00 9.16153669e-01
6.84157154e-03 -1.24502695e+00 4.28922325e-02 -1.22578023e-02
5.29611528e-01 4.36772453e-03 -1.14687979e-01 2.76248097e-01
2.79671341e-01 -1.00348580e+00 4.12235111e-01 5.79284988e-02
1.01459587e+00 -7.70189941e-01 8.66981030e-01 -2.29044687e-02
-1.45990181e+00 5.82278743e-02 -5.89694202e-01 -2.78366178e-01
1.22175865e-01 1.11426175e+00 -7.49173343e-01 5.86710811e-01
4.25904155e-01 6.43682003e-01 -2.96731740e-01 5.17516673e-01
-4.69078153e-01 5.53943634e-01 -3.34522992e-01 -4.29630429e-01
3.30315202e-01 -3.33992429e-02 7.57432163e-01 1.39059079e+00
3.38233322e-01 1.97175115e-01 2.46671692e-01 8.44892800e-01
7.66604096e-02 1.31730914e-01 -6.71622992e-01 -4.88702059e-01
7.76223719e-01 8.96887898e-01 -5.56965172e-01 -2.35069960e-01
-2.81867772e-01 7.72691071e-01 2.33276188e-01 2.44499251e-01
-2.44463161e-01 -7.78439999e-01 1.75174177e+00 -1.01609111e-01
4.16329622e-01 -2.51365840e-01 -2.95735806e-01 -1.04597557e+00
1.37547478e-01 -4.07104224e-01 5.02353847e-01 -5.00119805e-01
-1.68420053e+00 6.62818670e-01 -1.04452431e-01 -9.18080270e-01
-3.61657381e-01 -1.08434999e+00 -6.79375827e-01 9.47002947e-01
-1.53364122e+00 -5.57944059e-01 2.98458636e-01 2.70157576e-01
2.66956627e-01 -9.61044133e-02 1.39414847e+00 -6.51072115e-02
-1.19411759e-01 6.73566997e-01 -3.00492602e-03 2.50450432e-01
5.56083560e-01 -1.65810323e+00 3.07395130e-01 6.08254135e-01
5.54565549e-01 1.20211208e+00 1.07080460e+00 -5.10181248e-01
-1.34139442e+00 -7.12901711e-01 1.46320629e+00 -4.91284549e-01
8.71409178e-01 -2.74604499e-01 -8.28741074e-01 3.97186279e-01
1.14422806e-01 1.64610259e-02 1.17515945e+00 8.34643424e-01
-8.43310714e-01 -6.96268817e-03 -1.18969476e+00 5.46063900e-01
1.17638445e+00 -8.16091597e-01 -1.16340256e+00 -3.76525410e-02
1.02855253e+00 4.27609533e-02 -9.93324697e-01 -4.29069996e-02
6.50126576e-01 -7.22993791e-01 9.47764099e-01 -8.75096023e-01
1.56447545e-01 -1.31273210e-01 -7.18097568e-01 -1.59127486e+00
-2.25355044e-01 -3.68297458e-01 2.70893961e-01 7.71008074e-01
5.79401970e-01 -8.98766160e-01 2.59673804e-01 1.68359160e-01
3.21811140e-02 -9.00026917e-01 -1.24194849e+00 -8.86616468e-01
5.09828866e-01 -7.37234116e-01 8.56769323e-01 1.26657581e+00
5.52733958e-01 2.83454508e-01 3.24365050e-01 -3.95804718e-02
1.53838322e-01 1.82289511e-01 -4.45625186e-02 -1.58530426e+00
-5.21137156e-02 -6.93602443e-01 -8.86007726e-01 -8.49966049e-01
3.19676727e-01 -1.02749884e+00 -4.14306596e-02 -1.53824854e+00
-6.39548153e-02 -5.51444590e-01 -8.12791765e-01 5.74654043e-01
-5.43441534e-01 3.31078507e-02 1.10422142e-01 -3.46793920e-01
-3.43833506e-01 2.02846378e-01 8.30591023e-01 -1.77285373e-01
1.37024611e-01 -4.39160526e-01 -9.64016020e-01 5.06050229e-01
6.25417411e-01 -7.34071136e-01 -3.46615523e-01 -3.34916443e-01
8.14550102e-01 -3.54903102e-01 1.55169573e-02 -6.18173540e-01
-8.44372287e-02 -4.21080112e-01 -9.45844129e-02 5.27047850e-02
3.03745061e-01 -5.23109019e-01 -4.22386438e-01 4.44362789e-01
-3.48204702e-01 4.47063148e-01 1.30031571e-01 6.51837349e-01
-3.77340168e-01 -4.19572085e-01 4.64576840e-01 -4.68201451e-02
-1.24516201e+00 -2.42376849e-01 -3.91023606e-01 5.04849374e-01
6.27676606e-01 -2.65521824e-01 -1.48933142e-01 -6.53760284e-02
-3.87309641e-01 -1.30066024e-02 7.36238599e-01 7.40707994e-01
4.96587783e-01 -1.43722534e+00 -4.11483139e-01 1.61752179e-01
7.85606146e-01 -4.93729711e-01 -1.42528206e-01 2.86957830e-01
8.95564333e-02 3.43936056e-01 1.65844202e-01 -1.21203750e-01
-1.11292493e+00 6.35455191e-01 6.09358139e-02 -3.57593507e-01
-2.90135235e-01 1.04175448e+00 -1.23678394e-01 -4.38607246e-01
-2.15113372e-01 -3.80675018e-01 1.36304237e-02 2.88401127e-01
7.46068299e-01 7.99489394e-02 1.01760998e-01 -4.59742010e-01
-8.79798234e-01 5.22630095e-01 2.50644565e-01 -2.01705784e-01
1.31388521e+00 1.42947197e-01 -3.40266079e-01 6.97508395e-01
1.39903641e+00 1.08808585e-01 -3.84632319e-01 -2.92935491e-01
3.65438372e-01 -4.79381979e-01 -7.27077872e-02 -8.55302036e-01
-5.85411310e-01 6.98114514e-01 6.04557395e-01 4.27642643e-01
9.16316867e-01 1.96439102e-01 8.76116276e-01 5.02552807e-01
7.04366922e-01 -1.21992195e+00 -1.85932264e-01 9.50735986e-01
4.75474566e-01 -9.24404383e-01 -2.03387350e-01 -1.68496847e-01
-4.98937607e-01 9.82157648e-01 1.38020203e-01 -2.09500670e-01
9.64221001e-01 2.76124924e-01 2.25826547e-01 -2.66980857e-01
-6.85818315e-01 -7.53689051e-01 3.68974686e-01 7.95976579e-01
9.01253223e-01 3.22716475e-01 -9.70775783e-01 6.19464755e-01
-4.44542855e-01 -3.27462316e-01 3.51430327e-01 8.32106948e-01
-7.45332539e-01 -1.56878781e+00 8.67174417e-02 4.46408004e-01
-2.17372805e-01 -3.61666679e-01 -4.34499919e-01 7.43580222e-01
4.59327310e-01 1.11394811e+00 6.07278906e-02 -6.33663952e-01
2.63500273e-01 5.47798038e-01 4.11881506e-01 -9.71207023e-01
-5.21511614e-01 -6.18856668e-01 5.43208532e-02 -6.24015450e-01
-2.45821729e-01 -3.64299655e-01 -1.38065350e+00 -9.97875705e-02
-1.25975817e-01 3.68478954e-01 4.02329206e-01 1.27569902e+00
3.05549175e-01 2.80855626e-01 1.46736503e-01 -1.11463182e-01
-5.73216081e-01 -1.10943389e+00 -9.50563490e-01 9.83957231e-01
2.10162982e-01 -9.25991058e-01 -2.76548415e-01 -3.57803345e-01] | [10.34162712097168, 8.893814086914062] |
19bace99-89d7-451d-b770-aff4397fef88 | improving-word-embeddings-through-iterative | null | null | https://aclanthology.org/2020.coling-main.104 | https://aclanthology.org/2020.coling-main.104.pdf | Improving Word Embeddings through Iterative Refinement of Word- and Character-level Models | Embedding of rare and out-of-vocabulary (OOV) words is an important open NLP problem. A popular solution is to train a character-level neural network to reproduce the embeddings from a standard word embedding model. The trained network is then used to assign vectors to any input string, including OOV and rare words. We enhance this approach and introduce an algorithm that iteratively refines and improves both word- and character-level models. We demonstrate that our method outperforms the existing algorithms on 5 word similarity data sets, and that it can be successfully applied to job title normalization, an important problem in the e-recruitment domain that suffers from the OOV problem. | ['Slobodan Vucetic', 'Nemanja Djuric', 'Shanshan Zhang', 'Phong Ha'] | 2020-12-01 | null | null | null | coling-2020-8 | ['word-similarity'] | ['natural-language-processing'] | [ 5.50830364e-01 6.58842735e-03 -7.26241112e-01 -3.37707072e-01
-5.66428602e-01 -5.34362555e-01 5.46802282e-01 6.12206936e-01
-8.96417320e-01 4.64247495e-01 4.54462677e-01 -4.77240682e-01
-2.53479451e-01 -8.67332935e-01 -2.72231162e-01 -2.86834747e-01
4.48556930e-01 7.61405885e-01 1.50183961e-02 -4.02526170e-01
4.09872860e-01 7.55496442e-01 -1.53417313e+00 -7.01419860e-02
6.43701255e-01 3.56965929e-01 1.16590083e-01 7.83937812e-01
-8.38898122e-01 1.92300379e-01 -6.02329910e-01 -6.30865455e-01
9.64878201e-02 -2.24520326e-01 -8.59019399e-01 -4.31546032e-01
8.17257583e-01 1.38083965e-01 -6.88276291e-01 1.19465101e+00
9.80651319e-01 6.36570036e-01 1.07972610e+00 -1.08526611e+00
-1.18650234e+00 5.72103024e-01 -2.77178496e-01 5.60590267e-01
2.02542618e-02 -2.77113408e-01 1.38271892e+00 -9.37575817e-01
7.55599499e-01 1.21938872e+00 6.62610173e-01 7.09397256e-01
-1.26027775e+00 -6.08896017e-01 -2.17511490e-01 1.19596899e-01
-1.48475838e+00 -2.24293545e-01 4.95190889e-01 -4.66739088e-01
1.36688256e+00 3.09940159e-01 2.09234238e-01 1.03958058e+00
4.62084174e-01 3.24976206e-01 5.49196184e-01 -8.58227074e-01
-3.97219248e-02 1.66047826e-01 7.29391098e-01 4.27010804e-01
5.78879535e-01 -1.07590713e-01 -2.46107236e-01 -3.65069449e-01
5.51381111e-01 2.07519591e-01 8.41699615e-02 -3.01958591e-01
-9.31497753e-01 1.26736403e+00 3.18492092e-02 6.40069425e-01
-2.19938099e-01 2.47922674e-01 4.80225176e-01 2.80900717e-01
5.41147649e-01 9.40127552e-01 -2.95800447e-01 -2.41098270e-01
-9.25882101e-01 4.96212333e-01 7.89002836e-01 7.85191715e-01
5.25196970e-01 -5.12349606e-02 -7.99973667e-01 1.25228572e+00
1.87022358e-01 3.78680021e-01 9.67639029e-01 -7.46297359e-01
1.19359687e-01 3.55191410e-01 -9.69874859e-02 -1.03662014e+00
-3.03268373e-01 -4.95750308e-01 -6.62731171e-01 -1.08754814e-01
1.49374604e-01 2.27756530e-01 -1.14508593e+00 1.61565804e+00
1.03224799e-01 1.78514004e-01 -3.30680907e-02 4.30620372e-01
1.12482166e+00 7.08637118e-01 1.79586917e-01 -8.92762169e-02
1.71180439e+00 -8.48043084e-01 -1.29518878e+00 -5.45560300e-01
7.92145967e-01 -7.90522754e-01 1.16334033e+00 -2.34233141e-02
-8.70891869e-01 -6.69256628e-01 -1.07925546e+00 -3.71132791e-01
-8.88917148e-01 -9.89890322e-02 4.64376122e-01 9.37019527e-01
-8.21694613e-01 7.82241344e-01 -1.78393528e-01 -5.31351388e-01
3.37793082e-01 4.14355278e-01 -4.88629967e-01 -2.95253724e-01
-1.61029637e+00 1.12360942e+00 7.19271243e-01 -4.18188483e-01
-1.15116581e-01 -8.89176130e-01 -1.12939966e+00 3.61983120e-01
-4.30956855e-02 -5.56579173e-01 1.11100793e+00 -4.12914515e-01
-1.19877779e+00 8.29581976e-01 -4.30435747e-01 -1.69007659e-01
-2.87539363e-01 -2.98699904e-02 -5.84258556e-01 -1.83495879e-01
1.07435517e-01 4.77506399e-01 7.94941485e-01 -8.30606103e-01
-3.41114104e-01 -2.32855648e-01 -2.55337834e-01 1.17560811e-01
-1.08761847e+00 2.50843197e-01 -5.75158715e-01 -1.00314116e+00
-3.21705610e-01 -8.82052481e-01 -2.28075758e-01 -3.05820644e-01
1.23695306e-01 -9.20888960e-01 2.04359099e-01 -5.88212907e-01
1.69601893e+00 -2.02947974e+00 1.19408190e-01 3.25846404e-01
2.33736366e-01 5.61707020e-01 -5.76777458e-01 4.80973691e-01
-3.46953660e-01 3.12376589e-01 -1.03077225e-01 -3.16290170e-01
3.03064942e-01 5.27923048e-01 -2.58984506e-01 2.38040388e-01
1.81225106e-01 1.06344986e+00 -1.09179294e+00 -5.86844862e-01
-2.96428911e-02 4.05415952e-01 -4.40222174e-01 2.42638648e-01
1.56754330e-01 -4.42912161e-01 -4.02515056e-03 1.69282272e-01
5.46570420e-01 -3.90107632e-02 2.61957288e-01 1.18615851e-01
1.47445738e-01 4.48167682e-01 -1.26382518e+00 1.63214469e+00
-5.06775439e-01 7.81686723e-01 -3.17196548e-01 -1.17584467e+00
1.06526577e+00 2.69581854e-01 2.48701230e-01 -5.15013456e-01
3.22560757e-01 1.59369484e-01 5.15529923e-02 -5.33364058e-01
1.03628516e+00 -3.82263005e-01 -3.45324248e-01 4.88758206e-01
6.56056225e-01 -1.35948434e-01 4.02202398e-01 2.73938417e-01
1.12838769e+00 -4.30800050e-01 5.37684500e-01 -1.04616687e-01
4.46676940e-01 -2.36473650e-01 5.05257785e-01 1.05889583e+00
-2.17976987e-01 7.97821701e-01 4.83723313e-01 -1.79286316e-01
-1.19715405e+00 -1.02203274e+00 -4.81104225e-01 1.63938832e+00
-2.00980037e-01 -6.63842380e-01 -4.89110261e-01 -7.36119092e-01
5.27414799e-01 1.03828001e+00 -7.26040125e-01 -4.99791086e-01
-5.11105955e-01 -5.54835916e-01 5.67584336e-01 7.31468439e-01
-5.62307179e-01 -1.23283947e+00 2.99213469e-01 3.70534271e-01
-1.16211325e-01 -8.97274852e-01 -7.21264243e-01 4.69985247e-01
-6.61137819e-01 -5.28811634e-01 -7.57996321e-01 -1.25183058e+00
5.78626990e-01 2.13139281e-01 1.18784952e+00 1.25096932e-01
-4.65807438e-01 3.26442063e-01 -3.80142927e-01 -6.35160387e-01
-4.90400761e-01 5.53628623e-01 4.00307149e-01 -2.77767807e-01
1.14864635e+00 -2.64288396e-01 7.17639849e-02 -1.67483479e-01
-1.19775486e+00 -7.16656446e-01 5.18886089e-01 1.17101192e+00
6.19062781e-01 -1.57287791e-01 7.47838318e-01 -1.10340238e+00
1.32948971e+00 -3.80010992e-01 -2.12983415e-01 3.62640321e-01
-9.24021244e-01 3.45037162e-01 3.40650380e-01 -7.88108945e-01
-4.88969803e-01 1.44799631e-02 -7.17384040e-01 -3.20648581e-01
-1.33617342e-01 4.77596045e-01 -7.57922530e-02 -9.60836709e-02
5.74399829e-01 1.52999699e-01 -7.35383034e-02 -6.07288122e-01
7.70575464e-01 1.21797001e+00 5.98372042e-01 -2.25040346e-01
1.04542470e+00 -3.66106406e-02 -3.42070848e-01 -9.91947532e-01
-7.45592356e-01 -1.02422893e+00 -8.70827198e-01 2.25795880e-01
9.02064919e-01 -5.35607278e-01 -4.40158546e-01 -2.29631409e-01
-1.39518607e+00 1.44361109e-01 -6.04308844e-01 5.81381321e-01
-1.65608376e-01 5.05490243e-01 -5.08389473e-01 -4.69773024e-01
-3.76301676e-01 -7.02344358e-01 8.47607791e-01 3.23289603e-01
-6.42683327e-01 -1.03200781e+00 5.06918252e-01 2.34819263e-01
4.51869637e-01 -4.89566982e-01 1.21959996e+00 -1.32460511e+00
5.03116369e-01 -6.70447648e-01 -8.60407948e-02 4.45715278e-01
9.86015424e-02 -2.35180721e-01 -7.71694541e-01 -2.69132018e-01
-2.50467390e-01 -2.71273553e-01 1.01809752e+00 2.29778662e-01
1.29272807e+00 -2.08410900e-02 -4.17455554e-01 5.38938880e-01
1.16051877e+00 -2.07405314e-01 5.58279634e-01 4.11436319e-01
6.02088451e-01 5.39304376e-01 4.34330672e-01 1.16851822e-01
-1.08723812e-01 6.52477980e-01 3.52489837e-02 -8.02161694e-02
-1.12515181e-01 -2.30461255e-01 6.85981438e-02 1.13011014e+00
