id
large_stringlengths 9
16
| submitter
large_stringlengths 3
64
⌀ | authors
large_stringlengths 5
20.2k
| title
large_stringlengths 5
245
| comments
large_stringlengths 1
482
⌀ | journal-ref
large_stringlengths 4
382
⌀ | doi
large_stringlengths 9
151
⌀ | report-no
large_stringlengths 2
509
⌀ | categories
large_stringlengths 5
78
| license
large_stringclasses 9
values | abstract
large_stringlengths 83
3.55k
| versions
large listlengths 1
21
| update_date
timestamp[ms]date 2007-05-23 00:00:00
2025-05-30 00:00:00
| authors_parsed
large listlengths 1
1.35k
| prompt
large_stringlengths 166
3.64k
| label
stringclasses 2
values | prob
float64 0.5
0.73
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2301.07836 | Vaishaal Shankar | Floris Weers, Vaishaal Shankar, Angelos Katharopoulos, Yinfei Yang,
Tom Gunter | Masked Autoencoding Does Not Help Natural Language Supervision at Scale | Accepted at CVPR 2023 | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by/4.0/ | Self supervision and natural language supervision have emerged as two
exciting ways to train general purpose image encoders which excel at a variety
of downstream tasks. Recent works such as M3AE and SLIP have suggested that
these approaches can be effectively combined, but most notably their results
use small pre-training datasets (<50M samples) and don't effectively reflect
the large-scale regime (>100M examples) that is commonly used for these
approaches. Here we investigate whether a similar approach can be effective
when trained with a much larger amount of data. We find that a combination of
two state of the art approaches: masked auto-encoders, MAE and contrastive
language image pre-training, CLIP provides a benefit over CLIP when trained on
a corpus of 11.3M image-text pairs, but little to no benefit (as evaluated on a
suite of common vision tasks) over CLIP when trained on a large corpus of 1.4B
images. Our work provides some much needed clarity into the effectiveness (or
lack thereof) of self supervision for large-scale image-text training.
| [
{
"version": "v1",
"created": "Thu, 19 Jan 2023 01:05:18 GMT"
},
{
"version": "v2",
"created": "Fri, 20 Jan 2023 22:26:21 GMT"
},
{
"version": "v3",
"created": "Tue, 25 Apr 2023 01:47:15 GMT"
},
{
"version": "v4",
"created": "Mon, 15 May 2023 17:05:32 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Weers",
"Floris",
""
],
[
"Shankar",
"Vaishaal",
""
],
[
"Katharopoulos",
"Angelos",
""
],
[
"Yang",
"Yinfei",
""
],
[
"Gunter",
"Tom",
""
]
] | TITLE: Masked Autoencoding Does Not Help Natural Language Supervision at Scale
ABSTRACT: Self supervision and natural language supervision have emerged as two
exciting ways to train general purpose image encoders which excel at a variety
of downstream tasks. Recent works such as M3AE and SLIP have suggested that
these approaches can be effectively combined, but most notably their results
use small pre-training datasets (<50M samples) and don't effectively reflect
the large-scale regime (>100M examples) that is commonly used for these
approaches. Here we investigate whether a similar approach can be effective
when trained with a much larger amount of data. We find that a combination of
two state of the art approaches: masked auto-encoders, MAE and contrastive
language image pre-training, CLIP provides a benefit over CLIP when trained on
a corpus of 11.3M image-text pairs, but little to no benefit (as evaluated on a
suite of common vision tasks) over CLIP when trained on a large corpus of 1.4B
images. Our work provides some much needed clarity into the effectiveness (or
lack thereof) of self supervision for large-scale image-text training.
| no_new_dataset | 0.709416 |
2301.08143 | Chunxu Zhang | Chunxu Zhang, Guodong Long, Tianyi Zhou, Peng Yan, Zijian Zhang,
Chengqi Zhang, Bo Yang | Dual Personalization on Federated Recommendation | Accepted as a regular paper of IJCAI23 | null | null | null | cs.IR cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Federated recommendation is a new Internet service architecture that aims to
provide privacy-preserving recommendation services in federated settings.
Existing solutions are used to combine distributed recommendation algorithms
and privacy-preserving mechanisms. Thus it inherently takes the form of
heavyweight models at the server and hinders the deployment of on-device
intelligent models to end-users. This paper proposes a novel Personalized
Federated Recommendation (PFedRec) framework to learn many user-specific
lightweight models to be deployed on smart devices rather than a heavyweight
model on a server. Moreover, we propose a new dual personalization mechanism to
effectively learn fine-grained personalization on both users and items. The
overall learning process is formulated into a unified federated optimization
framework. Specifically, unlike previous methods that share exactly the same
item embeddings across users in a federated system, dual personalization allows
mild finetuning of item embeddings for each user to generate user-specific
views for item representations which can be integrated into existing federated
recommendation methods to gain improvements immediately. Experiments on
multiple benchmark datasets have demonstrated the effectiveness of PFedRec and
the dual personalization mechanism. Moreover, we provide visualizations and
in-depth analysis of the personalization techniques in item embedding, which
shed novel insights on the design of recommender systems in federated settings.
The code is available.
| [
{
"version": "v1",
"created": "Mon, 16 Jan 2023 05:26:07 GMT"
},
{
"version": "v2",
"created": "Sat, 13 May 2023 08:23:17 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Zhang",
"Chunxu",
""
],
[
"Long",
"Guodong",
""
],
[
"Zhou",
"Tianyi",
""
],
[
"Yan",
"Peng",
""
],
[
"Zhang",
"Zijian",
""
],
[
"Zhang",
"Chengqi",
""
],
[
"Yang",
"Bo",
""
]
] | TITLE: Dual Personalization on Federated Recommendation
ABSTRACT: Federated recommendation is a new Internet service architecture that aims to
provide privacy-preserving recommendation services in federated settings.
Existing solutions are used to combine distributed recommendation algorithms
and privacy-preserving mechanisms. Thus it inherently takes the form of
heavyweight models at the server and hinders the deployment of on-device
intelligent models to end-users. This paper proposes a novel Personalized
Federated Recommendation (PFedRec) framework to learn many user-specific
lightweight models to be deployed on smart devices rather than a heavyweight
model on a server. Moreover, we propose a new dual personalization mechanism to
effectively learn fine-grained personalization on both users and items. The
overall learning process is formulated into a unified federated optimization
framework. Specifically, unlike previous methods that share exactly the same
item embeddings across users in a federated system, dual personalization allows
mild finetuning of item embeddings for each user to generate user-specific
views for item representations which can be integrated into existing federated
recommendation methods to gain improvements immediately. Experiments on
multiple benchmark datasets have demonstrated the effectiveness of PFedRec and
the dual personalization mechanism. Moreover, we provide visualizations and
in-depth analysis of the personalization techniques in item embedding, which
shed novel insights on the design of recommender systems in federated settings.
The code is available.
| no_new_dataset | 0.709818 |
2301.09071 | Juncheng Li | Juncheng Li, Siliang Tang, Linchao Zhu, Wenqiao Zhang, Yi Yang,
Tat-Seng Chua, Fei Wu, Yueting Zhuang | Variational Cross-Graph Reasoning and Adaptive Structured Semantics
Learning for Compositional Temporal Grounding | arXiv admin note: substantial text overlap with arXiv:2203.13049 | IEEE Transactions on Pattern Analysis and Machine Intelligence
2023 | 10.1109/TPAMI.2023.3274139 | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Temporal grounding is the task of locating a specific segment from an
untrimmed video according to a query sentence. This task has achieved
significant momentum in the computer vision community as it enables activity
grounding beyond pre-defined activity classes by utilizing the semantic
diversity of natural language descriptions. The semantic diversity is rooted in
the principle of compositionality in linguistics, where novel semantics can be
systematically described by combining known words in novel ways (compositional
generalization). However, existing temporal grounding datasets are not
carefully designed to evaluate the compositional generalizability. To
systematically benchmark the compositional generalizability of temporal
grounding models, we introduce a new Compositional Temporal Grounding task and
construct two new dataset splits, i.e., Charades-CG and ActivityNet-CG. When
evaluating the state-of-the-art methods on our new dataset splits, we
empirically find that they fail to generalize to queries with novel
combinations of seen words. We argue that the inherent structured semantics
inside the videos and language is the crucial factor to achieve compositional
generalization. Based on this insight, we propose a variational cross-graph
reasoning framework that explicitly decomposes video and language into
hierarchical semantic graphs, respectively, and learns fine-grained semantic
correspondence between the two graphs. Furthermore, we introduce a novel
adaptive structured semantics learning approach to derive the
structure-informed and domain-generalizable graph representations, which
facilitate the fine-grained semantic correspondence reasoning between the two
graphs. Extensive experiments validate the superior compositional
generalizability of our approach.
| [
{
"version": "v1",
"created": "Sun, 22 Jan 2023 08:02:23 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Li",
"Juncheng",
""
],
[
"Tang",
"Siliang",
""
],
[
"Zhu",
"Linchao",
""
],
[
"Zhang",
"Wenqiao",
""
],
[
"Yang",
"Yi",
""
],
[
"Chua",
"Tat-Seng",
""
],
[
"Wu",
"Fei",
""
],
[
"Zhuang",
"Yueting",
""
]
] | TITLE: Variational Cross-Graph Reasoning and Adaptive Structured Semantics
Learning for Compositional Temporal Grounding
ABSTRACT: Temporal grounding is the task of locating a specific segment from an
untrimmed video according to a query sentence. This task has achieved
significant momentum in the computer vision community as it enables activity
grounding beyond pre-defined activity classes by utilizing the semantic
diversity of natural language descriptions. The semantic diversity is rooted in
the principle of compositionality in linguistics, where novel semantics can be
systematically described by combining known words in novel ways (compositional
generalization). However, existing temporal grounding datasets are not
carefully designed to evaluate the compositional generalizability. To
systematically benchmark the compositional generalizability of temporal
grounding models, we introduce a new Compositional Temporal Grounding task and
construct two new dataset splits, i.e., Charades-CG and ActivityNet-CG. When
evaluating the state-of-the-art methods on our new dataset splits, we
empirically find that they fail to generalize to queries with novel
combinations of seen words. We argue that the inherent structured semantics
inside the videos and language is the crucial factor to achieve compositional
generalization. Based on this insight, we propose a variational cross-graph
reasoning framework that explicitly decomposes video and language into
hierarchical semantic graphs, respectively, and learns fine-grained semantic
correspondence between the two graphs. Furthermore, we introduce a novel
adaptive structured semantics learning approach to derive the
structure-informed and domain-generalizable graph representations, which
facilitate the fine-grained semantic correspondence reasoning between the two
graphs. Extensive experiments validate the superior compositional
generalizability of our approach.
| new_dataset | 0.68297 |
2302.00070 | Ching-Yao Chuang | Ching-Yao Chuang, Varun Jampani, Yuanzhen Li, Antonio Torralba,
Stefanie Jegelka | Debiasing Vision-Language Models via Biased Prompts | null | null | null | null | cs.LG cs.CV | http://creativecommons.org/licenses/by/4.0/ | Machine learning models have been shown to inherit biases from their training
datasets. This can be particularly problematic for vision-language foundation
models trained on uncurated datasets scraped from the internet. The biases can
be amplified and propagated to downstream applications like zero-shot
classifiers and text-to-image generative models. In this study, we propose a
general approach for debiasing vision-language foundation models by projecting
out biased directions in the text embedding. In particular, we show that
debiasing only the text embedding with a calibrated projection matrix suffices
to yield robust classifiers and fair generative models. The proposed
closed-form solution enables easy integration into large-scale pipelines, and
empirical results demonstrate that our approach effectively reduces social bias
and spurious correlation in both discriminative and generative vision-language
models without the need for additional data or training.
| [
{
"version": "v1",
"created": "Tue, 31 Jan 2023 20:09:33 GMT"
},
{
"version": "v2",
"created": "Mon, 15 May 2023 07:51:14 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Chuang",
"Ching-Yao",
""
],
[
"Jampani",
"Varun",
""
],
[
"Li",
"Yuanzhen",
""
],
[
"Torralba",
"Antonio",
""
],
[
"Jegelka",
"Stefanie",
""
]
] | TITLE: Debiasing Vision-Language Models via Biased Prompts
ABSTRACT: Machine learning models have been shown to inherit biases from their training
datasets. This can be particularly problematic for vision-language foundation
models trained on uncurated datasets scraped from the internet. The biases can
be amplified and propagated to downstream applications like zero-shot
classifiers and text-to-image generative models. In this study, we propose a
general approach for debiasing vision-language foundation models by projecting
out biased directions in the text embedding. In particular, we show that
debiasing only the text embedding with a calibrated projection matrix suffices
to yield robust classifiers and fair generative models. The proposed
closed-form solution enables easy integration into large-scale pipelines, and
empirical results demonstrate that our approach effectively reduces social bias
and spurious correlation in both discriminative and generative vision-language
models without the need for additional data or training.
| no_new_dataset | 0.709422 |
2302.08508 | Romain Xu | Romain Xu-Darme (LSL, MRIM), Georges Qu\'enot (MRIM), Zakaria Chihani
(LSL), Marie-Christine Rousset (SLIDE) | Sanity checks and improvements for patch visualisation in
prototype-based image classification | null | null | null | null | cs.CV cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this work, we perform an in-depth analysis of the visualisation methods
implemented in two popular self-explaining models for visual classification
based on prototypes - ProtoPNet and ProtoTree. Using two fine-grained datasets
(CUB-200-2011 and Stanford Cars), we first show that such methods do not
correctly identify the regions of interest inside of the images, and therefore
do not reflect the model behaviour. Secondly, using a deletion metric, we
demonstrate quantitatively that saliency methods such as Smoothgrads or PRP
provide more faithful image patches. We also propose a new relevance metric
based on the segmentation of the object provided in some datasets (e.g.
CUB-200-2011) and show that the imprecise patch visualisations generated by
ProtoPNet and ProtoTree can create a false sense of bias that can be mitigated
by the use of more faithful methods. Finally, we discuss the implications of
our findings for other prototype-based models sharing the same visualisation
method.
| [
{
"version": "v1",
"created": "Fri, 20 Jan 2023 15:13:04 GMT"
},
{
"version": "v2",
"created": "Mon, 15 May 2023 09:09:19 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Xu-Darme",
"Romain",
"",
"LSL, MRIM"
],
[
"Quénot",
"Georges",
"",
"MRIM"
],
[
"Chihani",
"Zakaria",
"",
"LSL"
],
[
"Rousset",
"Marie-Christine",
"",
"SLIDE"
]
] | TITLE: Sanity checks and improvements for patch visualisation in
prototype-based image classification
ABSTRACT: In this work, we perform an in-depth analysis of the visualisation methods
implemented in two popular self-explaining models for visual classification
based on prototypes - ProtoPNet and ProtoTree. Using two fine-grained datasets
(CUB-200-2011 and Stanford Cars), we first show that such methods do not
correctly identify the regions of interest inside of the images, and therefore
do not reflect the model behaviour. Secondly, using a deletion metric, we
demonstrate quantitatively that saliency methods such as Smoothgrads or PRP
provide more faithful image patches. We also propose a new relevance metric
based on the segmentation of the object provided in some datasets (e.g.
CUB-200-2011) and show that the imprecise patch visualisations generated by
ProtoPNet and ProtoTree can create a false sense of bias that can be mitigated
by the use of more faithful methods. Finally, we discuss the implications of
our findings for other prototype-based models sharing the same visualisation
method.
| no_new_dataset | 0.712845 |
2302.09574 | Idan Achituve | Idan Achituve, Gal Chechik, Ethan Fetaya | Guided Deep Kernel Learning | null | null | null | null | cs.LG stat.ML | http://creativecommons.org/licenses/by/4.0/ | Combining Gaussian processes with the expressive power of deep neural
networks is commonly done nowadays through deep kernel learning (DKL).
Unfortunately, due to the kernel optimization process, this often results in
losing their Bayesian benefits. In this study, we present a novel approach for
learning deep kernels by utilizing infinite-width neural networks. We propose
to use the Neural Network Gaussian Process (NNGP) model as a guide to the DKL
model in the optimization process. Our approach harnesses the reliable
uncertainty estimation of the NNGPs to adapt the DKL target confidence when it
encounters novel data points. As a result, we get the best of both worlds, we
leverage the Bayesian behavior of the NNGP, namely its robustness to
overfitting, and accurate uncertainty estimation, while maintaining the
generalization abilities, scalability, and flexibility of deep kernels.
Empirically, we show on multiple benchmark datasets of varying sizes and
dimensionality, that our method is robust to overfitting, has good predictive
performance, and provides reliable uncertainty estimations.
| [
{
"version": "v1",
"created": "Sun, 19 Feb 2023 13:37:34 GMT"
},
{
"version": "v2",
"created": "Sun, 14 May 2023 14:24:07 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Achituve",
"Idan",
""
],
[
"Chechik",
"Gal",
""
],
[
"Fetaya",
"Ethan",
""
]
] | TITLE: Guided Deep Kernel Learning
ABSTRACT: Combining Gaussian processes with the expressive power of deep neural
networks is commonly done nowadays through deep kernel learning (DKL).
Unfortunately, due to the kernel optimization process, this often results in
losing their Bayesian benefits. In this study, we present a novel approach for
learning deep kernels by utilizing infinite-width neural networks. We propose
to use the Neural Network Gaussian Process (NNGP) model as a guide to the DKL
model in the optimization process. Our approach harnesses the reliable
uncertainty estimation of the NNGPs to adapt the DKL target confidence when it
encounters novel data points. As a result, we get the best of both worlds, we
leverage the Bayesian behavior of the NNGP, namely its robustness to
overfitting, and accurate uncertainty estimation, while maintaining the
generalization abilities, scalability, and flexibility of deep kernels.
Empirically, we show on multiple benchmark datasets of varying sizes and
dimensionality, that our method is robust to overfitting, has good predictive
performance, and provides reliable uncertainty estimations.
| no_new_dataset | 0.710226 |
2302.14297 | Xu Chen | Xu Chen, Erik G. Larsson, Kaibin Huang | On-the-Fly Communication-and-Computing for Distributed Tensor
Decomposition Over MIMO Channels | 31 pages, 13 figures | null | null | null | cs.DC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Distributed tensor decomposition (DTD) is a fundamental data-analytics
technique that extracts latent important properties from high-dimensional
multi-attribute datasets distributed over edge devices. Conventionally its
wireless implementation follows a one-shot approach that first computes local
results at devices using local data and then aggregates them to a server with
communication-efficient techniques such as over-the-air computation (AirComp)
for global computation. Such implementation is confronted with the issues of
limited storage-and-computation capacities and link interruption, which
motivates us to propose a framework of on-the-fly communication-and-computing
(FlyCom$^2$) in this work. The proposed framework enables streaming computation
with low complexity by leveraging a random sketching technique and achieves
progressive global aggregation through the integration of progressive uploading
and multiple-input-multiple-output (MIMO) AirComp. To develop FlyCom$^2$, an
on-the-fly sub-space estimator is designed to take real-time sketches
accumulated at the server to generate online estimates for the decomposition.
Its performance is evaluated by deriving both deterministic and probabilistic
error bounds using the perturbation theory and concentration of measure. Both
results reveal that the decomposition error is inversely proportional to the
population of sketching observations received by the server. To further rein in
the noise effect on the error, we propose a threshold-based scheme to select a
subset of sufficiently reliable received sketches for DTD at the server.
Experimental results validate the performance gain of the proposed selection
algorithm and show that compared to its one-shot counterparts, the proposed
FlyCom$^2$ achieves comparable (even better in the case of large eigen-gaps)
decomposition accuracy besides dramatically reducing devices' complexity costs.
| [
{
"version": "v1",
"created": "Tue, 28 Feb 2023 04:10:57 GMT"
},
{
"version": "v2",
"created": "Sat, 13 May 2023 06:07:26 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Chen",
"Xu",
""
],
[
"Larsson",
"Erik G.",
""
],
[
"Huang",
"Kaibin",
""
]
] | TITLE: On-the-Fly Communication-and-Computing for Distributed Tensor
Decomposition Over MIMO Channels
ABSTRACT: Distributed tensor decomposition (DTD) is a fundamental data-analytics
technique that extracts latent important properties from high-dimensional
multi-attribute datasets distributed over edge devices. Conventionally its
wireless implementation follows a one-shot approach that first computes local
results at devices using local data and then aggregates them to a server with
communication-efficient techniques such as over-the-air computation (AirComp)
for global computation. Such implementation is confronted with the issues of
limited storage-and-computation capacities and link interruption, which
motivates us to propose a framework of on-the-fly communication-and-computing
(FlyCom$^2$) in this work. The proposed framework enables streaming computation
with low complexity by leveraging a random sketching technique and achieves
progressive global aggregation through the integration of progressive uploading
and multiple-input-multiple-output (MIMO) AirComp. To develop FlyCom$^2$, an
on-the-fly sub-space estimator is designed to take real-time sketches
accumulated at the server to generate online estimates for the decomposition.
Its performance is evaluated by deriving both deterministic and probabilistic
error bounds using the perturbation theory and concentration of measure. Both
results reveal that the decomposition error is inversely proportional to the
population of sketching observations received by the server. To further rein in
the noise effect on the error, we propose a threshold-based scheme to select a
subset of sufficiently reliable received sketches for DTD at the server.
Experimental results validate the performance gain of the proposed selection
algorithm and show that compared to its one-shot counterparts, the proposed
FlyCom$^2$ achieves comparable (even better in the case of large eigen-gaps)
decomposition accuracy besides dramatically reducing devices' complexity costs.
| no_new_dataset | 0.710189 |
2302.14517 | Bogdan Kulynych | Bogdan Kulynych, Hsiang Hsu, Carmela Troncoso, Flavio P. Calmon | Arbitrary Decisions are a Hidden Cost of Differentially Private Training | To appear in ACM FAccT 2023 | null | 10.1145/3593013.3594103 | null | cs.LG cs.CR cs.CY stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Mechanisms used in privacy-preserving machine learning often aim to guarantee
differential privacy (DP) during model training. Practical DP-ensuring training
methods use randomization when fitting model parameters to privacy-sensitive
data (e.g., adding Gaussian noise to clipped gradients). We demonstrate that
such randomization incurs predictive multiplicity: for a given input example,
the output predicted by equally-private models depends on the randomness used
in training. Thus, for a given input, the predicted output can vary drastically
if a model is re-trained, even if the same training dataset is used. The
predictive-multiplicity cost of DP training has not been studied, and is
currently neither audited for nor communicated to model designers and
stakeholders. We derive a bound on the number of re-trainings required to
estimate predictive multiplicity reliably. We analyze--both theoretically and
through extensive experiments--the predictive-multiplicity cost of three
DP-ensuring algorithms: output perturbation, objective perturbation, and
DP-SGD. We demonstrate that the degree of predictive multiplicity rises as the
level of privacy increases, and is unevenly distributed across individuals and
demographic groups in the data. Because randomness used to ensure DP during
training explains predictions for some examples, our results highlight a
fundamental challenge to the justifiability of decisions supported by
differentially private models in high-stakes settings. We conclude that
practitioners should audit the predictive multiplicity of their DP-ensuring
algorithms before deploying them in applications of individual-level
consequence.
| [
{
"version": "v1",
"created": "Tue, 28 Feb 2023 12:13:43 GMT"
},
{
"version": "v2",
"created": "Mon, 15 May 2023 15:07:24 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Kulynych",
"Bogdan",
""
],
[
"Hsu",
"Hsiang",
""
],
[
"Troncoso",
"Carmela",
""
],
[
"Calmon",
"Flavio P.",
""
]
] | TITLE: Arbitrary Decisions are a Hidden Cost of Differentially Private Training
ABSTRACT: Mechanisms used in privacy-preserving machine learning often aim to guarantee
differential privacy (DP) during model training. Practical DP-ensuring training
methods use randomization when fitting model parameters to privacy-sensitive
data (e.g., adding Gaussian noise to clipped gradients). We demonstrate that
such randomization incurs predictive multiplicity: for a given input example,
the output predicted by equally-private models depends on the randomness used
in training. Thus, for a given input, the predicted output can vary drastically
if a model is re-trained, even if the same training dataset is used. The
predictive-multiplicity cost of DP training has not been studied, and is
currently neither audited for nor communicated to model designers and
stakeholders. We derive a bound on the number of re-trainings required to
estimate predictive multiplicity reliably. We analyze--both theoretically and
through extensive experiments--the predictive-multiplicity cost of three
DP-ensuring algorithms: output perturbation, objective perturbation, and
DP-SGD. We demonstrate that the degree of predictive multiplicity rises as the
level of privacy increases, and is unevenly distributed across individuals and
demographic groups in the data. Because randomness used to ensure DP during
training explains predictions for some examples, our results highlight a
fundamental challenge to the justifiability of decisions supported by
differentially private models in high-stakes settings. We conclude that
practitioners should audit the predictive multiplicity of their DP-ensuring
algorithms before deploying them in applications of individual-level
consequence.
| no_new_dataset | 0.710804 |
2303.09190 | RuiYang Ju | Rui-Yang Ju, Chih-Chia Chen, Jen-Shiun Chiang, Yu-Shian Lin, Wei-Han
Chen, Chun-Tse Chien | Resolution Enhancement Processing on Low Quality Images Using Swin
Transformer Based on Interval Dense Connection Strategy | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The Transformer-based method has demonstrated remarkable performance for
image super-resolution in comparison to the method based on the convolutional
neural networks (CNNs). However, using the self-attention mechanism like SwinIR
(Image Restoration Using Swin Transformer) to extract feature information from
images needs a significant amount of computational resources, which limits its
application on low computing power platforms. To improve the model feature
reuse, this research work proposes the Interval Dense Connection Strategy,
which connects different blocks according to the newly designed algorithm. We
apply this strategy to SwinIR and present a new model, which named SwinOIR
(Object Image Restoration Using Swin Transformer). For image super-resolution,
an ablation study is conducted to demonstrate the positive effect of the
Interval Dense Connection Strategy on the model performance. Furthermore, we
evaluate our model on various popular benchmark datasets, and compare it with
other state-of-the-art (SOTA) lightweight models. For example, SwinOIR obtains
a PSNR of 26.62 dB for x4 upscaling image super-resolution on Urban100 dataset,
which is 0.15 dB higher than the SOTA model SwinIR. For real-life application,
this work applies the lastest version of You Only Look Once (YOLOv8) model and
the proposed model to perform object detection and real-life image
super-resolution on low-quality images. This implementation code is publicly
available at https://github.com/Rubbbbbbbbby/SwinOIR.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 10:01:12 GMT"
},
{
"version": "v2",
"created": "Sat, 13 May 2023 06:54:14 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Ju",
"Rui-Yang",
""
],
[
"Chen",
"Chih-Chia",
""
],
[
"Chiang",
"Jen-Shiun",
""
],
[
"Lin",
"Yu-Shian",
""
],
[
"Chen",
"Wei-Han",
""
],
[
"Chien",
"Chun-Tse",
""
]
] | TITLE: Resolution Enhancement Processing on Low Quality Images Using Swin
Transformer Based on Interval Dense Connection Strategy
ABSTRACT: The Transformer-based method has demonstrated remarkable performance for
image super-resolution in comparison to the method based on the convolutional
neural networks (CNNs). However, using the self-attention mechanism like SwinIR
(Image Restoration Using Swin Transformer) to extract feature information from
images needs a significant amount of computational resources, which limits its
application on low computing power platforms. To improve the model feature
reuse, this research work proposes the Interval Dense Connection Strategy,
which connects different blocks according to the newly designed algorithm. We
apply this strategy to SwinIR and present a new model, which named SwinOIR
(Object Image Restoration Using Swin Transformer). For image super-resolution,
an ablation study is conducted to demonstrate the positive effect of the
Interval Dense Connection Strategy on the model performance. Furthermore, we
evaluate our model on various popular benchmark datasets, and compare it with
other state-of-the-art (SOTA) lightweight models. For example, SwinOIR obtains
a PSNR of 26.62 dB for x4 upscaling image super-resolution on Urban100 dataset,
which is 0.15 dB higher than the SOTA model SwinIR. For real-life application,
this work applies the lastest version of You Only Look Once (YOLOv8) model and
the proposed model to perform object detection and real-life image
super-resolution on low-quality images. This implementation code is publicly
available at https://github.com/Rubbbbbbbbby/SwinOIR.
| no_new_dataset | 0.710584 |
2303.15748 | Johannes Leuschner | Marco Nittscher, Michael Lameter, Riccardo Barbano, Johannes
Leuschner, Bangti Jin, Peter Maass | SVD-DIP: Overcoming the Overfitting Problem in DIP-based CT
Reconstruction | null | null | null | null | eess.IV cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The deep image prior (DIP) is a well-established unsupervised deep learning
method for image reconstruction; yet it is far from being flawless. The DIP
overfits to noise if not early stopped, or optimized via a regularized
objective. We build on the regularized fine-tuning of a pretrained DIP, by
adopting a novel strategy that restricts the learning to the adaptation of
singular values. The proposed SVD-DIP uses ad hoc convolutional layers whose
pretrained parameters are decomposed via the singular value decomposition.