2.09992528e-01 -5.29373169e-01 -7.88590133e-01 8.74023676e-01
-1.57451868e+00 -1.01290691e+00 -1.23722795e-02 1.79087722e+00
1.13303757e+00 2.61056781e-01 -2.67383844e-01 1.51040450e-01
7.53862083e-01 3.13618332e-01 -1.78674847e-01 -1.15482914e+00
-2.84966510e-02 9.29156005e-01 6.65930271e-01 4.86851126e-01
-1.09818375e+00 1.13565493e+00 7.23673487e+00 1.32670081e+00
-5.07076442e-01 4.06753093e-01 6.92397356e-02 -1.30592305e-02
-6.22645736e-01 -3.50431174e-01 -1.17464137e+00 2.34680161e-01
1.07336807e+00 -3.98794234e-01 2.86451548e-01 8.54581356e-01
-1.66624680e-01 5.68334877e-01 -9.88142133e-01 9.21679020e-01
5.52305758e-01 -1.37549937e+00 3.80883276e-01 2.15797499e-02
7.24935293e-01 -1.90398976e-01 3.17544565e-02 7.67669022e-01
3.01623732e-01 -1.51924431e+00 -3.87220830e-02 2.16042295e-01
1.00642955e+00 -9.10793781e-01 1.17929018e+00 2.22929046e-01
-9.78220463e-01 3.50631885e-02 -9.88539398e-01 3.57010178e-02
6.73551932e-02 4.49823081e-01 -7.14853644e-01 2.89274693e-01
4.35157299e-01 3.69887173e-01 -6.27710700e-01 7.60448277e-01
-2.58812279e-01 6.79719150e-01 6.07300084e-03 -3.37669492e-01
1.88574135e-01 -1.40594393e-01 3.35044473e-01 1.58244431e+00
1.66199878e-01 4.30548340e-02 -9.42805409e-02 7.06686616e-01
-5.54926813e-01 5.38569152e-01 -8.71447980e-01 -5.10636806e-01
5.69726050e-01 1.51547062e+00 -5.69135368e-01 -4.44838136e-01
-5.43445289e-01 1.09222794e+00 5.41934550e-01 2.54090428e-01
-6.43294871e-01 -1.35294855e+00 1.13617480e+00 -1.13887705e-01
5.80902040e-01 -1.09141700e-01 -1.18259765e-01 -8.59545410e-01
-3.01031381e-01 -7.41726398e-01 4.53329474e-01 -4.08049166e-01
-1.60416448e+00 4.11644936e-01 -1.57268792e-01 -8.05519879e-01
-2.15187192e-01 -9.20925498e-01 -5.63723087e-01 1.06500196e+00
-1.64542866e+00 -5.85247934e-01 6.72919378e-02 7.43363127e-02
5.31484008e-01 -3.59797865e-01 1.18840337e+00 4.85523283e-01
-5.52904248e-01 1.09171927e+00 7.32690096e-01 4.34546351e-01
1.05356205e+00 -1.20318818e+00 6.05684757e-01 5.42501211e-01
3.83744180e-01 9.80068684e-01 6.33346200e-01 -6.50763035e-01
-1.16128671e+00 -1.27270222e+00 1.95524597e+00 -6.36903286e-01
7.24491835e-01 -5.84365487e-01 -1.02932918e+00 6.30066454e-01
3.29539120e-01 1.61115453e-01 1.11679685e+00 2.75567293e-01
-2.81877071e-01 2.19156936e-01 -8.37320566e-01 5.49760997e-01
1.07649779e+00 -6.54315293e-01 -1.44524097e+00 5.29502213e-01
1.11907470e+00 -2.02650651e-01 -1.04156780e+00 3.39339823e-02
5.83107412e-01 -1.14736907e-01 1.33173466e+00 -1.27480352e+00
2.50367671e-01 2.17191026e-01 -7.77620450e-03 -1.44412613e+00
-8.69493842e-01 -4.17188406e-01 -1.14233308e-01 1.50337100e+00
4.05083984e-01 -5.40699899e-01 6.53799117e-01 3.21137369e-01
2.13984065e-02 -5.43688059e-01 -1.03989422e+00 -1.10466838e+00
5.94350696e-01 -5.12656152e-01 6.00882053e-01 1.18533051e+00
8.55714157e-02 6.24260485e-01 -2.43822888e-01 -2.05615878e-01
2.43515179e-01 -1.55058473e-01 4.15275216e-01 -1.56380641e+00
7.92552158e-02 -5.59383154e-01 -5.48859954e-01 -1.05837858e+00
7.52876937e-01 -1.54168677e+00 6.80780709e-02 -1.64545536e+00
2.11515069e-01 -8.62426981e-02 -7.08658218e-01 3.74422938e-01
-5.25805891e-01 5.66359460e-01 8.65437537e-02 -2.69217372e-01
-3.16647828e-01 5.29212296e-01 8.62410128e-01 -4.26697671e-01
-5.62477447e-02 -2.58783579e-01 -8.95778000e-01 3.94560575e-01
8.19930017e-01 -1.03922033e+00 -8.22413713e-02 -2.87429571e-01
2.90215284e-01 -6.07029319e-01 -2.03110114e-01 -5.63108325e-01
1.20556258e-01 -1.69039309e-01 3.67874026e-01 -5.86057305e-01
1.08929850e-01 -6.24494016e-01 -4.39749479e-01 4.44002926e-01
-8.47213686e-01 1.67567924e-01 1.03859298e-01 5.67701817e-01
-6.75413162e-02 -1.08862662e+00 4.88107800e-01 1.42227083e-01
-5.83717644e-01 2.61514425e-01 -6.30733907e-01 3.87643933e-01
6.48964465e-01 -2.22058639e-01 -1.03887707e-01 -1.20956011e-01
-4.67857540e-01 1.92932576e-01 1.46743461e-01 8.69119763e-01
7.01360941e-01 -1.76654768e+00 -7.77527630e-01 3.73718262e-01
5.47331333e-01 -6.39523029e-01 -2.26071373e-01 3.43431413e-01
-4.21890676e-01 5.97862303e-01 -9.40549672e-02 -8.90200287e-02
-1.39817452e+00 7.39600480e-01 -1.01357609e-01 -4.45797682e-01
-3.72148782e-01 9.05947745e-01 -2.94923604e-01 -9.48059738e-01
4.19283569e-01 -1.21071830e-01 -8.55127811e-01 3.22839856e-01
6.38564050e-01 3.11177522e-01 1.49100274e-01 -7.67363191e-01
-3.50761205e-01 6.15861118e-01 -6.47465065e-02 -4.53486517e-02
1.43930924e+00 7.37305507e-02 -1.72294945e-01 4.64766800e-01
1.62710702e+00 1.17550418e-01 -1.41160101e-01 -4.76505220e-01
8.74257684e-02 -4.97355223e-01 1.12689160e-01 -1.46067739e-01
-5.34216583e-01 1.04271841e+00 6.38875067e-01 -5.18545136e-03
4.69146490e-01 -8.59187469e-02 1.02713084e+00 6.09918952e-01
-4.01353896e-01 -1.44547009e+00 -2.02690493e-02 9.36022043e-01
4.30034369e-01 -1.10535455e+00 9.60514992e-02 -2.40925565e-01
-4.68107373e-01 1.41233671e+00 4.46075618e-01 -5.56607842e-02
5.95248163e-01 1.61724702e-01 -4.78243977e-02 -1.21108405e-01
-4.37036633e-01 -3.09247643e-01 6.50332272e-01 6.78787231e-01
6.77586198e-01 -1.64865136e-01 -1.00904691e+00 8.16847682e-01
-1.51581287e-01 -1.73574686e-01 4.07483190e-01 8.40673327e-01
-5.38087547e-01 -1.50361764e+00 -3.71538490e-01 8.69653344e-01
-6.28261149e-01 -5.15963852e-01 -4.46184248e-01 6.17316008e-01
1.80656657e-01 6.70741975e-01 3.87971252e-01 -3.68646562e-01
3.76583397e-01 6.74104929e-01 2.08329305e-01 -1.09041154e+00
-7.79773057e-01 -3.78060848e-01 6.50475025e-02 -2.52501398e-01
-4.43664566e-02 -3.73360932e-01 -1.14207971e+00 -3.09343845e-01
-5.64968884e-01 3.66528630e-01 6.72367394e-01 7.72427797e-01
2.88091928e-01 7.57153690e-01 3.55293900e-01 -5.38303852e-01
-8.85978222e-01 -1.20880163e+00 -7.95198798e-01 8.59238207e-01
1.06526710e-01 -5.93727112e-01 -4.09380674e-01 -1.84158459e-01] | [10.465869903564453, 8.758708953857422] |
8c928ec5-7df5-4421-9b49-32cba4d89e5e | macech-at-semeval-2021-task-5-toxic-spans | null | null | https://aclanthology.org/2021.semeval-1.137 | https://aclanthology.org/2021.semeval-1.137.pdf | macech at SemEval-2021 Task 5: Toxic Spans Detection | Toxic language is often present in online forums, especially when politics and other polarizing topics arise, and can lead to people becoming discouraged from joining or continuing conversations. In this paper, we use data consisting of comments with the indices of toxic text labelled to train an RNN to deter-mine which parts of the comments make them toxic, which could aid online moderators. We compare results using both the original dataset and an augmented set, as well as GRU versus LSTM RNN models. | ['Maggie Cech'] | 2021-08-01 | null | null | null | semeval-2021 | ['toxic-spans-detection'] | ['natural-language-processing'] | [-8.58958364e-02 3.71246934e-01 -3.43917668e-01 -1.56033531e-01
-6.52462482e-01 -7.65965521e-01 7.57309854e-01 2.18429968e-01
-3.48570347e-01 9.31103587e-01 1.27144814e+00 -7.89185286e-01
3.49238634e-01 -6.34270668e-01 -1.62403882e-01 -5.66222429e-01
2.89089587e-02 1.96126357e-01 -5.98554134e-01 -4.99218911e-01
3.62135500e-01 -1.17788240e-01 -8.36149037e-01 4.00768906e-01
9.52913582e-01 3.33869785e-01 -3.65813613e-01 3.63542020e-01
-4.08131719e-01 1.31583798e+00 -1.09315681e+00 -5.58075309e-01
1.33349627e-01 -5.47025502e-01 -7.74770916e-01 -8.64957199e-02
3.84063214e-01 -4.26988043e-02 -4.96514410e-01 9.40934718e-01
5.82910061e-01 3.07571709e-01 5.65216482e-01 -6.83563769e-01
-7.22481668e-01 1.34146464e+00 -5.24572492e-01 3.62935334e-01
5.42132795e-01 1.42804310e-01 1.19334948e+00 -5.20053446e-01
9.03776884e-01 1.57752907e+00 5.30420601e-01 6.21983349e-01
-1.24173808e+00 -6.97755218e-01 3.02427620e-01 -2.50941962e-01
-7.44049311e-01 -4.25594211e-01 9.37983215e-01 -6.65925384e-01
4.83127117e-01 1.76416755e-01 5.56864619e-01 2.06549907e+00
1.29331753e-01 9.19143558e-01 1.32894266e+00 -9.41248387e-02
-1.47457272e-01 2.83081770e-01 3.27737868e-01 2.24959925e-01
-5.00989556e-02 -2.18479261e-01 -3.62034142e-01 -7.28294313e-01
-1.20850921e-01 -1.12873793e-01 -3.04517955e-01 7.90023804e-01
-8.39642882e-01 1.51165032e+00 7.76470602e-01 6.62854612e-01
-4.78362232e-01 -1.11933701e-01 7.09675252e-01 4.92853999e-01
1.10096240e+00 7.67347157e-01 -1.13852046e-01 -4.54480648e-01
-6.22546315e-01 1.61151424e-01 1.25192797e+00 1.27103016e-01
5.37012935e-01 -2.92340107e-02 -3.58163774e-01 1.09004962e+00
2.80895025e-01 3.10614377e-01 5.16246378e-01 -7.81161010e-01
6.50046468e-01 8.03544879e-01 -8.78822897e-03 -1.18073893e+00
-5.04324973e-01 -5.27649224e-01 -7.90519476e-01 -1.91588193e-01
6.16631269e-01 -9.84676480e-01 -4.87241119e-01 1.55818188e+00
1.25293791e-01 -2.85352528e-01 -3.48750651e-01 5.52304268e-01
8.26911271e-01 9.64259744e-01 1.80372328e-01 -2.25991845e-01
9.98524487e-01 -7.04987586e-01 -1.04571807e+00 -1.93511561e-01
1.13669813e+00 -8.22472513e-01 1.06679273e+00 3.96292537e-01
-6.74851954e-01 1.83420815e-02 -6.67289078e-01 -3.35671484e-01
-5.82589805e-01 -5.62807322e-01 3.39402288e-01 7.92097688e-01
-9.44764376e-01 7.63771296e-01 -5.27409241e-02 -2.45310545e-01
4.37253594e-01 4.75145178e-03 -1.87866092e-02 3.26468945e-01
-1.72331929e+00 1.03478110e+00 -7.77465329e-02 3.20778549e-01
-4.52106446e-01 -5.03622651e-01 -6.96696579e-01 -3.00195485e-01
1.48403257e-01 -1.85252398e-01 1.45672119e+00 -1.20582342e+00
-1.43905962e+00 6.49733543e-01 2.28407606e-02 -2.47760803e-01
5.80272436e-01 -4.23473328e-01 -2.36903176e-01 -4.83464360e-01
1.22933172e-01 2.45137662e-01 6.94485068e-01 -1.04417729e+00
-1.96526393e-01 -3.51846635e-01 2.32312515e-01 2.64977306e-01
-5.91495275e-01 5.21193802e-01 3.59590739e-01 -4.55452174e-01
-4.46315438e-01 -8.96677494e-01 -4.64482129e-01 -5.70836782e-01
-1.16683304e+00 -7.26625085e-01 9.76698220e-01 -1.04472411e+00
1.52382624e+00 -1.88555861e+00 4.65119034e-02 2.41397426e-01
5.86388946e-01 2.19849154e-01 -7.40938261e-02 6.72550559e-01
2.95058899e-02 1.01726091e+00 3.64764869e-01 -2.85394311e-01
2.25107372e-02 8.42951089e-02 -2.20018744e-01 5.84677517e-01
-3.45448285e-01 5.20623147e-01 -9.49581265e-01 -1.09449349e-01
-2.69717366e-01 4.28993702e-01 -2.69118190e-01 2.82165743e-02
-3.66819173e-01 5.42832255e-01 -7.09840059e-01 4.84135330e-01
4.33780938e-01 -1.77280217e-01 3.18068653e-01 6.16334796e-01
-3.46958756e-01 1.35669971e+00 -2.75533885e-01 7.20583439e-01
-7.02254593e-01 1.21080554e+00 3.82003516e-01 -3.02966297e-01
8.14552069e-01 3.78992051e-01 2.55918428e-02 -6.71320617e-01
6.19885564e-01 4.07734737e-02 4.44765329e-01 -4.66063082e-01
5.75631797e-01 -1.75563723e-01 -2.65713364e-01 1.15882862e+00
-7.39147723e-01 3.92271310e-01 1.47910833e-01 4.57166433e-01
1.09863913e+00 -4.49090093e-01 2.61105180e-01 -2.26580381e-01
3.11909944e-01 -4.23355877e-01 4.19954181e-01 7.28818297e-01
-3.08382273e-01 1.77262858e-01 1.25754559e+00 -4.32664603e-01
-8.20359766e-01 -4.13426608e-01 3.96219455e-02 1.44715750e+00
-3.29070687e-01 -5.78250647e-01 -6.23263001e-01 -1.04183364e+00
-2.94664167e-02 9.54099834e-01 -8.98414731e-01 1.70113206e-01
-7.76612461e-01 -6.95744812e-01 4.30124611e-01 -6.86579645e-02
1.86952084e-01 -1.31057978e+00 4.68421318e-02 3.67795467e-01
-6.24561787e-01 -4.07777578e-01 -7.33909845e-01 1.76051393e-01
-6.01641476e-01 -9.69845653e-01 -5.68545341e-01 -3.82831305e-01
4.19985324e-01 3.34381849e-01 1.05117142e+00 4.53529000e-01
4.06516433e-01 -4.26856935e-01 -4.93838251e-01 -4.49176997e-01
-8.56296062e-01 5.80896676e-01 -2.41243169e-02 5.33284713e-03
3.52118880e-01 -6.12942100e-01 -3.71232748e-01 3.29848304e-02
-5.55057347e-01 -1.85478568e-01 -2.41881348e-02 7.21630216e-01
-7.31823087e-01 -5.28940797e-01 6.16919518e-01 -1.61603642e+00
1.28026009e+00 -1.12659991e+00 4.12002578e-03 -2.63144165e-01
-3.00607681e-01 -2.48885930e-01 8.93659651e-01 -4.17214245e-01
-1.03518450e+00 -6.26664817e-01 -3.57439876e-01 2.13939637e-01
2.40004972e-01 8.99338961e-01 -6.77743740e-03 2.58415192e-01
1.11744618e+00 -4.52195466e-01 -1.19392008e-01 -7.80532002e-01
3.42309445e-01 1.27217853e+00 -5.40591180e-01 -1.91559911e-01
8.43536973e-01 5.66558018e-02 -8.49982500e-01 -9.36144710e-01
-1.33646584e+00 -4.68959808e-01 -2.90744811e-01 -4.59282130e-01
6.61026001e-01 -6.50877416e-01 -6.90968633e-01 2.40132496e-01
-1.50693655e+00 -3.62211287e-01 2.02385858e-01 8.43508616e-02
3.06099504e-01 2.15083107e-01 -1.17128134e+00 -1.09583759e+00
-5.32756567e-01 -7.21409619e-01 3.22080582e-01 2.75307715e-01
-8.71726751e-01 -1.39060736e+00 3.54857117e-01 6.69186294e-01
5.16129971e-01 4.02707011e-01 9.12779152e-01 -1.06049848e+00
4.19277810e-02 -1.34436741e-01 9.81283374e-03 1.02952838e-01
2.11028084e-01 2.38608032e-01 -9.70869362e-01 -3.27219293e-02
-8.50296207e-03 -6.43549383e-01 8.67744029e-01 2.55120903e-01
9.24589157e-01 -1.15710521e+00 -2.37169787e-01 1.17360696e-01
7.28536069e-01 -1.89435836e-02 6.49231672e-01 4.28931862e-01
7.32532084e-01 9.59158957e-01 7.65195042e-02 5.17090440e-01
1.86212823e-01 2.78486371e-01 3.20236146e-01 -1.37410283e-01
3.76836985e-01 -5.89962244e-01 8.89954805e-01 1.05525064e+00
2.66427040e-01 -7.59797931e-01 -7.97520161e-01 4.22537267e-01
-1.63219059e+00 -9.91215765e-01 -5.70876598e-01 1.69449365e+00
9.37218010e-01 1.08951256e-01 3.83567184e-01 -1.99886151e-02
9.36557949e-01 8.87344003e-01 -3.62677574e-01 -9.65100348e-01
-3.43946666e-02 -2.10580721e-01 4.95542586e-01 7.21141636e-01
-1.18491626e+00 7.83958375e-01 7.15052795e+00 5.57191014e-01
-1.13387287e+00 3.76741350e-01 1.25408304e+00 -1.48331195e-01
-6.74050629e-01 -1.33575469e-01 -5.11962056e-01 6.77185178e-01
1.32302201e+00 -2.88215846e-01 3.01672637e-01 7.23248005e-01
8.77295554e-01 9.06045586e-02 -7.29849219e-01 4.90000963e-01
1.53870776e-01 -1.01975846e+00 -2.04222426e-01 4.23358440e-01
9.38081920e-01 3.13156068e-01 1.28163010e-01 4.78439778e-01
9.21118557e-01 -1.16595113e+00 3.84624302e-01 3.47453132e-02
2.50558585e-01 -4.78059858e-01 8.45727205e-01 5.48930049e-01
-1.33658350e-01 -1.99816987e-01 -1.08704038e-01 -6.67528808e-01
3.10821891e-01 8.50734651e-01 -1.01382029e+00 -2.38964744e-02
3.77670020e-01 1.00595510e+00 -4.01186973e-01 6.81661367e-01
-6.24646544e-01 1.13203895e+00 -2.74457872e-01 -6.28432691e-01
6.60181642e-01 -3.92269403e-01 7.99993277e-01 1.05826485e+00
3.18076536e-02 -2.31326908e-01 1.81044161e-01 7.72267342e-01
-7.99700618e-01 2.67544806e-01 -8.38262856e-01 -5.46372294e-01
4.01635915e-01 1.52221024e+00 -3.75261873e-01 -3.11396688e-01
-2.30483547e-01 5.00959396e-01 5.98716795e-01 5.39033055e-01
-4.88735646e-01 -7.44182318e-02 5.94347894e-01 3.70435148e-01
-4.59276080e-01 2.66709439e-02 -6.07272744e-01 -1.12121105e+00
-3.04864973e-01 -1.14658523e+00 4.38802898e-01 -2.62394994e-01
-1.72106433e+00 5.35060227e-01 -6.40069008e-01 -7.91384041e-01
-2.98104465e-01 -2.59575903e-01 -1.19479775e+00 9.95396674e-01
-1.22104096e+00 -7.44585454e-01 1.70077741e-01 -4.60240357e-02
5.34487963e-01 1.55510262e-01 5.36815166e-01 3.58347028e-01
-9.22469497e-01 5.33076823e-01 2.68885672e-01 4.15032625e-01
6.75156772e-01 -1.17719007e+00 4.91745949e-01 4.53323960e-01
-1.53571010e-01 8.00491452e-01 9.14011180e-01 -7.62700438e-01
-7.87246764e-01 -1.04206836e+00 1.59972882e+00 -6.16178989e-01
1.18349802e+00 -6.03575826e-01 -7.82074273e-01 7.99813449e-01
5.53663731e-01 -8.18395078e-01 1.05459452e+00 1.04216576e+00
-3.70653719e-01 4.44343418e-01 -8.67735982e-01 9.27303433e-01
9.86496091e-01 -7.54709899e-01 -5.26765108e-01 8.54019821e-01
6.51343346e-01 -2.39306800e-02 -5.82282662e-01 -4.56902742e-01
2.54171789e-01 -7.53676951e-01 2.57728249e-01 -8.78650129e-01
7.68441677e-01 5.10334134e-01 4.05656695e-01 -1.64346111e+00
-4.75370228e-01 -1.26997674e+00 5.35359001e-03 1.31762993e+00
1.09990668e+00 -5.25284708e-01 8.78622115e-01 7.55043149e-01
2.74868775e-02 -6.01685166e-01 -9.07679260e-01 -1.52802914e-01
5.50197721e-01 5.06780632e-02 -6.69565126e-02 1.37599993e+00
3.61085773e-01 9.18103993e-01 -8.46934199e-01 -7.22518981e-01
1.59967259e-01 -2.97614872e-01 6.19411051e-01 -1.31240308e+00
1.79033667e-01 -8.37821960e-01 2.80371755e-01 -1.01656747e+00
3.80752087e-01 -7.83004105e-01 4.01046537e-02 -1.55501580e+00
2.09831476e-01 -3.08262795e-01 2.83185765e-02 4.41721022e-01
-2.18460754e-01 1.80603534e-01 2.31633201e-01 1.77285776e-01
-5.24014592e-01 7.38096237e-01 1.25366473e+00 -3.52842987e-01
-5.87122917e-01 3.10885817e-01 -1.40731835e+00 8.51612747e-01
8.69655430e-01 -6.19467199e-01 3.21250036e-02 -1.85823981e-02
7.54837036e-01 -1.11415088e-01 -2.18605742e-01 -1.63933426e-01
-2.18134165e-01 -1.39726341e-01 4.37068306e-02 -5.01216114e-01
1.03924744e-01 -1.67066142e-01 -4.18311119e-01 4.75903571e-01
-9.74871516e-01 2.02966295e-02 -3.74668866e-01 3.89105797e-01
2.44778916e-02 -2.46598661e-01 6.04921222e-01 -1.90702677e-01
3.51116836e-01 -5.46162203e-03 -1.11574113e+00 3.00357789e-01
4.08107489e-01 -2.95601729e-02 -6.45921111e-01 -1.25419843e+00
-5.48295259e-01 3.72012675e-01 2.41282597e-01 5.91953814e-01
2.21995309e-01 -1.09372211e+00 -1.04708266e+00 -5.48452318e-01
-3.24851394e-01 -4.44206834e-01 1.37574553e-01 8.19206297e-01
-3.45192373e-01 2.05397859e-01 2.53388971e-01 2.20346287e-01
-1.15875673e+00 2.40748301e-01 2.81445414e-01 -4.93196011e-01
-6.01477087e-01 8.03735077e-01 1.35656044e-01 -7.13411570e-01
3.23642552e-01 -2.76683997e-02 -7.10134685e-01 6.02713823e-01
7.10181355e-01 5.34263849e-01 -3.44423413e-01 -6.78879201e-01
2.21460722e-02 -4.34915751e-01 -4.58784640e-01 1.60473231e-02
1.07734990e+00 -8.75851363e-02 -3.30704331e-01 8.50575030e-01
1.57814336e+00 3.86934519e-01 -5.76488197e-01 -3.60647172e-01
3.57167304e-01 -4.92311150e-01 4.53684255e-02 -8.53129208e-01
-8.76609504e-01 8.24037433e-01 -1.47591293e-01 7.86017418e-01
2.13996977e-01 -3.27666244e-03 1.17866027e+00 3.78789783e-01
-3.29493910e-01 -1.29439592e+00 8.84075686e-02 1.06429493e+00
1.08389568e+00 -1.18336952e+00 -5.72294220e-02 -2.06511751e-01
-5.83949387e-01 9.59923923e-01 6.38218701e-01 -1.27689438e-02
6.41831160e-01 -9.45072994e-02 6.41841650e-01 -3.01662177e-01
-1.04168940e+00 1.41503766e-01 -6.23873733e-02 2.20345363e-01
1.10115731e+00 4.13953774e-02 -9.03190613e-01 2.99525589e-01
-4.35395837e-01 -6.67223573e-01 1.09347630e+00 4.79324847e-01
-4.69152093e-01 -1.12663865e+00 -3.82776886e-01 8.63412678e-01
-9.39640701e-01 -2.65243948e-01 -1.36173499e+00 4.84896809e-01
-2.55304072e-02 1.29063368e+00 -1.35644317e-01 -5.66480339e-01
-1.04463093e-01 9.66344774e-02 -6.10096097e-01 -6.79661512e-01
-1.59103155e+00 1.46938145e-01 1.11997175e+00 -7.68638924e-02
-2.74969429e-01 -9.06260788e-01 -8.01141858e-01 -9.09214854e-01
-5.94052672e-01 5.26361108e-01 5.52732944e-01 9.70159292e-01
1.64602742e-01 2.63647020e-01 1.04870832e+00 -5.84339499e-01
-7.18086898e-01 -1.77395999e+00 -5.70763767e-01 4.74278659e-01
4.08384442e-01 -2.82110572e-01 -8.26750994e-01 -5.67472756e-01] | [8.761497497558594, 10.379647254943848] |
562f02f7-bb02-483c-a1c6-650b2f4fa929 | outcome-oriented-predictive-process | 1707.06766 | null | http://arxiv.org/abs/1707.06766v4 | http://arxiv.org/pdf/1707.06766v4.pdf | Outcome-Oriented Predictive Process Monitoring: Review and Benchmark | Predictive business process monitoring refers to the act of making
predictions about the future state of ongoing cases of a business process,
based on their incomplete execution traces and logs of historical (completed)
traces. Motivated by the increasingly pervasive availability of fine-grained
event data about business process executions, the problem of predictive process
monitoring has received substantial attention in the past years. In particular,
a considerable number of methods have been put forward to address the problem
of outcome-oriented predictive process monitoring, which refers to classifying
each ongoing case of a process according to a given set of possible categorical
outcomes - e.g., Will the customer complain or not? Will an order be delivered,
canceled or withdrawn? Unfortunately, different authors have used different
datasets, experimental settings, evaluation measures and baselines to assess
their proposals, resulting in poor comparability and an unclear picture of the
relative merits and applicability of different methods. To address this gap,
this article presents a systematic review and taxonomy of outcome-oriented
predictive process monitoring methods, and a comparative experimental
evaluation of eleven representative methods using a benchmark covering 24
predictive process monitoring tasks based on nine real-life event logs. | ['Marlon Dumas', 'Marcello La Rosa', 'Irene Teinemaa', 'Fabrizio Maria Maggi'] | 2017-07-21 | null | null | null | null | ['predictive-process-monitoring'] | ['time-series'] | [ 6.50066018e-01 1.48241147e-01 -5.55146821e-02 -4.38432962e-01
-3.39547306e-01 -3.79736513e-01 1.19067502e+00 9.67658699e-01
-8.57170373e-02 5.01929104e-01 3.04744065e-01 -3.12819481e-01
-7.25077152e-01 -8.13123405e-01 -3.59544642e-02 -3.51061136e-01
-3.47955048e-01 8.91393006e-01 3.88417184e-01 4.67153668e-01
4.65338707e-01 4.85001802e-01 -1.53643990e+00 4.92434978e-01
2.43907019e-01 1.17874467e+00 -2.30032846e-01 6.15674198e-01
-3.22298795e-01 1.38754356e+00 -5.28030336e-01 -4.45374340e-01
6.10556155e-02 -4.12886918e-01 -8.71583402e-01 3.40418905e-01
-4.32858020e-01 5.69277741e-02 -1.10820949e-01 7.43149161e-01
-1.12240516e-01 7.96954185e-02 5.43074429e-01 -1.66882038e+00
-1.80279657e-01 5.83886743e-01 -3.40401381e-01 5.07115901e-01
7.29079604e-01 4.16155726e-01 1.06643081e+00 -4.22682762e-01
6.42055571e-01 1.04678464e+00 5.87097585e-01 2.40873367e-01
-1.42847073e+00 -2.86476761e-01 3.53236079e-01 4.61044371e-01
-1.00038922e+00 -2.05988958e-01 4.82129633e-01 -4.25544977e-01
1.09339929e+00 3.86019647e-01 5.42075634e-01 1.06609201e+00
5.21713912e-01 6.45376980e-01 1.21185887e+00 -2.93749124e-01
5.64524114e-01 -1.80169538e-01 3.01450551e-01 -7.19850957e-02
4.47428972e-01 -3.28345969e-02 -5.95945954e-01 -6.79870427e-01
3.48219573e-01 6.52071059e-01 5.11180377e-03 -1.08430840e-01
-1.28586102e+00 3.66351575e-01 -4.70854491e-01 3.95184636e-01
-8.67805362e-01 -1.21121466e-01 6.22250795e-01 5.07687151e-01
3.29791844e-01 4.68006760e-01 -4.87309575e-01 -7.39525497e-01
-7.20721364e-01 5.88133514e-01 1.58600533e+00 8.17687273e-01
3.60972494e-01 -3.10941428e-01 -4.50564831e-01 2.37564057e-01
4.12534207e-01 -1.08509742e-01 4.06471938e-01 -8.05930674e-01
5.17606378e-01 8.85302842e-01 5.17589927e-01 -6.00803256e-01
-2.82991111e-01 1.26292512e-01 -6.52515650e-01 -2.58261114e-02
5.00806749e-01 2.38510594e-02 -5.21666169e-01 1.04163158e+00
-6.18699491e-02 1.41452298e-01 -2.18647584e-01 4.33351994e-01
4.38534692e-02 7.00973928e-01 5.49562156e-01 -7.59256601e-01
1.34523308e+00 -6.36193395e-01 -8.11267793e-01 -1.23259708e-01
-1.76325813e-01 -5.66221058e-01 6.00236833e-01 7.08059311e-01
-1.04024315e+00 -4.22228336e-01 -6.04856551e-01 7.96169937e-01
-6.36910349e-02 -5.86094439e-01 6.56433105e-01 5.20962477e-01
-6.90889657e-01 9.16153610e-01 -1.35688198e+00 -5.32907367e-01
2.16450244e-01 4.14158553e-02 -2.80624181e-01 -1.29516557e-01
-6.42593741e-01 8.09378862e-01 4.71576929e-01 1.78675894e-02
-9.10797060e-01 -4.93925035e-01 -2.91841000e-01 4.71706122e-01
6.97371125e-01 -3.81837726e-01 1.58124781e+00 -6.82847857e-01
-1.11926079e+00 4.55188304e-01 -3.91929001e-01 -5.82498372e-01
7.62250364e-01 -1.49775401e-01 -8.13828886e-01 -1.82080910e-01
-1.47581622e-01 -3.50423217e-01 6.23643875e-01 -9.29643393e-01
-1.42549324e+00 -7.05516458e-01 -3.51515740e-01 -4.35881838e-02
2.70642370e-01 4.79888648e-01 -2.05177665e-01 -2.95453668e-01
3.44542772e-01 -8.57556999e-01 -4.79503423e-01 -7.81311154e-01
-3.07517588e-01 -5.93570471e-01 5.53131342e-01 -4.59618747e-01
1.39110911e+00 -1.96512115e+00 -2.55280346e-01 2.53360987e-01
3.19393814e-01 -1.32921383e-01 3.40844601e-01 1.09716153e+00
-2.83973292e-02 3.84175211e-01 -1.99527159e-01 -3.54364753e-01
6.51013628e-02 2.42353931e-01 -6.72437847e-01 3.14705253e-01
3.60692859e-01 5.15227199e-01 -8.39401424e-01 -2.53007859e-01
3.59318525e-01 6.77547604e-02 2.40783334e-01 5.77537000e-01
-3.03800017e-01 6.10560954e-01 -6.07854545e-01 9.39777076e-01
1.55248106e-01 -1.66061506e-01 3.60732943e-01 5.37011802e-01
-2.09208727e-01 3.56291085e-01 -1.22061694e+00 7.57916093e-01
-5.09984642e-02 3.80638361e-01 -2.52921253e-01 -7.72013903e-01
9.65469718e-01 8.39856565e-01 8.72745275e-01 -6.08565927e-01
-3.08851570e-01 6.37150332e-02 -2.37332284e-02 -3.42030734e-01
4.26618308e-01 -4.34753388e-01 -7.32910931e-02 7.26191401e-01
-3.46051037e-01 2.28086799e-01 5.85196018e-01 -4.73792493e-01
1.98138642e+00 3.97864997e-01 8.63725245e-01 2.02162981e-01
6.30366564e-01 1.55131429e-01 9.15010273e-01 7.66949952e-01
-6.77154899e-01 3.37175936e-01 1.02838588e+00 -8.39516163e-01
-9.21283424e-01 -1.13106215e+00 1.88415229e-01 1.07135797e+00
5.02472371e-02 -6.09537423e-01 -1.11444958e-01 -4.85954285e-01
3.01091913e-02 9.09397006e-01 -5.98802030e-01 1.51469305e-01
-6.83591127e-01 -8.33583295e-01 1.86760351e-01 4.26786214e-01
1.95716053e-01 -1.63042653e+00 -9.61045682e-01 9.49627042e-01
2.10147966e-02 -1.09716928e+00 3.32979739e-01 3.36776674e-01
-1.24375463e+00 -1.45553124e+00 2.36458868e-01 -1.07685760e-01
1.73438311e-01 -1.64761662e-01 1.37870920e+00 -3.91574740e-01
7.93138221e-02 4.88970667e-01 -3.94090265e-01 -9.12729084e-01
-6.66201115e-01 -1.02598168e-01 -1.61235064e-01 4.26901489e-01