Optimizing the DIP then solely consists in the fine-tuning of the singular
values, while keeping the left and right singular vectors fixed. We thoroughly
validate the proposed method on real-measured $\mu$CT data of a lotus root as
well as two medical datasets (LoDoPaB and Mayo). We report significantly
improved stability of the DIP optimization, by overcoming the overfitting to
noise.
| [
{
"version": "v1",
"created": "Tue, 28 Mar 2023 06:08:32 GMT"
},
{
"version": "v2",
"created": "Mon, 8 May 2023 11:47:35 GMT"
},
{
"version": "v3",
"created": "Mon, 15 May 2023 16:30:10 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Nittscher",
"Marco",
""
],
[
"Lameter",
"Michael",
""
],
[
"Barbano",
"Riccardo",
""
],
[
"Leuschner",
"Johannes",
""
],
[
"Jin",
"Bangti",
""
],
[
"Maass",
"Peter",
""
]
] | TITLE: SVD-DIP: Overcoming the Overfitting Problem in DIP-based CT
Reconstruction
ABSTRACT: The deep image prior (DIP) is a well-established unsupervised deep learning
method for image reconstruction; yet it is far from being flawless. The DIP
overfits to noise if not early stopped, or optimized via a regularized
objective. We build on the regularized fine-tuning of a pretrained DIP, by
adopting a novel strategy that restricts the learning to the adaptation of
singular values. The proposed SVD-DIP uses ad hoc convolutional layers whose
pretrained parameters are decomposed via the singular value decomposition.
Optimizing the DIP then solely consists in the fine-tuning of the singular
values, while keeping the left and right singular vectors fixed. We thoroughly
validate the proposed method on real-measured $\mu$CT data of a lotus root as
well as two medical datasets (LoDoPaB and Mayo). We report significantly
improved stability of the DIP optimization, by overcoming the overfitting to
noise.
| no_new_dataset | 0.707803 |
2303.17559 | Ji Yuanfeng | Yuanfeng Ji, Zhe Chen, Enze Xie, Lanqing Hong, Xihui Liu, Zhaoqiang
Liu, Tong Lu, Zhenguo Li, Ping Luo | DDP: Diffusion Model for Dense Visual Prediction | Added controlnet exp | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | We propose a simple, efficient, yet powerful framework for dense visual
predictions based on the conditional diffusion pipeline. Our approach follows a
"noise-to-map" generative paradigm for prediction by progressively removing
noise from a random Gaussian distribution, guided by the image. The method,
called DDP, efficiently extends the denoising diffusion process into the modern
perception pipeline. Without task-specific design and architecture
customization, DDP is easy to generalize to most dense prediction tasks, e.g.,
semantic segmentation and depth estimation. In addition, DDP shows attractive
properties such as dynamic inference and uncertainty awareness, in contrast to
previous single-step discriminative methods. We show top results on three
representative tasks with six diverse benchmarks, without tricks, DDP achieves
state-of-the-art or competitive performance on each task compared to the
specialist counterparts. For example, semantic segmentation (83.9 mIoU on
Cityscapes), BEV map segmentation (70.6 mIoU on nuScenes), and depth estimation
(0.05 REL on KITTI). We hope that our approach will serve as a solid baseline
and facilitate future research
| [
{
"version": "v1",
"created": "Thu, 30 Mar 2023 17:26:50 GMT"
},
{
"version": "v2",
"created": "Sat, 13 May 2023 11:38:59 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Ji",
"Yuanfeng",
""
],
[
"Chen",
"Zhe",
""
],
[
"Xie",
"Enze",
""
],
[
"Hong",
"Lanqing",
""
],
[
"Liu",
"Xihui",
""
],
[
"Liu",
"Zhaoqiang",
""
],
[
"Lu",
"Tong",
""
],
[
"Li",
"Zhenguo",
""
],
[
"Luo",
"Ping",
""
]
] | TITLE: DDP: Diffusion Model for Dense Visual Prediction
ABSTRACT: We propose a simple, efficient, yet powerful framework for dense visual
predictions based on the conditional diffusion pipeline. Our approach follows a
"noise-to-map" generative paradigm for prediction by progressively removing
noise from a random Gaussian distribution, guided by the image. The method,
called DDP, efficiently extends the denoising diffusion process into the modern
perception pipeline. Without task-specific design and architecture
customization, DDP is easy to generalize to most dense prediction tasks, e.g.,
semantic segmentation and depth estimation. In addition, DDP shows attractive
properties such as dynamic inference and uncertainty awareness, in contrast to
previous single-step discriminative methods. We show top results on three
representative tasks with six diverse benchmarks, without tricks, DDP achieves
state-of-the-art or competitive performance on each task compared to the
specialist counterparts. For example, semantic segmentation (83.9 mIoU on
Cityscapes), BEV map segmentation (70.6 mIoU on nuScenes), and depth estimation
(0.05 REL on KITTI). We hope that our approach will serve as a solid baseline
and facilitate future research
| no_new_dataset | 0.709636 |
2304.02931 | Natalia Kowalczyk | Natalia Kowalczyk and Jacek Rumi\'nski | Mask Detection and Classification in Thermal Face Images | null | IEEE Access, vol. 11, (2023) pp. 43349-43359 | 10.1109/ACCESS.2023.3272214 | null | cs.CV cs.LG eess.IV | http://creativecommons.org/licenses/by/4.0/ | Face masks are recommended to reduce the transmission of many viruses,
especially SARS-CoV-2. Therefore, the automatic detection of whether there is a
mask on the face, what type of mask is worn, and how it is worn is an important
research topic. In this work, the use of thermal imaging was considered to
analyze the possibility of detecting (localizing) a mask on the face, as well
as to check whether it is possible to classify the type of mask on the face.
The previously proposed dataset of thermal images was extended and annotated
with the description of a type of mask and a location of a mask within a face.
Different deep learning models were adapted. The best model for face mask
detection turned out to be the Yolov5 model in the "nano" version, reaching mAP
higher than 97% and precision of about 95%. High accuracy was also obtained for
mask type classification. The best results were obtained for the convolutional
neural network model built on an autoencoder initially trained in the thermal
image reconstruction problem. The pretrained encoder was used to train a
classifier which achieved an accuracy of 91%.
| [
{
"version": "v1",
"created": "Thu, 6 Apr 2023 08:41:31 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Kowalczyk",
"Natalia",
""
],
[
"Rumiński",
"Jacek",
""
]
] | TITLE: Mask Detection and Classification in Thermal Face Images
ABSTRACT: Face masks are recommended to reduce the transmission of many viruses,
especially SARS-CoV-2. Therefore, the automatic detection of whether there is a
mask on the face, what type of mask is worn, and how it is worn is an important
research topic. In this work, the use of thermal imaging was considered to
analyze the possibility of detecting (localizing) a mask on the face, as well
as to check whether it is possible to classify the type of mask on the face.
The previously proposed dataset of thermal images was extended and annotated
with the description of a type of mask and a location of a mask within a face.
Different deep learning models were adapted. The best model for face mask
detection turned out to be the Yolov5 model in the "nano" version, reaching mAP
higher than 97% and precision of about 95%. High accuracy was also obtained for
mask type classification. The best results were obtained for the convolutional
neural network model built on an autoencoder initially trained in the thermal
image reconstruction problem. The pretrained encoder was used to train a
classifier which achieved an accuracy of 91%.
| new_dataset | 0.693213 |
2304.04276 | Yehonatan Fridman | Yehonatan Fridman, Guy Tamir, Gal Oren | Portability and Scalability of OpenMP Offloading on State-of-the-art
Accelerators | 13 pages, 5 Figures, 5 Tables | null | null | null | cs.DC | http://creativecommons.org/licenses/by/4.0/ | Over the last decade, most of the increase in computing power has been gained
by advances in accelerated many-core architectures, mainly in the form of
GPGPUs. While accelerators achieve phenomenal performances in various computing
tasks, their utilization requires code adaptations and transformations. Thus,
OpenMP, the most common standard for multi-threading in scientific computing
applications, introduced offloading capabilities between host (CPUs) and
accelerators since v4.0, with increasing support in the successive v4.5, v5.0,
v5.1, and the latest v5.2 versions. Recently, two state-of-the-art GPUs -- the
Intel Ponte Vecchio Max 1100 and the NVIDIA A100 GPUs -- were released to the
market, with the oneAPI and NVHPC compilers for offloading, correspondingly. In
this work, we present early performance results of OpenMP offloading
capabilities to these devices while specifically analyzing the portability of
advanced directives (using SOLLVE's OMPVV test suite) and the scalability of
the hardware in representative scientific mini-app (the LULESH benchmark). Our
results show that the coverage for version 4.5 is nearly complete in both
latest NVHPC and oneAPI tools. However, we observed a lack of support in
versions 5.0, 5.1, and 5.2, which is particularly noticeable when using NVHPC.
From the performance perspective, we found that the PVC1100 and A100 are
relatively comparable on the LULESH benchmark. While the A100 is slightly
better due to faster memory bandwidth, the PVC1100 reaches the next problem
size (400^3) scalably due to the larger memory size.
| [
{
"version": "v1",
"created": "Sun, 9 Apr 2023 16:40:20 GMT"
},
{
"version": "v2",
"created": "Sun, 14 May 2023 04:47:18 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Fridman",
"Yehonatan",
""
],
[
"Tamir",
"Guy",
""
],
[
"Oren",
"Gal",
""
]
] | TITLE: Portability and Scalability of OpenMP Offloading on State-of-the-art
Accelerators
ABSTRACT: Over the last decade, most of the increase in computing power has been gained
by advances in accelerated many-core architectures, mainly in the form of
GPGPUs. While accelerators achieve phenomenal performances in various computing
tasks, their utilization requires code adaptations and transformations. Thus,
OpenMP, the most common standard for multi-threading in scientific computing
applications, introduced offloading capabilities between host (CPUs) and
accelerators since v4.0, with increasing support in the successive v4.5, v5.0,
v5.1, and the latest v5.2 versions. Recently, two state-of-the-art GPUs -- the
Intel Ponte Vecchio Max 1100 and the NVIDIA A100 GPUs -- were released to the
market, with the oneAPI and NVHPC compilers for offloading, correspondingly. In
this work, we present early performance results of OpenMP offloading
capabilities to these devices while specifically analyzing the portability of
advanced directives (using SOLLVE's OMPVV test suite) and the scalability of
the hardware in representative scientific mini-app (the LULESH benchmark). Our
results show that the coverage for version 4.5 is nearly complete in both
latest NVHPC and oneAPI tools. However, we observed a lack of support in
versions 5.0, 5.1, and 5.2, which is particularly noticeable when using NVHPC.
From the performance perspective, we found that the PVC1100 and A100 are
relatively comparable on the LULESH benchmark. While the A100 is slightly
better due to faster memory bandwidth, the PVC1100 reaches the next problem
size (400^3) scalably due to the larger memory size.
| no_new_dataset | 0.707771 |
2304.05642 | Chi Liu | Chi Liu, Haochun Wang, Nuwa Xi, Sendong Zhao, Bing Qin | Global Prompt Cell: A Portable Control Module for Effective Prompt
Tuning | null | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | As a novel approach to tuning pre-trained models, prompt tuning involves
freezing the parameters in downstream tasks while inserting trainable
embeddings into inputs in the first layer. However, previous methods have
mainly focused on the initialization of prompt embeddings. The strategy of
training and utilizing prompt embeddings in a reasonable way has become a
limiting factor in the effectiveness of prompt tuning. To address this issue,
we introduce the Global Prompt Cell (GPC), a portable control module for prompt
tuning that selectively preserves prompt information across all encoder layers.
Our experimental results demonstrate a 5.8% improvement on SuperGLUE datasets
compared to vanilla prompt tuning.
| [
{
"version": "v1",
"created": "Wed, 12 Apr 2023 06:46:33 GMT"
},
{
"version": "v2",
"created": "Sat, 13 May 2023 07:45:59 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Liu",
"Chi",
""
],
[
"Wang",
"Haochun",
""
],
[
"Xi",
"Nuwa",
""
],
[
"Zhao",
"Sendong",
""
],
[
"Qin",
"Bing",
""
]
] | TITLE: Global Prompt Cell: A Portable Control Module for Effective Prompt
Tuning
ABSTRACT: As a novel approach to tuning pre-trained models, prompt tuning involves
freezing the parameters in downstream tasks while inserting trainable
embeddings into inputs in the first layer. However, previous methods have
mainly focused on the initialization of prompt embeddings. The strategy of
training and utilizing prompt embeddings in a reasonable way has become a
limiting factor in the effectiveness of prompt tuning. To address this issue,
we introduce the Global Prompt Cell (GPC), a portable control module for prompt
tuning that selectively preserves prompt information across all encoder layers.
Our experimental results demonstrate a 5.8% improvement on SuperGLUE datasets
compared to vanilla prompt tuning.
| no_new_dataset | 0.709604 |
2304.08742 | Suraj Nair | Maximilian Du, Suraj Nair, Dorsa Sadigh, Chelsea Finn | Behavior Retrieval: Few-Shot Imitation Learning by Querying Unlabeled
Datasets | null | null | null | null | cs.RO cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Enabling robots to learn novel visuomotor skills in a data-efficient manner
remains an unsolved problem with myriad challenges. A popular paradigm for
tackling this problem is through leveraging large unlabeled datasets that have
many behaviors in them and then adapting a policy to a specific task using a
small amount of task-specific human supervision (i.e. interventions or
demonstrations). However, how best to leverage the narrow task-specific
supervision and balance it with offline data remains an open question. Our key
insight in this work is that task-specific data not only provides new data for
an agent to train on but can also inform the type of prior data the agent
should use for learning. Concretely, we propose a simple approach that uses a
small amount of downstream expert data to selectively query relevant behaviors
from an offline, unlabeled dataset (including many sub-optimal behaviors). The
agent is then jointly trained on the expert and queried data. We observe that
our method learns to query only the relevant transitions to the task, filtering
out sub-optimal or task-irrelevant data. By doing so, it is able to learn more
effectively from the mix of task-specific and offline data compared to naively
mixing the data or only using the task-specific data. Furthermore, we find that
our simple querying approach outperforms more complex goal-conditioned methods
by 20% across simulated and real robotic manipulation tasks from images. See
https://sites.google.com/view/behaviorretrieval for videos and code.
| [
{
"version": "v1",
"created": "Tue, 18 Apr 2023 05:42:53 GMT"
},
{
"version": "v2",
"created": "Sat, 13 May 2023 00:05:05 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Du",
"Maximilian",
""
],
[
"Nair",
"Suraj",
""
],
[
"Sadigh",
"Dorsa",
""
],
[
"Finn",
"Chelsea",
""
]
] | TITLE: Behavior Retrieval: Few-Shot Imitation Learning by Querying Unlabeled
Datasets
ABSTRACT: Enabling robots to learn novel visuomotor skills in a data-efficient manner
remains an unsolved problem with myriad challenges. A popular paradigm for
tackling this problem is through leveraging large unlabeled datasets that have
many behaviors in them and then adapting a policy to a specific task using a
small amount of task-specific human supervision (i.e. interventions or
demonstrations). However, how best to leverage the narrow task-specific
supervision and balance it with offline data remains an open question. Our key
insight in this work is that task-specific data not only provides new data for
an agent to train on but can also inform the type of prior data the agent
should use for learning. Concretely, we propose a simple approach that uses a
small amount of downstream expert data to selectively query relevant behaviors
from an offline, unlabeled dataset (including many sub-optimal behaviors). The
agent is then jointly trained on the expert and queried data. We observe that
our method learns to query only the relevant transitions to the task, filtering
out sub-optimal or task-irrelevant data. By doing so, it is able to learn more
effectively from the mix of task-specific and offline data compared to naively
mixing the data or only using the task-specific data. Furthermore, we find that
our simple querying approach outperforms more complex goal-conditioned methods
by 20% across simulated and real robotic manipulation tasks from images. See
https://sites.google.com/view/behaviorretrieval for videos and code.
| no_new_dataset | 0.709598 |
2304.10848 | Benjamin Doerr | Benjamin Doerr, Taha El Ghazi El Houssaini, Amirhossein Rajabi, and
Carsten Witt | How Well Does the Metropolis Algorithm Cope With Local Optima? | To appear in the proceedings of GECCO 2023. With appendix containing
all proofs. 28 pages | null | 10.1145/3583131.3590390 | null | cs.NE cs.AI cs.DS | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The Metropolis algorithm (MA) is a classic stochastic local search heuristic.
It avoids getting stuck in local optima by occasionally accepting inferior
solutions. To better and in a rigorous manner understand this ability, we
conduct a mathematical runtime analysis of the MA on the CLIFF benchmark. Apart
from one local optimum, cliff functions are monotonically increasing towards
the global optimum. Consequently, to optimize a cliff function, the MA only
once needs to accept an inferior solution. Despite seemingly being an ideal
benchmark for the MA to profit from its main working principle, our
mathematical runtime analysis shows that this hope does not come true. Even
with the optimal temperature (the only parameter of the MA), the MA optimizes
most cliff functions less efficiently than simple elitist evolutionary
algorithms (EAs), which can only leave the local optimum by generating a
superior solution possibly far away. This result suggests that our
understanding of why the MA is often very successful in practice is not yet
complete. Our work also suggests to equip the MA with global mutation
operators, an idea supported by our preliminary experiments.
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 09:44:08 GMT"
},
{
"version": "v2",
"created": "Mon, 15 May 2023 08:55:28 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Doerr",
"Benjamin",
""
],
[
"Houssaini",
"Taha El Ghazi El",
""
],
[
"Rajabi",
"Amirhossein",
""
],
[
"Witt",
"Carsten",
""
]
] | TITLE: How Well Does the Metropolis Algorithm Cope With Local Optima?
ABSTRACT: The Metropolis algorithm (MA) is a classic stochastic local search heuristic.
It avoids getting stuck in local optima by occasionally accepting inferior
solutions. To better and in a rigorous manner understand this ability, we
conduct a mathematical runtime analysis of the MA on the CLIFF benchmark. Apart
from one local optimum, cliff functions are monotonically increasing towards
the global optimum. Consequently, to optimize a cliff function, the MA only
once needs to accept an inferior solution. Despite seemingly being an ideal
benchmark for the MA to profit from its main working principle, our
mathematical runtime analysis shows that this hope does not come true. Even
with the optimal temperature (the only parameter of the MA), the MA optimizes
most cliff functions less efficiently than simple elitist evolutionary
algorithms (EAs), which can only leave the local optimum by generating a
superior solution possibly far away. This result suggests that our
understanding of why the MA is often very successful in practice is not yet
complete. Our work also suggests to equip the MA with global mutation
operators, an idea supported by our preliminary experiments.
| no_new_dataset | 0.708584 |
2304.14601 | Jinhao Duan | Jinhao Duan, Quanfu Fan, Hao Cheng, Xiaoshuang Shi, Kaidi Xu | Improve Video Representation with Temporal Adversarial Augmentation | To be appeared in IJCAI 2023 | null | null | null | cs.CV cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recent works reveal that adversarial augmentation benefits the generalization
of neural networks (NNs) if used in an appropriate manner. In this paper, we
introduce Temporal Adversarial Augmentation (TA), a novel video augmentation
technique that utilizes temporal attention. Unlike conventional adversarial
augmentation, TA is specifically designed to shift the attention distributions
of neural networks with respect to video clips by maximizing a temporal-related
loss function. We demonstrate that TA will obtain diverse temporal views, which
significantly affect the focus of neural networks. Training with these examples
remedies the flaw of unbalanced temporal information perception and enhances
the ability to defend against temporal shifts, ultimately leading to better
generalization. To leverage TA, we propose Temporal Video Adversarial
Fine-tuning (TAF) framework for improving video representations. TAF is a
model-agnostic, generic, and interpretability-friendly training strategy. We
evaluate TAF with four powerful models (TSM, GST, TAM, and TPN) over three
challenging temporal-related benchmarks (Something-something V1&V2 and
diving48). Experimental results demonstrate that TAF effectively improves the
test accuracy of these models with notable margins without introducing
additional parameters or computational costs. As a byproduct, TAF also improves
the robustness under out-of-distribution (OOD) settings. Code is available at
https://github.com/jinhaoduan/TAF.
| [
{
"version": "v1",
"created": "Fri, 28 Apr 2023 03:06:37 GMT"
},
{
"version": "v2",
"created": "Sun, 14 May 2023 19:25:16 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Duan",
"Jinhao",
""
],
[
"Fan",
"Quanfu",
""
],
[
"Cheng",
"Hao",
""
],
[
"Shi",
"Xiaoshuang",
""
],
[
"Xu",
"Kaidi",
""
]
] | TITLE: Improve Video Representation with Temporal Adversarial Augmentation
ABSTRACT: Recent works reveal that adversarial augmentation benefits the generalization
of neural networks (NNs) if used in an appropriate manner. In this paper, we
introduce Temporal Adversarial Augmentation (TA), a novel video augmentation
technique that utilizes temporal attention. Unlike conventional adversarial
augmentation, TA is specifically designed to shift the attention distributions
of neural networks with respect to video clips by maximizing a temporal-related
loss function. We demonstrate that TA will obtain diverse temporal views, which
significantly affect the focus of neural networks. Training with these examples
remedies the flaw of unbalanced temporal information perception and enhances
the ability to defend against temporal shifts, ultimately leading to better
generalization. To leverage TA, we propose Temporal Video Adversarial
Fine-tuning (TAF) framework for improving video representations. TAF is a
model-agnostic, generic, and interpretability-friendly training strategy. We
evaluate TAF with four powerful models (TSM, GST, TAM, and TPN) over three
challenging temporal-related benchmarks (Something-something V1&V2 and
diving48). Experimental results demonstrate that TAF effectively improves the
test accuracy of these models with notable margins without introducing
additional parameters or computational costs. As a byproduct, TAF also improves
the robustness under out-of-distribution (OOD) settings. Code is available at
https://github.com/jinhaoduan/TAF.
| no_new_dataset | 0.710998 |
2305.00675 | Kunyang Zhou | Kunyang Zhou and Rui Zhou | End-to-End Lane detection with One-to-Several Transformer | code: https://github.com/zkyseu/O2SFormer | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Although lane detection methods have shown impressive performance in
real-world scenarios, most of methods require post-processing which is not
robust enough. Therefore, end-to-end detectors like DEtection TRansformer(DETR)
have been introduced in lane detection.However, one-to-one label assignment in
DETR can degrade the training efficiency due to label semantic conflicts.
Besides, positional query in DETR is unable to provide explicit positional
prior, making it difficult to be optimized. In this paper, we present the
One-to-Several Transformer(O2SFormer). We first propose the one-to-several
label assignment, which combines one-to-many and one-to-one label assignment to
solve label semantic conflicts while keeping end-to-end detection. To overcome
the difficulty in optimizing one-to-one assignment. We further propose the
layer-wise soft label which dynamically adjusts the positive weight of positive
lane anchors in different decoder layers. Finally, we design the dynamic
anchor-based positional query to explore positional prior by incorporating lane
anchors into positional query. Experimental results show that O2SFormer with
ResNet50 backbone achieves 77.83% F1 score on CULane dataset, outperforming
existing Transformer-based and CNN-based detectors. Futhermore, O2SFormer
converges 12.5x faster than DETR for the ResNet18 backbone.
| [
{
"version": "v1",
"created": "Mon, 1 May 2023 06:07:11 GMT"
},
{
"version": "v2",
"created": "Tue, 2 May 2023 12:13:04 GMT"
},
{
"version": "v3",
"created": "Tue, 9 May 2023 14:30:16 GMT"
},
{
"version": "v4",
"created": "Sat, 13 May 2023 04:42:42 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Zhou",
"Kunyang",
""
],
[
"Zhou",
"Rui",
""
]
] | TITLE: End-to-End Lane detection with One-to-Several Transformer
ABSTRACT: Although lane detection methods have shown impressive performance in
real-world scenarios, most of methods require post-processing which is not
robust enough. Therefore, end-to-end detectors like DEtection TRansformer(DETR)
have been introduced in lane detection.However, one-to-one label assignment in
DETR can degrade the training efficiency due to label semantic conflicts.