8.40145648e-01 -6.54402614e-01 -3.76837134e-01 3.39911610e-01
-8.35616469e-01 -2.53378242e-01 6.36562288e-01 2.15677738e-01
7.71538138e-01 3.00783515e-01 5.72640300e-01 -8.84260833e-01
1.07628584e+00 -6.63224995e-01 -4.08830971e-01 4.40262556e-01
-1.20639920e+00 -1.69350430e-02 5.37794769e-01 -2.74331063e-01
-1.33571315e+00 -1.40808627e-01 2.43064553e-01 -2.36275420e-01
-7.23015070e-01 5.62297404e-01 -6.94247261e-02 9.56137538e-01
1.35074899e-01 4.26876038e-01 -5.55967018e-02 -3.17053705e-01
-2.25428417e-01 2.48295009e-01 3.89489442e-01 -4.35074180e-01
4.94235009e-01 6.72951043e-01 1.11982673e-01 -3.38702261e-01
-2.85527378e-01 -9.27591681e-01 -3.75281990e-01 -3.52128237e-01
5.52604258e-01 -4.08556044e-01 -9.96634722e-01 2.69214571e-01
-9.09129381e-01 -5.69647960e-02 -6.08124316e-01 2.18475848e-01
-7.54788160e-01 5.88019975e-02 -8.79055440e-01 -1.19271088e+00
-4.43108022e-01 -9.06973183e-01 7.62903392e-01 5.91750778e-02
-9.39262629e-01 -8.75872433e-01 2.32504413e-01 2.47813389e-01
4.87591833e-01 5.18716991e-01 8.20954680e-01 -1.34216750e+00
-6.11001849e-01 -8.69917154e-01 1.44001037e-01 4.73550335e-02
3.85969907e-01 7.59370178e-02 -6.42183125e-01 -7.49150589e-02
4.70118701e-01 4.42709714e-01 2.13559285e-01 6.33675084e-02
1.03116751e+00 -3.25247765e-01 -6.05684161e-01 -6.06621690e-02
1.35490203e+00 7.07077861e-01 6.74355805e-01 7.85437465e-01
2.93647677e-01 8.85993838e-01 8.57589841e-01 8.13366354e-01
2.30829597e-01 1.68846637e-01 3.89433682e-01 7.70697773e-01
5.56631505e-01 -2.02900544e-01 2.37816736e-01 3.69763255e-01
-5.93886018e-01 -1.33353218e-01 -1.15213466e+00 4.58810478e-01
-2.04923916e+00 -1.50083971e+00 -4.87843007e-01 2.23187280e+00
2.84174472e-01 4.53823328e-01 9.45534259e-02 4.49585348e-01
6.41550064e-01 1.69216484e-01 -4.23668146e-01 -4.17500526e-01
2.79655993e-01 -1.17922738e-01 3.37312132e-01 -1.18605904e-01
-1.07135403e+00 8.48249048e-02 6.05032587e+00 2.66354561e-01
-7.05431759e-01 1.09041668e-01 7.30386734e-01 -4.39404361e-02
1.44560160e-02 3.49891126e-01 -8.05286288e-01 5.64438343e-01
1.47484851e+00 -4.53569114e-01 1.34366333e-01 8.17739129e-01
6.42438233e-01 -2.24292472e-01 -1.72318518e+00 7.79295921e-01
-3.80979091e-01 -1.17044973e+00 -1.57162473e-01 2.98269957e-01
5.78130484e-01 -8.40605702e-03 -5.84800005e-01 3.92072350e-01
3.83526325e-01 -1.11404908e+00 8.91601384e-01 9.43237126e-01
-1.02945073e-02 -6.34005189e-01 8.45722139e-01 5.56522667e-01
-1.20136237e+00 -6.10411167e-01 2.63042837e-01 -5.58460593e-01
4.76109177e-01 5.47931969e-01 -1.10210276e+00 5.84680676e-01
7.94668674e-01 5.25492907e-01 -2.17659056e-01 1.01670659e+00
-3.11333001e-01 1.05592787e+00 -4.00428884e-02 5.16504236e-02
4.82362509e-02 -3.38352323e-01 6.13561511e-01 1.05315936e+00
2.22881570e-01 -1.92805141e-01 2.70085990e-01 6.90434337e-01
3.77700984e-01 -5.23041636e-02 -3.56133252e-01 -7.74457753e-02
4.17909533e-01 1.14171922e+00 -8.47996593e-01 -3.67410719e-01
-5.89976788e-01 4.60921943e-01 -8.59052390e-02 2.19868287e-01
-5.61913431e-01 1.59281403e-01 7.05107868e-01 7.15762973e-01
8.70911926e-02 -1.66858122e-01 -6.20742202e-01 -7.07694948e-01
2.15873003e-01 -8.39238524e-01 7.26035833e-01 -4.67529565e-01
-1.66940904e+00 5.37995875e-01 -8.80315900e-03 -1.09732163e+00
-5.08452117e-01 -1.46347046e-01 -8.53239536e-01 9.26385343e-01
-1.13756788e+00 -7.37894058e-01 -3.50380063e-01 1.72221556e-01
7.36754954e-01 -2.37378348e-02 7.10561633e-01 -2.46086851e-01
-3.90674263e-01 -6.61450088e-01 -2.96050142e-02 -1.98465779e-01
5.74426413e-01 -1.37724352e+00 5.30029178e-01 6.91769302e-01
-1.11690357e-01 4.30140316e-01 9.77653146e-01 -1.01447725e+00
-1.49220622e+00 -1.19943261e+00 1.22965622e+00 -9.57536280e-01
9.94470239e-01 1.72117218e-01 -1.29648340e+00 1.00437450e+00
-8.23569205e-03 -2.28815511e-01 7.32787073e-01 1.50312454e-01
1.05737805e-01 -3.46540034e-01 -1.11200595e+00 4.30166215e-01
9.41819847e-01 -2.70674527e-01 -1.01517081e+00 -1.19304255e-01
2.53697991e-01 1.69446975e-01 -1.21812129e+00 4.80718702e-01
2.62259394e-01 -1.08002663e+00 5.23065925e-01 -5.90689123e-01
4.02084976e-01 -2.47126326e-01 2.66232565e-02 -9.65880096e-01
-3.97710502e-01 -7.68683493e-01 -5.28443635e-01 1.29430497e+00
1.21023893e-01 -7.25130975e-01 7.35540926e-01 1.10944581e+00
1.56169876e-01 -6.91318214e-01 -1.04134476e+00 -5.52620411e-01
-6.38223648e-01 -7.68589556e-01 9.00199056e-01 8.25311244e-01
2.79742569e-01 2.00119942e-01 -1.13951266e-01 1.07493907e-01
5.18684685e-01 2.81485766e-01 6.43631160e-01 -1.77119291e+00
-8.72638449e-02 -5.38980484e-01 -4.81587440e-01 -3.04075956e-01
-3.15266937e-01 -3.92699987e-01 -1.30613213e-02 -1.88133669e+00
3.43321025e-01 -1.68205395e-01 -5.30560553e-01 4.07534897e-01
-2.60541081e-01 -4.45590913e-01 2.21492454e-01 9.26493466e-01
-8.41884255e-01 2.60721684e-01 8.18795204e-01 1.28572285e-01
-3.73745352e-01 5.89127421e-01 -4.94625747e-01 8.13147843e-01
8.54862332e-01 -5.43761849e-01 -2.87688494e-01 3.35101962e-01
4.33693856e-01 6.38055861e-01 2.72596270e-01 -1.06189656e+00
4.24468756e-01 -5.52995086e-01 2.89814383e-01 -4.12927300e-01
2.38873437e-03 -1.04823816e+00 8.50120664e-01 7.05292165e-01
-5.29390335e-01 6.07472897e-01 -2.72016823e-01 1.18472564e+00
-7.03832209e-01 -2.02969238e-01 1.46761507e-01 -3.09813142e-01
-1.04509795e+00 4.73331183e-01 -6.96433127e-01 -2.46466473e-01
1.39195693e+00 -5.11778414e-01 4.88979463e-03 -2.18296364e-01
-9.45132434e-01 1.01263635e-01 1.40150696e-01 6.01305425e-01
2.73875475e-01 -8.04681301e-01 -7.75783122e-01 -1.07285254e-01
2.50332505e-01 7.66657740e-02 -9.55498964e-02 9.71308172e-01
-2.69952983e-01 4.69067514e-01 -1.64225161e-01 -4.14063007e-01
-1.14314198e+00 6.22441947e-01 4.82947156e-02 -9.40235734e-01
-7.55794764e-01 -6.98237047e-02 -2.79609799e-01 -2.90398914e-02
-3.72484960e-02 -4.64885622e-01 -3.06860924e-01 1.27908170e-01
6.82801664e-01 7.68150866e-01 1.46830142e-01 -9.63370651e-02
-3.16961050e-01 -4.84793723e-01 9.32637081e-02 -1.52397454e-01
1.45965159e+00 -1.74931809e-01 -3.80988717e-01 1.11062372e+00
2.44490415e-01 -2.66466975e-01 -1.51221907e+00 -3.28829110e-01
1.15076387e+00 -4.08899099e-01 -6.00504041e-01 -8.30122173e-01
-4.25050288e-01 3.69870275e-01 1.68548748e-01 1.01407099e+00
9.66990411e-01 2.19168961e-01 2.04266205e-01 8.93892720e-02
6.89720392e-01 -1.06217396e+00 -3.06768179e-01 5.73026478e-01
8.12659264e-01 -9.78502333e-01 -4.78578210e-02 -3.78151059e-01
-8.86738241e-01 1.01000333e+00 3.22956413e-01 2.04554200e-01
6.01335883e-01 1.75848633e-01 -3.21792394e-01 -2.59030581e-01
-1.23820329e+00 2.79927611e-01 -2.49108717e-01 5.86981714e-01
5.01008570e-01 4.87571806e-01 -3.65249574e-01 7.39980876e-01
1.80508662e-02 3.29327315e-01 5.32531202e-01 1.38093531e+00
-2.66357124e-01 -9.93077636e-01 -6.17842793e-01 9.62120414e-01
-8.63310993e-01 4.33680803e-01 -2.66097307e-01 6.67561054e-01
-8.61751661e-02 1.26467800e+00 3.24370146e-01 1.69059876e-02
8.96364689e-01 6.24162138e-01 9.59631279e-02 -6.11618876e-01
-7.85121620e-01 -2.23336250e-01 2.94234753e-01 -5.41233122e-01
-3.65514129e-01 -1.34443259e+00 -1.02141726e+00 -3.81233186e-01
2.48671368e-01 7.98917487e-02 5.06935835e-01 1.17898047e+00
1.99595764e-01 6.79217100e-01 4.43505913e-01 -4.84620780e-01
-9.83122289e-01 -1.23205614e+00 -6.05199456e-01 8.63548160e-01
-1.49104223e-01 -3.56433034e-01 -1.79226726e-01 2.72098362e-01] | [8.59414005279541, 5.9924468994140625] |
11604daf-67a1-4e78-815e-0248dafbe076 | a-graph-based-framework-for-complex-system | 2302.06473 | null | https://arxiv.org/abs/2302.06473v1 | https://arxiv.org/pdf/2302.06473v1.pdf | A Graph-based Framework for Complex System Simulating and Diagnosis with Automatic Reconfiguration | Fault detection has a long tradition: the necessity to provide the most accurate diagnosis possible for a process plant criticality is somehow intrinsic in its functioning. Continuous monitoring is a possible way for early detection. However, it is somehow fundamental to be able to actually simulate failures. Reproducing the issues remotely allows to quantify in advance their consequences, causing literally no real damage. Within this context, signed directed graphs have played an essential role within the years, managing to model with a relatively simple theory diverse elements of an industrial network, as well as the logic relations between them.\\ In this work we present a quantitative approach, employing directed graphs to the simulation and automatic reconfiguration of a fault in a network. To model the typical operation of industrial plants, we propose several additions with respect to the standard graphs: 1. a quantitative measure to control the overall residual capacity, 2. nodes of different categories - and then different behaviors - and 3. a fault propagation procedure based on the predecessors and the redundancy of the system. The obtained graph is able to mimic the behaviour of the real target plant when one or more faults occur. Additionally, we also implement a generative approach capable to activate a particular category of nodes in order to contain the issue propagation, equipping the network with the capability of reconfigure itself and resulting then in a mathematical tool useful not only for simulating and monitoring, but also to design and optimize complex plants. The final asset of the system is provided in output with its complete diagnostics, and a detailed description of the steps that have been carried out to obtain the final realization. | ['Gianluigi Rozza', 'Nicola Demo', 'Martina Teruzzi'] | 2023-02-10 | null | null | null | null | ['fault-detection'] | ['miscellaneous'] | [ 2.05063775e-01 4.99175638e-01 5.19120634e-01 7.19709098e-02
4.63362843e-01 -4.30498004e-01 6.46010339e-01 6.52017176e-01
-2.10187361e-02 7.28007734e-01 -7.23776639e-01 -2.80021220e-01
-7.57856131e-01 -1.21336544e+00 -4.61496443e-01 -7.63839364e-01
-4.93467510e-01 8.50204110e-01 3.45151931e-01 -4.68032777e-01
-8.96623209e-02 1.10342133e+00 -1.73079216e+00 -1.25613511e-01
7.76617110e-01 7.48555541e-01 4.06900018e-01 3.52927506e-01
-4.80612256e-02 6.25745773e-01 -8.47273886e-01 3.83678637e-02
-2.86005903e-03 -7.80871749e-01 -7.43197262e-01 6.42781258e-01
-6.08704507e-01 1.34025618e-01 2.40608245e-01 9.54588890e-01
-8.41232315e-02 -6.69589937e-02 8.17196846e-01 -1.30164170e+00
2.31824368e-01 8.74732375e-01 1.26392543e-01 -2.08328784e-01
4.69277948e-01 3.90462071e-01 5.43055058e-01 -2.55063802e-01
5.36201239e-01 8.96714449e-01 1.15083389e-01 6.50734678e-02
-1.47493505e+00 8.70327652e-02 -9.18808952e-02 2.38112062e-01
-1.27295601e+00 -2.57833377e-02 6.57423317e-01 -6.32174015e-01
4.45573330e-01 3.01913738e-01 1.00459957e+00 8.40613604e-01
3.54450345e-01 -6.65752068e-02 1.09701216e+00 -5.36291420e-01
5.46847582e-01 2.06389993e-01 4.50936146e-02 4.71962631e-01
7.34370291e-01 7.02999979e-02 2.76526749e-01 1.80159032e-01
6.29334569e-01 -7.87297487e-02 -5.10490894e-01 -4.58520830e-01
-7.13434041e-01 3.57098639e-01 5.08478999e-01 1.26459610e+00
-4.91655469e-01 8.82217139e-02 2.71747470e-01 3.04691970e-01
-9.95689407e-02 5.72297871e-01 -3.00383210e-01 8.68656784e-02
-5.52894413e-01 7.04799443e-02 1.17508006e+00 3.91427308e-01
6.95995629e-01 2.35565662e-01 4.36513573e-01 -6.82196692e-02
1.15716554e-01 1.04391515e-01 1.77313149e-01 -3.84430408e-01
-1.86186031e-01 1.04426861e+00 -5.15417233e-02 -1.13272226e+00
-6.06763780e-01 -8.23404610e-01 -1.00051510e+00 8.72213840e-01
3.97775978e-01 5.40020578e-02 -3.11514169e-01 1.59283900e+00
2.52088040e-01 1.26008645e-01 -3.77547778e-02 4.79501665e-01
-3.46967191e-01 5.19493163e-01 -2.14699343e-01 -5.08113980e-01
1.20467222e+00 4.41558566e-03 -6.67633116e-01 1.84070051e-01
5.08879423e-01 -4.79619116e-01 6.91362858e-01 9.33775246e-01
-9.27514493e-01 -4.49764401e-01 -1.22222722e+00 9.22121823e-01
-4.60979074e-01 1.57944426e-01 1.50844470e-01 4.97388780e-01
-1.07023013e+00 1.18426478e+00 -6.74070418e-01 -3.62592101e-01
-2.69169956e-01 2.55903631e-01 -3.82607341e-01 1.00416190e-03
-1.09230340e+00 1.12853932e+00 8.48301172e-01 6.91657245e-01
-8.66353810e-01 -2.62422413e-01 -5.37561238e-01 5.29761910e-01
6.49821579e-01 -6.33184612e-01 7.31469572e-01 -9.47705865e-01
-1.24412000e+00 3.32406491e-01 4.09817338e-01 -4.97360855e-01
9.40472841e-01 4.31340694e-01 -4.77826923e-01 1.77678734e-01
-3.19595456e-01 -7.43231699e-02 7.38974690e-01 -1.55187428e+00
-2.70116836e-01 -1.52037889e-01 2.34559909e-01 -2.95411468e-01
-2.40409255e-01 -5.49007654e-01 1.47513837e-01 -1.90977052e-01
1.31133601e-01 -6.27590418e-01 -3.06650311e-01 -1.93494126e-01
-5.41455150e-01 4.35944498e-02 6.58893168e-01 -3.01444709e-01
1.01755297e+00 -1.99190056e+00 6.95455194e-01 6.51011288e-01
9.24334452e-02 2.40847468e-01 2.98936456e-01 9.17439461e-01
-4.71117467e-01 -3.22645418e-02 -5.23284316e-01 5.41860843e-03
-6.57442287e-02 3.91379476e-01 6.72434717e-02 4.50266510e-01
3.69319469e-01 1.25164077e-01 -7.49103606e-01 -2.70686954e-01
5.55542469e-01 3.75701606e-01 -9.02277753e-02 4.99766581e-02
-5.09707272e-01 5.64413249e-01 -5.37654817e-01 1.49244845e-01
3.68169874e-01 1.28445432e-01 6.75146937e-01 -1.25778854e-01
-3.09170932e-01 -2.90299654e-01 -1.55370414e+00 8.90688181e-01
-5.40157020e-01 2.82569509e-02 3.48921925e-01 -1.26064825e+00
1.29252958e+00 6.93559587e-01 7.36959279e-01 -3.87237370e-01
4.91336673e-01 5.04650831e-01 2.31224652e-02 -2.75770366e-01
-1.28360698e-02 2.91673206e-02 1.51831135e-01 1.67396784e-01
-1.44642085e-01 -3.76970202e-01 6.62492514e-01 -1.00486740e-01
1.39755738e+00 -1.76408529e-01 3.44210267e-01 -3.69095623e-01
1.22103512e+00 -2.09620863e-01 1.02294683e-01 8.72600079e-02
4.20456171e-01 -1.13867812e-01 1.22180653e+00 -1.69553816e-01
-8.22265029e-01 -8.00191283e-01 -8.35111514e-02 -2.77242362e-01
-1.12533905e-01 4.30413298e-02 -6.93718672e-01 -4.76835608e-01
2.54608169e-02 7.19841301e-01 -6.26296759e-01 -4.23414826e-01
-3.35250467e-01 -3.03973466e-01 1.85656309e-01 -3.39368939e-01
4.35037851e-01 -1.01310861e+00 -7.43873358e-01 3.69660228e-01
1.59614414e-01 -7.01832175e-01 7.08221555e-01 5.52854478e-01
-1.13759840e+00 -1.47058141e+00 -2.12139025e-01 -3.80291730e-01
7.91256964e-01 -2.48322070e-01 9.67426836e-01 5.90381682e-01
-5.92623711e-01 1.43100053e-01 -5.58452547e-01 -2.61559300e-02
-1.26243246e+00 -5.64878248e-02 -2.22116522e-02 -1.78301707e-02
-6.03821158e-01 -8.18737745e-01 -1.56770825e-01 3.97331983e-01
-1.52568686e+00 -5.10212779e-01 7.02429235e-01 4.39162582e-01
4.13308442e-02 7.35520005e-01 5.50985515e-01 -7.49884963e-01
5.23695529e-01 -2.48837546e-01 -9.80313659e-01 2.18430191e-01
-8.51983190e-01 1.24156840e-01 1.12641323e+00 -7.33436123e-02
-7.12917268e-01 2.49391481e-01 -2.88210839e-01 -1.16955705e-01
-5.10669291e-01 5.70460439e-01 -7.10353255e-01 6.89038634e-02
6.17800653e-01 -1.37715284e-02 1.33456022e-01 -4.58384126e-01
2.17736781e-01 3.54888737e-02 2.63387799e-01 -4.65316832e-01
1.07340801e+00 5.70401587e-02 8.15411031e-01 -7.71035671e-01
2.03579783e-01 2.45137624e-02 -7.08239079e-01 -6.55367136e-01
4.39402580e-01 -1.46139398e-01 -1.10173094e+00 4.73211646e-01
-1.33333158e+00 -2.38413021e-01 -7.48631179e-01 1.78373590e-01
-4.03397828e-01 3.44382554e-01 -4.14607108e-01 -8.44729543e-01
2.75663316e-01 -1.19964755e+00 4.48953539e-01 -1.15126349e-01
9.96097624e-02 -1.01802886e+00 6.20725416e-02 -4.26400751e-01
1.33915260e-01 5.94574153e-01 1.08712101e+00 -4.68715340e-01
-6.37473404e-01 -4.60099280e-01 2.77501404e-01 6.19818151e-01
3.60855103e-01 4.74639654e-01 -5.68587124e-01 -2.93666750e-01
3.40600222e-01 4.61894423e-01 3.82704943e-01 3.36552970e-02
6.12548828e-01 -7.67130777e-02 -3.01340878e-01 -1.10457338e-01
1.86136389e+00 3.24802816e-01 7.15593815e-01 9.24111754e-02
7.20513612e-02 8.59707713e-01 7.42841959e-01 5.16081214e-01
-4.12290841e-01 9.07471180e-01 1.10932553e+00 -1.13163618e-02
2.20824137e-01 3.11785817e-01 3.98499399e-01 3.95065606e-01
-1.73149407e-01 -4.31537956e-01 -7.15792894e-01 1.31026521e-01
-1.54298794e+00 -9.64480996e-01 -7.39335716e-01 2.38544679e+00
2.62054831e-01 6.55971467e-01 2.78675072e-02 1.04351735e+00
8.56959224e-01 -7.37144724e-02 1.00583673e-01 -4.25284505e-01
1.03116915e-01 1.82881936e-01 9.84456614e-02 6.62605345e-01
-5.48491538e-01 1.09241351e-01 4.52333593e+00 3.42673451e-01
-1.08292842e+00 -2.47432262e-01 1.31952524e-01 4.87730116e-01
-2.04774827e-01 2.92626023e-01 -2.91551799e-01 4.47772384e-01
9.80340779e-01 -1.95260003e-01 3.34288239e-01 7.15588391e-01
4.67639238e-01 -4.32991058e-01 -1.10001647e+00 1.94611102e-01
-1.85401455e-01 -7.20341086e-01 -1.30822584e-01 3.02590609e-01
2.39272445e-01 -7.92888105e-01 -5.08246541e-01 -2.58926302e-01
-2.34888509e-01 -6.52822912e-01 8.27272236e-01 9.13042426e-01
3.61349791e-01 -8.31006765e-01 8.56061876e-01 4.93206710e-01
-1.19933546e+00 -2.15697899e-01 4.01884355e-02 2.64003128e-02
5.86804152e-01 1.16863084e+00 -9.36493039e-01 1.31595612e+00
8.31357203e-03 2.33596131e-01 -4.17864263e-01 1.02994812e+00
-3.42649072e-01 2.52370924e-01 -3.25583547e-01 -1.60597622e-01
-9.27386209e-02 -4.03676838e-01 7.61601567e-01 7.24045873e-01
6.21297657e-01 -4.35485929e-01 -1.08873434e-01 1.14341831e+00
4.49101001e-01 -5.22531569e-02 -7.42857277e-01 -1.97542063e-03
2.97451377e-01 1.30862057e+00 -1.25551105e+00 -1.40335560e-01
3.11609179e-01 8.25267315e-01 -1.39863342e-01 -1.95494965e-02
-8.00629616e-01 -4.79913741e-01 4.70520020e-01 3.96665812e-01
5.97384945e-02 -5.21456838e-01 -6.61910162e-04 -5.93560755e-01
5.24138063e-02 -5.61740756e-01 -1.31632000e-01 -6.74400151e-01
-7.33694375e-01 8.45510960e-01 3.47132189e-03 -1.28240538e+00
-6.12098634e-01 -6.17609859e-01 -4.77753699e-01 8.81906331e-01
-1.02986276e+00 -6.32768333e-01 -4.00679171e-01 4.66559142e-01
7.64017776e-02 3.95662248e-01 8.68393183e-01 4.43516880e-01
-7.31111169e-01 -4.78405744e-01 -1.64201722e-01 -4.28605467e-01
3.18633854e-01 -1.36164165e+00 -1.27204224e-01 1.20084918e+00
-6.87699243e-02 2.65448660e-01 1.18534601e+00 -8.50572288e-01
-1.19269478e+00 -6.15356445e-01 8.87405813e-01 1.18315041e-01
7.44324088e-01 -1.26667559e-01 -9.00876045e-01 4.61523443e-01
1.68744132e-01 -3.24618518e-01 -2.97272921e-01 -3.29125255e-01
5.14254212e-01 -5.01707137e-01 -1.13418400e+00 3.12421888e-01
4.83125806e-01 -5.93174249e-02 -3.52424473e-01 2.64085680e-01
4.57559764e-01 4.74150851e-03 -1.00844026e+00 2.74977297e-01
-5.05772009e-02 -1.46205020e+00 6.37253106e-01 9.65245962e-02
2.51292050e-01 -5.66976786e-01 6.95791245e-01 -1.36955130e+00
-2.17077166e-01 -3.83139253e-01 1.78474665e-01 1.39508212e+00
1.82295665e-01 -9.06409740e-01 3.55659276e-01 6.57492653e-02
-3.86293530e-01 -5.18320203e-01 -8.92660916e-01 -7.17007816e-01
-5.31952858e-01 -2.28403836e-01 5.14625072e-01 7.68301189e-01
9.38008875e-02 1.46424863e-02 2.50827730e-01 4.93749440e-01
4.64268833e-01 -1.70612484e-01 3.27367306e-01 -1.73282659e+00
-7.08498240e-01 -6.15397990e-01 -7.80468822e-01 -6.59359321e-02
-8.50628614e-02 -5.76793015e-01 1.16384467e-02 -1.51937044e+00
-7.49566495e-01 -4.34495330e-01 -1.93573870e-02 1.51761577e-01
4.52598363e-01 -2.80360103e-01 8.46233517e-02 1.09134562e-01
1.83513552e-01 2.61447787e-01 9.92814243e-01 2.03056499e-01
-2.63317507e-02 3.92354399e-01 -2.39366083e-03 4.74921107e-01
6.22776389e-01 -3.50953817e-01 -6.03582919e-01 2.73739725e-01
5.34336030e-01 4.28926975e-01 6.63337529e-01 -1.53839362e+00
1.09783679e-01 1.65057406e-01 1.99364230e-01 -2.33236611e-01
2.68922567e-01 -1.63107073e+00 9.44515765e-01 1.17982495e+00
1.11187279e-01 1.38566479e-01 -6.00166619e-02 5.02198875e-01
-3.27698857e-01 -8.56440306e-01 6.56578243e-01 -1.21145830e-01
-4.48066533e-01 -2.53624469e-01 -7.41745889e-01 -7.43900836e-01
1.39952672e+00 -1.06834963e-01 -1.64373651e-01 4.02727339e-04
-1.17577696e+00 9.62256640e-02 6.23031557e-01 -1.71955198e-01
2.49552488e-01 -6.28671408e-01 -2.66058683e-01 1.04430236e-01
-1.20073795e-01 -1.54145375e-01 4.78568584e-01 8.54267120e-01
-9.67211425e-01 2.91036010e-01 -4.89780545e-01 -4.63220298e-01
-1.14074576e+00 8.90152216e-01 5.57827830e-01 -3.74959320e-01
-2.74986178e-01 5.05138002e-02 -5.41406453e-01 2.11038470e-01
-1.31796688e-01 -7.91448712e-01 -4.10345376e-01 2.14639440e-01
-9.02855769e-03 4.37349230e-01 4.97385800e-01 -2.19215870e-01
-3.11366588e-01 3.53158504e-01 7.25746274e-01 -1.29609317e-01
1.19817305e+00 -6.26289621e-02 -5.58325529e-01 5.42336285e-01
5.21803141e-01 -2.23887525e-03 -1.01229656e+00 6.15446568e-01
2.08664089e-01 -6.04823753e-02 -1.53663605e-01 -7.14299262e-01
-9.27418113e-01 5.95442414e-01 3.98240358e-01 1.26955748e+00
1.37587893e+00 -3.00633818e-01 -1.04068168e-01 1.85638085e-01
8.14061105e-01 -5.44521630e-01 -1.25799567e-01 1.19975433e-01
8.56687605e-01 -2.39809468e-01 6.50901943e-02 -8.36843133e-01
-9.05973166e-02 1.77092481e+00 9.77276936e-02 -9.53222215e-02
6.09969735e-01 6.17847025e-01 -4.45784807e-01 -2.88326412e-01
-5.06442845e-01 -4.30792630e-01 -2.84737825e-01 3.82729769e-01
2.17055053e-01 -7.63012245e-02 -8.84737313e-01 2.24282011e-01
2.58627415e-01 3.11097145e-01 8.07481945e-01 8.53795886e-01
-4.78350222e-01 -1.70876086e+00 -6.76801801e-01 1.03307933e-01
1.49919137e-01 5.25386810e-01 -4.56316531e-01 1.20380533e+00
4.54970300e-01 8.75374496e-01 -7.99106359e-02 -3.85618627e-01
9.44329858e-01 1.96611568e-01 4.72572982e-01 -4.66443121e-01
-7.36101449e-01 -1.76988930e-01 1.65563509e-01 -5.02368271e-01
-2.88719833e-01 -6.08091056e-01 -1.17299390e+00 -2.51756787e-01
-4.51885521e-01 4.85113680e-01 1.08195329e+00 9.37629461e-01
2.82496624e-02 1.19532633e+00 8.90988886e-01 -8.70720625e-01
-6.00846052e-01 -8.42470407e-01 -1.14948606e+00 2.16923617e-02
7.31330551e-03 -7.00070918e-01 -5.89883804e-01 -1.96589053e-01] | [6.446704387664795, 2.3410909175872803] |
85c6daeb-7d7b-4076-8856-9dc14640333f | crime-prediction-through-urban-metrics-and | 1712.03834 | null | http://arxiv.org/abs/1712.03834v2 | http://arxiv.org/pdf/1712.03834v2.pdf | Crime prediction through urban metrics and statistical learning | Understanding the causes of crime is a longstanding issue in researcher's
agenda. While it is a hard task to extract causality from data, several linear
models have been proposed to predict crime through the existing correlations
between crime and urban metrics. However, because of non-Gaussian distributions
and multicollinearity in urban indicators, it is common to find controversial
conclusions about the influence of some urban indicators on crime. Machine
learning ensemble-based algorithms can handle well such problems. Here, we use
a random forest regressor to predict crime and quantify the influence of urban
indicators on homicides. Our approach can have up to 97% of accuracy on crime
prediction, and the importance of urban indicators is ranked and clustered in
groups of equal influence, which are robust under slightly changes in the data
sample analyzed. Our results determine the rank of importance of urban
indicators to predict crime, unveiling that unemployment and illiteracy are the
most important variables for describing homicides in Brazilian cities. We
further believe that our approach helps in producing more robust conclusions
regarding the effects of urban indicators on crime, having potential
applications for guiding public policies for crime control. | ['Haroldo V. Ribeiro', 'Luiz G. A. Alves', 'Francisco A. Rodrigues'] | 2017-12-08 | null | null | null | null | ['crime-prediction'] | ['miscellaneous'] | [-1.89762354e-01 -3.08162093e-01 -5.67080975e-01 -3.12539607e-01
-4.92866844e-01 -2.19419718e-01 5.95317125e-01 6.40516400e-01
-7.02630222e-01 8.36369455e-01 8.44853878e-01 -9.19098079e-01
-4.81377959e-01 -1.27012610e+00 -3.12091768e-01 -5.88525176e-01
2.15724781e-01 1.56068563e-01 -2.57791191e-01 -9.69046056e-02
4.74251062e-01 5.56206346e-01 -1.20762181e+00 -4.82023247e-02
1.23390639e+00 7.00299814e-02 1.18017197e-01 2.63418436e-01
-3.25262547e-03 8.78880024e-01 -4.61020082e-01 -4.25585508e-01
-1.52225778e-01 -2.66203582e-01 -6.55858517e-01 -3.61238360e-01
7.85304084e-02 -2.77570218e-01 -5.99270426e-02 8.05555403e-01
3.69581848e-01 -1.64736897e-01 1.10418940e+00 -8.67451668e-01
-6.99696064e-01 7.40100920e-01 -4.72596943e-01 4.08892781e-01
6.03844106e-01 -4.22524568e-03 8.19961488e-01 -7.27525890e-01
3.35500419e-01 1.08026111e+00 6.40873313e-01 -6.09400384e-02
-1.23142684e+00 -8.12439144e-01 -8.18378665e-03 3.93229008e-01
-1.43714106e+00 -5.97788513e-01 5.50175607e-01 -1.04011059e+00
8.02503884e-01 3.60512316e-01 6.77511871e-01 6.60613894e-01
1.38887644e-01 2.01501995e-01 1.31426418e+00 -5.29472709e-01
5.13503514e-02 2.68603355e-01 3.96618396e-01 3.06640595e-01
8.25835407e-01 1.22097358e-01 -2.32411101e-01 -2.51129657e-01
4.81795996e-01 2.83159971e-01 7.38215372e-02 3.19473475e-01
-8.98988605e-01 1.19197643e+00 3.52159232e-01 5.52109838e-01
-4.43515033e-01 -1.18107207e-01 1.88500986e-01 -2.92603076e-02
8.16546917e-01 3.87303799e-01 -2.19361827e-01 -4.24053162e-01
-9.84945536e-01 2.80752420e-01 2.80219823e-01 -4.07774374e-02
7.71372199e-01 -9.33282226e-02 7.98643753e-02 8.35851192e-01
9.55236107e-02 8.85247767e-01 -6.75620064e-02 -6.70912564e-01
8.23079050e-01 7.82178938e-01 -2.06843726e-02 -1.63115489e+00
-5.48221648e-01 -1.82682097e-01 -8.74859750e-01 1.96292162e-01
8.44090760e-01 -3.36359590e-01 -3.95774037e-01 1.44768417e+00