Besides, positional query in DETR is unable to provide explicit positional
prior, making it difficult to be optimized. In this paper, we present the
One-to-Several Transformer(O2SFormer). We first propose the one-to-several
label assignment, which combines one-to-many and one-to-one label assignment to
solve label semantic conflicts while keeping end-to-end detection. To overcome
the difficulty in optimizing one-to-one assignment. We further propose the
layer-wise soft label which dynamically adjusts the positive weight of positive
lane anchors in different decoder layers. Finally, we design the dynamic
anchor-based positional query to explore positional prior by incorporating lane
anchors into positional query. Experimental results show that O2SFormer with
ResNet50 backbone achieves 77.83% F1 score on CULane dataset, outperforming
existing Transformer-based and CNN-based detectors. Futhermore, O2SFormer
converges 12.5x faster than DETR for the ResNet18 backbone.
| no_new_dataset | 0.710465 |
2305.03319 | Irene Li | Irene Li, Aosong Feng, Dragomir Radev, Rex Ying | HiPool: Modeling Long Documents Using Graph Neural Networks | null | ACL 2023 main proceedings | null | null | cs.CL | http://creativecommons.org/publicdomain/zero/1.0/ | Encoding long sequences in Natural Language Processing (NLP) is a challenging
problem. Though recent pretraining language models achieve satisfying
performances in many NLP tasks, they are still restricted by a pre-defined
maximum length, making them challenging to be extended to longer sequences. So
some recent works utilize hierarchies to model long sequences. However, most of
them apply sequential models for upper hierarchies, suffering from long
dependency issues. In this paper, we alleviate these issues through a
graph-based method. We first chunk the sequence with a fixed length to model
the sentence-level information. We then leverage graphs to model intra- and
cross-sentence correlations with a new attention mechanism. Additionally, due
to limited standard benchmarks for long document classification (LDC), we
propose a new challenging benchmark, totaling six datasets with up to 53k
samples and 4034 average tokens' length. Evaluation shows our model surpasses
competitive baselines by 2.6% in F1 score, and 4.8% on the longest sequence
dataset. Our method is shown to outperform hierarchical sequential models with
better performance and scalability, especially for longer sequences.
| [
{
"version": "v1",
"created": "Fri, 5 May 2023 06:58:24 GMT"
},
{
"version": "v2",
"created": "Mon, 15 May 2023 03:48:36 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Li",
"Irene",
""
],
[
"Feng",
"Aosong",
""
],
[
"Radev",
"Dragomir",
""
],
[
"Ying",
"Rex",
""
]
] | TITLE: HiPool: Modeling Long Documents Using Graph Neural Networks
ABSTRACT: Encoding long sequences in Natural Language Processing (NLP) is a challenging
problem. Though recent pretraining language models achieve satisfying
performances in many NLP tasks, they are still restricted by a pre-defined
maximum length, making them challenging to be extended to longer sequences. So
some recent works utilize hierarchies to model long sequences. However, most of
them apply sequential models for upper hierarchies, suffering from long
dependency issues. In this paper, we alleviate these issues through a
graph-based method. We first chunk the sequence with a fixed length to model
the sentence-level information. We then leverage graphs to model intra- and
cross-sentence correlations with a new attention mechanism. Additionally, due
to limited standard benchmarks for long document classification (LDC), we
propose a new challenging benchmark, totaling six datasets with up to 53k
samples and 4034 average tokens' length. Evaluation shows our model surpasses
competitive baselines by 2.6% in F1 score, and 4.8% on the longest sequence
dataset. Our method is shown to outperform hierarchical sequential models with
better performance and scalability, especially for longer sequences.
| no_new_dataset | 0.70825 |
2305.03902 | Xiaoyu Guo | Xiaoyu Guo, Xiang Wei, Qi Su, Huiqin Zhao and Shunli Zhang | Prompt What You Need: Enhancing Segmentation in Rainy Scenes with
Anchor-based Prompting | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Semantic segmentation in rainy scenes is a challenging task due to the
complex environment, class distribution imbalance, and limited annotated data.
To address these challenges, we propose a novel framework that utilizes
semi-supervised learning and pre-trained segmentation foundation model to
achieve superior performance. Specifically, our framework leverages the
semi-supervised model as the basis for generating raw semantic segmentation
results, while also serving as a guiding force to prompt pre-trained foundation
model to compensate for knowledge gaps with entropy-based anchors. In addition,
to minimize the impact of irrelevant segmentation masks generated by the
pre-trained foundation model, we also propose a mask filtering and fusion
mechanism that optimizes raw semantic segmentation results based on the
principle of minimum risk. The proposed framework achieves superior
segmentation performance on the Rainy WCity dataset and is awarded the first
prize in the sub-track of STRAIN in ICME 2023 Grand Challenges.
| [
{
"version": "v1",
"created": "Sat, 6 May 2023 02:44:54 GMT"
},
{
"version": "v2",
"created": "Fri, 12 May 2023 18:55:49 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Guo",
"Xiaoyu",
""
],
[
"Wei",
"Xiang",
""
],
[
"Su",
"Qi",
""
],
[
"Zhao",
"Huiqin",
""
],
[
"Zhang",
"Shunli",
""
]
] | TITLE: Prompt What You Need: Enhancing Segmentation in Rainy Scenes with
Anchor-based Prompting
ABSTRACT: Semantic segmentation in rainy scenes is a challenging task due to the
complex environment, class distribution imbalance, and limited annotated data.
To address these challenges, we propose a novel framework that utilizes
semi-supervised learning and pre-trained segmentation foundation model to
achieve superior performance. Specifically, our framework leverages the
semi-supervised model as the basis for generating raw semantic segmentation
results, while also serving as a guiding force to prompt pre-trained foundation
model to compensate for knowledge gaps with entropy-based anchors. In addition,
to minimize the impact of irrelevant segmentation masks generated by the
pre-trained foundation model, we also propose a mask filtering and fusion
mechanism that optimizes raw semantic segmentation results based on the
principle of minimum risk. The proposed framework achieves superior
segmentation performance on the Rainy WCity dataset and is awarded the first
prize in the sub-track of STRAIN in ICME 2023 Grand Challenges.
| no_new_dataset | 0.712639 |
2305.04490 | Aobo Kong | Aobo Kong, Shiwan Zhao, Hao Chen, Qicheng Li, Yong Qin, Ruiqi Sun and
Xiaoyan Bai | PromptRank: Unsupervised Keyphrase Extraction Using Prompt | ACL 2023 main conference | null | null | null | cs.IR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The keyphrase extraction task refers to the automatic selection of phrases
from a given document to summarize its core content. State-of-the-art (SOTA)
performance has recently been achieved by embedding-based algorithms, which
rank candidates according to how similar their embeddings are to document
embeddings. However, such solutions either struggle with the document and
candidate length discrepancies or fail to fully utilize the pre-trained
language model (PLM) without further fine-tuning. To this end, in this paper,
we propose a simple yet effective unsupervised approach, PromptRank, based on
the PLM with an encoder-decoder architecture. Specifically, PromptRank feeds
the document into the encoder and calculates the probability of generating the
candidate with a designed prompt by the decoder. We extensively evaluate the
proposed PromptRank on six widely used benchmarks. PromptRank outperforms the
SOTA approach MDERank, improving the F1 score relatively by 34.18%, 24.87%, and
17.57% for 5, 10, and 15 returned results, respectively. This demonstrates the
great potential of using prompt for unsupervised keyphrase extraction. We
release our code at https://github.com/HLT-NLP/PromptRank.
| [
{
"version": "v1",
"created": "Mon, 8 May 2023 06:34:26 GMT"
},
{
"version": "v2",
"created": "Mon, 15 May 2023 06:00:21 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Kong",
"Aobo",
""
],
[
"Zhao",
"Shiwan",
""
],
[
"Chen",
"Hao",
""
],
[
"Li",
"Qicheng",
""
],
[
"Qin",
"Yong",
""
],
[
"Sun",
"Ruiqi",
""
],
[
"Bai",
"Xiaoyan",
""
]
] | TITLE: PromptRank: Unsupervised Keyphrase Extraction Using Prompt
ABSTRACT: The keyphrase extraction task refers to the automatic selection of phrases
from a given document to summarize its core content. State-of-the-art (SOTA)
performance has recently been achieved by embedding-based algorithms, which
rank candidates according to how similar their embeddings are to document
embeddings. However, such solutions either struggle with the document and
candidate length discrepancies or fail to fully utilize the pre-trained
language model (PLM) without further fine-tuning. To this end, in this paper,
we propose a simple yet effective unsupervised approach, PromptRank, based on
the PLM with an encoder-decoder architecture. Specifically, PromptRank feeds
the document into the encoder and calculates the probability of generating the
candidate with a designed prompt by the decoder. We extensively evaluate the
proposed PromptRank on six widely used benchmarks. PromptRank outperforms the
SOTA approach MDERank, improving the F1 score relatively by 34.18%, 24.87%, and
17.57% for 5, 10, and 15 returned results, respectively. This demonstrates the
great potential of using prompt for unsupervised keyphrase extraction. We
release our code at https://github.com/HLT-NLP/PromptRank.
| no_new_dataset | 0.710025 |
2305.04582 | Leonhard Hennig | Leonhard Hennig, Philippe Thomas, Sebastian M\"oller | MultiTACRED: A Multilingual Version of the TAC Relation Extraction
Dataset | Accepted at ACL 2023 | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Relation extraction (RE) is a fundamental task in information extraction,
whose extension to multilingual settings has been hindered by the lack of
supervised resources comparable in size to large English datasets such as
TACRED (Zhang et al., 2017). To address this gap, we introduce the MultiTACRED
dataset, covering 12 typologically diverse languages from 9 language families,
which is created by machine-translating TACRED instances and automatically
projecting their entity annotations. We analyze translation and annotation
projection quality, identify error categories, and experimentally evaluate
fine-tuned pretrained mono- and multilingual language models in common transfer
learning scenarios. Our analyses show that machine translation is a viable
strategy to transfer RE instances, with native speakers judging more than 83%
of the translated instances to be linguistically and semantically acceptable.
We find monolingual RE model performance to be comparable to the English
original for many of the target languages, and that multilingual models trained
on a combination of English and target language data can outperform their
monolingual counterparts. However, we also observe a variety of translation and
annotation projection errors, both due to the MT systems and linguistic
features of the target languages, such as pronoun-dropping, compounding and
inflection, that degrade dataset quality and RE model performance.
| [
{
"version": "v1",
"created": "Mon, 8 May 2023 09:48:21 GMT"
},
{
"version": "v2",
"created": "Mon, 15 May 2023 07:24:58 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Hennig",
"Leonhard",
""
],
[
"Thomas",
"Philippe",
""
],
[
"Möller",
"Sebastian",
""
]
] | TITLE: MultiTACRED: A Multilingual Version of the TAC Relation Extraction
Dataset
ABSTRACT: Relation extraction (RE) is a fundamental task in information extraction,
whose extension to multilingual settings has been hindered by the lack of
supervised resources comparable in size to large English datasets such as
TACRED (Zhang et al., 2017). To address this gap, we introduce the MultiTACRED
dataset, covering 12 typologically diverse languages from 9 language families,
which is created by machine-translating TACRED instances and automatically
projecting their entity annotations. We analyze translation and annotation
projection quality, identify error categories, and experimentally evaluate
fine-tuned pretrained mono- and multilingual language models in common transfer
learning scenarios. Our analyses show that machine translation is a viable
strategy to transfer RE instances, with native speakers judging more than 83%
of the translated instances to be linguistically and semantically acceptable.
We find monolingual RE model performance to be comparable to the English
original for many of the target languages, and that multilingual models trained
on a combination of English and target language data can outperform their
monolingual counterparts. However, we also observe a variety of translation and
annotation projection errors, both due to the MT systems and linguistic
features of the target languages, such as pronoun-dropping, compounding and
inflection, that degrade dataset quality and RE model performance.
| new_dataset | 0.716807 |
2305.05280 | Han Wu | Han Wu, Mingjie Zhan, Haochen Tan, Zhaohui Hou, Ding Liang, and Linqi
Song | VCSUM: A Versatile Chinese Meeting Summarization Dataset | Findings of ACL 2023 (long paper). GitHub:
https://github.com/hahahawu/VCSum | null | null | null | cs.CL cs.AI | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Compared to news and chat summarization, the development of meeting
summarization is hugely decelerated by the limited data. To this end, we
introduce a versatile Chinese meeting summarization dataset, dubbed VCSum,
consisting of 239 real-life meetings, with a total duration of over 230 hours.
We claim our dataset is versatile because we provide the annotations of topic
segmentation, headlines, segmentation summaries, overall meeting summaries, and
salient sentences for each meeting transcript. As such, the dataset can adapt
to various summarization tasks or methods, including segmentation-based
summarization, multi-granularity summarization and retrieval-then-generate
summarization. Our analysis confirms the effectiveness and robustness of VCSum.
We also provide a set of benchmark models regarding different downstream
summarization tasks on VCSum to facilitate further research. The dataset and
code will be released at https://github.com/hahahawu/VCSum.
| [
{
"version": "v1",
"created": "Tue, 9 May 2023 09:07:15 GMT"
},
{
"version": "v2",
"created": "Mon, 15 May 2023 09:30:39 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Wu",
"Han",
""
],
[
"Zhan",
"Mingjie",
""
],
[
"Tan",
"Haochen",
""
],
[
"Hou",
"Zhaohui",
""
],
[
"Liang",
"Ding",
""
],
[
"Song",
"Linqi",
""
]
] | TITLE: VCSUM: A Versatile Chinese Meeting Summarization Dataset
ABSTRACT: Compared to news and chat summarization, the development of meeting
summarization is hugely decelerated by the limited data. To this end, we
introduce a versatile Chinese meeting summarization dataset, dubbed VCSum,
consisting of 239 real-life meetings, with a total duration of over 230 hours.
We claim our dataset is versatile because we provide the annotations of topic
segmentation, headlines, segmentation summaries, overall meeting summaries, and
salient sentences for each meeting transcript. As such, the dataset can adapt
to various summarization tasks or methods, including segmentation-based
summarization, multi-granularity summarization and retrieval-then-generate
summarization. Our analysis confirms the effectiveness and robustness of VCSum.
We also provide a set of benchmark models regarding different downstream
summarization tasks on VCSum to facilitate further research. The dataset and
code will be released at https://github.com/hahahawu/VCSum.
| new_dataset | 0.71484 |
2305.06247 | Hanwen Deng | Hanwen Deng, Weijia Zhang, Min-Ling Zhang | Rethinking the Value of Labels for Instance-Dependent Label Noise
Learning | 20 pages,2 figures | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Label noise widely exists in large-scale datasets and significantly
degenerates the performances of deep learning algorithms. Due to the
non-identifiability of the instance-dependent noise transition matrix, most
existing algorithms address the problem by assuming the noisy label generation
process to be independent of the instance features. Unfortunately, noisy labels
in real-world applications often depend on both the true label and the
features. In this work, we tackle instance-dependent label noise with a novel
deep generative model that avoids explicitly modeling the noise transition
matrix. Our algorithm leverages casual representation learning and
simultaneously identifies the high-level content and style latent factors from
the data. By exploiting the supervision information of noisy labels with
structural causal models, our empirical evaluations on a wide range of
synthetic and real-world instance-dependent label noise datasets demonstrate
that the proposed algorithm significantly outperforms the state-of-the-art
counterparts.
| [
{
"version": "v1",
"created": "Wed, 10 May 2023 15:29:07 GMT"
},
{
"version": "v2",
"created": "Sun, 14 May 2023 04:14:48 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Deng",
"Hanwen",
""
],
[
"Zhang",
"Weijia",
""
],
[
"Zhang",
"Min-Ling",
""
]
] | TITLE: Rethinking the Value of Labels for Instance-Dependent Label Noise
Learning
ABSTRACT: Label noise widely exists in large-scale datasets and significantly
degenerates the performances of deep learning algorithms. Due to the
non-identifiability of the instance-dependent noise transition matrix, most
existing algorithms address the problem by assuming the noisy label generation
process to be independent of the instance features. Unfortunately, noisy labels
in real-world applications often depend on both the true label and the
features. In this work, we tackle instance-dependent label noise with a novel
deep generative model that avoids explicitly modeling the noise transition
matrix. Our algorithm leverages casual representation learning and
simultaneously identifies the high-level content and style latent factors from
the data. By exploiting the supervision information of noisy labels with
structural causal models, our empirical evaluations on a wide range of
synthetic and real-world instance-dependent label noise datasets demonstrate
that the proposed algorithm significantly outperforms the state-of-the-art
counterparts.
| no_new_dataset | 0.711005 |
2305.06784 | Xiaoli Tang | Xiaoli Tang, Han Yu | Utility-Maximizing Bidding Strategy for Data Consumers in Auction-based
Federated Learning | null | null | null | null | cs.LG cs.AI cs.GT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Auction-based Federated Learning (AFL) has attracted extensive research
interest due to its ability to motivate data owners to join FL through economic
means. Existing works assume that only one data consumer and multiple data
owners exist in an AFL marketplace (i.e., a monopoly market). Therefore, data
owners bid to join the data consumer for FL. However, this assumption is not
realistic in practical AFL marketplaces in which multiple data consumers can
compete to attract data owners to join their respective FL tasks. In this
paper, we bridge this gap by proposing a first-of-its-kind utility-maximizing
bidding strategy for data consumers in federated learning (Fed-Bidder). It
enables multiple FL data consumers to compete for data owners via AFL
effectively and efficiently by providing with utility estimation capabilities
which can accommodate diverse forms of winning functions, each reflecting
different market dynamics. Extensive experiments based on six commonly adopted
benchmark datasets show that Fed-Bidder is significantly more advantageous
compared to four state-of-the-art approaches.
| [
{
"version": "v1",
"created": "Thu, 11 May 2023 13:16:36 GMT"
},
{
"version": "v2",
"created": "Mon, 15 May 2023 01:49:56 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Tang",
"Xiaoli",
""
],
[
"Yu",
"Han",
""
]
] | TITLE: Utility-Maximizing Bidding Strategy for Data Consumers in Auction-based
Federated Learning
ABSTRACT: Auction-based Federated Learning (AFL) has attracted extensive research
interest due to its ability to motivate data owners to join FL through economic
means. Existing works assume that only one data consumer and multiple data
owners exist in an AFL marketplace (i.e., a monopoly market). Therefore, data
owners bid to join the data consumer for FL. However, this assumption is not
realistic in practical AFL marketplaces in which multiple data consumers can
compete to attract data owners to join their respective FL tasks. In this
paper, we bridge this gap by proposing a first-of-its-kind utility-maximizing
bidding strategy for data consumers in federated learning (Fed-Bidder). It
enables multiple FL data consumers to compete for data owners via AFL
effectively and efficiently by providing with utility estimation capabilities
which can accommodate diverse forms of winning functions, each reflecting
different market dynamics. Extensive experiments based on six commonly adopted
benchmark datasets show that Fed-Bidder is significantly more advantageous
compared to four state-of-the-art approaches.
| no_new_dataset | 0.711588 |
2305.06830 | Shuowen Zhang | Chan Xu and Shuowen Zhang | MIMO Radar Transmit Signal Optimization for Target Localization
Exploiting Prior Information | This is the longer version of a paper to appear in IEEE International
Symposium on Information Theory (ISIT), 2023 | null | null | null | cs.IT eess.SP math.IT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper, we consider a multiple-input multiple-output (MIMO) radar
system for localizing a target based on its reflected echo signals.
Specifically, we aim to estimate the random and unknown angle information of
the target, by exploiting its prior distribution information. First, we
characterize the estimation performance by deriving the posterior Cram\'er-Rao
bound (PCRB), which quantifies a lower bound of the estimation mean-squared
error (MSE). Since the PCRB is in a complicated form, we derive a tight upper
bound of it to approximate the estimation performance. Based on this, we
analytically show that by exploiting the prior distribution information, the
PCRB is always no larger than the Cram\'er-Rao bound (CRB) averaged over random
angle realizations without prior information exploitation. Next, we formulate
the transmit signal optimization problem to minimize the PCRB upper bound. We
show that the optimal sample covariance matrix has a rank-one structure, and
derive the optimal signal solution in closed form. Numerical results show that
our proposed design achieves significantly improved PCRB performance compared
to various benchmark schemes.
| [
{
"version": "v1",
"created": "Thu, 11 May 2023 14:23:05 GMT"
},
{
"version": "v2",
"created": "Mon, 15 May 2023 05:03:42 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Xu",
"Chan",
""
],
[
"Zhang",
"Shuowen",
""
]
] | TITLE: MIMO Radar Transmit Signal Optimization for Target Localization
Exploiting Prior Information
ABSTRACT: In this paper, we consider a multiple-input multiple-output (MIMO) radar
system for localizing a target based on its reflected echo signals.
Specifically, we aim to estimate the random and unknown angle information of
the target, by exploiting its prior distribution information. First, we
characterize the estimation performance by deriving the posterior Cram\'er-Rao
bound (PCRB), which quantifies a lower bound of the estimation mean-squared
error (MSE). Since the PCRB is in a complicated form, we derive a tight upper
bound of it to approximate the estimation performance. Based on this, we
analytically show that by exploiting the prior distribution information, the
PCRB is always no larger than the Cram\'er-Rao bound (CRB) averaged over random
angle realizations without prior information exploitation. Next, we formulate
the transmit signal optimization problem to minimize the PCRB upper bound. We
show that the optimal sample covariance matrix has a rank-one structure, and
derive the optimal signal solution in closed form. Numerical results show that
our proposed design achieves significantly improved PCRB performance compared
to various benchmark schemes.
| no_new_dataset | 0.708572 |
2305.07205 | Gopi Krishna Jha | Gopi Krishna Jha, Anthony Thomas, Nilesh Jain, Sameh Gobriel, Tajana
Rosing, Ravi Iyer | Mem-Rec: Memory Efficient Recommendation System using Alternative
Representation | null | null | null | null | cs.IR cs.AI cs.LG | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Deep learning-based recommendation systems (e.g., DLRMs) are widely used AI
models to provide high-quality personalized recommendations. Training data used
for modern recommendation systems commonly includes categorical features taking
on tens-of-millions of possible distinct values. These categorical tokens are
typically assigned learned vector representations, that are stored in large
embedding tables, on the order of 100s of GB. Storing and accessing these
tables represent a substantial burden in commercial deployments. Our work
proposes MEM-REC, a novel alternative representation approach for embedding
tables. MEM-REC leverages bloom filters and hashing methods to encode
categorical features using two cache-friendly embedding tables. The first table
(token embedding) contains raw embeddings (i.e. learned vector representation),
and the second table (weight embedding), which is much smaller, contains
weights to scale these raw embeddings to provide better discriminative
capability to each data point. We provide a detailed architecture, design and
analysis of MEM-REC addressing trade-offs in accuracy and computation
requirements, in comparison with state-of-the-art techniques. We show that
MEM-REC can not only maintain the recommendation quality and significantly
reduce the memory footprint for commercial scale recommendation models but can
also improve the embedding latency. In particular, based on our results,
MEM-REC compresses the MLPerf CriteoTB benchmark DLRM model size by 2900x and
performs up to 3.4x faster embeddings while achieving the same AUC as that of
the full uncompressed model.
| [
{
"version": "v1",
"created": "Fri, 12 May 2023 02:36:07 GMT"
},
{
"version": "v2",
"created": "Mon, 15 May 2023 01:50:51 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Jha",
"Gopi Krishna",
""
],
[
"Thomas",
"Anthony",
""
],
[
"Jain",
"Nilesh",
""
],
[
"Gobriel",
"Sameh",
""
],
[
"Rosing",
"Tajana",
""
],
[
"Iyer",
"Ravi",
""
]
] | TITLE: Mem-Rec: Memory Efficient Recommendation System using Alternative
Representation
ABSTRACT: Deep learning-based recommendation systems (e.g., DLRMs) are widely used AI
models to provide high-quality personalized recommendations. Training data used
for modern recommendation systems commonly includes categorical features taking
on tens-of-millions of possible distinct values. These categorical tokens are
typically assigned learned vector representations, that are stored in large
embedding tables, on the order of 100s of GB. Storing and accessing these
tables represent a substantial burden in commercial deployments. Our work
proposes MEM-REC, a novel alternative representation approach for embedding
tables. MEM-REC leverages bloom filters and hashing methods to encode
categorical features using two cache-friendly embedding tables. The first table
(token embedding) contains raw embeddings (i.e. learned vector representation),
and the second table (weight embedding), which is much smaller, contains
weights to scale these raw embeddings to provide better discriminative
capability to each data point. We provide a detailed architecture, design and
analysis of MEM-REC addressing trade-offs in accuracy and computation
requirements, in comparison with state-of-the-art techniques. We show that
MEM-REC can not only maintain the recommendation quality and significantly
reduce the memory footprint for commercial scale recommendation models but can
also improve the embedding latency. In particular, based on our results,
MEM-REC compresses the MLPerf CriteoTB benchmark DLRM model size by 2900x and
performs up to 3.4x faster embeddings while achieving the same AUC as that of
the full uncompressed model.
| no_new_dataset | 0.711249 |
2305.07670 | Deivendran P | P. Deivendran, S. Selvakanmani, S. Jegadeesan, V. Vinoth Kumar | Liver Infection Prediction Analysis using Machine Learning to Evaluate
Analytical Performance in Neural Networks by Optimization Techniques | null | null | 10.14445/22315381/IJETT-V71I3P240 | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | Liver infection is a common disease, which poses a great threat to human
health, but there is still able to identify an optimal technique that can be
used on large-level screening. This paper deals with ML algorithms using
different data sets and predictive analyses. Therefore, machine ML can be
utilized in different diseases for integrating a piece of pattern for
visualization. This paper deals with various machine learning algorithms on
different liver illness datasets to evaluate the analytical performance using
different types of parameters and optimization techniques. The selected
classification algorithms analyze the difference in results and find out the
most excellent categorization models for liver disease. Machine learning
optimization is the procedure of modifying hyperparameters in arrange to employ
one of the optimization approaches to minimise the cost function. To set the
hyperparameter, include a number of Phosphotase,Direct Billirubin, Protiens,
Albumin and Albumin Globulin. Since it describes the difference linking the
predictable parameter's true importance and the model's prediction, it is
crucial to minimise the cost function.
| [
{
"version": "v1",
"created": "Thu, 11 May 2023 14:40:39 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Deivendran",
"P.",
""
],
[
"Selvakanmani",
"S.",
""
],
[
"Jegadeesan",
"S.",
""
],
[
"Kumar",
"V. Vinoth",
""
]
] | TITLE: Liver Infection Prediction Analysis using Machine Learning to Evaluate
Analytical Performance in Neural Networks by Optimization Techniques
ABSTRACT: Liver infection is a common disease, which poses a great threat to human
health, but there is still able to identify an optimal technique that can be
used on large-level screening. This paper deals with ML algorithms using
different data sets and predictive analyses. Therefore, machine ML can be
utilized in different diseases for integrating a piece of pattern for
visualization. This paper deals with various machine learning algorithms on
different liver illness datasets to evaluate the analytical performance using
different types of parameters and optimization techniques. The selected
classification algorithms analyze the difference in results and find out the
most excellent categorization models for liver disease. Machine learning
optimization is the procedure of modifying hyperparameters in arrange to employ
one of the optimization approaches to minimise the cost function. To set the
hyperparameter, include a number of Phosphotase,Direct Billirubin, Protiens,
Albumin and Albumin Globulin. Since it describes the difference linking the
predictable parameter's true importance and the model's prediction, it is
crucial to minimise the cost function.
| no_new_dataset | 0.707398 |
2305.07685 | Lisa K\"uhnel | Lisa K\"uhnel, Julian Schneider, Ines Perrar, Tim Adams, Fabian
Prasser, Ute N\"othlings, Holger Fr\"ohlich, Juliane Fluck | Synthetic data generation for a longitudinal cohort study -- Evaluation,
method extension and reproduction of published data analysis results | null | null | null | null | stat.ME cs.LG | http://creativecommons.org/licenses/by/4.0/ | Access to individual-level health data is essential for gaining new insights
and advancing science. In particular, modern methods based on artificial
intelligence rely on the availability of and access to large datasets. In the
health sector, access to individual-level data is often challenging due to
privacy concerns. A promising alternative is the generation of fully synthetic
data, i.e. data generated through a randomised process that have similar
statistical properties as the original data, but do not have a one-to-one
correspondence with the original individual-level records. In this study, we
use a state-of-the-art synthetic data generation method and perform in-depth
quality analyses of the generated data for a specific use case in the field of
nutrition. We demonstrate the need for careful analyses of synthetic data that
go beyond descriptive statistics and provide valuable insights into how to
realise the full potential of synthetic datasets. By extending the methods, but
also by thoroughly analysing the effects of sampling from a trained model, we
are able to largely reproduce significant real-world analysis results in the
chosen use case.
| [
{
"version": "v1",
"created": "Fri, 12 May 2023 13:13:55 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Kühnel",
"Lisa",
""
],
[
"Schneider",
"Julian",
""
],
[
"Perrar",
"Ines",
""
],
[
"Adams",
"Tim",
""
],
[
"Prasser",
"Fabian",
""
],
[
"Nöthlings",
"Ute",
""
],
[
"Fröhlich",
"Holger",
""
],
[
"Fluck",
"Juliane",
""
]
] | TITLE: Synthetic data generation for a longitudinal cohort study -- Evaluation,
method extension and reproduction of published data analysis results
ABSTRACT: Access to individual-level health data is essential for gaining new insights
and advancing science. In particular, modern methods based on artificial
intelligence rely on the availability of and access to large datasets. In the
health sector, access to individual-level data is often challenging due to
privacy concerns. A promising alternative is the generation of fully synthetic
data, i.e. data generated through a randomised process that have similar
statistical properties as the original data, but do not have a one-to-one
correspondence with the original individual-level records. In this study, we
use a state-of-the-art synthetic data generation method and perform in-depth
quality analyses of the generated data for a specific use case in the field of
nutrition. We demonstrate the need for careful analyses of synthetic data that
go beyond descriptive statistics and provide valuable insights into how to
realise the full potential of synthetic datasets. By extending the methods, but
also by thoroughly analysing the effects of sampling from a trained model, we
are able to largely reproduce significant real-world analysis results in the
chosen use case.
| no_new_dataset | 0.710998 |
2305.07713 | Zhe Liu | Zhe Liu, Xiaoqing Ye, Zhikang Zou, Xinwei He, Xiao Tan, Errui Ding,
Jingdong Wang, Xiang Bai | Multi-Modal 3D Object Detection by Box Matching | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Multi-modal 3D object detection has received growing attention as the
information from different sensors like LiDAR and cameras are complementary.