-5.13197668e-02 -1.96363121e-01 -3.60845804e-01 7.14020431e-01
3.95452470e-01 4.52657670e-01 4.15502459e-01 -3.19887757e-01
9.66956019e-01 -4.05782601e-03 -5.28527856e-01 -1.96524367e-01
7.88133204e-01 -4.79593843e-01 7.92919219e-01 7.97708929e-02
-7.54445255e-01 -3.64862949e-01 -1.78613827e-01 1.73033357e-01
-4.45735544e-01 -9.43510234e-02 8.70709240e-01 1.02653718e+00
-7.27171361e-01 4.77113366e-01 -8.66272926e-01 -3.61391187e-01
4.90458578e-01 1.57180324e-01 -3.05279374e-01 -7.21968636e-02
-8.77296329e-01 1.18259561e+00 4.85895388e-02 -8.98291245e-02
9.67149809e-02 -4.62142855e-01 -9.06896532e-01 -3.00643072e-02
-1.31834313e-01 -3.39922607e-01 2.48753771e-01 -7.81113684e-01
-4.41633135e-01 6.19054973e-01 -5.97088099e-01 3.78387123e-02
1.78724095e-01 2.30509460e-01 -4.94875848e-01 -2.12124974e-01
8.08573484e-01 -1.61965657e-02 3.74659449e-02 -9.36706781e-01
-7.12238729e-01 -9.78923321e-01 -2.12936569e-02 -7.48425126e-02
-3.91239703e-01 6.74377024e-01 3.40300620e-01 -4.71569568e-01
1.43091157e-01 -6.13653898e-01 -5.23174763e-01 -7.76463032e-01
-1.60340890e-01 -3.75126660e-01 2.78467506e-01 -8.67863774e-01
1.93088353e+00 -1.92511380e+00 -2.90829450e-01 3.93202990e-01
1.96697131e-01 1.19117208e-01 2.72203445e-01 4.09307241e-01
-1.07674643e-01 3.81448388e-01 -2.35816985e-01 6.61905389e-03
-2.07269609e-01 6.68988302e-02 -1.98740378e-01 7.11481869e-01
3.49265784e-01 7.69690812e-01 -1.01305747e+00 -5.64032197e-01
6.46452546e-01 5.06100059e-01 -6.26081288e-01 -3.77775759e-01
6.76411867e-01 3.89378339e-01 -4.23475295e-01 5.24518311e-01
6.73309326e-01 2.28356674e-01 -8.56014714e-02 6.47900820e-01
-6.41784549e-01 7.85112739e-01 -1.07268119e+00 6.55438244e-01
-2.65410334e-01 1.02892017e+00 -1.95379987e-01 -1.11158442e+00
1.04887354e+00 1.93121955e-01 4.98695105e-01 -8.29906821e-01
-1.17793813e-01 1.65480837e-01 1.48973331e-01 -5.95776081e-01
6.24135852e-01 -1.72299027e-01 -8.23469684e-02 5.65217197e-01
-5.60659051e-01 -9.88241099e-03 4.00994241e-01 -1.43860117e-01
9.19143200e-01 -3.76552969e-01 4.09941435e-01 -3.24840367e-01
3.07470232e-01 -3.94397266e-02 7.18906462e-01 5.97346485e-01
-8.50518718e-02 3.69610578e-01 7.46315658e-01 -7.31096566e-01
-8.26339245e-01 -8.97178233e-01 -6.52334511e-01 7.69009531e-01
-2.92166650e-01 -1.76119477e-01 -4.66560602e-01 -2.71359235e-01
-3.96067314e-02 1.00724018e+00 -4.96521592e-01 1.23879693e-01
-4.15700465e-01 -1.41566312e+00 3.39664251e-01 3.14175010e-01
2.27358326e-01 -8.60001147e-01 -7.22210407e-01 1.53453201e-01
-3.79993320e-01 -6.77551210e-01 3.75531584e-01 -5.94687723e-02
-8.56233299e-01 -1.41367567e+00 -3.85181367e-01 -1.49524659e-01
7.98377752e-01 4.76924509e-01 1.09019113e+00 3.35068792e-01
9.04446468e-02 1.74979381e-02 -4.58877474e-01 -6.18722320e-01
-2.23761305e-01 1.52027830e-01 -3.00361291e-02 -6.67413175e-02
8.58123541e-01 -6.37530386e-01 -2.82413572e-01 -3.09402831e-02
-6.10687435e-01 -1.90897658e-01 3.14873755e-02 2.56053448e-01
-1.01475213e-02 3.04985434e-01 6.03869140e-01 -8.64921987e-01
7.08481252e-01 -1.00543392e+00 -3.19562972e-01 -1.55545369e-01
-8.76463532e-01 -2.87610203e-01 4.62277383e-01 -2.87941024e-02
-9.20641303e-01 -2.58584946e-01 -1.04536086e-01 5.35154581e-01
-6.55957878e-01 6.98850036e-01 1.04479238e-01 4.20313507e-01
7.38514423e-01 -2.42865786e-01 -4.70953166e-01 -2.49937028e-01
-1.93841740e-01 6.88524067e-01 -1.12484165e-01 -6.78650141e-01
7.44895935e-01 5.49541414e-01 -2.05186363e-02 -1.01072121e+00
-6.33560598e-01 -6.52380884e-01 -8.18419099e-01 -3.30573916e-01
1.15020573e+00 -7.46992707e-01 -6.64060175e-01 2.36690924e-01
-1.24588680e+00 -9.91061106e-02 1.99151158e-01 9.08622324e-01
-2.86143366e-02 5.49003966e-02 -2.38771945e-01 -1.38579607e+00
1.30705535e-01 -1.16327679e+00 5.74313581e-01 5.26234023e-02
-7.01275766e-01 -1.24271798e+00 5.10852635e-01 6.74989104e-01
2.76258498e-01 6.68637335e-01 1.05071402e+00 -1.44730315e-01
-1.65384278e-01 1.38914213e-02 -4.19939488e-01 -1.87667146e-01
3.16795617e-01 2.51137972e-01 -8.63675475e-01 2.42539465e-01
-3.74037236e-01 2.51295954e-01 1.01619267e+00 7.59238660e-01
7.31009662e-01 -4.40601557e-01 -3.46429527e-01 2.99317211e-01
1.44246709e+00 2.06855610e-01 6.45050466e-01 6.25682831e-01
6.45220816e-01 9.92744386e-01 2.96903104e-01 4.38457668e-01
7.90720761e-01 6.31811559e-01 3.22507083e-01 -3.75524253e-01
2.71020800e-01 -1.31314635e-01 3.35849494e-01 4.98565406e-01
-7.97439516e-01 4.42175418e-01 -1.46954525e+00 8.22739363e-01
-1.63960910e+00 -1.60322666e+00 -1.04024899e+00 2.13641405e+00
4.55542982e-01 2.43236907e-02 4.73389983e-01 4.82052505e-01
5.75001657e-01 1.24625236e-01 1.74623594e-01 -7.01887846e-01
-2.21349940e-01 3.41957301e-01 6.33927584e-01 7.99238801e-01
-8.63801837e-01 7.16664672e-01 7.08968830e+00 4.10531014e-01
-9.57681596e-01 -6.15996905e-02 9.81310785e-01 4.90359589e-02
-4.64228302e-01 2.35600352e-01 -6.27768517e-01 6.91143274e-01
8.81454051e-01 -2.18524337e-01 2.54637480e-01 5.97176731e-01
1.13131344e+00 -5.58407605e-01 -4.10929173e-01 4.34326649e-01
-1.70050576e-01 -1.09748125e+00 -3.52725178e-01 5.57811618e-01
9.51798499e-01 -7.43733048e-02 -4.39224951e-02 1.43160552e-01
6.27713561e-01 -1.32981312e+00 6.20849371e-01 6.81083262e-01
7.81414628e-01 -1.08914876e+00 7.85970986e-01 5.14231086e-01
-8.94915342e-01 -2.67681301e-01 -3.92272085e-01 -1.25520372e+00
8.67111310e-02 1.06380200e+00 -8.31956625e-01 1.21640638e-01
6.99398160e-01 7.01315343e-01 -6.29642904e-01 7.73940265e-01
-3.45525771e-01 1.02862084e+00 -6.45932853e-02 -8.98247287e-02
2.02316418e-01 -6.48060322e-01 5.76907955e-02 1.22555172e+00
3.83558363e-01 3.29728097e-01 -3.30606818e-01 9.85168040e-01
5.61947048e-01 2.48537198e-01 -1.17910302e+00 2.09084615e-01
4.53697950e-01 7.51208723e-01 -7.48141229e-01 2.03510132e-02
-7.10176587e-01 1.83676302e-01 5.14707208e-01 2.91170299e-01
-5.79838932e-01 -1.29894361e-01 9.01111960e-01 5.15556157e-01
-3.35408449e-01 -3.89228523e-01 -8.84655178e-01 -1.14048815e+00
-1.67753324e-01 -4.85518336e-01 1.61688894e-01 -3.26553345e-01
-7.29695141e-01 1.61496643e-02 8.78049433e-02 -6.62215948e-01
-3.39427054e-01 -2.76808977e-01 -8.38023245e-01 9.91018772e-01
-1.33377123e+00 -8.39129806e-01 1.87402889e-01 5.38679600e-01
1.11077920e-01 -2.35141702e-02 8.76181483e-01 2.27494657e-01
-7.06231236e-01 9.63969976e-02 2.88765162e-01 4.52699214e-01
2.19066948e-01 -1.05596209e+00 7.05178380e-02 1.05543065e+00
2.03818381e-01 5.81043422e-01 5.57668269e-01 -8.43268394e-01
-4.44304764e-01 -9.30682003e-01 1.97732759e+00 -7.69447327e-01
6.98399603e-01 -4.06486727e-03 -4.95013833e-01 5.72744429e-01
-3.19502473e-01 -7.23207951e-01 9.53412592e-01 7.74064898e-01
-1.64090782e-01 -6.89092875e-02 -1.02671278e+00 6.57871783e-01
7.53632128e-01 -4.47170854e-01 -4.52412307e-01 1.61940262e-01
1.29044250e-01 4.39896911e-01 -7.98687875e-01 -5.52213825e-02
4.78030294e-01 -1.47913599e+00 8.47755611e-01 -4.30144697e-01
8.22246611e-01 1.74246728e-01 2.92404722e-02 -1.26832318e+00
-7.63418257e-01 1.95254892e-01 7.22726464e-01 1.25386572e+00
6.33848786e-01 -6.68841302e-01 7.54546583e-01 1.12079740e+00
2.83310682e-01 -6.49261355e-01 -1.10506022e+00 -4.12505239e-01
5.61422110e-01 -1.00545108e+00 7.17944086e-01 1.34879732e+00
3.84756863e-01 2.23692566e-01 -1.38913348e-01 1.43196220e-02
5.50083041e-01 -9.35269147e-02 8.52782428e-01 -1.37864709e+00
3.14593166e-01 -7.61591613e-01 -4.37927455e-01 -4.16965447e-02
4.19395745e-01 -6.66129649e-01 -7.19372392e-01 -1.74045765e+00
6.02826774e-01 -8.79455328e-01 1.07281893e-01 3.86556387e-01
-4.58463699e-01 1.10424973e-01 3.20737809e-01 1.48967309e-02
-1.30957395e-01 6.25452921e-02 7.76811421e-01 -2.00757027e-01
-4.88190234e-01 2.90294886e-01 -7.99429357e-01 9.96113241e-01
9.87922609e-01 -5.69893777e-01 -3.51591595e-02 -6.87366962e-01
5.89837193e-01 6.41979948e-02 4.01214302e-01 -8.42801988e-01
9.18812007e-02 -9.19639170e-01 3.79021704e-01 -3.54450285e-01
-3.18925991e-03 -9.29118037e-01 2.13444129e-01 4.77293342e-01
-4.87770773e-02 1.80184945e-01 -1.04068071e-01 -1.75660551e-01
-1.06487721e-01 -2.64613211e-01 3.22632849e-01 1.42093459e-02
-1.27582671e-02 1.33933708e-01 -7.90456355e-01 -1.39105976e-01
7.84484446e-01 -5.63758910e-01 -1.38880715e-01 -3.85354161e-01
-4.21862215e-01 -1.34557918e-01 4.77158308e-01 1.20283403e-01
4.98397827e-01 -1.27976787e+00 -1.12036049e+00 1.06676333e-01
-1.15203440e-01 -4.48784858e-01 -1.28574774e-01 9.28667545e-01
-4.27400291e-01 7.82504439e-01 -1.02370724e-01 -5.65551482e-02
-1.23146713e+00 4.02916878e-01 3.03546563e-02 -4.23372567e-01
-1.50755256e-01 1.50165722e-01 -1.14567071e-01 -3.52844030e-01
-2.73652494e-01 -6.83766827e-02 -5.61811209e-01 2.50809729e-01
5.35995901e-01 1.22992885e+00 -2.46030137e-01 -1.08308017e+00
-3.63049358e-01 5.65808773e-01 4.89274085e-01 -4.71754819e-02
1.72230291e+00 -1.89040646e-01 -4.07961220e-01 5.38532376e-01
9.18230534e-01 4.12745893e-01 -5.54167390e-01 2.27522999e-01
2.65028894e-01 -7.38254368e-01 -6.07189238e-02 -4.87966985e-01
-8.33294272e-01 8.01617205e-01 6.07411563e-02 3.36614281e-01
1.10743630e+00 -5.83352298e-02 2.34118819e-01 4.00336795e-02
3.77301604e-01 -1.16339266e+00 -6.05998099e-01 5.06110191e-01
5.19355059e-01 -1.48071694e+00 7.88004175e-02 -1.79338723e-01
-2.32238397e-01 1.00400031e+00 1.68905526e-01 -7.82149807e-02
7.54213989e-01 1.24014117e-01 -7.22931698e-02 -7.23086968e-02
-3.45200986e-01 -4.35145736e-01 -3.68942283e-02 7.67985404e-01
8.02115619e-01 6.84442461e-01 -9.83739376e-01 3.90865713e-01
-5.93480647e-01 -3.59543175e-01 6.57085299e-01 8.44012946e-02
-5.37933946e-01 -9.65406775e-01 -1.05095422e+00 7.36066043e-01
-8.50555897e-01 -3.59952688e-01 -4.73612130e-01 9.29199815e-01
5.35220265e-01 1.55463004e+00 3.58520895e-01 -2.18380213e-01
9.60017368e-02 -1.69584602e-01 2.22813841e-02 -4.62160677e-01
-6.21069133e-01 -4.00680840e-01 2.18512163e-01 -3.03932250e-01
-5.00059783e-01 -1.07187760e+00 -8.45522702e-01 -9.15290594e-01
-3.21854711e-01 2.67726094e-01 6.10241294e-01 1.20659840e+00
-1.10494316e-01 -7.60912895e-04 8.05148959e-01 -5.18150508e-01
2.05537662e-01 -1.02764285e+00 -6.18571103e-01 4.12263542e-01
2.46988460e-01 -5.19833922e-01 -3.95685792e-01 -1.37256235e-01] | [6.725536823272705, 1.983161211013794] |
a6ff8d33-dfb4-4cfb-bed7-58c926a18b92 | disentangling-aesthetic-and-technical-effects | 2211.04894 | null | https://arxiv.org/abs/2211.04894v3 | https://arxiv.org/pdf/2211.04894v3.pdf | Exploring Video Quality Assessment on User Generated Contents from Aesthetic and Technical Perspectives | The rapid increase in user-generated-content (UGC) videos calls for the development of effective video quality assessment (VQA) algorithms. However, the objective of the UGC-VQA problem is still ambiguous and can be viewed from two perspectives: the technical perspective, measuring the perception of distortions; and the aesthetic perspective, which relates to preference and recommendation on contents. To understand how these two perspectives affect overall subjective opinions in UGC-VQA, we conduct a large-scale subjective study to collect human quality opinions on overall quality of videos as well as perceptions from aesthetic and technical perspectives. The collected Disentangled Video Quality Database (DIVIDE-3k) confirms that human quality opinions on UGC videos are universally and inevitably affected by both aesthetic and technical perspectives. In light of this, we propose the Disentangled Objective Video Quality Evaluator (DOVER) to learn the quality of UGC videos based on the two perspectives. The DOVER proves state-of-the-art performance in UGC-VQA under very high efficiency. With perspective opinions in DIVIDE-3k, we further propose DOVER++, the first approach to provide reliable clear-cut quality evaluations from a single aesthetic or technical perspective. Code at https://github.com/VQAssessment/DOVER. | ['Weisi Lin', 'Jingwen Hou', 'Chaofeng Chen', 'Liang Liao', 'Erli Zhang', 'Qiong Yan', 'Wenxiu Sun', 'Annan Wang', 'HaoNing Wu'] | 2022-11-09 | null | null | null | null | ['video-generation'] | ['computer-vision'] | [-2.38825917e-01 -3.87785345e-01 -9.93864238e-02 -2.32310265e-01
-9.04745996e-01 -8.40458214e-01 1.37813196e-01 -1.47262439e-01
-2.12468356e-02 3.06898445e-01 4.93840337e-01 -2.16821972e-02
-1.82487458e-01 -5.57509363e-01 -5.11992216e-01 -6.75354838e-01
-7.78381675e-02 -2.45067313e-01 -3.94308344e-02 -2.82354534e-01
3.18284571e-01 8.93563628e-02 -1.56948185e+00 5.13653696e-01
1.11198628e+00 1.22372699e+00 -8.59464630e-02 9.21176374e-01
4.61733371e-01 5.97539008e-01 -5.02501130e-01 -1.19165719e+00
4.71723795e-01 -5.06957233e-01 -4.78648424e-01 2.11817488e-01