Most fusion methods for 3D detection rely on an accurate alignment and
calibration between 3D point clouds and RGB images. However, such an assumption
is not reliable in a real-world self-driving system, as the alignment between
different modalities is easily affected by asynchronous sensors and disturbed
sensor placement. We propose a novel {F}usion network by {B}ox {M}atching
(FBMNet) for multi-modal 3D detection, which provides an alternative way for
cross-modal feature alignment by learning the correspondence at the bounding
box level to free up the dependency of calibration during inference. With the
learned assignments between 3D and 2D object proposals, the fusion for
detection can be effectively performed by combing their ROI features. Extensive
experiments on the nuScenes dataset demonstrate that our method is much more
stable in dealing with challenging cases such as asynchronous sensors,
misaligned sensor placement, and degenerated camera images than existing fusion
methods. We hope that our FBMNet could provide an available solution to dealing
with these challenging cases for safety in real autonomous driving scenarios.
Codes will be publicly available at https://github.com/happinesslz/FBMNet.
| [
{
"version": "v1",
"created": "Fri, 12 May 2023 18:08:51 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Liu",
"Zhe",
""
],
[
"Ye",
"Xiaoqing",
""
],
[
"Zou",
"Zhikang",
""
],
[
"He",
"Xinwei",
""
],
[
"Tan",
"Xiao",
""
],
[
"Ding",
"Errui",
""
],
[
"Wang",
"Jingdong",
""
],
[
"Bai",
"Xiang",
""
]
] | TITLE: Multi-Modal 3D Object Detection by Box Matching
ABSTRACT: Multi-modal 3D object detection has received growing attention as the
information from different sensors like LiDAR and cameras are complementary.
Most fusion methods for 3D detection rely on an accurate alignment and
calibration between 3D point clouds and RGB images. However, such an assumption
is not reliable in a real-world self-driving system, as the alignment between
different modalities is easily affected by asynchronous sensors and disturbed
sensor placement. We propose a novel {F}usion network by {B}ox {M}atching
(FBMNet) for multi-modal 3D detection, which provides an alternative way for
cross-modal feature alignment by learning the correspondence at the bounding
box level to free up the dependency of calibration during inference. With the
learned assignments between 3D and 2D object proposals, the fusion for
detection can be effectively performed by combing their ROI features. Extensive
experiments on the nuScenes dataset demonstrate that our method is much more
stable in dealing with challenging cases such as asynchronous sensors,
misaligned sensor placement, and degenerated camera images than existing fusion
methods. We hope that our FBMNet could provide an available solution to dealing
with these challenging cases for safety in real autonomous driving scenarios.
Codes will be publicly available at https://github.com/happinesslz/FBMNet.
| no_new_dataset | 0.708364 |
2305.07716 | Georgia Chalvatzaki | Georgia Chalvatzaki, Ali Younes, Daljeet Nandha, An Le, Leonardo F. R.
Ribeiro, and Iryna Gurevych | Learning to Reason over Scene Graphs: A Case Study of Finetuning GPT-2
into a Robot Language Model for Grounded Task Planning | 21 pages, 6 figures | null | null | null | cs.RO cs.AI | http://creativecommons.org/licenses/by/4.0/ | Long-horizon task planning is essential for the development of intelligent
assistive and service robots. In this work, we investigate the applicability of
a smaller class of large language models (LLMs), specifically GPT-2, in robotic
task planning by learning to decompose tasks into subgoal specifications for a
planner to execute sequentially. Our method grounds the input of the LLM on the
domain that is represented as a scene graph, enabling it to translate human
requests into executable robot plans, thereby learning to reason over
long-horizon tasks, as encountered in the ALFRED benchmark. We compare our
approach with classical planning and baseline methods to examine the
applicability and generalizability of LLM-based planners. Our findings suggest
that the knowledge stored in an LLM can be effectively grounded to perform
long-horizon task planning, demonstrating the promising potential for the
future application of neuro-symbolic planning methods in robotics.
| [
{
"version": "v1",
"created": "Fri, 12 May 2023 18:14:32 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Chalvatzaki",
"Georgia",
""
],
[
"Younes",
"Ali",
""
],
[
"Nandha",
"Daljeet",
""
],
[
"Le",
"An",
""
],
[
"Ribeiro",
"Leonardo F. R.",
""
],
[
"Gurevych",
"Iryna",
""
]
] | TITLE: Learning to Reason over Scene Graphs: A Case Study of Finetuning GPT-2
into a Robot Language Model for Grounded Task Planning
ABSTRACT: Long-horizon task planning is essential for the development of intelligent
assistive and service robots. In this work, we investigate the applicability of
a smaller class of large language models (LLMs), specifically GPT-2, in robotic
task planning by learning to decompose tasks into subgoal specifications for a
planner to execute sequentially. Our method grounds the input of the LLM on the
domain that is represented as a scene graph, enabling it to translate human
requests into executable robot plans, thereby learning to reason over
long-horizon tasks, as encountered in the ALFRED benchmark. We compare our
approach with classical planning and baseline methods to examine the
applicability and generalizability of LLM-based planners. Our findings suggest
that the knowledge stored in an LLM can be effectively grounded to perform
long-horizon task planning, demonstrating the promising potential for the
future application of neuro-symbolic planning methods in robotics.
| no_new_dataset | 0.710007 |
2305.07717 | Souad Taouti | Souad Taouti, Hadda Cherroun and Djelloul Ziadi | Parallel Tree Kernel Computation | 9 pages, 4 figures | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Tree kernels are fundamental tools that have been leveraged in many
applications, particularly those based on machine learning for Natural Language
Processing tasks. In this paper, we devise a parallel implementation of the
sequential algorithm for the computation of some tree kernels of two finite
sets of trees (Ouali-Sebti, 2015). Our comparison is narrowed on a sequential
implementation of SubTree kernel computation. This latter is mainly reduced to
an intersection of weighted tree automata. Our approach relies on the nature of
the data parallelism source inherent in this computation by deploying the
MapReduce paradigm. One of the key benefits of our approach is its versatility
in being adaptable to a wide range of substructure tree kernel-based learning
methods. To evaluate the efficacy of our parallel approach, we conducted a
series of experiments that compared it against the sequential version using a
diverse set of synthetic tree language datasets that were manually crafted for
our analysis. The reached results clearly demonstrate that the proposed
parallel algorithm outperforms the sequential one in terms of latency.
| [
{
"version": "v1",
"created": "Fri, 12 May 2023 18:16:45 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Taouti",
"Souad",
""
],
[
"Cherroun",
"Hadda",
""
],
[
"Ziadi",
"Djelloul",
""
]
] | TITLE: Parallel Tree Kernel Computation
ABSTRACT: Tree kernels are fundamental tools that have been leveraged in many
applications, particularly those based on machine learning for Natural Language
Processing tasks. In this paper, we devise a parallel implementation of the
sequential algorithm for the computation of some tree kernels of two finite
sets of trees (Ouali-Sebti, 2015). Our comparison is narrowed on a sequential
implementation of SubTree kernel computation. This latter is mainly reduced to
an intersection of weighted tree automata. Our approach relies on the nature of
the data parallelism source inherent in this computation by deploying the
MapReduce paradigm. One of the key benefits of our approach is its versatility
in being adaptable to a wide range of substructure tree kernel-based learning
methods. To evaluate the efficacy of our parallel approach, we conducted a
series of experiments that compared it against the sequential version using a
diverse set of synthetic tree language datasets that were manually crafted for
our analysis. The reached results clearly demonstrate that the proposed
parallel algorithm outperforms the sequential one in terms of latency.
| no_new_dataset | 0.704999 |
2305.07731 | Truong Son Hy | Viet Bach Nguyen, Truong Son Hy, Long Tran-Thanh, Nhung Nghiem | Predicting COVID-19 pandemic by spatio-temporal graph neural networks: A
New Zealand's study | null | null | null | null | cs.LG physics.soc-ph | http://creativecommons.org/licenses/by/4.0/ | Modeling and simulations of pandemic dynamics play an essential role in
understanding and addressing the spreading of highly infectious diseases such
as COVID-19. In this work, we propose a novel deep learning architecture named
Attention-based Multiresolution Graph Neural Networks (ATMGNN) that learns to
combine the spatial graph information, i.e. geographical data, with the
temporal information, i.e. timeseries data of number of COVID-19 cases, to
predict the future dynamics of the pandemic. The key innovation is that our
method can capture the multiscale structures of the spatial graph via a
learning to cluster algorithm in a data-driven manner. This allows our
architecture to learn to pick up either local or global signals of a pandemic,
and model both the long-range spatial and temporal dependencies. Importantly,
we collected and assembled a new dataset for New Zealand. We established a
comprehensive benchmark of statistical methods, temporal architectures, graph
neural networks along with our spatio-temporal model. We also incorporated
socioeconomic cross-sectional data to further enhance our prediction. Our
proposed model have shown highly robust predictions and outperformed all other
baselines in various metrics for our new dataset of New Zealand along with
existing datasets of England, France, Italy and Spain. For a future work, we
plan to extend our work for real-time prediction and global scale. Our data and
source code are publicly available at https://github.com/HySonLab/pandemic_tgnn
| [
{
"version": "v1",
"created": "Fri, 12 May 2023 19:00:17 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Nguyen",
"Viet Bach",
""
],
[
"Hy",
"Truong Son",
""
],
[
"Tran-Thanh",
"Long",
""
],
[
"Nghiem",
"Nhung",
""
]
] | TITLE: Predicting COVID-19 pandemic by spatio-temporal graph neural networks: A
New Zealand's study
ABSTRACT: Modeling and simulations of pandemic dynamics play an essential role in
understanding and addressing the spreading of highly infectious diseases such
as COVID-19. In this work, we propose a novel deep learning architecture named
Attention-based Multiresolution Graph Neural Networks (ATMGNN) that learns to
combine the spatial graph information, i.e. geographical data, with the
temporal information, i.e. timeseries data of number of COVID-19 cases, to
predict the future dynamics of the pandemic. The key innovation is that our
method can capture the multiscale structures of the spatial graph via a
learning to cluster algorithm in a data-driven manner. This allows our
architecture to learn to pick up either local or global signals of a pandemic,
and model both the long-range spatial and temporal dependencies. Importantly,
we collected and assembled a new dataset for New Zealand. We established a
comprehensive benchmark of statistical methods, temporal architectures, graph
neural networks along with our spatio-temporal model. We also incorporated
socioeconomic cross-sectional data to further enhance our prediction. Our
proposed model have shown highly robust predictions and outperformed all other
baselines in various metrics for our new dataset of New Zealand along with
existing datasets of England, France, Italy and Spain. For a future work, we
plan to extend our work for real-time prediction and global scale. Our data and
source code are publicly available at https://github.com/HySonLab/pandemic_tgnn
| new_dataset | 0.716615 |
2305.07762 | Keyu Zhu | Keyu Zhu, Nabeel Gillani, Pascal Van Hentenryck | Impacts of Differential Privacy on Fostering more Racially and
Ethnically Diverse Elementary Schools | null | null | null | null | cs.CY | http://creativecommons.org/licenses/by/4.0/ | In the face of increasingly severe privacy threats in the era of data and AI,
the US Census Bureau has recently adopted differential privacy, the de facto
standard of privacy protection for the 2020 Census release. Enforcing
differential privacy involves adding carefully calibrated random noise to
sensitive demographic information prior to its release. This change has the
potential to impact policy decisions like political redistricting and other
high-stakes practices, partly because tremendous federal funds and resources
are allocated according to datasets (like Census data) released by the US
government. One under-explored yet important application of such data is the
redrawing of school attendance boundaries to foster less demographically
segregated schools. In this study, we ask: how differential privacy might
impact diversity-promoting boundaries in terms of resulting levels of
segregation, student travel times, and school switching requirements?
Simulating alternative boundaries using differentially-private student counts
across 67 Georgia districts, we find that increasing data privacy requirements
decreases the extent to which alternative boundaries might reduce segregation
and foster more diverse and integrated schools, largely by reducing the number
of students who would switch schools under boundary changes. Impacts on travel
times are minimal. These findings point to a privacy-diversity tradeoff local
educational policymakers may face in forthcoming years, particularly as
computational methods are increasingly poised to facilitate attendance boundary
redrawings in the pursuit of less segregated schools.
| [
{
"version": "v1",
"created": "Fri, 12 May 2023 21:06:15 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Zhu",
"Keyu",
""
],
[
"Gillani",
"Nabeel",
""
],
[
"Van Hentenryck",
"Pascal",
""
]
] | TITLE: Impacts of Differential Privacy on Fostering more Racially and
Ethnically Diverse Elementary Schools
ABSTRACT: In the face of increasingly severe privacy threats in the era of data and AI,
the US Census Bureau has recently adopted differential privacy, the de facto
standard of privacy protection for the 2020 Census release. Enforcing
differential privacy involves adding carefully calibrated random noise to
sensitive demographic information prior to its release. This change has the
potential to impact policy decisions like political redistricting and other
high-stakes practices, partly because tremendous federal funds and resources
are allocated according to datasets (like Census data) released by the US
government. One under-explored yet important application of such data is the
redrawing of school attendance boundaries to foster less demographically
segregated schools. In this study, we ask: how differential privacy might
impact diversity-promoting boundaries in terms of resulting levels of
segregation, student travel times, and school switching requirements?
Simulating alternative boundaries using differentially-private student counts
across 67 Georgia districts, we find that increasing data privacy requirements
decreases the extent to which alternative boundaries might reduce segregation
and foster more diverse and integrated schools, largely by reducing the number
of students who would switch schools under boundary changes. Impacts on travel
times are minimal. These findings point to a privacy-diversity tradeoff local
educational policymakers may face in forthcoming years, particularly as
computational methods are increasingly poised to facilitate attendance boundary
redrawings in the pursuit of less segregated schools.
| no_new_dataset | 0.706349 |
2305.07763 | Yuheng Wang | Yuheng Wang, Paul Fodor, Michael Kifer | Knowledge Authoring for Rules and Actions | null | null | null | null | cs.CL cs.AI | http://creativecommons.org/licenses/by/4.0/ | Knowledge representation and reasoning (KRR) systems describe and reason with
complex concepts and relations in the form of facts and rules. Unfortunately,
wide deployment of KRR systems runs into the problem that domain experts have
great difficulty constructing correct logical representations of their domain
knowledge. Knowledge engineers can help with this construction process, but
there is a deficit of such specialists. The earlier Knowledge Authoring Logic
Machine (KALM) based on Controlled Natural Language (CNL) was shown to have
very high accuracy for authoring facts and questions. More recently, KALMFL, a
successor of KALM, replaced CNL with factual English, which is much less
restrictive and requires very little training from users. However, KALMFL has
limitations in representing certain types of knowledge, such as authoring rules
for multi-step reasoning or understanding actions with timestamps. To address
these limitations, we propose KALMRA to enable authoring of rules and actions.
Our evaluation using the UTI guidelines benchmark shows that KALMRA achieves a
high level of correctness (100%) on rule authoring. When used for authoring and
reasoning with actions, KALMRA achieves more than 99.3% correctness on the bAbI
benchmark, demonstrating its effectiveness in more sophisticated KRR jobs.
Finally, we illustrate the logical reasoning capabilities of KALMRA by drawing
attention to the problems faced by the recently made famous AI, ChatGPT.
| [
{
"version": "v1",
"created": "Fri, 12 May 2023 21:08:35 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Wang",
"Yuheng",
""
],
[
"Fodor",
"Paul",
""
],
[
"Kifer",
"Michael",
""
]
] | TITLE: Knowledge Authoring for Rules and Actions
ABSTRACT: Knowledge representation and reasoning (KRR) systems describe and reason with
complex concepts and relations in the form of facts and rules. Unfortunately,
wide deployment of KRR systems runs into the problem that domain experts have
great difficulty constructing correct logical representations of their domain
knowledge. Knowledge engineers can help with this construction process, but
there is a deficit of such specialists. The earlier Knowledge Authoring Logic
Machine (KALM) based on Controlled Natural Language (CNL) was shown to have
very high accuracy for authoring facts and questions. More recently, KALMFL, a
successor of KALM, replaced CNL with factual English, which is much less
restrictive and requires very little training from users. However, KALMFL has
limitations in representing certain types of knowledge, such as authoring rules
for multi-step reasoning or understanding actions with timestamps. To address
these limitations, we propose KALMRA to enable authoring of rules and actions.
Our evaluation using the UTI guidelines benchmark shows that KALMRA achieves a
high level of correctness (100%) on rule authoring. When used for authoring and
reasoning with actions, KALMRA achieves more than 99.3% correctness on the bAbI
benchmark, demonstrating its effectiveness in more sophisticated KRR jobs.
Finally, we illustrate the logical reasoning capabilities of KALMRA by drawing
attention to the problems faced by the recently made famous AI, ChatGPT.
| no_new_dataset | 0.707859 |
2305.07767 | Ryan Boldi | Ryan Boldi and Lee Spector | Can the Problem-Solving Benefits of Quality Diversity Be Obtained
Without Explicit Diversity Maintenance? | GECCO 2023 Workshop on Quality Diversity Algorithm Benchmarks | null | null | null | cs.NE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | When using Quality Diversity (QD) optimization to solve hard exploration or
deceptive search problems, we assume that diversity is extrinsically valuable.
This means that diversity is important to help us reach an objective, but is
not an objective in itself. Often, in these domains, practitioners benchmark
their QD algorithms against single objective optimization frameworks. In this
paper, we argue that the correct comparison should be made to
\emph{multi-objective} optimization frameworks. This is because single
objective optimization frameworks rely on the aggregation of sub-objectives,
which could result in decreased information that is crucial for maintaining
diverse populations automatically. In order to facilitate a fair comparison
between quality diversity and multi-objective optimization, we present a method
that utilizes dimensionality reduction to automatically determine a set of
behavioral descriptors for an individual, as well as a set of objectives for an
individual to solve. Using the former, one can generate solutions using
standard quality diversity optimization techniques, and using the latter, one
can generate solutions using standard multi-objective optimization techniques.
This allows for a level comparison between these two classes of algorithms,
without requiring domain and algorithm specific modifications to facilitate a
comparison.
| [
{
"version": "v1",
"created": "Fri, 12 May 2023 21:24:04 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Boldi",
"Ryan",
""
],
[
"Spector",
"Lee",
""
]
] | TITLE: Can the Problem-Solving Benefits of Quality Diversity Be Obtained
Without Explicit Diversity Maintenance?
ABSTRACT: When using Quality Diversity (QD) optimization to solve hard exploration or
deceptive search problems, we assume that diversity is extrinsically valuable.
This means that diversity is important to help us reach an objective, but is
not an objective in itself. Often, in these domains, practitioners benchmark
their QD algorithms against single objective optimization frameworks. In this
paper, we argue that the correct comparison should be made to
\emph{multi-objective} optimization frameworks. This is because single
objective optimization frameworks rely on the aggregation of sub-objectives,
which could result in decreased information that is crucial for maintaining
diverse populations automatically. In order to facilitate a fair comparison
between quality diversity and multi-objective optimization, we present a method
that utilizes dimensionality reduction to automatically determine a set of
behavioral descriptors for an individual, as well as a set of objectives for an
individual to solve. Using the former, one can generate solutions using
standard quality diversity optimization techniques, and using the latter, one
can generate solutions using standard multi-objective optimization techniques.
This allows for a level comparison between these two classes of algorithms,
without requiring domain and algorithm specific modifications to facilitate a
comparison.
| no_new_dataset | 0.709787 |
2305.07783 | Haisheng Fu | Binglin Li and Jie Liang and Haisheng Fu and Jingning Han | ROI-based Deep Image Compression with Swin Transformers | This paper has been accepted by ICASSP 2023 | null | null | null | cs.CV eess.IV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Encoding the Region Of Interest (ROI) with better quality than the background
has many applications including video conferencing systems, video surveillance
and object-oriented vision tasks. In this paper, we propose a ROI-based image
compression framework with Swin transformers as main building blocks for the
autoencoder network. The binary ROI mask is integrated into different layers of
the network to provide spatial information guidance. Based on the ROI mask, we
can control the relative importance of the ROI and non-ROI by modifying the
corresponding Lagrange multiplier $ \lambda $ for different regions.
Experimental results show our model achieves higher ROI PSNR than other methods
and modest average PSNR for human evaluation. When tested on models pre-trained
with original images, it has superior object detection and instance
segmentation performance on the COCO validation dataset.
| [
{
"version": "v1",
"created": "Fri, 12 May 2023 22:05:44 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Li",
"Binglin",
""
],
[
"Liang",
"Jie",
""
],
[
"Fu",
"Haisheng",
""
],
[
"Han",
"Jingning",
""
]
] | TITLE: ROI-based Deep Image Compression with Swin Transformers
ABSTRACT: Encoding the Region Of Interest (ROI) with better quality than the background
has many applications including video conferencing systems, video surveillance
and object-oriented vision tasks. In this paper, we propose a ROI-based image
compression framework with Swin transformers as main building blocks for the
autoencoder network. The binary ROI mask is integrated into different layers of
the network to provide spatial information guidance. Based on the ROI mask, we
can control the relative importance of the ROI and non-ROI by modifying the
corresponding Lagrange multiplier $ \lambda $ for different regions.
Experimental results show our model achieves higher ROI PSNR than other methods
and modest average PSNR for human evaluation. When tested on models pre-trained
with original images, it has superior object detection and instance
segmentation performance on the COCO validation dataset.
| no_new_dataset | 0.710986 |
2305.07815 | Md Adnan Arefeen | Md Adnan Arefeen, Zhouyu Li, Md Yusuf Sarwar Uddin, Anupam Das | MetaMorphosis: Task-oriented Privacy Cognizant Feature Generation for
Multi-task Learning | Preprint version, 22 pages. Keywords: Multi-task learning, neural
networks, collaborative intelligence, differential privacy, task privacy | null | 10.1145/3576842.3582372 | null | cs.CV cs.CR cs.DC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | With the growth of computer vision applications, deep learning, and edge
computing contribute to ensuring practical collaborative intelligence (CI) by
distributing the workload among edge devices and the cloud. However, running
separate single-task models on edge devices is inefficient regarding the
required computational resource and time. In this context, multi-task learning
allows leveraging a single deep learning model for performing multiple tasks,
such as semantic segmentation and depth estimation on incoming video frames.
This single processing pipeline generates common deep features that are shared
among multi-task modules. However, in a collaborative intelligence scenario,
generating common deep features has two major issues. First, the deep features
may inadvertently contain input information exposed to the downstream modules
(violating input privacy). Second, the generated universal features expose a
piece of collective information than what is intended for a certain task, in
which features for one task can be utilized to perform another task (violating
task privacy). This paper proposes a novel deep learning-based
privacy-cognizant feature generation process called MetaMorphosis that limits
inference capability to specific tasks at hand. To achieve this, we propose a
channel squeeze-excitation based feature metamorphosis module, Cross-SEC, to
achieve distinct attention of all tasks and a de-correlation loss function with
differential-privacy to train a deep learning model that produces distinct
privacy-aware features as an output for the respective tasks. With extensive
experimentation on four datasets consisting of diverse images related to scene
understanding and facial attributes, we show that MetaMorphosis outperforms
recent adversarial learning and universal feature generation methods by
guaranteeing privacy requirements in an efficient way for image and video
analytics.
| [
{
"version": "v1",
"created": "Sat, 13 May 2023 01:59:07 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Arefeen",
"Md Adnan",
""
],
[
"Li",
"Zhouyu",
""
],
[
"Uddin",
"Md Yusuf Sarwar",
""
],
[
"Das",
"Anupam",
""
]
] | TITLE: MetaMorphosis: Task-oriented Privacy Cognizant Feature Generation for
Multi-task Learning
ABSTRACT: With the growth of computer vision applications, deep learning, and edge
computing contribute to ensuring practical collaborative intelligence (CI) by
distributing the workload among edge devices and the cloud. However, running
separate single-task models on edge devices is inefficient regarding the
required computational resource and time. In this context, multi-task learning
allows leveraging a single deep learning model for performing multiple tasks,
such as semantic segmentation and depth estimation on incoming video frames.
This single processing pipeline generates common deep features that are shared
among multi-task modules. However, in a collaborative intelligence scenario,
generating common deep features has two major issues. First, the deep features
may inadvertently contain input information exposed to the downstream modules
(violating input privacy). Second, the generated universal features expose a
piece of collective information than what is intended for a certain task, in
which features for one task can be utilized to perform another task (violating
task privacy). This paper proposes a novel deep learning-based
privacy-cognizant feature generation process called MetaMorphosis that limits
inference capability to specific tasks at hand. To achieve this, we propose a
channel squeeze-excitation based feature metamorphosis module, Cross-SEC, to
achieve distinct attention of all tasks and a de-correlation loss function with
differential-privacy to train a deep learning model that produces distinct
privacy-aware features as an output for the respective tasks. With extensive
experimentation on four datasets consisting of diverse images related to scene
understanding and facial attributes, we show that MetaMorphosis outperforms
recent adversarial learning and universal feature generation methods by
guaranteeing privacy requirements in an efficient way for image and video
analytics.
| no_new_dataset | 0.710208 |
2305.07816 | Huihui Fang Miss | Huihui Fang, Fei Li, Junde Wu, Huazhu Fu, Xu Sun, Jos\'e Ignacio
Orlando, Hrvoje Bogunovi\'c, Xiulan Zhang, Yanwu Xu | PALM: Open Fundus Photograph Dataset with Pathologic Myopia Recognition
and Anatomical Structure Annotation | 10 pages, 6 figures | null | null | null | eess.IV cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Pathologic myopia (PM) is a common blinding retinal degeneration suffered by
highly myopic population. Early screening of this condition can reduce the
damage caused by the associated fundus lesions and therefore prevent vision
loss. Automated diagnostic tools based on artificial intelligence methods can
benefit this process by aiding clinicians to identify disease signs or to
screen mass populations using color fundus photographs as inputs. This paper
provides insights about PALM, our open fundus imaging dataset for pathological
myopia recognition and anatomical structure annotation. Our databases comprises
1200 images with associated labels for the pathologic myopia category and
manual annotations of the optic disc, the position of the fovea and
delineations of lesions such as patchy retinal atrophy (including peripapillary
atrophy) and retinal detachment. In addition, this paper elaborates on other
details such as the labeling process used to construct the database, the
quality and characteristics of the samples and provides other relevant usage
notes.
| [
{
"version": "v1",
"created": "Sat, 13 May 2023 02:00:06 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Fang",
"Huihui",
""
],
[
"Li",
"Fei",
""
],
[
"Wu",
"Junde",
""
],
[
"Fu",
"Huazhu",
""
],
[
"Sun",
"Xu",
""
],
[
"Orlando",
"José Ignacio",
""
],
[
"Bogunović",
"Hrvoje",
""
],
[
"Zhang",
"Xiulan",
""
],
[
"Xu",
"Yanwu",
""
]
] | TITLE: PALM: Open Fundus Photograph Dataset with Pathologic Myopia Recognition
and Anatomical Structure Annotation
ABSTRACT: Pathologic myopia (PM) is a common blinding retinal degeneration suffered by
highly myopic population. Early screening of this condition can reduce the
damage caused by the associated fundus lesions and therefore prevent vision
loss. Automated diagnostic tools based on artificial intelligence methods can
benefit this process by aiding clinicians to identify disease signs or to
screen mass populations using color fundus photographs as inputs. This paper
provides insights about PALM, our open fundus imaging dataset for pathological
myopia recognition and anatomical structure annotation. Our databases comprises
1200 images with associated labels for the pathologic myopia category and
manual annotations of the optic disc, the position of the fovea and
delineations of lesions such as patchy retinal atrophy (including peripapillary
atrophy) and retinal detachment. In addition, this paper elaborates on other
details such as the labeling process used to construct the database, the
quality and characteristics of the samples and provides other relevant usage
notes.
| new_dataset | 0.714864 |
2305.07840 | Yunsheng Ma | Yunsheng Ma, Wenqian Ye, Xu Cao, Amr Abdelraouf, Kyungtae Han, Rohit
Gupta, Ziran Wang | CEMFormer: Learning to Predict Driver Intentions from In-Cabin and
External Cameras via Spatial-Temporal Transformers | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Driver intention prediction seeks to anticipate drivers' actions by analyzing
their behaviors with respect to surrounding traffic environments. Existing
approaches primarily focus on late-fusion techniques, and neglect the
importance of maintaining consistency between predictions and prevailing
driving contexts. In this paper, we introduce a new framework called Cross-View
Episodic Memory Transformer (CEMFormer), which employs spatio-temporal
transformers to learn unified memory representations for an improved driver
intention prediction. Specifically, we develop a spatial-temporal encoder to
integrate information from both in-cabin and external camera views, along with
episodic memory representations to continuously fuse historical data.
Furthermore, we propose a novel context-consistency loss that incorporates
driving context as an auxiliary supervision signal to improve prediction
performance. Comprehensive experiments on the Brain4Cars dataset demonstrate
that CEMFormer consistently outperforms existing state-of-the-art methods in
driver intention prediction.
| [
{
"version": "v1",
"created": "Sat, 13 May 2023 05:27:36 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Ma",
"Yunsheng",
""
],
[
"Ye",
"Wenqian",
""
],
[
"Cao",
"Xu",
""
],
[
"Abdelraouf",
"Amr",
""
],
[
"Han",
"Kyungtae",
""
],
[
"Gupta",
"Rohit",
""
],
[
"Wang",
"Ziran",
""
]
] | TITLE: CEMFormer: Learning to Predict Driver Intentions from In-Cabin and
External Cameras via Spatial-Temporal Transformers
ABSTRACT: Driver intention prediction seeks to anticipate drivers' actions by analyzing
their behaviors with respect to surrounding traffic environments. Existing
approaches primarily focus on late-fusion techniques, and neglect the
importance of maintaining consistency between predictions and prevailing
driving contexts. In this paper, we introduce a new framework called Cross-View
Episodic Memory Transformer (CEMFormer), which employs spatio-temporal
transformers to learn unified memory representations for an improved driver
intention prediction. Specifically, we develop a spatial-temporal encoder to
integrate information from both in-cabin and external camera views, along with
episodic memory representations to continuously fuse historical data.
Furthermore, we propose a novel context-consistency loss that incorporates
driving context as an auxiliary supervision signal to improve prediction
performance. Comprehensive experiments on the Brain4Cars dataset demonstrate
that CEMFormer consistently outperforms existing state-of-the-art methods in
driver intention prediction.
| no_new_dataset | 0.706209 |
2305.07853 | Kuanxu Hou | Hao Zhuang, Xinjie Huang, Kuanxu Hou, Delei Kong, Chenming Hu, Zheng
Fang | EV-MGRFlowNet: Motion-Guided Recurrent Network for Unsupervised
Event-based Optical Flow with Hybrid Motion-Compensation Loss | 11 pages, 7 figures | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Event cameras offer promising properties, such as high temporal resolution
and high dynamic range. These benefits have been utilized into many machine
vision tasks, especially optical flow estimation. Currently, most existing
event-based works use deep learning to estimate optical flow. However, their
networks have not fully exploited prior hidden states and motion flows.
Additionally, their supervision strategy has not fully leveraged the geometric
constraints of event data to unlock the potential of networks. In this paper,
we propose EV-MGRFlowNet, an unsupervised event-based optical flow estimation
pipeline with motion-guided recurrent networks using a hybrid
motion-compensation loss. First, we propose a feature-enhanced recurrent
encoder network (FERE-Net) which fully utilizes prior hidden states to obtain
multi-level motion features. Then, we propose a flow-guided decoder network
(FGD-Net) to integrate prior motion flows. Finally, we design a hybrid
motion-compensation loss (HMC-Loss) to strengthen geometric constraints for the
more accurate alignment of events. Experimental results show that our method
outperforms the current state-of-the-art (SOTA) method on the MVSEC dataset,
with an average reduction of approximately 22.71% in average endpoint error
(AEE). To our knowledge, our method ranks first among unsupervised
learning-based methods.
| [
{
"version": "v1",
"created": "Sat, 13 May 2023 07:08:48 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Zhuang",
"Hao",
""
],
[
"Huang",
"Xinjie",
""
],
[
"Hou",
"Kuanxu",
""
],
[
"Kong",
"Delei",
""
],
[
"Hu",
"Chenming",
""
],
[
"Fang",
"Zheng",
""
]
] | TITLE: EV-MGRFlowNet: Motion-Guided Recurrent Network for Unsupervised
Event-based Optical Flow with Hybrid Motion-Compensation Loss
ABSTRACT: Event cameras offer promising properties, such as high temporal resolution
and high dynamic range. These benefits have been utilized into many machine
vision tasks, especially optical flow estimation. Currently, most existing
event-based works use deep learning to estimate optical flow. However, their
networks have not fully exploited prior hidden states and motion flows.
Additionally, their supervision strategy has not fully leveraged the geometric
constraints of event data to unlock the potential of networks. In this paper,
we propose EV-MGRFlowNet, an unsupervised event-based optical flow estimation
pipeline with motion-guided recurrent networks using a hybrid
motion-compensation loss. First, we propose a feature-enhanced recurrent
encoder network (FERE-Net) which fully utilizes prior hidden states to obtain
multi-level motion features. Then, we propose a flow-guided decoder network
(FGD-Net) to integrate prior motion flows. Finally, we design a hybrid
motion-compensation loss (HMC-Loss) to strengthen geometric constraints for the
more accurate alignment of events. Experimental results show that our method
outperforms the current state-of-the-art (SOTA) method on the MVSEC dataset,
with an average reduction of approximately 22.71% in average endpoint error
(AEE). To our knowledge, our method ranks first among unsupervised
learning-based methods.
| no_new_dataset | 0.71039 |
2305.07892 | Jun Shu | Jun Shu, Xiang Yuan, Deyu Meng, Zongben Xu | DAC-MR: Data Augmentation Consistency Based Meta-Regularization for
Meta-Learning | 27 pages | null | null | null | cs.LG cs.AI cs.CV | http://creativecommons.org/licenses/by/4.0/ | Meta learning recently has been heavily researched and helped advance the
contemporary machine learning. However, achieving well-performing meta-learning
model requires a large amount of training tasks with high-quality meta-data
representing the underlying task generalization goal, which is sometimes
difficult and expensive to obtain for real applications. Current
meta-data-driven meta-learning approaches, however, are fairly hard to train
satisfactory meta-models with imperfect training tasks. To address this issue,
we suggest a meta-knowledge informed meta-learning (MKIML) framework to improve
meta-learning by additionally integrating compensated meta-knowledge into
meta-learning process. We preliminarily integrate meta-knowledge into
meta-objective via using an appropriate meta-regularization (MR) objective to
regularize capacity complexity of the meta-model function class to facilitate
better generalization on unseen tasks. As a practical implementation, we
introduce data augmentation consistency to encode invariance as meta-knowledge
for instantiating MR objective, denoted by DAC-MR. The proposed DAC-MR is
hopeful to learn well-performing meta-models from training tasks with noisy,
sparse or unavailable meta-data. We theoretically demonstrate that DAC-MR can
be treated as a proxy meta-objective used to evaluate meta-model without
high-quality meta-data. Besides, meta-data-driven meta-loss objective combined
with DAC-MR is capable of achieving better meta-level generalization. 10
meta-learning tasks with different network architectures and benchmarks
substantiate the capability of our DAC-MR on aiding meta-model learning. Fine
performance of DAC-MR are obtained across all settings, and are well-aligned
with our theoretical insights. This implies that our DAC-MR is
problem-agnostic, and hopeful to be readily applied to extensive meta-learning
problems and tasks.
| [
{
"version": "v1",
"created": "Sat, 13 May 2023 11:01:47 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Shu",
"Jun",
""
],
[
"Yuan",
"Xiang",
""
],
[
"Meng",
"Deyu",
""
],
[
"Xu",
"Zongben",
""
]
] | TITLE: DAC-MR: Data Augmentation Consistency Based Meta-Regularization for
Meta-Learning
ABSTRACT: Meta learning recently has been heavily researched and helped advance the
contemporary machine learning. However, achieving well-performing meta-learning
model requires a large amount of training tasks with high-quality meta-data
representing the underlying task generalization goal, which is sometimes
difficult and expensive to obtain for real applications. Current
meta-data-driven meta-learning approaches, however, are fairly hard to train
satisfactory meta-models with imperfect training tasks. To address this issue,
we suggest a meta-knowledge informed meta-learning (MKIML) framework to improve
meta-learning by additionally integrating compensated meta-knowledge into
meta-learning process. We preliminarily integrate meta-knowledge into
meta-objective via using an appropriate meta-regularization (MR) objective to
regularize capacity complexity of the meta-model function class to facilitate
better generalization on unseen tasks. As a practical implementation, we
introduce data augmentation consistency to encode invariance as meta-knowledge
for instantiating MR objective, denoted by DAC-MR. The proposed DAC-MR is
hopeful to learn well-performing meta-models from training tasks with noisy,
sparse or unavailable meta-data. We theoretically demonstrate that DAC-MR can
be treated as a proxy meta-objective used to evaluate meta-model without
high-quality meta-data. Besides, meta-data-driven meta-loss objective combined
with DAC-MR is capable of achieving better meta-level generalization. 10
meta-learning tasks with different network architectures and benchmarks
substantiate the capability of our DAC-MR on aiding meta-model learning. Fine
performance of DAC-MR are obtained across all settings, and are well-aligned
with our theoretical insights. This implies that our DAC-MR is
problem-agnostic, and hopeful to be readily applied to extensive meta-learning
problems and tasks.
| no_new_dataset | 0.709384 |
2305.07910 | Han Fang | Han Fang and Zhifei Yang and Xianghao Zang and Chao Ban and Hao Sun | Mask to reconstruct: Cooperative Semantics Completion for Video-text
Retrieval | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Recently, masked video modeling has been widely explored and significantly
improved the model's understanding ability of visual regions at a local level.
However, existing methods usually adopt random masking and follow the same
reconstruction paradigm to complete the masked regions, which do not leverage
the correlations between cross-modal content. In this paper, we present Mask
for Semantics Completion (MASCOT) based on semantic-based masked modeling.
Specifically, after applying attention-based video masking to generate
high-informed and low-informed masks, we propose Informed Semantics Completion
to recover masked semantics information. The recovery mechanism is achieved by
aligning the masked content with the unmasked visual regions and corresponding
textual context, which makes the model capture more text-related details at a
patch level. Additionally, we shift the emphasis of reconstruction from
irrelevant backgrounds to discriminative parts to ignore regions with
low-informed masks. Furthermore, we design dual-mask co-learning to incorporate
video cues under different masks and learn more aligned video representation.
Our MASCOT performs state-of-the-art performance on four major text-video
retrieval benchmarks, including MSR-VTT, LSMDC, ActivityNet, and DiDeMo.
Extensive ablation studies demonstrate the effectiveness of the proposed
schemes.
| [
{
"version": "v1",
"created": "Sat, 13 May 2023 12:31:37 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Fang",
"Han",
""
],
[
"Yang",
"Zhifei",
""
],
[
"Zang",
"Xianghao",
""
],
[
"Ban",
"Chao",
""
],
[
"Sun",
"Hao",
""
]
] | TITLE: Mask to reconstruct: Cooperative Semantics Completion for Video-text
Retrieval
ABSTRACT: Recently, masked video modeling has been widely explored and significantly
improved the model's understanding ability of visual regions at a local level.
However, existing methods usually adopt random masking and follow the same
reconstruction paradigm to complete the masked regions, which do not leverage
the correlations between cross-modal content. In this paper, we present Mask
for Semantics Completion (MASCOT) based on semantic-based masked modeling.
Specifically, after applying attention-based video masking to generate
high-informed and low-informed masks, we propose Informed Semantics Completion
to recover masked semantics information. The recovery mechanism is achieved by
aligning the masked content with the unmasked visual regions and corresponding
textual context, which makes the model capture more text-related details at a
patch level. Additionally, we shift the emphasis of reconstruction from
irrelevant backgrounds to discriminative parts to ignore regions with
low-informed masks. Furthermore, we design dual-mask co-learning to incorporate
video cues under different masks and learn more aligned video representation.
Our MASCOT performs state-of-the-art performance on four major text-video
retrieval benchmarks, including MSR-VTT, LSMDC, ActivityNet, and DiDeMo.
Extensive ablation studies demonstrate the effectiveness of the proposed
schemes.
| no_new_dataset | 0.710038 |
2305.07927 | Chulun Zhou | Chulun Zhou, Yunlong Liang, Fandong Meng, Jinan Xu, Jinsong Su and Jie
Zhou | RC3: Regularized Contrastive Cross-lingual Cross-modal Pre-training | accepted to ACL 2023 Findings | null | null | null | cs.CL | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Multilingual vision-language (V&L) pre-training has achieved remarkable
progress in learning universal representations across different modalities and
languages. In spite of recent success, there still remain challenges limiting
further improvements of V&L pre-trained models in multilingual settings.
Particularly, current V&L pre-training methods rely heavily on strictly-aligned
multilingual image-text pairs generated from English-centric datasets through
machine translation. However, the cost of collecting and translating such
strictly-aligned datasets is usually unbearable. In this paper, we propose
Regularized Contrastive Cross-lingual Cross-modal (RC^3) pre-training, which
further exploits more abundant weakly-aligned multilingual image-text pairs.
Specifically, we design a regularized cross-lingual visio-textual contrastive
learning objective that constrains the representation proximity of
weakly-aligned visio-textual inputs according to textual relevance. Besides,
existing V&L pre-training approaches mainly deal with visual inputs by either
region-of-interest (ROI) features or patch embeddings. We flexibly integrate
the two forms of visual features into our model for pre-training and downstream
multi-modal tasks. Extensive experiments on 5 downstream multi-modal tasks
across 6 languages demonstrate the effectiveness of our proposed method over
competitive contrast models with stronger zero-shot capability.
| [
{
"version": "v1",
"created": "Sat, 13 May 2023 14:41:05 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Zhou",
"Chulun",
""
],
[
"Liang",
"Yunlong",
""
],
[
"Meng",
"Fandong",
""
],
[
"Xu",
"Jinan",
""
],
[
"Su",
"Jinsong",
""
],
[
"Zhou",
"Jie",
""
]
] | TITLE: RC3: Regularized Contrastive Cross-lingual Cross-modal Pre-training
ABSTRACT: Multilingual vision-language (V&L) pre-training has achieved remarkable
progress in learning universal representations across different modalities and
languages. In spite of recent success, there still remain challenges limiting
further improvements of V&L pre-trained models in multilingual settings.
Particularly, current V&L pre-training methods rely heavily on strictly-aligned
multilingual image-text pairs generated from English-centric datasets through
machine translation. However, the cost of collecting and translating such
strictly-aligned datasets is usually unbearable. In this paper, we propose
Regularized Contrastive Cross-lingual Cross-modal (RC^3) pre-training, which
further exploits more abundant weakly-aligned multilingual image-text pairs.
Specifically, we design a regularized cross-lingual visio-textual contrastive
learning objective that constrains the representation proximity of
weakly-aligned visio-textual inputs according to textual relevance. Besides,
existing V&L pre-training approaches mainly deal with visual inputs by either
region-of-interest (ROI) features or patch embeddings. We flexibly integrate
the two forms of visual features into our model for pre-training and downstream
multi-modal tasks. Extensive experiments on 5 downstream multi-modal tasks
across 6 languages demonstrate the effectiveness of our proposed method over
competitive contrast models with stronger zero-shot capability.
| no_new_dataset | 0.708452 |
2305.07928 | Qianglong Chen | Qianglong Chen, Feng Ji, Feng-Lin Li, Guohai Xu, Ming Yan, Ji Zhang
and Yin Zhang | AMTSS: An Adaptive Multi-Teacher Single-Student Knowledge Distillation
Framework For Multilingual Language Inference | null | null | null | null | cs.CL cs.AI | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Knowledge distillation is of key importance to launching multilingual
pre-trained language models for real applications. To support cost-effective
language inference in multilingual settings, we propose AMTSS, an adaptive
multi-teacher single-student distillation framework, which allows distilling
knowledge from multiple teachers to a single student. We first introduce an
adaptive learning strategy and teacher importance weight, which enables a
student to effectively learn from max-margin teachers and easily adapt to new
languages. Moreover, we present a shared student encoder with different
projection layers in support of multiple languages, which contributes to
largely reducing development and machine cost. Experimental results show that
AMTSS gains competitive results on the public XNLI dataset and the realistic
industrial dataset AliExpress (AE) in the E-commerce scenario.
| [
{
"version": "v1",
"created": "Sat, 13 May 2023 14:42:30 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Chen",
"Qianglong",
""
],
[
"Ji",
"Feng",
""
],
[
"Li",
"Feng-Lin",
""
],
[
"Xu",
"Guohai",
""
],
[
"Yan",
"Ming",
""
],
[
"Zhang",
"Ji",
""
],
[
"Zhang",
"Yin",
""
]
] | TITLE: AMTSS: An Adaptive Multi-Teacher Single-Student Knowledge Distillation
Framework For Multilingual Language Inference
ABSTRACT: Knowledge distillation is of key importance to launching multilingual
pre-trained language models for real applications. To support cost-effective
language inference in multilingual settings, we propose AMTSS, an adaptive
multi-teacher single-student distillation framework, which allows distilling
knowledge from multiple teachers to a single student. We first introduce an
adaptive learning strategy and teacher importance weight, which enables a
student to effectively learn from max-margin teachers and easily adapt to new
languages. Moreover, we present a shared student encoder with different
projection layers in support of multiple languages, which contributes to
largely reducing development and machine cost. Experimental results show that
AMTSS gains competitive results on the public XNLI dataset and the realistic
industrial dataset AliExpress (AE) in the E-commerce scenario.
| no_new_dataset | 0.706994 |
2305.07943 | Xinyu Lin | Xinyu Lin, Yingjie Zhou, Xun Zhang, Yipeng Liu, and Ce Zhu | Illumination-insensitive Binary Descriptor for Visual Measurement Based
on Local Inter-patch Invariance | Accepted by IEEE Transactions on Instrumentation and Measurement | IEEE Transactions on Instrumentation and Measurement 2023 | 10.1109/TIM.2023.3273211 | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Binary feature descriptors have been widely used in various visual
measurement tasks, particularly those with limited computing resources and
storage capacities. Existing binary descriptors may not perform well for
long-term visual measurement tasks due to their sensitivity to illumination
variations. It can be observed that when image illumination changes
dramatically, the relative relationship among local patches mostly remains
intact. Based on the observation, consequently, this study presents an
illumination-insensitive binary (IIB) descriptor by leveraging the local
inter-patch invariance exhibited in multiple spatial granularities to deal with
unfavorable illumination variations. By taking advantage of integral images for
local patch feature computation, a highly efficient IIB descriptor is achieved.
It can encode scalable features in multiple spatial granularities, thus
facilitating a computationally efficient hierarchical matching from coarse to
fine. Moreover, the IIB descriptor can also apply to other types of image data,
such as depth maps and semantic segmentation results, when available in some
applications. Numerical experiments on both natural and synthetic datasets
reveal that the proposed IIB descriptor outperforms state-of-the-art binary
descriptors and some testing float descriptors. The proposed IIB descriptor has
also been successfully employed in a demo system for long-term visual
localization. The code of the IIB descriptor will be publicly available.
| [
{
"version": "v1",
"created": "Sat, 13 May 2023 15:15:18 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Lin",
"Xinyu",
""
],
[
"Zhou",
"Yingjie",
""
],
[
"Zhang",
"Xun",
""
],
[
"Liu",
"Yipeng",
""
],
[
"Zhu",
"Ce",
""
]
] | TITLE: Illumination-insensitive Binary Descriptor for Visual Measurement Based
on Local Inter-patch Invariance
ABSTRACT: Binary feature descriptors have been widely used in various visual
measurement tasks, particularly those with limited computing resources and
storage capacities. Existing binary descriptors may not perform well for
long-term visual measurement tasks due to their sensitivity to illumination
variations. It can be observed that when image illumination changes
dramatically, the relative relationship among local patches mostly remains
intact. Based on the observation, consequently, this study presents an
illumination-insensitive binary (IIB) descriptor by leveraging the local
inter-patch invariance exhibited in multiple spatial granularities to deal with
unfavorable illumination variations. By taking advantage of integral images for
local patch feature computation, a highly efficient IIB descriptor is achieved.
It can encode scalable features in multiple spatial granularities, thus
facilitating a computationally efficient hierarchical matching from coarse to
fine. Moreover, the IIB descriptor can also apply to other types of image data,
such as depth maps and semantic segmentation results, when available in some
applications. Numerical experiments on both natural and synthetic datasets
reveal that the proposed IIB descriptor outperforms state-of-the-art binary
descriptors and some testing float descriptors. The proposed IIB descriptor has
also been successfully employed in a demo system for long-term visual
localization. The code of the IIB descriptor will be publicly available.
| no_new_dataset | 0.709422 |
2305.07958 | Thiago D. Sim\~ao | Patrick Wienh\"oft, Marnix Suilen, Thiago D. Sim\~ao, Clemens
Dubslaff, Christel Baier, Nils Jansen | More for Less: Safe Policy Improvement With Stronger Performance
Guarantees | Accecpted at IJCAI 2023 | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by/4.0/ | In an offline reinforcement learning setting, the safe policy improvement
(SPI) problem aims to improve the performance of a behavior policy according to
which sample data has been generated. State-of-the-art approaches to SPI
require a high number of samples to provide practical probabilistic guarantees
on the improved policy's performance. We present a novel approach to the SPI
problem that provides the means to require less data for such guarantees.
Specifically, to prove the correctness of these guarantees, we devise implicit
transformations on the data set and the underlying environment model that serve
as theoretical foundations to derive tighter improvement bounds for SPI. Our
empirical evaluation, using the well-established SPI with baseline
bootstrapping (SPIBB) algorithm, on standard benchmarks shows that our method
indeed significantly reduces the sample complexity of the SPIBB algorithm.
| [
{
"version": "v1",
"created": "Sat, 13 May 2023 16:22:21 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Wienhöft",
"Patrick",
""
],
[
"Suilen",
"Marnix",
""
],
[
"Simão",
"Thiago D.",
""
],
[
"Dubslaff",
"Clemens",
""
],
[
"Baier",
"Christel",
""
],
[
"Jansen",
"Nils",
""
]
] | TITLE: More for Less: Safe Policy Improvement With Stronger Performance
Guarantees
ABSTRACT: In an offline reinforcement learning setting, the safe policy improvement
(SPI) problem aims to improve the performance of a behavior policy according to
which sample data has been generated. State-of-the-art approaches to SPI
require a high number of samples to provide practical probabilistic guarantees
on the improved policy's performance. We present a novel approach to the SPI
problem that provides the means to require less data for such guarantees.
Specifically, to prove the correctness of these guarantees, we devise implicit
transformations on the data set and the underlying environment model that serve
as theoretical foundations to derive tighter improvement bounds for SPI. Our
empirical evaluation, using the well-established SPI with baseline
bootstrapping (SPIBB) algorithm, on standard benchmarks shows that our method
indeed significantly reduces the sample complexity of the SPIBB algorithm.
| no_new_dataset | 0.708042 |
2305.07959 | Tommaso Aldinucci | Tommaso Aldinucci | A Novel Memetic Strategy for Optimized Learning of Classification Trees | null | null | null | null | cs.LG stat.CO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Given the increasing interest in interpretable machine learning,
classification trees have again attracted the attention of the scientific
community because of their glass-box structure. These models are usually built
using greedy procedures, solving subproblems to find cuts in the feature space
that minimize some impurity measures. In contrast to this standard greedy
approach and to the recent advances in the definition of the learning problem
through MILP-based exact formulations, in this paper we propose a novel
evolutionary algorithm for the induction of classification trees that exploits
a memetic approach that is able to handle datasets with thousands of points.
Our procedure combines the exploration of the feasible space of solutions with
local searches to obtain structures with generalization capabilities that are
competitive with the state-of-the-art methods.
| [
{
"version": "v1",
"created": "Sat, 13 May 2023 16:29:10 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Aldinucci",
"Tommaso",
""
]
] | TITLE: A Novel Memetic Strategy for Optimized Learning of Classification Trees
ABSTRACT: Given the increasing interest in interpretable machine learning,
classification trees have again attracted the attention of the scientific
community because of their glass-box structure. These models are usually built
using greedy procedures, solving subproblems to find cuts in the feature space
that minimize some impurity measures. In contrast to this standard greedy
approach and to the recent advances in the definition of the learning problem
through MILP-based exact formulations, in this paper we propose a novel
evolutionary algorithm for the induction of classification trees that exploits
a memetic approach that is able to handle datasets with thousands of points.