3.42476666e-01 -6.33996189e-01 -2.86188513e-01 1.33677185e+00
5.99317133e-01 -2.15225339e-01 6.13206923e-01 -1.66930711e+00
-1.15683794e+00 2.09988609e-01 -4.02546555e-01 4.71204743e-02
7.12664485e-01 6.61656797e-01 1.42833400e+00 -8.83175969e-01
4.70860928e-01 1.17940903e+00 2.20075324e-01 3.74088734e-01
-9.04308796e-01 -4.00459975e-01 2.68572103e-02 7.36326039e-01
-1.17313540e+00 -3.47099364e-01 9.46187854e-01 -5.49834967e-01
2.22710595e-01 5.35270870e-01 8.40512335e-01 1.31357217e+00
1.52140468e-01 8.78537118e-01 1.15299451e+00 -8.48330855e-02
5.41528404e-01 3.57049704e-01 -2.93935955e-01 4.85949397e-01
3.00203413e-01 3.15069616e-01 -5.79185605e-01 -6.71908539e-03
6.38598680e-01 -3.61941278e-01 -6.76655412e-01 -3.28793049e-01
-9.08993721e-01 6.53353095e-01 6.66630864e-02 -8.83130804e-02
-2.78363675e-01 7.05011860e-02 4.84588623e-01 7.25367129e-01
2.92286932e-01 6.49484694e-01 -4.33183998e-01 -5.61820567e-01
-7.59071767e-01 1.53350234e-01 6.03426576e-01 1.13811386e+00
4.25329208e-01 6.41563758e-02 -1.85142487e-01 7.13338852e-01
6.08352661e-01 1.05051315e+00 -3.22766751e-02 -1.50049078e+00
4.32800919e-01 4.45412308e-01 3.38566095e-01 -1.51845622e+00
1.88173458e-01 -4.48158443e-01 -5.13500333e-01 5.87287962e-01
1.35415614e-01 8.91943797e-02 -2.91081935e-01 1.47274625e+00
1.76231656e-03 -4.24534619e-01 -1.40168026e-01 1.50163686e+00
9.06561792e-01 7.11892247e-01 -1.63099781e-01 -3.80248487e-01
1.48621392e+00 -8.19504917e-01 -1.07018471e+00 2.83779114e-01
2.30847478e-01 -1.01202059e+00 1.41716886e+00 9.08649206e-01
-1.35599673e+00 -5.13112903e-01 -1.31054354e+00 -1.82650844e-03
-6.50517792e-02 2.59499311e-01 -1.60555274e-03 1.05907691e+00
-1.14073288e+00 4.71793175e-01 -3.28292340e-01 -1.27701778e-02
3.80253226e-01 -7.28091598e-02 -2.40298152e-01 -2.88960218e-01
-1.21865237e+00 7.73780406e-01 -2.27558583e-01 -1.13873258e-02
-1.21565723e+00 -6.20664656e-01 -5.63017666e-01 5.98300900e-03
6.26985550e-01 -5.28737128e-01 1.09263706e+00 -1.31699347e+00
-1.69335556e+00 4.92510736e-01 2.12967202e-01 2.19085023e-01
5.89454651e-01 -3.21562320e-01 -7.85239160e-01 6.57391608e-01
-1.18401192e-01 2.24034905e-01 1.03736269e+00 -1.70782685e+00
-4.69136685e-01 -2.58276880e-01 4.43665266e-01 2.36647546e-01
-1.46708906e-01 -2.60887351e-02 -6.66316509e-01 -6.89288437e-01
-1.80432066e-01 -6.65963113e-01 4.12080973e-01 4.14751500e-01
2.36376598e-02 7.97438323e-02 6.37496948e-01 -9.10912097e-01
1.49198449e+00 -2.11667895e+00 3.36224794e-01 9.40668955e-02
5.30613184e-01 3.05396259e-01 -4.87960488e-01 3.91058683e-01
7.11331293e-02 4.36752260e-01 3.59291464e-01 -1.77860156e-01
2.50592262e-01 1.34680979e-02 1.69871584e-01 4.97609317e-01
3.20283204e-01 8.36585522e-01 -1.18027592e+00 -5.86400628e-01
3.99809666e-02 2.84300089e-01 -1.04804242e+00 4.06052977e-01
-1.79796368e-02 2.01816648e-01 -2.82670379e-01 9.89748359e-01
8.54756594e-01 -1.07259072e-01 5.70969023e-02 -8.52042198e-01
-1.74434409e-01 -2.20068067e-01 -1.03323019e+00 1.41518009e+00
-3.39032978e-01 7.39081323e-01 1.25686720e-01 -5.74254990e-01
5.19255817e-01 7.41846740e-01 3.66436243e-01 -1.02283847e+00
3.39566052e-01 2.63331652e-01 -5.25906757e-02 -9.63806927e-01
5.23599684e-01 -6.31637499e-02 8.21447968e-02 1.46006525e-01
4.00358766e-01 -1.36960909e-01 2.94449180e-01 4.72346634e-01
7.68255532e-01 3.10781803e-02 3.36477220e-01 -2.42710501e-01
4.87606049e-01 -4.03132200e-01 5.99705875e-01 3.73851061e-01
-9.05460238e-01 8.57197583e-01 8.20150733e-01 -1.75182179e-01
-1.10774505e+00 -1.22362280e+00 3.24565887e-01 9.20400083e-01
4.84261453e-01 -6.15594387e-01 -7.98929155e-01 -5.90499580e-01
-3.59841615e-01 5.08704245e-01 -5.56056321e-01 -1.74816713e-01
1.32094175e-02 -6.39459938e-02 3.46982390e-01 2.21179038e-01
4.25864220e-01 -7.21377552e-01 -3.77726674e-01 -2.10580871e-01
-6.70568705e-01 -1.19640827e+00 -7.91296661e-01 -7.56708026e-01
-5.00214219e-01 -1.33241069e+00 -8.09766173e-01 -6.01275302e-02
2.87835151e-01 4.34251964e-01 1.33056009e+00 1.41439468e-01
5.99363260e-02 8.55527461e-01 -9.44190979e-01 -4.69069295e-02
-3.38946402e-01 -6.26516700e-01 1.70659140e-01 1.87029421e-01
6.88634366e-02 -4.91112649e-01 -1.02743685e+00 7.75710583e-01
-9.47337329e-01 1.61703993e-02 5.80168128e-01 3.21287721e-01
3.28226477e-01 1.95338115e-01 3.58173430e-01 -2.15855330e-01
5.60167491e-01 -3.48067552e-01 -4.91875708e-01 3.88154984e-01
-6.93117261e-01 -2.75287509e-01 5.59070706e-01 -3.00797284e-01
-1.04881930e+00 -7.71436036e-01 -1.09052807e-01 -5.56139469e-01
3.08563858e-02 3.49180371e-01 -7.62978852e-01 4.64497656e-02
5.98310351e-01 -7.62028396e-02 -2.17980027e-01 -1.10231400e-01
4.94394183e-01 6.72181547e-01 4.37484473e-01 -6.78027213e-01
8.78310502e-01 3.65443677e-01 -2.67891318e-01 -6.81277215e-01
-6.08323693e-01 -4.63786483e-01 -2.09121436e-01 -9.85054791e-01
1.03045166e+00 -1.06824291e+00 -9.80700076e-01 4.57740068e-01
-1.16082454e+00 -4.80628163e-02 -8.16905499e-02 4.23263311e-01
-6.24555945e-01 6.98421001e-01 -5.31957567e-01 -9.42655087e-01
-2.04796314e-01 -1.55160069e+00 9.98129010e-01 7.93532953e-02
9.01712943e-03 -8.79090726e-01 -1.37463853e-01 7.67528534e-01
2.92172164e-01 4.90677282e-02 7.91116774e-01 1.62690520e-01
-7.21013904e-01 -7.52048641e-02 -3.74166995e-01 8.07075977e-01
-7.04804882e-02 3.00610155e-01 -8.00821960e-01 -3.58086467e-01
7.49623999e-02 -3.31860304e-01 1.48130924e-01 2.59468764e-01
8.46714556e-01 -4.40666616e-01 7.71684647e-01 4.80991364e-01
1.68859208e+00 1.87380955e-01 1.05519128e+00 1.39385939e-01
4.39692438e-01 5.64212859e-01 8.40462089e-01 7.74071217e-01
2.39383504e-01 7.86504269e-01 7.62253881e-01 1.00526288e-01
-1.08318880e-01 -2.67353058e-01 7.92768359e-01 1.21317625e+00
-6.31881237e-01 -7.70526230e-01 -4.60020304e-01 2.45790258e-01
-1.50612962e+00 -9.66191590e-01 -3.08066100e-01 2.06141043e+00
4.88469064e-01 -1.48617486e-02 1.40037313e-01 4.03895766e-01
5.51739395e-01 1.07218310e-01 -2.69807547e-01 -3.85340631e-01
-1.78317577e-01 -3.96614134e-01 1.73212871e-01 4.06854272e-01
-6.50504529e-01 3.69568437e-01 6.01063538e+00 1.10292327e+00
-8.18484664e-01 1.61162168e-01 6.21965170e-01 -2.82035649e-01
-7.72381306e-01 -9.99613479e-02 -1.37365395e-02 6.50460660e-01
6.69656515e-01 -2.10693419e-01 6.07752979e-01 7.04213798e-01
8.23195636e-01 -1.59453705e-01 -1.01840866e+00 1.48047352e+00
1.46832377e-01 -9.85061407e-01 3.31502467e-01 1.70963332e-01
7.53470719e-01 -5.64508796e-01 5.76750338e-01 1.52849168e-01
-2.13246703e-01 -7.37468183e-01 1.10470402e+00 5.67145050e-01
9.62786376e-01 -5.16552448e-01 8.94011021e-01 -1.15524217e-01
-1.08358276e+00 5.23127727e-02 -1.89107224e-01 -1.64908059e-02
2.75028944e-01 7.01777518e-01 8.63847286e-02 7.89969683e-01
7.00312316e-01 6.49942756e-01 -6.39270008e-01 8.85307550e-01
-3.26971889e-01 7.84160972e-01 4.18869972e-01 9.88662690e-02
-5.32473736e-02 -4.31730568e-01 6.52240694e-01 8.01408887e-01
4.75484461e-01 5.55413723e-01 -2.93488890e-01 8.61408591e-01
7.34586865e-02 2.38565505e-01 -1.78310215e-01 -3.76490265e-01
2.52535999e-01 1.24430668e+00 -2.86692172e-01 -2.71788597e-01
-6.05837226e-01 1.07793164e+00 -1.97704926e-01 5.44690847e-01
-1.17528021e+00 2.90669594e-03 7.31181741e-01 7.61640742e-02
5.09711541e-02 -4.95897010e-02 -2.86832303e-01 -1.53513825e+00
2.18095168e-01 -1.39156699e+00 1.12405621e-01 -1.19450688e+00
-1.32451344e+00 5.19012094e-01 -1.65457189e-01 -1.90348613e+00
3.84865880e-01 -8.98881018e-01 -4.53921497e-01 5.08015275e-01
-1.41752875e+00 -8.00798059e-01 -4.51954275e-01 5.31128585e-01
5.70173740e-01 -2.70786174e-02 3.04576278e-01 5.95936477e-01
-3.99959803e-01 6.80602491e-01 -1.44085810e-02 -1.56685159e-01
8.89137924e-01 -1.14300585e+00 -1.59651488e-01 8.50721359e-01
-2.92439133e-01 2.89138317e-01 1.06825852e+00 -3.30506086e-01
-1.59451652e+00 -5.81098676e-01 3.86239499e-01 -4.18225110e-01
5.78393519e-01 -4.62589599e-02 -5.26123524e-01 -1.26220807e-01
3.54962289e-01 -2.19384328e-01 1.09090042e+00 -1.49810702e-01
-6.24490738e-01 -1.18232347e-01 -1.19143367e+00 6.93584919e-01
1.13300705e+00 -6.56559527e-01 -2.02209234e-01 -1.89552292e-01
8.82004499e-01 1.27650946e-01 -1.16110551e+00 2.29878768e-01
8.13247025e-01 -1.48664606e+00 6.98799253e-01 -1.72941029e-01
1.01985753e+00 -3.56977791e-01 -6.81243539e-01 -1.45909345e+00
-4.93032008e-01 -4.32238013e-01 -6.49768189e-02 1.14663744e+00
3.27701956e-01 -1.88427448e-01 3.78749222e-01 4.78141844e-01
-1.20449848e-01 -7.70913541e-01 -5.58365941e-01 -7.16927469e-01
-3.07172924e-01 -7.85182595e-01 4.00814295e-01 7.01769292e-01
2.13288873e-01 1.23704433e-01 -8.46521616e-01 2.97870398e-01
5.90211391e-01 -2.22344071e-01 5.25222778e-01 -8.81347775e-01
-5.12111425e-01 -4.58536834e-01 -6.43806815e-01 -1.04616857e+00
-5.76765239e-01 -2.89319277e-01 -1.24294870e-01 -1.31759560e+00
3.53961885e-01 1.98469609e-01 -1.88725322e-01 -2.64190972e-01
-1.87679574e-01 2.89317280e-01 6.22941852e-01 1.20839022e-01
-1.12344003e+00 8.93669248e-01 1.76532745e+00 -8.45543742e-02
2.14367375e-01 -3.23266923e-01 -7.77601004e-01 5.36661148e-01
5.68183184e-01 -8.85620713e-02 -5.98461688e-01 -5.76586783e-01
8.38311017e-01 3.91308367e-01 3.75827551e-01 -9.01065946e-01
-2.49481618e-01 -2.78236896e-01 4.25811186e-02 -3.09687734e-01
2.88238734e-01 -8.97531748e-01 5.36213480e-02 2.76670575e-01
-1.81148455e-01 1.97269976e-01 -1.85956821e-01 5.84541798e-01
-3.67799878e-01 -7.99724236e-02 7.68731534e-01 -1.15122825e-01
-6.87786698e-01 3.21488172e-01 -3.72055769e-01 1.95869938e-01
6.60686374e-01 -2.86536217e-01 -5.37893295e-01 -9.54964221e-01
-6.18084013e-01 2.61502415e-01 6.63806081e-01 4.35839653e-01
9.12484050e-01 -1.59844768e+00 -7.92499006e-01 -1.05772577e-01
4.12692457e-01 -9.11207736e-01 7.62141109e-01 9.08075809e-01
-6.17528439e-01 1.05874561e-01 -3.41035694e-01 -3.83437306e-01
-1.27673852e+00 8.40833008e-01 9.86379907e-02 4.38474976e-02
1.43840954e-01 5.58732450e-01 2.72692055e-01 3.41542028e-02
8.73778686e-02 -1.80793479e-01 -2.53314763e-01 1.26152828e-01
6.20726228e-01 8.05652857e-01 -4.45960201e-02 -8.72366548e-01
-1.77740380e-01 6.04400218e-01 4.42237794e-01 -3.55537266e-01
8.82687688e-01 -6.05120480e-01 1.64135367e-01 2.86825746e-01
1.55835950e+00 1.94542408e-01 -1.20853794e+00 2.04858184e-01
-4.15028632e-01 -1.00374377e+00 2.19657812e-02 -1.05663407e+00
-1.32665074e+00 8.35096717e-01 8.69362712e-01 2.64647931e-01
1.53296673e+00 -2.41453618e-01 6.38626754e-01 -1.80270895e-01
5.60985625e-01 -1.26071084e+00 7.50395298e-01 -4.90912497e-02
1.32571304e+00 -1.34288502e+00 -2.44560558e-02 -6.63492203e-01
-1.10661292e+00 1.01030350e+00 4.54588860e-01 6.27597272e-02
5.79146445e-01 -2.03244120e-01 3.44059825e-01 -2.45748863e-01
-8.95189643e-01 -1.07404120e-01 7.54863501e-01 6.89715683e-01
4.82047021e-01 2.88727373e-01 -5.14665067e-01 7.40232229e-01
-1.00902304e-01 -1.04658611e-01 8.86937737e-01 4.46955174e-01
-2.39126399e-01 -8.18709671e-01 -3.48738611e-01 -5.06933555e-02
-3.78727138e-01 7.97644481e-02 -1.53407782e-01 3.90318185e-01
3.36235493e-01 1.58081305e+00 -4.49215561e-01 -7.23727226e-01
5.42862296e-01 -5.29565156e-01 4.25548166e-01 4.53842580e-02
-1.45045474e-01 2.21615240e-01 1.33049026e-01 -1.08449745e+00
-5.47110975e-01 -3.19269300e-01 -5.85090756e-01 -5.70323408e-01
-3.19204897e-01 1.92537144e-01 5.60103893e-01 6.61649704e-01
2.08477914e-01 4.01681364e-01 8.55203867e-01 -6.67021036e-01
-4.68075991e-01 -5.52666724e-01 -6.99205279e-01 9.07275975e-01
3.53490591e-01 -5.58943272e-01 -6.27790749e-01 1.05189741e-01] | [11.756007194519043, -1.8192503452301025] |
548483db-bb36-4d7d-a5a3-47ec9692731a | energy-based-detection-of-adverse-weather | 2305.16129 | null | https://arxiv.org/abs/2305.16129v3 | https://arxiv.org/pdf/2305.16129v3.pdf | Energy-based Detection of Adverse Weather Effects in LiDAR Data | Autonomous vehicles rely on LiDAR sensors to perceive the environment. Adverse weather conditions like rain, snow, and fog negatively affect these sensors, reducing their reliability by introducing unwanted noise in the measurements. In this work, we tackle this problem by proposing a novel approach for detecting adverse weather effects in LiDAR data. We reformulate this problem as an outlier detection task and use an energy-based framework to detect outliers in point clouds. More specifically, our method learns to associate low energy scores with inlier points and high energy scores with outliers allowing for robust detection of adverse weather effects. In extensive experiments, we show that our method performs better in adverse weather detection and has higher robustness to unseen weather effects than previous state-of-the-art methods. Furthermore, we show how our method can be used to perform simultaneous outlier detection and semantic segmentation. Finally, to help expand the research field of LiDAR perception in adverse weather, we release the SemanticSpray dataset, which contains labeled vehicle spray data in highway-like scenarios. The dataset is available at https://semantic-spray-dataset.github.io . | ['Klaus Dietmayer', 'Daniel Meissner', 'Marc Walessa', 'Johannes Kopp', 'Vinzenz Dallabetta', 'Aldi Piroli'] | 2023-05-25 | null | null | null | null | ['outlier-detection'] | ['methodology'] | [ 1.89618208e-02 -2.64645278e-01 -4.46098372e-02 -6.71868443e-01