Our procedure combines the exploration of the feasible space of solutions with
local searches to obtain structures with generalization capabilities that are
competitive with the state-of-the-art methods.
| no_new_dataset | 0.708616 |
2305.07960 | Ozer Can Devecioglu | Ozer Can Devecioglu, Serkan Kiranyaz, Amer Elhmes, Sadok Sassi, Turker
Ince, Onur Avci, Mohammad Hesam Soleimani-Babakamali, Ertugrul Taciroglu, and
Moncef Gabbouj | Sound-to-Vibration Transformation for Sensorless Motor Health Monitoring | null | null | null | null | cs.SD cs.HC eess.AS | http://creativecommons.org/licenses/by/4.0/ | Automatic sensor-based detection of motor failures such as bearing faults is
crucial for predictive maintenance in various industries. Numerous
methodologies have been developed over the years to detect bearing faults.
Despite the appearance of numerous different approaches for diagnosing faults
in motors have been proposed, vibration-based methods have become the de facto
standard and the most commonly used techniques. However, acquiring reliable
vibration signals, especially from rotating machinery, can sometimes be
infeasibly difficult due to challenging installation and operational conditions
(e.g., variations on accelerometer locations on the motor body), which will not
only alter the signal patterns significantly but may also induce severe
artifacts. Moreover, sensors are costly and require periodic maintenance to
sustain a reliable signal acquisition. To address these drawbacks and void the
need for vibration sensors, in this study, we propose a novel
sound-to-vibration transformation method that can synthesize realistic
vibration signals directly from the sound measurements regardless of the
working conditions, fault type, and fault severity. As a result, using this
transformation, the data acquired by a simple sound recorder, e.g., a mobile
phone, can be transformed into the vibration signal, which can then be used for
fault detection by a pre-trained model. The proposed method is extensively
evaluated over the benchmark Qatar University Dual-Machine Bearing Fault
Benchmark dataset (QU-DMBF), which encapsulates sound and vibration data from
two different machines operating under various conditions. Experimental results
show that this novel approach can synthesize such realistic vibration signals
that can directly be used for reliable and highly accurate motor health
monitoring.
| [
{
"version": "v1",
"created": "Sat, 13 May 2023 16:37:18 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Devecioglu",
"Ozer Can",
""
],
[
"Kiranyaz",
"Serkan",
""
],
[
"Elhmes",
"Amer",
""
],
[
"Sassi",
"Sadok",
""
],
[
"Ince",
"Turker",
""
],
[
"Avci",
"Onur",
""
],
[
"Soleimani-Babakamali",
"Mohammad Hesam",
""
],
[
"Taciroglu",
"Ertugrul",
""
],
[
"Gabbouj",
"Moncef",
""
]
] | TITLE: Sound-to-Vibration Transformation for Sensorless Motor Health Monitoring
ABSTRACT: Automatic sensor-based detection of motor failures such as bearing faults is
crucial for predictive maintenance in various industries. Numerous
methodologies have been developed over the years to detect bearing faults.
Despite the appearance of numerous different approaches for diagnosing faults
in motors have been proposed, vibration-based methods have become the de facto
standard and the most commonly used techniques. However, acquiring reliable
vibration signals, especially from rotating machinery, can sometimes be
infeasibly difficult due to challenging installation and operational conditions
(e.g., variations on accelerometer locations on the motor body), which will not
only alter the signal patterns significantly but may also induce severe
artifacts. Moreover, sensors are costly and require periodic maintenance to
sustain a reliable signal acquisition. To address these drawbacks and void the
need for vibration sensors, in this study, we propose a novel
sound-to-vibration transformation method that can synthesize realistic
vibration signals directly from the sound measurements regardless of the
working conditions, fault type, and fault severity. As a result, using this
transformation, the data acquired by a simple sound recorder, e.g., a mobile
phone, can be transformed into the vibration signal, which can then be used for
fault detection by a pre-trained model. The proposed method is extensively
evaluated over the benchmark Qatar University Dual-Machine Bearing Fault
Benchmark dataset (QU-DMBF), which encapsulates sound and vibration data from
two different machines operating under various conditions. Experimental results
show that this novel approach can synthesize such realistic vibration signals
that can directly be used for reliable and highly accurate motor health
monitoring.
| no_new_dataset | 0.643525 |
2305.07972 | Agam Shah | Agam Shah and Suvan Paturi and Sudheer Chava | Trillion Dollar Words: A New Financial Dataset, Task & Market Analysis | ACL 2023 (main) | null | null | null | cs.CL q-fin.CP | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Monetary policy pronouncements by Federal Open Market Committee (FOMC) are a
major driver of financial market returns. We construct the largest tokenized
and annotated dataset of FOMC speeches, meeting minutes, and press conference
transcripts in order to understand how monetary policy influences financial
markets. In this study, we develop a novel task of hawkish-dovish
classification and benchmark various pre-trained language models on the
proposed dataset. Using the best-performing model (RoBERTa-large), we construct
a measure of monetary policy stance for the FOMC document release days. To
evaluate the constructed measure, we study its impact on the treasury market,
stock market, and macroeconomic indicators. Our dataset, models, and code are
publicly available on Huggingface and GitHub under CC BY-NC 4.0 license.
| [
{
"version": "v1",
"created": "Sat, 13 May 2023 17:32:39 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Shah",
"Agam",
""
],
[
"Paturi",
"Suvan",
""
],
[
"Chava",
"Sudheer",
""
]
] | TITLE: Trillion Dollar Words: A New Financial Dataset, Task & Market Analysis
ABSTRACT: Monetary policy pronouncements by Federal Open Market Committee (FOMC) are a
major driver of financial market returns. We construct the largest tokenized
and annotated dataset of FOMC speeches, meeting minutes, and press conference
transcripts in order to understand how monetary policy influences financial
markets. In this study, we develop a novel task of hawkish-dovish
classification and benchmark various pre-trained language models on the
proposed dataset. Using the best-performing model (RoBERTa-large), we construct
a measure of monetary policy stance for the FOMC document release days. To
evaluate the constructed measure, we study its impact on the treasury market,
stock market, and macroeconomic indicators. Our dataset, models, and code are
publicly available on Huggingface and GitHub under CC BY-NC 4.0 license.
| new_dataset | 0.712988 |
2305.07979 | Yun Guo | Yun Guo, Xueyao Xiao, Xiaoxiong Wang, Yi Li, Yi Chang, Luxin Yan | A Two-Stage Real Image Deraining Method for GT-RAIN Challenge CVPR 2023
Workshop UG$^{\textbf{2}}$+ Track 3 | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by-sa/4.0/ | In this technical report, we briefly introduce the solution of our team
HUST\li VIE for GT-Rain Challenge in CVPR 2023 UG$^{2}$+ Track 3. In this task,
we propose an efficient two-stage framework to reconstruct a clear image from
rainy frames. Firstly, a low-rank based video deraining method is utilized to
generate pseudo GT, which fully takes the advantage of multi and aligned rainy
frames. Secondly, a transformer-based single image deraining network Uformer is
implemented to pre-train on large real rain dataset and then fine-tuned on
pseudo GT to further improve image restoration. Moreover, in terms of visual
pleasing effect, a comprehensive image processor module is utilized at the end
of pipeline. Our overall framework is elaborately designed and able to handle
both heavy rainy and foggy sequences provided in the final testing phase.
Finally, we rank 1st on the average structural similarity (SSIM) and rank 2nd
on the average peak signal-to-noise ratio (PSNR). Our code is available at
https://github.com/yunguo224/UG2_Deraining.
| [
{
"version": "v1",
"created": "Sat, 13 May 2023 18:30:27 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Guo",
"Yun",
""
],
[
"Xiao",
"Xueyao",
""
],
[
"Wang",
"Xiaoxiong",
""
],
[
"Li",
"Yi",
""
],
[
"Chang",
"Yi",
""
],
[
"Yan",
"Luxin",
""
]
] | TITLE: A Two-Stage Real Image Deraining Method for GT-RAIN Challenge CVPR 2023
Workshop UG$^{\textbf{2}}$+ Track 3
ABSTRACT: In this technical report, we briefly introduce the solution of our team
HUST\li VIE for GT-Rain Challenge in CVPR 2023 UG$^{2}$+ Track 3. In this task,
we propose an efficient two-stage framework to reconstruct a clear image from
rainy frames. Firstly, a low-rank based video deraining method is utilized to
generate pseudo GT, which fully takes the advantage of multi and aligned rainy
frames. Secondly, a transformer-based single image deraining network Uformer is
implemented to pre-train on large real rain dataset and then fine-tuned on
pseudo GT to further improve image restoration. Moreover, in terms of visual
pleasing effect, a comprehensive image processor module is utilized at the end
of pipeline. Our overall framework is elaborately designed and able to handle
both heavy rainy and foggy sequences provided in the final testing phase.
Finally, we rank 1st on the average structural similarity (SSIM) and rank 2nd
on the average peak signal-to-noise ratio (PSNR). Our code is available at
https://github.com/yunguo224/UG2_Deraining.
| no_new_dataset | 0.710377 |
2305.08017 | Cara Van Uden | Cara Van Uden and Jeremy Irvin and Mars Huang and Nathan Dean and
Jason Carr and Andrew Ng and Curtis Langlotz | How to Train Your CheXDragon: Training Chest X-Ray Models for Transfer
to Novel Tasks and Healthcare Systems | 13 pages, 12 figures | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Self-supervised learning (SSL) enables label efficient training for machine
learning models. This is essential for domains such as medical imaging, where
labels are costly and time-consuming to curate. However, the most effective
supervised or SSL strategy for transferring models to different healthcare
systems or novel tasks is not well understood. In this work, we systematically
experiment with a variety of supervised and self-supervised pretraining
strategies using multimodal datasets of medical images (chest X-rays) and text
(radiology reports). We then evaluate their performance on data from two
external institutions with diverse sets of tasks. In addition, we experiment
with different transfer learning strategies to effectively adapt these
pretrained models to new tasks and healthcare systems. Our empirical results
suggest that multimodal SSL gives substantial gains over unimodal SSL in
performance across new healthcare systems and tasks, comparable to models
pretrained with full supervision. We demonstrate additional performance gains
with models further adapted to the new dataset and task, using multimodal
domain-adaptive pretraining (DAPT), linear probing then finetuning (LP-FT), and
both methods combined. We offer suggestions for alternative models to use in
scenarios where not all of these additions are feasible. Our results provide
guidance for improving the generalization of medical image interpretation
models to new healthcare systems and novel tasks.
| [
{
"version": "v1",
"created": "Sat, 13 May 2023 22:33:09 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Van Uden",
"Cara",
""
],
[
"Irvin",
"Jeremy",
""
],
[
"Huang",
"Mars",
""
],
[
"Dean",
"Nathan",
""
],
[
"Carr",
"Jason",
""
],
[
"Ng",
"Andrew",
""
],
[
"Langlotz",
"Curtis",
""
]
] | TITLE: How to Train Your CheXDragon: Training Chest X-Ray Models for Transfer
to Novel Tasks and Healthcare Systems
ABSTRACT: Self-supervised learning (SSL) enables label efficient training for machine
learning models. This is essential for domains such as medical imaging, where
labels are costly and time-consuming to curate. However, the most effective
supervised or SSL strategy for transferring models to different healthcare
systems or novel tasks is not well understood. In this work, we systematically
experiment with a variety of supervised and self-supervised pretraining
strategies using multimodal datasets of medical images (chest X-rays) and text
(radiology reports). We then evaluate their performance on data from two
external institutions with diverse sets of tasks. In addition, we experiment
with different transfer learning strategies to effectively adapt these
pretrained models to new tasks and healthcare systems. Our empirical results
suggest that multimodal SSL gives substantial gains over unimodal SSL in
performance across new healthcare systems and tasks, comparable to models
pretrained with full supervision. We demonstrate additional performance gains
with models further adapted to the new dataset and task, using multimodal
domain-adaptive pretraining (DAPT), linear probing then finetuning (LP-FT), and
both methods combined. We offer suggestions for alternative models to use in
scenarios where not all of these additions are feasible. Our results provide
guidance for improving the generalization of medical image interpretation
models to new healthcare systems and novel tasks.
| no_new_dataset | 0.709818 |
2305.08025 | Amani Abusafia | Jiajie Li, Amani Abusafia, Abdallah Lakhdari, and Athman Bouguettaya | Activity-based Profiling for Energy Harvesting Estimation | 2 pages, 2 Figures,This is an accepted posterpaper at the ACM/IEEE
International Conference on Information Processing in Sensor Networks (IPSN)
2023 | null | 10.1145/3583120.3589833 | null | cs.DC | http://creativecommons.org/licenses/by/4.0/ | We propose a novel activity-based profiling framework to estimate IoT users'
harvested energy based on their daily activities. Energy is harvested from
natural sources such as the kinetic movement of IoT users. The profiling
framework captures the users' physical activity data to define activity-based
profiles. These profiles are utilized to estimate the harvested energy by IoT
users. We train and evaluate our framework based on a real Fitbit dataset.
| [
{
"version": "v1",
"created": "Sat, 13 May 2023 23:49:11 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Li",
"Jiajie",
""
],
[
"Abusafia",
"Amani",
""
],
[
"Lakhdari",
"Abdallah",
""
],
[
"Bouguettaya",
"Athman",
""
]
] | TITLE: Activity-based Profiling for Energy Harvesting Estimation
ABSTRACT: We propose a novel activity-based profiling framework to estimate IoT users'
harvested energy based on their daily activities. Energy is harvested from
natural sources such as the kinetic movement of IoT users. The profiling
framework captures the users' physical activity data to define activity-based
profiles. These profiles are utilized to estimate the harvested energy by IoT
users. We train and evaluate our framework based on a real Fitbit dataset.
| no_new_dataset | 0.71145 |
2305.08031 | Muhammad Huzaifa | Raza Imam, Muhammad Huzaifa, and Mohammed El-Amine Azz | On enhancing the robustness of Vision Transformers: Defensive Diffusion | Our code is publicly available at
https://github.com/Muhammad-Huzaifaa/Defensive_Diffusion | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by/4.0/ | Privacy and confidentiality of medical data are of utmost importance in
healthcare settings. ViTs, the SOTA vision model, rely on large amounts of
patient data for training, which raises concerns about data security and the
potential for unauthorized access. Adversaries may exploit vulnerabilities in
ViTs to extract sensitive patient information and compromising patient privacy.
This work address these vulnerabilities to ensure the trustworthiness and
reliability of ViTs in medical applications. In this work, we introduced a
defensive diffusion technique as an adversarial purifier to eliminate
adversarial noise introduced by attackers in the original image. By utilizing
the denoising capabilities of the diffusion model, we employ a reverse
diffusion process to effectively eliminate the adversarial noise from the
attack sample, resulting in a cleaner image that is then fed into the ViT
blocks. Our findings demonstrate the effectiveness of the diffusion model in
eliminating attack-agnostic adversarial noise from images. Additionally, we
propose combining knowledge distillation with our framework to obtain a
lightweight student model that is both computationally efficient and robust
against gray box attacks. Comparison of our method with a SOTA baseline method,
SEViT, shows that our work is able to outperform the baseline. Extensive
experiments conducted on a publicly available Tuberculosis X-ray dataset
validate the computational efficiency and improved robustness achieved by our
proposed architecture.
| [
{
"version": "v1",
"created": "Sun, 14 May 2023 00:17:33 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Imam",
"Raza",
""
],
[
"Huzaifa",
"Muhammad",
""
],
[
"Azz",
"Mohammed El-Amine",
""
]
] | TITLE: On enhancing the robustness of Vision Transformers: Defensive Diffusion
ABSTRACT: Privacy and confidentiality of medical data are of utmost importance in
healthcare settings. ViTs, the SOTA vision model, rely on large amounts of
patient data for training, which raises concerns about data security and the
potential for unauthorized access. Adversaries may exploit vulnerabilities in
ViTs to extract sensitive patient information and compromising patient privacy.
This work address these vulnerabilities to ensure the trustworthiness and
reliability of ViTs in medical applications. In this work, we introduced a
defensive diffusion technique as an adversarial purifier to eliminate
adversarial noise introduced by attackers in the original image. By utilizing
the denoising capabilities of the diffusion model, we employ a reverse
diffusion process to effectively eliminate the adversarial noise from the
attack sample, resulting in a cleaner image that is then fed into the ViT
blocks. Our findings demonstrate the effectiveness of the diffusion model in
eliminating attack-agnostic adversarial noise from images. Additionally, we
propose combining knowledge distillation with our framework to obtain a
lightweight student model that is both computationally efficient and robust
against gray box attacks. Comparison of our method with a SOTA baseline method,
SEViT, shows that our work is able to outperform the baseline. Extensive
experiments conducted on a publicly available Tuberculosis X-ray dataset
validate the computational efficiency and improved robustness achieved by our
proposed architecture.
| no_new_dataset | 0.71022 |
2305.08048 | Huayi Tang | Huayi Tang and Yong Liu | Towards Understanding the Generalization of Graph Neural Networks | null | null | null | null | cs.LG cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Graph neural networks (GNNs) are the most widely adopted model in
graph-structured data oriented learning and representation. Despite their
extraordinary success in real-world applications, understanding their working
mechanism by theory is still on primary stage. In this paper, we move towards
this goal from the perspective of generalization. To be specific, we first
establish high probability bounds of generalization gap and gradients in
transductive learning with consideration of stochastic optimization. After
that, we provide high probability bounds of generalization gap for popular
GNNs. The theoretical results reveal the architecture specific factors
affecting the generalization gap. Experimental results on benchmark datasets
show the consistency between theoretical results and empirical evidence. Our
results provide new insights in understanding the generalization of GNNs.
| [
{
"version": "v1",
"created": "Sun, 14 May 2023 03:05:14 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Tang",
"Huayi",
""
],
[
"Liu",
"Yong",
""
]
] | TITLE: Towards Understanding the Generalization of Graph Neural Networks
ABSTRACT: Graph neural networks (GNNs) are the most widely adopted model in
graph-structured data oriented learning and representation. Despite their
extraordinary success in real-world applications, understanding their working
mechanism by theory is still on primary stage. In this paper, we move towards
this goal from the perspective of generalization. To be specific, we first
establish high probability bounds of generalization gap and gradients in
transductive learning with consideration of stochastic optimization. After
that, we provide high probability bounds of generalization gap for popular
GNNs. The theoretical results reveal the architecture specific factors
affecting the generalization gap. Experimental results on benchmark datasets
show the consistency between theoretical results and empirical evidence. Our
results provide new insights in understanding the generalization of GNNs.
| no_new_dataset | 0.710666 |
2305.08053 | Shenghui Zhong | Miao Zhang, Yiqing Shen and Shenghui Zhong | SCRNet: a Retinex Structure-based Low-light Enhancement Model Guided by
Spatial Consistency | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Images captured under low-light conditions are often plagued by several
challenges, including diminished contrast, increased noise, loss of fine
details, and unnatural color reproduction. These factors can significantly
hinder the performance of computer vision tasks such as object detection and
image segmentation. As a result, improving the quality of low-light images is
of paramount importance for practical applications in the computer vision
domain.To effectively address these challenges, we present a novel low-light
image enhancement model, termed Spatial Consistency Retinex Network (SCRNet),
which leverages the Retinex-based structure and is guided by the principle of
spatial consistency.Specifically, our proposed model incorporates three levels
of consistency: channel level, semantic level, and texture level, inspired by
the principle of spatial consistency.These levels of consistency enable our
model to adaptively enhance image features, ensuring more accurate and visually
pleasing results.Extensive experimental evaluations on various low-light image
datasets demonstrate that our proposed SCRNet outshines existing
state-of-the-art methods, highlighting the potential of SCRNet as an effective
solution for enhancing low-light images.
| [
{
"version": "v1",
"created": "Sun, 14 May 2023 03:32:19 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Zhang",
"Miao",
""
],
[
"Shen",
"Yiqing",
""
],
[
"Zhong",
"Shenghui",
""
]
] | TITLE: SCRNet: a Retinex Structure-based Low-light Enhancement Model Guided by
Spatial Consistency
ABSTRACT: Images captured under low-light conditions are often plagued by several
challenges, including diminished contrast, increased noise, loss of fine
details, and unnatural color reproduction. These factors can significantly
hinder the performance of computer vision tasks such as object detection and
image segmentation. As a result, improving the quality of low-light images is
of paramount importance for practical applications in the computer vision
domain.To effectively address these challenges, we present a novel low-light
image enhancement model, termed Spatial Consistency Retinex Network (SCRNet),
which leverages the Retinex-based structure and is guided by the principle of
spatial consistency.Specifically, our proposed model incorporates three levels
of consistency: channel level, semantic level, and texture level, inspired by
the principle of spatial consistency.These levels of consistency enable our
model to adaptively enhance image features, ensuring more accurate and visually
pleasing results.Extensive experimental evaluations on various low-light image
datasets demonstrate that our proposed SCRNet outshines existing
state-of-the-art methods, highlighting the potential of SCRNet as an effective
solution for enhancing low-light images.
| no_new_dataset | 0.71227 |
2305.08059 | Chenyang Lyu | Chenyang Lyu, Tianbo Ji, Yvette Graham, Jennifer Foster | Semantic-aware Dynamic Retrospective-Prospective Reasoning for
Event-level Video Question Answering | null | null | null | null | cs.CV cs.AI cs.CL | http://creativecommons.org/licenses/by/4.0/ | Event-Level Video Question Answering (EVQA) requires complex reasoning across
video events to obtain the visual information needed to provide optimal
answers. However, despite significant progress in model performance, few
studies have focused on using the explicit semantic connections between the
question and visual information especially at the event level. There is need
for using such semantic connections to facilitate complex reasoning across
video frames. Therefore, we propose a semantic-aware dynamic
retrospective-prospective reasoning approach for video-based question
answering. Specifically, we explicitly use the Semantic Role Labeling (SRL)
structure of the question in the dynamic reasoning process where we decide to
move to the next frame based on which part of the SRL structure (agent, verb,
patient, etc.) of the question is being focused on. We conduct experiments on a
benchmark EVQA dataset - TrafficQA. Results show that our proposed approach
achieves superior performance compared to previous state-of-the-art models. Our
code will be made publicly available for research use.
| [
{
"version": "v1",
"created": "Sun, 14 May 2023 03:57:11 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Lyu",
"Chenyang",
""
],
[
"Ji",
"Tianbo",
""
],
[
"Graham",
"Yvette",
""
],
[
"Foster",
"Jennifer",
""
]
] | TITLE: Semantic-aware Dynamic Retrospective-Prospective Reasoning for
Event-level Video Question Answering
ABSTRACT: Event-Level Video Question Answering (EVQA) requires complex reasoning across
video events to obtain the visual information needed to provide optimal
answers. However, despite significant progress in model performance, few
studies have focused on using the explicit semantic connections between the
question and visual information especially at the event level. There is need
for using such semantic connections to facilitate complex reasoning across
video frames. Therefore, we propose a semantic-aware dynamic
retrospective-prospective reasoning approach for video-based question
answering. Specifically, we explicitly use the Semantic Role Labeling (SRL)
structure of the question in the dynamic reasoning process where we decide to
move to the next frame based on which part of the SRL structure (agent, verb,
patient, etc.) of the question is being focused on. We conduct experiments on a
benchmark EVQA dataset - TrafficQA. Results show that our proposed approach
achieves superior performance compared to previous state-of-the-art models. Our
code will be made publicly available for research use.
| no_new_dataset | 0.708377 |
2305.08067 | Shangeth Rajaa | Shangeth Rajaa | Improving End-to-End SLU performance with Prosodic Attention and
Distillation | Submitted to InterSpeech 2023 | null | null | null | cs.CL cs.LG cs.SD eess.AS | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Most End-to-End SLU methods depend on the pretrained ASR or language model
features for intent prediction. However, other essential information in speech,
such as prosody, is often ignored. Recent research has shown improved results
in classifying dialogue acts by incorporating prosodic information. The margins
of improvement in these methods are minimal as the neural models ignore
prosodic features. In this work, we propose prosody-attention, which uses the
prosodic features differently to generate attention maps across time frames of
the utterance. Then we propose prosody-distillation to explicitly learn the
prosodic information in the acoustic encoder rather than concatenating the
implicit prosodic features. Both the proposed methods improve the baseline
results, and the prosody-distillation method gives an intent classification
accuracy improvement of 8\% and 2\% on SLURP and STOP datasets over the prosody
baseline.
| [
{
"version": "v1",
"created": "Sun, 14 May 2023 04:38:20 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Rajaa",
"Shangeth",
""
]
] | TITLE: Improving End-to-End SLU performance with Prosodic Attention and
Distillation
ABSTRACT: Most End-to-End SLU methods depend on the pretrained ASR or language model
features for intent prediction. However, other essential information in speech,
such as prosody, is often ignored. Recent research has shown improved results
in classifying dialogue acts by incorporating prosodic information. The margins
of improvement in these methods are minimal as the neural models ignore
prosodic features. In this work, we propose prosody-attention, which uses the
prosodic features differently to generate attention maps across time frames of
the utterance. Then we propose prosody-distillation to explicitly learn the
prosodic information in the acoustic encoder rather than concatenating the
implicit prosodic features. Both the proposed methods improve the baseline
results, and the prosody-distillation method gives an intent classification
accuracy improvement of 8\% and 2\% on SLURP and STOP datasets over the prosody
baseline.
| no_new_dataset | 0.709636 |
2305.08075 | Maniratnam Mandal | Maniratnam Mandal and Imran Khan | Analyzing Compression Techniques for Computer Vision | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by-sa/4.0/ | Compressing deep networks is highly desirable for practical use-cases in
computer vision applications. Several techniques have been explored in the
literature, and research has been done in finding efficient strategies for
combining them. For this project, we aimed to explore three different basic
compression techniques - knowledge distillation, pruning, and quantization for
small-scale recognition tasks. Along with the basic methods, we also test the
efficacy of combining them in a sequential manner. We analyze them using MNIST
and CIFAR-10 datasets and present the results along with few observations
inferred from them.
| [
{
"version": "v1",
"created": "Sun, 14 May 2023 05:17:32 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Mandal",
"Maniratnam",
""
],
[
"Khan",
"Imran",
""
]
] | TITLE: Analyzing Compression Techniques for Computer Vision
ABSTRACT: Compressing deep networks is highly desirable for practical use-cases in
computer vision applications. Several techniques have been explored in the
literature, and research has been done in finding efficient strategies for
combining them. For this project, we aimed to explore three different basic
compression techniques - knowledge distillation, pruning, and quantization for
small-scale recognition tasks. Along with the basic methods, we also test the
efficacy of combining them in a sequential manner. We analyze them using MNIST
and CIFAR-10 datasets and present the results along with few observations
inferred from them.
| no_new_dataset | 0.710635 |
2305.08115 | Samuel Ackerman | Samuel Ackerman, Axel Bendavid, Eitan Farchi, Orna Raz | Automatic Generation of Attention Rules For Containment of Machine
Learning Model Errors | null | null | null | null | cs.LG stat.AP | http://creativecommons.org/licenses/by/4.0/ | Machine learning (ML) solutions are prevalent in many applications. However,
many challenges exist in making these solutions business-grade. For instance,
maintaining the error rate of the underlying ML models at an acceptably low
level. Typically, the true relationship between feature inputs and the target
feature to be predicted is uncertain, and hence statistical in nature. The
approach we propose is to separate the observations that are the most likely to
be predicted incorrectly into 'attention sets'. These can directly aid model
diagnosis and improvement, and be used to decide on alternative courses of
action for these problematic observations. We present several algorithms
(`strategies') for determining optimal rules to separate these observations. In
particular, we prefer strategies that use feature-based slicing because they
are human-interpretable, model-agnostic, and require minimal supplementary
inputs or knowledge. In addition, we show that these strategies outperform
several common baselines, such as selecting observations with prediction
confidence below a threshold. To evaluate strategies, we introduce metrics to
measure various desired qualities, such as their performance, stability, and
generalizability to unseen data; the strategies are evaluated on several
publicly-available datasets. We use TOPSIS, a Multiple Criteria Decision Making
method, to aggregate these metrics into a single quality score for each
strategy, to allow comparison.
| [
{
"version": "v1",
"created": "Sun, 14 May 2023 10:15:35 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Ackerman",
"Samuel",
""
],
[
"Bendavid",
"Axel",
""
],
[
"Farchi",
"Eitan",
""
],
[
"Raz",
"Orna",
""
]
] | TITLE: Automatic Generation of Attention Rules For Containment of Machine
Learning Model Errors
ABSTRACT: Machine learning (ML) solutions are prevalent in many applications. However,
many challenges exist in making these solutions business-grade. For instance,
maintaining the error rate of the underlying ML models at an acceptably low
level. Typically, the true relationship between feature inputs and the target
feature to be predicted is uncertain, and hence statistical in nature. The
approach we propose is to separate the observations that are the most likely to
be predicted incorrectly into 'attention sets'. These can directly aid model
diagnosis and improvement, and be used to decide on alternative courses of
action for these problematic observations. We present several algorithms
(`strategies') for determining optimal rules to separate these observations. In
particular, we prefer strategies that use feature-based slicing because they
are human-interpretable, model-agnostic, and require minimal supplementary
inputs or knowledge. In addition, we show that these strategies outperform
several common baselines, such as selecting observations with prediction
confidence below a threshold. To evaluate strategies, we introduce metrics to
measure various desired qualities, such as their performance, stability, and
generalizability to unseen data; the strategies are evaluated on several
publicly-available datasets. We use TOPSIS, a Multiple Criteria Decision Making
method, to aggregate these metrics into a single quality score for each
strategy, to allow comparison.
| no_new_dataset | 0.706975 |
2305.08139 | Eitam Sheetrit | Eitam Sheetrit, Menachem Brief, Oren Elisha | Predicting Unplanned Readmissions in the Intensive Care Unit: A
Multimodality Evaluation | null | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A hospital readmission is when a patient who was discharged from the hospital
is admitted again for the same or related care within a certain period.
Hospital readmissions are a significant problem in the healthcare domain, as
they lead to increased hospitalization costs, decreased patient satisfaction,
and increased risk of adverse outcomes such as infections, medication errors,
and even death. The problem of hospital readmissions is particularly acute in
intensive care units (ICUs), due to the severity of the patients' conditions,
and the substantial risk of complications. Predicting Unplanned Readmissions in
ICUs is a challenging task, as it involves analyzing different data modalities,
such as static data, unstructured free text, sequences of diagnoses and
procedures, and multivariate time-series. Here, we investigate the
effectiveness of each data modality separately, then alongside with others,
using state-of-the-art machine learning approaches in time-series analysis and
natural language processing. Using our evaluation process, we are able to
determine the contribution of each data modality, and for the first time in the
context of readmission, establish a hierarchy of their predictive value.
Additionally, we demonstrate the impact of Temporal Abstractions in enhancing
the performance of time-series approaches to readmission prediction. Due to
conflicting definitions in the literature, we also provide a clear definition
of the term Unplanned Readmission to enhance reproducibility and consistency of
future research and to prevent any potential misunderstandings that could
result from diverse interpretations of the term. Our experimental results on a
large benchmark clinical data set show that Discharge Notes written by
physicians, have better capabilities for readmission prediction than all other
modalities.
| [
{
"version": "v1",
"created": "Sun, 14 May 2023 12:20:13 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Sheetrit",
"Eitam",
""
],
[
"Brief",
"Menachem",
""
],
[
"Elisha",
"Oren",
""
]
] | TITLE: Predicting Unplanned Readmissions in the Intensive Care Unit: A
Multimodality Evaluation
ABSTRACT: A hospital readmission is when a patient who was discharged from the hospital
is admitted again for the same or related care within a certain period.
Hospital readmissions are a significant problem in the healthcare domain, as
they lead to increased hospitalization costs, decreased patient satisfaction,
and increased risk of adverse outcomes such as infections, medication errors,
and even death. The problem of hospital readmissions is particularly acute in
intensive care units (ICUs), due to the severity of the patients' conditions,
and the substantial risk of complications. Predicting Unplanned Readmissions in
ICUs is a challenging task, as it involves analyzing different data modalities,
such as static data, unstructured free text, sequences of diagnoses and
procedures, and multivariate time-series. Here, we investigate the
effectiveness of each data modality separately, then alongside with others,
using state-of-the-art machine learning approaches in time-series analysis and
natural language processing. Using our evaluation process, we are able to
determine the contribution of each data modality, and for the first time in the
context of readmission, establish a hierarchy of their predictive value.
Additionally, we demonstrate the impact of Temporal Abstractions in enhancing
the performance of time-series approaches to readmission prediction. Due to
conflicting definitions in the literature, we also provide a clear definition
of the term Unplanned Readmission to enhance reproducibility and consistency of
future research and to prevent any potential misunderstandings that could
result from diverse interpretations of the term. Our experimental results on a
large benchmark clinical data set show that Discharge Notes written by
physicians, have better capabilities for readmission prediction than all other
modalities.
| no_new_dataset | 0.706823 |
2305.08146 | Jipeng Qiang | Jipeng Qiang, Kang Liu, Yun Li, Yunhao Yuan, Yi Zhu | ParaLS: Lexical Substitution via Pretrained Paraphraser | null | ACL 2023 | null | null | cs.CL cs.AI | http://creativecommons.org/publicdomain/zero/1.0/ | Lexical substitution (LS) aims at finding appropriate substitutes for a
target word in a sentence. Recently, LS methods based on pretrained language
models have made remarkable progress, generating potential substitutes for a
target word through analysis of its contextual surroundings. However, these
methods tend to overlook the preservation of the sentence's meaning when
generating the substitutes. This study explores how to generate the substitute
candidates from a paraphraser, as the generated paraphrases from a paraphraser
contain variations in word choice and preserve the sentence's meaning. Since we
cannot directly generate the substitutes via commonly used decoding strategies,
we propose two simple decoding strategies that focus on the variations of the
target word during decoding. Experimental results show that our methods
outperform state-of-the-art LS methods based on pre-trained language models on
three benchmarks.
| [
{
"version": "v1",
"created": "Sun, 14 May 2023 12:49:16 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Qiang",
"Jipeng",
""
],
[
"Liu",
"Kang",
""
],
[
"Li",
"Yun",
""
],
[
"Yuan",
"Yunhao",
""
],
[
"Zhu",
"Yi",
""
]
] | TITLE: ParaLS: Lexical Substitution via Pretrained Paraphraser
ABSTRACT: Lexical substitution (LS) aims at finding appropriate substitutes for a
target word in a sentence. Recently, LS methods based on pretrained language
models have made remarkable progress, generating potential substitutes for a
target word through analysis of its contextual surroundings. However, these
methods tend to overlook the preservation of the sentence's meaning when
generating the substitutes. This study explores how to generate the substitute
candidates from a paraphraser, as the generated paraphrases from a paraphraser
contain variations in word choice and preserve the sentence's meaning. Since we
cannot directly generate the substitutes via commonly used decoding strategies,
we propose two simple decoding strategies that focus on the variations of the
target word during decoding. Experimental results show that our methods
outperform state-of-the-art LS methods based on pre-trained language models on
three benchmarks.
| no_new_dataset | 0.707379 |
2305.08152 | Yulun Du | Yulun Du and Lydia Chilton | STORYWARS: A Dataset and Instruction Tuning Baselines for Collaborative
Story Understanding and Generation | ACL 2023 | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Collaborative stories, which are texts created through the collaborative
efforts of multiple authors with different writing styles and intentions, pose
unique challenges for NLP models. Understanding and generating such stories
remains an underexplored area due to the lack of open-domain corpora. To
address this, we introduce STORYWARS, a new dataset of over 40,000
collaborative stories written by 9,400 different authors from an online
platform. We design 12 task types, comprising 7 understanding and 5 generation
task types, on STORYWARS, deriving 101 diverse story-related tasks in total as
a multi-task benchmark covering all fully-supervised, few-shot, and zero-shot
scenarios. Furthermore, we present our instruction-tuned model, INSTRUCTSTORY,
for the story tasks showing that instruction tuning, in addition to achieving
superior results in zero-shot and few-shot scenarios, can also obtain the best
performance on the fully-supervised tasks in STORYWARS, establishing strong
multi-task benchmark performances on STORYWARS.
| [
{
"version": "v1",
"created": "Sun, 14 May 2023 13:09:27 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Du",
"Yulun",
""
],
[
"Chilton",
"Lydia",
""
]
] | TITLE: STORYWARS: A Dataset and Instruction Tuning Baselines for Collaborative
Story Understanding and Generation
ABSTRACT: Collaborative stories, which are texts created through the collaborative
efforts of multiple authors with different writing styles and intentions, pose
unique challenges for NLP models. Understanding and generating such stories
remains an underexplored area due to the lack of open-domain corpora. To
address this, we introduce STORYWARS, a new dataset of over 40,000
collaborative stories written by 9,400 different authors from an online
platform. We design 12 task types, comprising 7 understanding and 5 generation
task types, on STORYWARS, deriving 101 diverse story-related tasks in total as
a multi-task benchmark covering all fully-supervised, few-shot, and zero-shot
scenarios. Furthermore, we present our instruction-tuned model, INSTRUCTSTORY,
for the story tasks showing that instruction tuning, in addition to achieving
superior results in zero-shot and few-shot scenarios, can also obtain the best
performance on the fully-supervised tasks in STORYWARS, establishing strong
multi-task benchmark performances on STORYWARS.
| new_dataset | 0.715405 |
2305.08164 | Zenan Huang | Zenan Huang, Haobo Wang, Junbo Zhao, Nenggan Zheng | Latent Processes Identification From Multi-View Time Series | 15 pages, 9 figures, accepted by IJCAI-23 | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Understanding the dynamics of time series data typically requires identifying
the unique latent factors for data generation, \textit{a.k.a.}, latent
processes identification. Driven by the independent assumption, existing works
have made great progress in handling single-view data. However, it is a
non-trivial problem that extends them to multi-view time series data because of
two main challenges: (i) the complex data structure, such as temporal
dependency, can result in violation of the independent assumption; (ii) the
factors from different views are generally overlapped and are hard to be
aggregated to a complete set. In this work, we propose a novel framework MuLTI
that employs the contrastive learning technique to invert the data generative
process for enhanced identifiability. Additionally, MuLTI integrates a
permutation mechanism that merges corresponding overlapped variables by the
establishment of an optimal transport formula. Extensive experimental results
on synthetic and real-world datasets demonstrate the superiority of our method
in recovering identifiable latent variables on multi-view time series.
| [
{
"version": "v1",
"created": "Sun, 14 May 2023 14:21:58 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Huang",
"Zenan",
""
],
[
"Wang",
"Haobo",
""
],
[
"Zhao",
"Junbo",
""
],
[
"Zheng",
"Nenggan",
""
]
] | TITLE: Latent Processes Identification From Multi-View Time Series
ABSTRACT: Understanding the dynamics of time series data typically requires identifying
the unique latent factors for data generation, \textit{a.k.a.}, latent
processes identification. Driven by the independent assumption, existing works
have made great progress in handling single-view data. However, it is a
non-trivial problem that extends them to multi-view time series data because of
two main challenges: (i) the complex data structure, such as temporal
dependency, can result in violation of the independent assumption; (ii) the
factors from different views are generally overlapped and are hard to be
aggregated to a complete set. In this work, we propose a novel framework MuLTI
that employs the contrastive learning technique to invert the data generative
process for enhanced identifiability. Additionally, MuLTI integrates a
permutation mechanism that merges corresponding overlapped variables by the
establishment of an optimal transport formula. Extensive experimental results
on synthetic and real-world datasets demonstrate the superiority of our method
in recovering identifiable latent variables on multi-view time series.
| no_new_dataset | 0.709636 |
2305.08173 | Gaurish Thakkar Mr | Gaurish Thakkar, Nives Mikelic Preradovic and Marko Tadi\'c | Croatian Film Review Dataset (Cro-FiReDa): A Sentiment Annotated Dataset
of Film Reviews | null | LTC 2023 | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | This paper introduces Cro-FiReDa, a sentiment-annotated dataset for Croatian
in the domain of movie reviews. The dataset, which contains over 10,000
sentences, has been annotated at the sentence level. In addition to presenting
the overall annotation process, we also present benchmark results based on the
transformer-based fine-tuning approach
| [
{
"version": "v1",
"created": "Sun, 14 May 2023 14:46:12 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Thakkar",
"Gaurish",
""
],
[
"Preradovic",
"Nives Mikelic",
""
],
[
"Tadić",
"Marko",
""
]
] | TITLE: Croatian Film Review Dataset (Cro-FiReDa): A Sentiment Annotated Dataset
of Film Reviews
ABSTRACT: This paper introduces Cro-FiReDa, a sentiment-annotated dataset for Croatian
in the domain of movie reviews. The dataset, which contains over 10,000
sentences, has been annotated at the sentence level. In addition to presenting
the overall annotation process, we also present benchmark results based on the
transformer-based fine-tuning approach
| new_dataset | 0.712389 |
2305.08186 | Tian Feng | Lehao Yang, Long Li, Qihao Chen, Jiling Zhang, Tian Feng, Wei Zhang | Street Layout Design via Conditional Adversarial Learning | null | null | null | null | cs.GR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Designing high-quality urban street layouts has long been in high demand, but
entangles notable challenges. Conventional methods based on deep generative
models are yet to fill the gap on integrating both natural and socioeconomic
factors in the design loop. In this paper, we propose a novel urban street
layout design method based on conditional adversarial learning. Specifically, a
conditional generative adversarial network trained on a real-world dataset
synthesizes street layout images from the feature map, into which an
autoencoder fuses a set of natural and socioeconomic data for a region of
interest; The following extraction module generates high-quality street layout
graphs corresponding to the synthesized images. Experiments and evaluations
suggest that the proposed method outputs various urban street layouts that are
visually and structurally alike their real-world counterparts, which can be
used to support the creation of high-quality urban virtual environments.
| [
{
"version": "v1",
"created": "Sun, 14 May 2023 15:39:38 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Yang",
"Lehao",
""
],
[
"Li",
"Long",
""
],
[
"Chen",
"Qihao",
""
],
[
"Zhang",
"Jiling",
""
],
[
"Feng",
"Tian",
""
],
[
"Zhang",
"Wei",
""
]
] | TITLE: Street Layout Design via Conditional Adversarial Learning
ABSTRACT: Designing high-quality urban street layouts has long been in high demand, but
entangles notable challenges. Conventional methods based on deep generative
models are yet to fill the gap on integrating both natural and socioeconomic
factors in the design loop. In this paper, we propose a novel urban street
layout design method based on conditional adversarial learning. Specifically, a
conditional generative adversarial network trained on a real-world dataset
synthesizes street layout images from the feature map, into which an
autoencoder fuses a set of natural and socioeconomic data for a region of
interest; The following extraction module generates high-quality street layout
graphs corresponding to the synthesized images. Experiments and evaluations
suggest that the proposed method outputs various urban street layouts that are
visually and structurally alike their real-world counterparts, which can be
used to support the creation of high-quality urban virtual environments.
| no_new_dataset | 0.712607 |
2305.08187 | Gaurish Thakkar Mr | Gaurish Thakkar, Nives Mikelic Preradovi\'c, Marko Tadi\'c | CroSentiNews 2.0: A Sentence-Level News Sentiment Corpus | null | Slavic NLP 2023 | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | This article presents a sentence-level sentiment dataset for the Croatian
news domain. In addition to the 3K annotated texts already present, our dataset
contains 14.5K annotated sentence occurrences that have been tagged with 5
classes. We provide baseline scores in addition to the annotation process and
inter-annotator agreement.
| [
{
"version": "v1",
"created": "Sun, 14 May 2023 15:53:54 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Thakkar",
"Gaurish",
""
],
[
"Preradović",
"Nives Mikelic",
""
],
[
"Tadić",
"Marko",
""
]
] | TITLE: CroSentiNews 2.0: A Sentence-Level News Sentiment Corpus
ABSTRACT: This article presents a sentence-level sentiment dataset for the Croatian
news domain. In addition to the 3K annotated texts already present, our dataset
contains 14.5K annotated sentence occurrences that have been tagged with 5
classes. We provide baseline scores in addition to the annotation process and
inter-annotator agreement.
| new_dataset | 0.711481 |
2305.08191 | Guillaume Berger | Antoine Mercier and Guillaume Berger and Sunny Panchal and Florian
Letsch and Cornelius Boehm and Nahua Kang and Ingo Bax and Roland Memisevic | Is end-to-end learning enough for fitness activity recognition? | 9 pages, 4 figures, 4 tables | null | null | null | cs.CV cs.LG | http://creativecommons.org/licenses/by/4.0/ | End-to-end learning has taken hold of many computer vision tasks, in
particular, related to still images, with task-specific optimization yielding
very strong performance. Nevertheless, human-centric action recognition is
still largely dominated by hand-crafted pipelines, and only individual
components are replaced by neural networks that typically operate on individual
frames. As a testbed to study the relevance of such pipelines, we present a new
fully annotated video dataset of fitness activities. Any recognition
capabilities in this domain are almost exclusively a function of human poses
and their temporal dynamics, so pose-based solutions should perform well. We
show that, with this labelled data, end-to-end learning on raw pixels can
compete with state-of-the-art action recognition pipelines based on pose
estimation. We also show that end-to-end learning can support temporally
fine-grained tasks such as real-time repetition counting.
| [
{
"version": "v1",
"created": "Sun, 14 May 2023 16:00:03 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Mercier",
"Antoine",
""
],
[
"Berger",
"Guillaume",
""
],
[
"Panchal",
"Sunny",
""
],
[
"Letsch",
"Florian",
""
],
[
"Boehm",
"Cornelius",
""
],
[
"Kang",
"Nahua",
""
],
[
"Bax",
"Ingo",
""
],
[
"Memisevic",
"Roland",
""
]
] | TITLE: Is end-to-end learning enough for fitness activity recognition?
ABSTRACT: End-to-end learning has taken hold of many computer vision tasks, in
particular, related to still images, with task-specific optimization yielding
very strong performance. Nevertheless, human-centric action recognition is
still largely dominated by hand-crafted pipelines, and only individual
components are replaced by neural networks that typically operate on individual
frames. As a testbed to study the relevance of such pipelines, we present a new
fully annotated video dataset of fitness activities. Any recognition
capabilities in this domain are almost exclusively a function of human poses
and their temporal dynamics, so pose-based solutions should perform well. We
show that, with this labelled data, end-to-end learning on raw pixels can
compete with state-of-the-art action recognition pipelines based on pose
estimation. We also show that end-to-end learning can support temporally
fine-grained tasks such as real-time repetition counting.
| new_dataset | 0.716857 |
2305.08200 | Jiyue Jiang | Jiyue Jiang, Sheng Wang, Qintong Li, Lingpeng Kong, Chuan Wu | A Cognitive Stimulation Dialogue System with Multi-source Knowledge
Fusion for Elders with Cognitive Impairment | Accepted by ACL 2023 | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | When communicating with elders with cognitive impairment, cognitive
stimulation (CS) help to maintain the cognitive health of elders. Data sparsity
is the main challenge in building CS-based dialogue systems, particularly in
the Chinese language. To fill this gap, we construct a Chinese CS conversation
(CSConv) dataset, which contains about 2.6K groups of dialogues with CS
principles and emotional support strategy labels. Making chit chat while
providing emotional support is overlooked by the majority of existing cognitive
dialogue systems. In this paper, we propose a multi-source knowledge fusion
method for CS dialogue (CSD), to generate open-ended responses guided by the CS
principle and emotional support strategy. We first use a progressive mask
method based on external knowledge to learn encoders as effective classifiers,
which is the prerequisite to predict the CS principle and emotional support
strategy of the target response. Then a decoder interacts with the perceived CS
principle and emotional support strategy to generate responses. Extensive
experiments conducted on the CSConv dataset demonstrate the effectiveness of
the proposed method, while there is still a large space for improvement
compared to human performance.
| [
{
"version": "v1",
"created": "Sun, 14 May 2023 16:52:20 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Jiang",
"Jiyue",
""
],
[
"Wang",
"Sheng",
""
],
[
"Li",
"Qintong",
""
],
[
"Kong",
"Lingpeng",
""
],
[
"Wu",
"Chuan",
""
]
] | TITLE: A Cognitive Stimulation Dialogue System with Multi-source Knowledge
Fusion for Elders with Cognitive Impairment
ABSTRACT: When communicating with elders with cognitive impairment, cognitive
stimulation (CS) help to maintain the cognitive health of elders. Data sparsity
is the main challenge in building CS-based dialogue systems, particularly in
the Chinese language. To fill this gap, we construct a Chinese CS conversation
(CSConv) dataset, which contains about 2.6K groups of dialogues with CS
principles and emotional support strategy labels. Making chit chat while
providing emotional support is overlooked by the majority of existing cognitive
dialogue systems. In this paper, we propose a multi-source knowledge fusion
method for CS dialogue (CSD), to generate open-ended responses guided by the CS
principle and emotional support strategy. We first use a progressive mask
method based on external knowledge to learn encoders as effective classifiers,
which is the prerequisite to predict the CS principle and emotional support
strategy of the target response. Then a decoder interacts with the perceived CS
principle and emotional support strategy to generate responses. Extensive
experiments conducted on the CSConv dataset demonstrate the effectiveness of
the proposed method, while there is still a large space for improvement
compared to human performance.
| new_dataset | 0.714777 |
2305.08215 | Arif Mahmood | Hira Yaseen and Arif Mahmood | Learning Structure Aware Deep Spectral Embedding | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Spectral Embedding (SE) has often been used to map data points from
non-linear manifolds to linear subspaces for the purpose of classification and
clustering. Despite significant advantages, the subspace structure of data in
the original space is not preserved in the embedding space. To address this
issue subspace clustering has been proposed by replacing the SE graph affinity
with a self-expression matrix. It works well if the data lies in a union of
linear subspaces however, the performance may degrade in real-world
applications where data often spans non-linear manifolds. To address this
problem we propose a novel structure-aware deep spectral embedding by combining
a spectral embedding loss and a structure preservation loss. To this end, a
deep neural network architecture is proposed that simultaneously encodes both
types of information and aims to generate structure-aware spectral embedding.
The subspace structure of the input data is encoded by using attention-based
self-expression learning. The proposed algorithm is evaluated on six publicly
available real-world datasets. The results demonstrate the excellent clustering
performance of the proposed algorithm compared to the existing state-of-the-art
methods. The proposed algorithm has also exhibited better generalization to
unseen data points and it is scalable to larger datasets without requiring
significant computational resources.
| [
{
"version": "v1",
"created": "Sun, 14 May 2023 18:18:05 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Yaseen",
"Hira",
""
],
[
"Mahmood",
"Arif",
""
]
] | TITLE: Learning Structure Aware Deep Spectral Embedding
ABSTRACT: Spectral Embedding (SE) has often been used to map data points from
non-linear manifolds to linear subspaces for the purpose of classification and
clustering. Despite significant advantages, the subspace structure of data in
the original space is not preserved in the embedding space. To address this
issue subspace clustering has been proposed by replacing the SE graph affinity
with a self-expression matrix. It works well if the data lies in a union of
linear subspaces however, the performance may degrade in real-world
applications where data often spans non-linear manifolds. To address this
problem we propose a novel structure-aware deep spectral embedding by combining
a spectral embedding loss and a structure preservation loss. To this end, a
deep neural network architecture is proposed that simultaneously encodes both
types of information and aims to generate structure-aware spectral embedding.