-5.60665429e-01 -5.20461559e-01 2.76094973e-01 4.80679244e-01
-4.28776652e-01 5.31911254e-01 -2.51449376e-01 -7.70394951e-02
3.32460463e-01 -1.09067762e+00 -1.07015300e+00 -5.49439967e-01
-1.01830430e-01 3.43343258e-01 6.37391627e-01 -7.41803125e-02
2.32916668e-01 6.76565051e-01 -2.00250125e+00 -1.97769418e-01
1.37558544e+00 8.48904312e-01 4.02372032e-01 3.66331488e-01
5.87407351e-02 8.68563652e-02 -5.64141273e-01 -7.65145943e-02
3.43658477e-01 5.81990071e-02 1.34072974e-01 -1.96797654e-01
7.46730387e-01 -2.90078670e-01 -7.76668787e-02 1.34768188e+00
4.13844168e-01 2.13784114e-01 6.44593120e-01 -1.75077021e+00
-2.42703229e-01 -1.51043177e-01 -6.07788384e-01 2.15620041e-01
2.14610890e-01 2.72008449e-01 7.47045398e-01 -1.22953916e+00
3.57284635e-01 1.24470437e+00 7.35286832e-01 2.41471067e-01
-9.42057669e-01 -9.15721416e-01 3.45109314e-01 6.31586611e-01
-1.31459785e+00 -4.36420023e-01 7.02860534e-01 -3.17016661e-01
6.87200725e-01 2.17772737e-01 6.56788290e-01 9.39337075e-01
2.03502491e-01 7.60699809e-01 8.44972312e-01 1.98683232e-01
7.34830856e-01 -2.05670521e-01 -9.12628919e-02 5.52478969e-01
8.20223689e-01 3.78258765e-01 -6.38502598e-01 -1.58328980e-01
-1.10767610e-01 2.99383968e-01 -1.49603635e-01 -4.71185178e-01
-7.81157970e-01 8.22276533e-01 9.31237817e-01 -3.11195701e-01
-4.18000937e-01 3.21480423e-01 9.88166928e-02 1.99047700e-01
8.17665398e-01 -7.09678754e-02 -2.85516053e-01 1.25825301e-01
-9.61314142e-01 4.91564810e-01 4.29951012e-01 1.00064909e+00
1.05310214e+00 2.77994256e-02 2.48901203e-01 7.00548708e-01
7.03312039e-01 1.23745513e+00 -1.71234682e-01 -1.04683089e+00
4.02001053e-01 4.22518492e-01 1.70522168e-01 -9.26611066e-01
-4.73914921e-01 -8.08753297e-02 -5.16438961e-01 6.55935287e-01
-6.12232760e-02 7.46167824e-02 -1.33666408e+00 1.49033165e+00
5.51515877e-01 9.49467003e-01 1.88255712e-01 9.94873941e-01
8.78257215e-01 5.98537862e-01 -6.40432015e-02 -1.17291488e-01
9.60311770e-01 -5.45033395e-01 -8.96113217e-01 -4.19173241e-01
5.04577696e-01 -6.33829176e-01 1.12160885e+00 3.94868881e-01
-6.22287750e-01 -3.59649241e-01 -1.24672711e+00 1.40395895e-01
-7.31203377e-01 -2.96624124e-01 2.75712430e-01 5.38596213e-01
-9.07897651e-01 3.83603841e-01 -1.02078485e+00 -5.06268620e-01
5.51333249e-01 1.15103230e-01 -7.32871064e-04 -4.04863060e-01
-9.11064029e-01 1.03145826e+00 1.19932003e-01 8.21573064e-02
-7.94204116e-01 -8.42392802e-01 -1.13197052e+00 -4.74627823e-01
5.18134177e-01 -4.26377922e-01 1.11549318e+00 -1.30673811e-01
-8.44429791e-01 5.98074973e-01 -7.00025678e-01 -5.96789062e-01
5.69002092e-01 -5.14884114e-01 -5.56616426e-01 -4.57574055e-02
4.79627222e-01 7.22404063e-01 6.86615109e-01 -1.73925650e+00
-7.67854035e-01 -6.75094187e-01 -3.37954938e-01 7.08640069e-02
1.09347120e-01 -3.32951546e-01 -4.56939012e-01 -3.68813008e-01
3.95481855e-01 -1.03414237e+00 -4.20266479e-01 2.39931092e-01
-2.71108478e-01 -2.42794707e-01 1.39793432e+00 -1.05857007e-01
7.11451113e-01 -2.23376679e+00 -4.46236908e-01 4.18156624e-01
2.10525468e-01 9.06511545e-02 -1.11502014e-01 1.50646091e-01
3.83518368e-01 2.02181205e-01 -6.04372025e-01 -5.97158253e-01
8.16018954e-02 7.79245734e-01 -4.75118428e-01 8.15213799e-01
4.09978718e-01 6.77299201e-01 -1.16193020e+00 -4.29518789e-01
6.09672010e-01 4.74428922e-01 -4.50121671e-01 1.12413839e-01
-3.73609394e-01 4.88485664e-01 -4.29770261e-01 1.13022554e+00
1.47525561e+00 7.64969468e-01 -5.24675131e-01 -6.48401529e-02
-2.24570721e-01 7.19132274e-02 -1.17194617e+00 1.45020270e+00
-3.52697760e-01 6.95295215e-01 2.10247114e-01 -6.26448631e-01
8.87253821e-01 -1.70198053e-01 4.73214149e-01 -9.95386004e-01
-1.37973741e-01 4.77415591e-01 -4.33614254e-01 -7.05544174e-01
7.19224274e-01 1.35356203e-01 -1.03666134e-01 -1.19049110e-01
-5.12633443e-01 -5.90559900e-01 1.20691314e-01 2.26061106e-01
1.01943088e+00 -1.66255925e-02 -1.49346039e-01 9.16569754e-02
3.40282582e-02 6.18872873e-04 9.61600602e-01 6.46141171e-01
-4.45323259e-01 6.25496209e-01 -1.67570710e-01 -1.13639168e-01
-7.31774628e-01 -1.45318258e+00 -3.69720608e-01 6.29027486e-01
6.90258443e-01 -2.88889408e-01 -3.57402742e-01 -6.65649891e-01
6.25413895e-01 9.63850677e-01 -3.38641912e-01 -2.70572633e-01
-4.73749906e-01 -7.79114783e-01 4.88787919e-01 6.02854192e-01
5.42867005e-01 -7.56216884e-01 -7.14924514e-01 -1.72670167e-02
-3.34925592e-01 -1.42572200e+00 -3.86764854e-03 2.53954083e-01
-9.22586143e-01 -9.74011838e-01 9.92805697e-03 -1.55871853e-01
5.19787431e-01 6.58196986e-01 1.07579648e+00 3.02387714e-01
-3.32823843e-01 3.17547530e-01 -3.04837763e-01 -1.21641374e+00
-1.17666079e-02 -3.86548936e-01 4.14026886e-01 -2.62605608e-01
5.28533936e-01 -6.59783483e-01 -6.84446037e-01 3.70837480e-01
-8.84472728e-01 -7.12247252e-01 2.75233358e-01 1.35286450e-01
8.47665131e-01 5.31237796e-02 2.79459894e-01 -4.78281438e-01
1.65138513e-01 -9.11677837e-01 -7.96012342e-01 -5.51927745e-01
-6.15455210e-01 -2.59136647e-01 2.48819456e-01 2.23904178e-01
-6.97908103e-01 3.83344114e-01 -9.90351215e-02 -5.53757429e-01
-4.62495089e-01 2.24424124e-01 -4.09046561e-01 -1.96583375e-01
5.90860069e-01 -2.55638212e-01 -4.06334549e-01 -3.49034637e-01
4.64882225e-01 5.66629469e-01 8.39253068e-01 -2.56291777e-01
1.38716388e+00 1.06468642e+00 1.49274826e-01 -1.01392055e+00
-8.26959193e-01 -8.60541821e-01 -4.95276660e-01 -5.15677452e-01
7.57811427e-01 -1.18188322e+00 -4.23199803e-01 3.78772587e-01
-1.02058184e+00 -2.14813158e-01 -2.55757272e-01 4.11841720e-01
-4.82847393e-01 4.62468296e-01 1.29598737e-01 -1.19893825e+00
2.15190098e-01 -9.45870280e-01 1.34580886e+00 2.82722771e-01
-8.50327387e-02 -6.62498593e-01 1.89750835e-01 3.05624664e-01
1.32095382e-01 5.95081210e-01 2.44552106e-01 -2.35766932e-01
-8.87893260e-01 -5.26985861e-02 -1.54674366e-01 1.37108311e-01
-2.68626846e-02 2.12799311e-01 -1.30076873e+00 -1.54923514e-01
-8.35901126e-02 -4.11565825e-02 1.48366070e+00 3.50651413e-01
9.77972150e-01 4.90043946e-02 -6.31583393e-01 7.21363485e-01
1.41248012e+00 -1.24328114e-01 6.25080764e-01 4.26553518e-01
6.93456531e-01 4.57410663e-01 1.28151929e+00 3.01388741e-01
7.07691491e-01 7.25152612e-01 1.36155999e+00 -2.58679569e-01
-3.52334231e-02 -1.63877845e-01 4.79274213e-01 2.02113241e-01
1.14403315e-01 -4.46291268e-01 -9.36441720e-01 8.80737841e-01
-2.15366793e+00 -8.66175532e-01 -7.87910759e-01 2.26870704e+00
2.11439416e-01 1.32015839e-01 -5.84706403e-02 6.74731284e-02
5.11413217e-01 2.65099972e-01 -7.89734542e-01 -3.82463872e-01
-1.88094825e-01 2.68202331e-02 1.00699341e+00 5.50962627e-01
-1.19528949e+00 9.97681618e-01 5.74329758e+00 4.16979581e-01
-8.46958995e-01 3.02804619e-01 -3.17515656e-02 -2.30443746e-01
-5.57582080e-01 -1.23555519e-01 -7.16887593e-01 6.95376277e-01
9.53545570e-01 1.05977401e-01 -5.74316569e-02 8.25905442e-01
8.47385406e-01 -6.10217929e-01 -8.34585667e-01 7.57688940e-01
-5.29301837e-02 -7.78994679e-01 -3.15241665e-01 1.14533558e-01
6.51284993e-01 8.39889407e-01 1.01494893e-01 5.10264710e-02
4.84174639e-01 -9.12269175e-01 7.02199519e-01 4.32360142e-01
3.18665475e-01 -7.94007003e-01 6.74286008e-01 3.39758068e-01
-1.40813005e+00 -8.97218585e-02 -6.82519615e-01 -8.42407271e-02
2.90608078e-01 1.32633722e+00 -8.26811492e-01 6.03858948e-01
1.15233338e+00 9.50545967e-01 -4.48256791e-01 1.54613078e+00
-5.10321736e-01 5.43589890e-01 -9.11167979e-01 2.49152303e-01
1.41850561e-01 -3.56433183e-01 9.21659172e-01 1.19027519e+00
7.14028656e-01 -5.56539707e-02 5.03620982e-01 7.54028857e-01
-9.60070416e-02 -1.63181692e-01 -1.23260856e+00 6.72338367e-01
7.49726236e-01 1.09831154e+00 -5.85877538e-01 -3.37561518e-01
-1.85954049e-01 8.21772635e-01 -7.14831948e-02 4.82043535e-01
-1.07499135e+00 -4.43648808e-02 1.20166671e+00 6.28026873e-02
2.45511502e-01 -5.40859282e-01 -6.71173453e-01 -1.09749103e+00
4.78467792e-01 -2.26225108e-01 9.53943580e-02 -8.61906946e-01
-1.11944032e+00 -1.49235232e-02 -7.18033910e-02 -1.53848112e+00
1.84666976e-01 -3.35890710e-01 -9.51996684e-01 4.87081945e-01
-2.24678874e+00 -8.23335111e-01 -9.25297201e-01 5.38936377e-01
5.17178953e-01 3.42224658e-01 4.49158937e-01 4.18561459e-01
-3.67459238e-01 3.78731713e-02 1.64527789e-01 -3.95661891e-01
8.37308228e-01 -1.33231139e+00 6.12979591e-01 1.22959709e+00
8.79242048e-02 -8.05855170e-02 1.06592846e+00 -1.03383696e+00
-1.24627662e+00 -1.83600736e+00 6.30524635e-01 -5.19046843e-01
4.09893990e-01 -3.50787491e-01 -9.71643984e-01 4.66008216e-01
-5.57262823e-02 2.37361029e-01 3.90924871e-01 -2.95739442e-01
-3.05299729e-01 -2.02504233e-01 -1.52726853e+00 5.44988036e-01
1.31965339e+00 -2.08304331e-01 -4.03904974e-01 5.11148930e-01
6.82278752e-01 -5.32315016e-01 -3.66435140e-01 1.08598638e+00
1.92573234e-01 -1.13419437e+00 9.46228147e-01 -2.12834142e-02
-7.63858706e-02 -8.82520556e-01 -4.51587200e-01 -1.60664380e+00
7.69982114e-02 -1.27353281e-01 -2.79025614e-01 9.67903256e-01
3.81604105e-01 -8.49417031e-01 7.33470976e-01 1.42290816e-01
-6.15959823e-01 -2.90397316e-01 -1.24477458e+00 -1.01020551e+00
-1.79684296e-01 -1.10030174e+00 6.11210942e-01 6.04463577e-01
-5.41657329e-01 -2.93339580e-01 2.94802012e-03 9.73377764e-01
1.03471720e+00 -2.58357693e-02 8.84650230e-01 -1.60563028e+00
6.24594152e-01 -1.31478384e-01 -7.04079628e-01 -7.47699201e-01
2.62247324e-01 -7.22644806e-01 6.71863675e-01 -1.64125311e+00
-1.72078997e-01 -4.96810585e-01 -2.15334341e-01 3.77807349e-01
-2.57945627e-01 7.67576516e-01 1.30178198e-01 1.26653075e-01
-6.49468303e-01 8.16731036e-01 5.65310001e-01 -3.26568276e-01
-1.12921767e-01 1.88236132e-01 -2.37924024e-01 1.05681002e+00
1.09338415e+00 -8.30206573e-01 -2.86389023e-01 -4.97401655e-01
3.19125503e-01 -7.56533504e-01 7.10480034e-01 -1.39258468e+00
1.79701254e-01 -3.23370188e-01 2.90486395e-01 -1.22431469e+00
5.05961835e-01 -1.09061587e+00 -2.03780264e-01 4.28558975e-01
4.05478954e-01 1.60428256e-01 3.45866829e-01 9.29999888e-01
-1.69382468e-01 -9.46410894e-02 7.69843876e-01 1.72404692e-01
-9.41451728e-01 4.63637292e-01 -5.92361987e-01 -5.69651723e-02
1.27733910e+00 -3.99813116e-01 -3.39305073e-01 -3.92163694e-01
-5.09850919e-01 9.25512135e-01 9.11941171e-01 6.28556788e-01
1.00959396e+00 -1.34711814e+00 -7.45189369e-01 2.65155584e-01
5.60517967e-01 3.00925195e-01 8.63197371e-02 8.47355008e-01
-2.30855867e-01 -1.50058717e-01 2.66998291e-01 -1.14671648e+00
-1.33883202e+00 1.96148768e-01 1.71765089e-01 4.43016231e-01
-6.32734597e-01 6.22154891e-01 -1.04207620e-01 -6.51476979e-01
2.95690596e-01 -7.74386406e-01 1.39583066e-01 9.56538469e-02
2.38624930e-01 5.08845508e-01 4.18856680e-01 -7.16667712e-01
-6.48011327e-01 8.20076823e-01 4.30827856e-01 2.33240183e-02
1.20776045e+00 -3.15516084e-01 9.18881074e-02 6.60797119e-01
9.90569949e-01 2.15795025e-01 -1.12753952e+00 -2.83064451e-02
7.10785910e-02 -6.19374156e-01 2.13372707e-01 -4.99273807e-01
-1.06199241e+00 7.73793161e-01 1.02777612e+00 2.07293823e-01
1.07538307e+00 6.81503788e-02 1.05624795e+00 5.39611518e-01
2.95626789e-01 -1.34589088e+00 -1.69856101e-01 6.30580246e-01
6.95023954e-01 -1.63407350e+00 -2.96076499e-02 -7.33629227e-01
-3.09988946e-01 8.14303815e-01 6.14463508e-01 -2.19825774e-01
6.94719136e-01 5.51118195e-01 3.42783004e-01 -4.51732159e-01
-5.30177534e-01 -7.45994627e-01 -2.91269124e-02 9.19033051e-01
-1.42965913e-01 2.82162964e-01 8.83595347e-02 -5.49664311e-02
-3.63613605e-01 -1.25113815e-01 7.15261996e-01 9.50030982e-01
-9.52697337e-01 -7.15426326e-01 -6.20434761e-01 3.59710902e-01
-5.25451116e-02 2.52829753e-02 -4.43343163e-01 4.14055347e-01
5.78511238e-01 1.34692359e+00 2.46020064e-01 -1.95076779e-01
7.51467168e-01 -5.83458357e-02 -9.87176434e-04 -5.03692567e-01
6.79667145e-02 -1.84722140e-01 6.26630783e-02 -1.11536825e+00
-5.89316785e-01 -7.22466469e-01 -1.67268157e+00 -1.49217054e-01
-2.44842127e-01 -4.54501174e-02 1.13593471e+00 7.80165970e-01
4.37808573e-01 3.30576688e-01 7.61979938e-01 -1.10266745e+00
-1.27131268e-01 -6.02506936e-01 -3.74736249e-01 2.31317043e-01
7.36707091e-01 -1.00617909e+00 -5.52283227e-01 -2.71148503e-01] | [7.839077949523926, -2.3771867752075195] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.