The subspace structure of the input data is encoded by using attention-based
self-expression learning. The proposed algorithm is evaluated on six publicly
available real-world datasets. The results demonstrate the excellent clustering
performance of the proposed algorithm compared to the existing state-of-the-art
methods. The proposed algorithm has also exhibited better generalization to
unseen data points and it is scalable to larger datasets without requiring
significant computational resources.
| no_new_dataset | 0.710848 |
2305.08227 | Hendrik Schr\"oter | Hendrik Schr\"oter, Tobias Rosenkranz, Alberto N. Escalante-B.,
Andreas Maier | DeepFilterNet: Perceptually Motivated Real-Time Speech Enhancement | Accepted as show and tell demo to interspeech 2023 | null | null | null | eess.AS cs.CL cs.SD | http://creativecommons.org/licenses/by-sa/4.0/ | 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.
| [
{
"version": "v1",
"created": "Sun, 14 May 2023 19:09:35 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Schröter",
"Hendrik",
""
],
[
"Rosenkranz",
"Tobias",
""
],
[
"Escalante-B.",
"Alberto N.",
""
],
[
"Maier",
"Andreas",
""
]
] | TITLE: DeepFilterNet: Perceptually Motivated Real-Time Speech Enhancement
ABSTRACT: 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.
| no_new_dataset | 0.708055 |
2305.08264 | Santiago Miret | Yu Song, Santiago Miret, Bang Liu | MatSci-NLP: Evaluating Scientific Language Models on Materials Science
Language Tasks Using Text-to-Schema Modeling | null | null | null | null | cs.CL cond-mat.mtrl-sci cs.AI | http://creativecommons.org/licenses/by/4.0/ | We present MatSci-NLP, a natural language benchmark for evaluating the
performance of natural language processing (NLP) models on materials science
text. We construct the benchmark from publicly available materials science text
data to encompass seven different NLP tasks, including conventional NLP tasks
like named entity recognition and relation classification, as well as NLP tasks
specific to materials science, such as synthesis action retrieval which relates
to creating synthesis procedures for materials. We study various BERT-based
models pretrained on different scientific text corpora on MatSci-NLP to
understand the impact of pretraining strategies on understanding materials
science text. Given the scarcity of high-quality annotated data in the
materials science domain, we perform our fine-tuning experiments with limited
training data to encourage the generalize across MatSci-NLP tasks. Our
experiments in this low-resource training setting show that language models
pretrained on scientific text outperform BERT trained on general text. MatBERT,
a model pretrained specifically on materials science journals, generally
performs best for most tasks. Moreover, we propose a unified text-to-schema for
multitask learning on \benchmark and compare its performance with traditional
fine-tuning methods. In our analysis of different training methods, we find
that our proposed text-to-schema methods inspired by question-answering
consistently outperform single and multitask NLP fine-tuning methods. The code
and datasets are publicly available at
\url{https://github.com/BangLab-UdeM-Mila/NLP4MatSci-ACL23}.
| [
{
"version": "v1",
"created": "Sun, 14 May 2023 22:01:24 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Song",
"Yu",
""
],
[
"Miret",
"Santiago",
""
],
[
"Liu",
"Bang",
""
]
] | TITLE: MatSci-NLP: Evaluating Scientific Language Models on Materials Science
Language Tasks Using Text-to-Schema Modeling
ABSTRACT: We present MatSci-NLP, a natural language benchmark for evaluating the
performance of natural language processing (NLP) models on materials science
text. We construct the benchmark from publicly available materials science text
data to encompass seven different NLP tasks, including conventional NLP tasks
like named entity recognition and relation classification, as well as NLP tasks
specific to materials science, such as synthesis action retrieval which relates
to creating synthesis procedures for materials. We study various BERT-based
models pretrained on different scientific text corpora on MatSci-NLP to
understand the impact of pretraining strategies on understanding materials
science text. Given the scarcity of high-quality annotated data in the
materials science domain, we perform our fine-tuning experiments with limited
training data to encourage the generalize across MatSci-NLP tasks. Our
experiments in this low-resource training setting show that language models
pretrained on scientific text outperform BERT trained on general text. MatBERT,
a model pretrained specifically on materials science journals, generally
performs best for most tasks. Moreover, we propose a unified text-to-schema for
multitask learning on \benchmark and compare its performance with traditional
fine-tuning methods. In our analysis of different training methods, we find
that our proposed text-to-schema methods inspired by question-answering
consistently outperform single and multitask NLP fine-tuning methods. The code
and datasets are publicly available at
\url{https://github.com/BangLab-UdeM-Mila/NLP4MatSci-ACL23}.
| new_dataset | 0.602471 |
2305.08273 | Yanping Zheng | Yanping Zheng, Zhewei Wei, Jiajun Liu | Decoupled Graph Neural Networks for Large Dynamic Graphs | null | null | null | null | cs.LG cs.SI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Real-world graphs, such as social networks, financial transactions, and
recommendation systems, often demonstrate dynamic behavior. This phenomenon,
known as graph stream, involves the dynamic changes of nodes and the emergence
and disappearance of edges. To effectively capture both the structural and
temporal aspects of these dynamic graphs, dynamic graph neural networks have
been developed. However, existing methods are usually tailored to process
either continuous-time or discrete-time dynamic graphs, and cannot be
generalized from one to the other. In this paper, we propose a decoupled graph
neural network for large dynamic graphs, including a unified dynamic
propagation that supports efficient computation for both continuous and
discrete dynamic graphs. Since graph structure-related computations are only
performed during the propagation process, the prediction process for the
downstream task can be trained separately without expensive graph computations,
and therefore any sequence model can be plugged-in and used. As a result, our
algorithm achieves exceptional scalability and expressiveness. We evaluate our
algorithm on seven real-world datasets of both continuous-time and
discrete-time dynamic graphs. The experimental results demonstrate that our
algorithm achieves state-of-the-art performance in both kinds of dynamic
graphs. Most notably, the scalability of our algorithm is well illustrated by
its successful application to large graphs with up to over a billion temporal
edges and over a hundred million nodes.
| [
{
"version": "v1",
"created": "Sun, 14 May 2023 23:00:10 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Zheng",
"Yanping",
""
],
[
"Wei",
"Zhewei",
""
],
[
"Liu",
"Jiajun",
""
]
] | TITLE: Decoupled Graph Neural Networks for Large Dynamic Graphs
ABSTRACT: Real-world graphs, such as social networks, financial transactions, and
recommendation systems, often demonstrate dynamic behavior. This phenomenon,
known as graph stream, involves the dynamic changes of nodes and the emergence
and disappearance of edges. To effectively capture both the structural and
temporal aspects of these dynamic graphs, dynamic graph neural networks have
been developed. However, existing methods are usually tailored to process
either continuous-time or discrete-time dynamic graphs, and cannot be
generalized from one to the other. In this paper, we propose a decoupled graph
neural network for large dynamic graphs, including a unified dynamic
propagation that supports efficient computation for both continuous and
discrete dynamic graphs. Since graph structure-related computations are only
performed during the propagation process, the prediction process for the
downstream task can be trained separately without expensive graph computations,
and therefore any sequence model can be plugged-in and used. As a result, our
algorithm achieves exceptional scalability and expressiveness. We evaluate our
algorithm on seven real-world datasets of both continuous-time and
discrete-time dynamic graphs. The experimental results demonstrate that our
algorithm achieves state-of-the-art performance in both kinds of dynamic
graphs. Most notably, the scalability of our algorithm is well illustrated by
its successful application to large graphs with up to over a billion temporal
edges and over a hundred million nodes.
| no_new_dataset | 0.71027 |
2305.08296 | Dafei Qin | Dafei Qin, Jun Saito, Noam Aigerman, Thibault Groueix, Taku Komura | Neural Face Rigging for Animating and Retargeting Facial Meshes in the
Wild | SIGGRAPH 2023(Conference Track), 13 pages, 15 figures | null | null | null | cs.GR cs.AI | http://creativecommons.org/licenses/by/4.0/ | We propose an end-to-end deep-learning approach for automatic rigging and
retargeting of 3D models of human faces in the wild. Our approach, called
Neural Face Rigging (NFR), holds three key properties:
(i) NFR's expression space maintains human-interpretable editing parameters
for artistic controls;
(ii) NFR is readily applicable to arbitrary facial meshes with different
connectivity and expressions;
(iii) NFR can encode and produce fine-grained details of complex expressions
performed by arbitrary subjects.
To the best of our knowledge, NFR is the first approach to provide realistic
and controllable deformations of in-the-wild facial meshes, without the manual
creation of blendshapes or correspondence. We design a deformation autoencoder
and train it through a multi-dataset training scheme, which benefits from the
unique advantages of two data sources: a linear 3DMM with interpretable control
parameters as in FACS, and 4D captures of real faces with fine-grained details.
Through various experiments, we show NFR's ability to automatically produce
realistic and accurate facial deformations across a wide range of existing
datasets as well as noisy facial scans in-the-wild, while providing
artist-controlled, editable parameters.
| [
{
"version": "v1",
"created": "Mon, 15 May 2023 01:51:08 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Qin",
"Dafei",
""
],
[
"Saito",
"Jun",
""
],
[
"Aigerman",
"Noam",
""
],
[
"Groueix",
"Thibault",
""
],
[
"Komura",
"Taku",
""
]
] | TITLE: Neural Face Rigging for Animating and Retargeting Facial Meshes in the
Wild
ABSTRACT: We propose an end-to-end deep-learning approach for automatic rigging and
retargeting of 3D models of human faces in the wild. Our approach, called
Neural Face Rigging (NFR), holds three key properties:
(i) NFR's expression space maintains human-interpretable editing parameters
for artistic controls;
(ii) NFR is readily applicable to arbitrary facial meshes with different
connectivity and expressions;
(iii) NFR can encode and produce fine-grained details of complex expressions
performed by arbitrary subjects.
To the best of our knowledge, NFR is the first approach to provide realistic
and controllable deformations of in-the-wild facial meshes, without the manual
creation of blendshapes or correspondence. We design a deformation autoencoder
and train it through a multi-dataset training scheme, which benefits from the
unique advantages of two data sources: a linear 3DMM with interpretable control
parameters as in FACS, and 4D captures of real faces with fine-grained details.
Through various experiments, we show NFR's ability to automatically produce
realistic and accurate facial deformations across a wide range of existing
datasets as well as noisy facial scans in-the-wild, while providing
artist-controlled, editable parameters.
| no_new_dataset | 0.707443 |
2305.08300 | Zexue He | Zexue He, An Yan, Amilcare Gentili, Julian McAuley, Chun-Nan Hsu | "Nothing Abnormal": Disambiguating Medical Reports via Contrastive
Knowledge Infusion | Accepted to AAAI 2023. 13 pages including 4-page supplementary
materials | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Sharing medical reports is essential for patient-centered care. A recent line
of work has focused on automatically generating reports with NLP methods.
However, different audiences have different purposes when writing/reading
medical reports -- for example, healthcare professionals care more about
pathology, whereas patients are more concerned with the diagnosis ("Is there
any abnormality?"). The expectation gap results in a common situation where
patients find their medical reports to be ambiguous and therefore unsure about
the next steps. In this work, we explore the audience expectation gap in
healthcare and summarize common ambiguities that lead patients to be confused
about their diagnosis into three categories: medical jargon, contradictory
findings, and misleading grammatical errors. Based on our analysis, we define a
disambiguation rewriting task to regenerate an input to be unambiguous while
preserving information about the original content. We further propose a
rewriting algorithm based on contrastive pretraining and perturbation-based
rewriting. In addition, we create two datasets, OpenI-Annotated based on chest
reports and VA-Annotated based on general medical reports, with available
binary labels for ambiguity and abnormality presence annotated by radiology
specialists. Experimental results on these datasets show that our proposed
algorithm effectively rewrites input sentences in a less ambiguous way with
high content fidelity. Our code and annotated data are released to facilitate
future research.
| [
{
"version": "v1",
"created": "Mon, 15 May 2023 02:01:20 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"He",
"Zexue",
""
],
[
"Yan",
"An",
""
],
[
"Gentili",
"Amilcare",
""
],
[
"McAuley",
"Julian",
""
],
[
"Hsu",
"Chun-Nan",
""
]
] | TITLE: "Nothing Abnormal": Disambiguating Medical Reports via Contrastive
Knowledge Infusion
ABSTRACT: Sharing medical reports is essential for patient-centered care. A recent line
of work has focused on automatically generating reports with NLP methods.
However, different audiences have different purposes when writing/reading
medical reports -- for example, healthcare professionals care more about
pathology, whereas patients are more concerned with the diagnosis ("Is there
any abnormality?"). The expectation gap results in a common situation where
patients find their medical reports to be ambiguous and therefore unsure about
the next steps. In this work, we explore the audience expectation gap in
healthcare and summarize common ambiguities that lead patients to be confused
about their diagnosis into three categories: medical jargon, contradictory
findings, and misleading grammatical errors. Based on our analysis, we define a
disambiguation rewriting task to regenerate an input to be unambiguous while
preserving information about the original content. We further propose a
rewriting algorithm based on contrastive pretraining and perturbation-based
rewriting. In addition, we create two datasets, OpenI-Annotated based on chest
reports and VA-Annotated based on general medical reports, with available
binary labels for ambiguity and abnormality presence annotated by radiology
specialists. Experimental results on these datasets show that our proposed
algorithm effectively rewrites input sentences in a less ambiguous way with
high content fidelity. Our code and annotated data are released to facilitate
future research.
| new_dataset | 0.714634 |
2305.08302 | Aboli Marathe | Aboli Marathe, Sanjana Prabhu | t-RAIN: Robust generalization under weather-aliasing label shift attacks | Accepted at Affective Behavior Analysis in-the-wild (ABAW) at CVPR
2023 | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by/4.0/ | In the classical supervised learning settings, classifiers are fit with the
assumption of balanced label distributions and produce remarkable results on
the same. In the real world, however, these assumptions often bend and in turn
adversely impact model performance. Identifying bad learners in skewed target
distributions is even more challenging. Thus achieving model robustness under
these "label shift" settings is an important task in autonomous perception. In
this paper, we analyze the impact of label shift on the task of multi-weather
classification for autonomous vehicles. We use this information as a prior to
better assess pedestrian detection in adverse weather. We model the
classification performance as an indicator of robustness under 4 label shift
scenarios and study the behavior of multiple classes of models. We propose
t-RAIN a similarity mapping technique for synthetic data augmentation using
large scale generative models and evaluate the performance on DAWN dataset.
This mapping boosts model test accuracy by 2.1, 4.4, 1.9, 2.7 % in no-shift,
fog, snow, dust shifts respectively. We present state-of-the-art pedestrian
detection results on real and synthetic weather domains with best performing
82.69 AP (snow) and 62.31 AP (fog) respectively.
| [
{
"version": "v1",
"created": "Mon, 15 May 2023 02:05:56 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Marathe",
"Aboli",
""
],
[
"Prabhu",
"Sanjana",
""
]
] | TITLE: t-RAIN: Robust generalization under weather-aliasing label shift attacks
ABSTRACT: In the classical supervised learning settings, classifiers are fit with the
assumption of balanced label distributions and produce remarkable results on
the same. In the real world, however, these assumptions often bend and in turn
adversely impact model performance. Identifying bad learners in skewed target
distributions is even more challenging. Thus achieving model robustness under
these "label shift" settings is an important task in autonomous perception. In
this paper, we analyze the impact of label shift on the task of multi-weather
classification for autonomous vehicles. We use this information as a prior to
better assess pedestrian detection in adverse weather. We model the
classification performance as an indicator of robustness under 4 label shift
scenarios and study the behavior of multiple classes of models. We propose
t-RAIN a similarity mapping technique for synthetic data augmentation using
large scale generative models and evaluate the performance on DAWN dataset.
This mapping boosts model test accuracy by 2.1, 4.4, 1.9, 2.7 % in no-shift,
fog, snow, dust shifts respectively. We present state-of-the-art pedestrian
detection results on real and synthetic weather domains with best performing
82.69 AP (snow) and 62.31 AP (fog) respectively.
| no_new_dataset | 0.711857 |
2305.08316 | Ziyuan Zhao | Ziyuan Zhao, Peisheng Qian, Xulei Yang, Zeng Zeng, Cuntai Guan, Wai
Leong Tam, Xiaoli Li | SemiGNN-PPI: Self-Ensembling Multi-Graph Neural Network for Efficient
and Generalizable Protein-Protein Interaction Prediction | Accepted by IJCAI 2023 | null | null | null | q-bio.MN cs.AI cs.CE cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Protein-protein interactions (PPIs) are crucial in various biological
processes and their study has significant implications for drug development and
disease diagnosis. Existing deep learning methods suffer from significant
performance degradation under complex real-world scenarios due to various
factors, e.g., label scarcity and domain shift. In this paper, we propose a
self-ensembling multigraph neural network (SemiGNN-PPI) that can effectively
predict PPIs while being both efficient and generalizable. In SemiGNN-PPI, we
not only model the protein correlations but explore the label dependencies by
constructing and processing multiple graphs from the perspectives of both
features and labels in the graph learning process. We further marry GNN with
Mean Teacher to effectively leverage unlabeled graph-structured PPI data for
self-ensemble graph learning. We also design multiple graph consistency
constraints to align the student and teacher graphs in the feature embedding
space, enabling the student model to better learn from the teacher model by
incorporating more relationships. Extensive experiments on PPI datasets of
different scales with different evaluation settings demonstrate that
SemiGNN-PPI outperforms state-of-the-art PPI prediction methods, particularly
in challenging scenarios such as training with limited annotations and testing
on unseen data.
| [
{
"version": "v1",
"created": "Mon, 15 May 2023 03:06:44 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Zhao",
"Ziyuan",
""
],
[
"Qian",
"Peisheng",
""
],
[
"Yang",
"Xulei",
""
],
[
"Zeng",
"Zeng",
""
],
[
"Guan",
"Cuntai",
""
],
[
"Tam",
"Wai Leong",
""
],
[
"Li",
"Xiaoli",
""
]
] | TITLE: SemiGNN-PPI: Self-Ensembling Multi-Graph Neural Network for Efficient
and Generalizable Protein-Protein Interaction Prediction
ABSTRACT: Protein-protein interactions (PPIs) are crucial in various biological
processes and their study has significant implications for drug development and
disease diagnosis. Existing deep learning methods suffer from significant
performance degradation under complex real-world scenarios due to various
factors, e.g., label scarcity and domain shift. In this paper, we propose a
self-ensembling multigraph neural network (SemiGNN-PPI) that can effectively
predict PPIs while being both efficient and generalizable. In SemiGNN-PPI, we
not only model the protein correlations but explore the label dependencies by
constructing and processing multiple graphs from the perspectives of both
features and labels in the graph learning process. We further marry GNN with
Mean Teacher to effectively leverage unlabeled graph-structured PPI data for
self-ensemble graph learning. We also design multiple graph consistency
constraints to align the student and teacher graphs in the feature embedding
space, enabling the student model to better learn from the teacher model by
incorporating more relationships. Extensive experiments on PPI datasets of
different scales with different evaluation settings demonstrate that
SemiGNN-PPI outperforms state-of-the-art PPI prediction methods, particularly
in challenging scenarios such as training with limited annotations and testing
on unseen data.
| no_new_dataset | 0.709988 |
2305.08318 | Shuhang Tan | Shuhang Tan, Hengyu Liu, Zhiling Wang | CMSG Cross-Media Semantic-Graph Feature Matching Algorithm for
Autonomous Vehicle Relocalization | null | null | null | null | cs.CV cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Relocalization is the basis of map-based localization algorithms. Camera and
LiDAR map-based methods are pervasive since their robustness under different
scenarios. Generally, mapping and localization using the same sensor have
better accuracy since matching features between the same type of data is
easier. However, due to the camera's lack of 3D information and the high cost
of LiDAR, cross-media methods are developing, which combined live image data
and Lidar map. Although matching features between different media is
challenging, we believe cross-media is the tendency for AV relocalization since
its low cost and accuracy can be comparable to the same-sensor-based methods.
In this paper, we propose CMSG, a novel cross-media algorithm for AV
relocalization tasks. Semantic features are utilized for better interpretation
the correlation between point clouds and image features. What's more,
abstracted semantic graph nodes are introduced, and a graph network
architecture is integrated to better extract the similarity of semantic
features. Validation experiments are conducted on the KITTI odometry dataset.
Our results show that CMSG can have comparable or even better accuracy compared
to current single-sensor-based methods at a speed of 25 FPS on NVIDIA 1080 Ti
GPU.
| [
{
"version": "v1",
"created": "Mon, 15 May 2023 03:08:10 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Tan",
"Shuhang",
""
],
[
"Liu",
"Hengyu",
""
],
[
"Wang",
"Zhiling",
""
]
] | TITLE: CMSG Cross-Media Semantic-Graph Feature Matching Algorithm for
Autonomous Vehicle Relocalization
ABSTRACT: Relocalization is the basis of map-based localization algorithms. Camera and
LiDAR map-based methods are pervasive since their robustness under different
scenarios. Generally, mapping and localization using the same sensor have
better accuracy since matching features between the same type of data is
easier. However, due to the camera's lack of 3D information and the high cost
of LiDAR, cross-media methods are developing, which combined live image data
and Lidar map. Although matching features between different media is
challenging, we believe cross-media is the tendency for AV relocalization since
its low cost and accuracy can be comparable to the same-sensor-based methods.
In this paper, we propose CMSG, a novel cross-media algorithm for AV
relocalization tasks. Semantic features are utilized for better interpretation
the correlation between point clouds and image features. What's more,
abstracted semantic graph nodes are introduced, and a graph network
architecture is integrated to better extract the similarity of semantic
features. Validation experiments are conducted on the KITTI odometry dataset.
Our results show that CMSG can have comparable or even better accuracy compared
to current single-sensor-based methods at a speed of 25 FPS on NVIDIA 1080 Ti
GPU.
| no_new_dataset | 0.711387 |
2305.08328 | Penghui Wei | Penghui Wei, Hongjian Dou, Shaoguo Liu, Rongjun Tang, Li Liu, Liang
Wang, Bo Zheng | FedAds: A Benchmark for Privacy-Preserving CVR Estimation with Vertical
Federated Learning | SIGIR 2023, Resource Track | null | null | null | cs.IR cs.LG | http://creativecommons.org/licenses/by/4.0/ | Conversion rate (CVR) estimation aims to predict the probability of
conversion event after a user has clicked an ad. Typically, online publisher
has user browsing interests and click feedbacks, while demand-side advertising
platform collects users' post-click behaviors such as dwell time and conversion
decisions. To estimate CVR accurately and protect data privacy better, vertical
federated learning (vFL) is a natural solution to combine two sides' advantages
for training models, without exchanging raw data. Both CVR estimation and
applied vFL algorithms have attracted increasing research attentions. However,
standardized and systematical evaluations are missing: due to the lack of
standardized datasets, existing studies adopt public datasets to simulate a vFL
setting via hand-crafted feature partition, which brings challenges to fair
comparison. We introduce FedAds, the first benchmark for CVR estimation with
vFL, to facilitate standardized and systematical evaluations for vFL
algorithms. It contains a large-scale real world dataset collected from
Alibaba's advertising platform, as well as systematical evaluations for both
effectiveness and privacy aspects of various vFL algorithms. Besides, we also
explore to incorporate unaligned data in vFL to improve effectiveness, and
develop perturbation operations to protect privacy well. We hope that future
research work in vFL and CVR estimation benefits from the FedAds benchmark.
| [
{
"version": "v1",
"created": "Mon, 15 May 2023 03:34:42 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Wei",
"Penghui",
""
],
[
"Dou",
"Hongjian",
""
],
[
"Liu",
"Shaoguo",
""
],
[
"Tang",
"Rongjun",
""
],
[
"Liu",
"Li",
""
],
[
"Wang",
"Liang",
""
],
[
"Zheng",
"Bo",
""
]
] | TITLE: FedAds: A Benchmark for Privacy-Preserving CVR Estimation with Vertical
Federated Learning
ABSTRACT: Conversion rate (CVR) estimation aims to predict the probability of
conversion event after a user has clicked an ad. Typically, online publisher
has user browsing interests and click feedbacks, while demand-side advertising
platform collects users' post-click behaviors such as dwell time and conversion
decisions. To estimate CVR accurately and protect data privacy better, vertical
federated learning (vFL) is a natural solution to combine two sides' advantages
for training models, without exchanging raw data. Both CVR estimation and
applied vFL algorithms have attracted increasing research attentions. However,
standardized and systematical evaluations are missing: due to the lack of
standardized datasets, existing studies adopt public datasets to simulate a vFL
setting via hand-crafted feature partition, which brings challenges to fair
comparison. We introduce FedAds, the first benchmark for CVR estimation with
vFL, to facilitate standardized and systematical evaluations for vFL
algorithms. It contains a large-scale real world dataset collected from
Alibaba's advertising platform, as well as systematical evaluations for both
effectiveness and privacy aspects of various vFL algorithms. Besides, we also
explore to incorporate unaligned data in vFL to improve effectiveness, and
develop perturbation operations to protect privacy well. We hope that future
research work in vFL and CVR estimation benefits from the FedAds benchmark.
| no_new_dataset | 0.699075 |
2305.08336 | Chenhao Li | Chenhao Li, Trung Thanh Ngo, Hajime Nagahara | Inverse Rendering of Translucent Objects using Physical and Neural
Renderers | Accepted to CVPR2023 | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | In this work, we propose an inverse rendering model that estimates 3D shape,
spatially-varying reflectance, homogeneous subsurface scattering parameters,
and an environment illumination jointly from only a pair of captured images of
a translucent object. In order to solve the ambiguity problem of inverse
rendering, we use a physically-based renderer and a neural renderer for scene
reconstruction and material editing. Because two renderers are differentiable,
we can compute a reconstruction loss to assist parameter estimation. To enhance
the supervision of the proposed neural renderer, we also propose an augmented
loss. In addition, we use a flash and no-flash image pair as the input. To
supervise the training, we constructed a large-scale synthetic dataset of
translucent objects, which consists of 117K scenes. Qualitative and
quantitative results on both synthetic and real-world datasets demonstrated the
effectiveness of the proposed model.
| [
{
"version": "v1",
"created": "Mon, 15 May 2023 04:03:11 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Li",
"Chenhao",
""
],
[
"Ngo",
"Trung Thanh",
""
],
[
"Nagahara",
"Hajime",
""
]
] | TITLE: Inverse Rendering of Translucent Objects using Physical and Neural
Renderers
ABSTRACT: In this work, we propose an inverse rendering model that estimates 3D shape,
spatially-varying reflectance, homogeneous subsurface scattering parameters,
and an environment illumination jointly from only a pair of captured images of
a translucent object. In order to solve the ambiguity problem of inverse
rendering, we use a physically-based renderer and a neural renderer for scene
reconstruction and material editing. Because two renderers are differentiable,
we can compute a reconstruction loss to assist parameter estimation. To enhance
the supervision of the proposed neural renderer, we also propose an augmented
loss. In addition, we use a flash and no-flash image pair as the input. To
supervise the training, we constructed a large-scale synthetic dataset of
translucent objects, which consists of 117K scenes. Qualitative and
quantitative results on both synthetic and real-world datasets demonstrated the
effectiveness of the proposed model.
| new_dataset | 0.713032 |
2305.08344 | Wei-I Lin | Wei-I Lin, Gang Niu, Hsuan-Tien Lin, Masashi Sugiyama | Enhancing Label Sharing Efficiency in Complementary-Label Learning with
Label Augmentation | null | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | Complementary-label Learning (CLL) is a form of weakly supervised learning
that trains an ordinary classifier using only complementary labels, which are
the classes that certain instances do not belong to. While existing CLL studies
typically use novel loss functions or training techniques to solve this
problem, few studies focus on how complementary labels collectively provide
information to train the ordinary classifier. In this paper, we fill the gap by
analyzing the implicit sharing of complementary labels on nearby instances
during training. Our analysis reveals that the efficiency of implicit label
sharing is closely related to the performance of existing CLL models. Based on
this analysis, we propose a novel technique that enhances the sharing
efficiency via complementary-label augmentation, which explicitly propagates
additional complementary labels to each instance. We carefully design the
augmentation process to enrich the data with new and accurate complementary
labels, which provide CLL models with fresh and valuable information to enhance
the sharing efficiency. We then verify our proposed technique by conducting
thorough experiments on both synthetic and real-world datasets. Our results
confirm that complementary-label augmentation can systematically improve
empirical performance over state-of-the-art CLL models.
| [
{
"version": "v1",
"created": "Mon, 15 May 2023 04:43:14 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Lin",
"Wei-I",
""
],
[
"Niu",
"Gang",
""
],
[
"Lin",
"Hsuan-Tien",
""
],
[
"Sugiyama",
"Masashi",
""
]
] | TITLE: Enhancing Label Sharing Efficiency in Complementary-Label Learning with
Label Augmentation
ABSTRACT: Complementary-label Learning (CLL) is a form of weakly supervised learning
that trains an ordinary classifier using only complementary labels, which are
the classes that certain instances do not belong to. While existing CLL studies
typically use novel loss functions or training techniques to solve this
problem, few studies focus on how complementary labels collectively provide
information to train the ordinary classifier. In this paper, we fill the gap by
analyzing the implicit sharing of complementary labels on nearby instances
during training. Our analysis reveals that the efficiency of implicit label
sharing is closely related to the performance of existing CLL models. Based on
this analysis, we propose a novel technique that enhances the sharing
efficiency via complementary-label augmentation, which explicitly propagates
additional complementary labels to each instance. We carefully design the
augmentation process to enrich the data with new and accurate complementary
labels, which provide CLL models with fresh and valuable information to enhance
the sharing efficiency. We then verify our proposed technique by conducting
thorough experiments on both synthetic and real-world datasets. Our results
confirm that complementary-label augmentation can systematically improve
empirical performance over state-of-the-art CLL models.
| no_new_dataset | 0.709573 |
2305.08347 | Zhifeng Li | Zhifeng Li and Bowei Zou and Yifan Fan and Yu Hong | KEPR: Knowledge Enhancement and Plausibility Ranking for Generative
Commonsense Question Answering | IJCNN 2023 | null | null | null | cs.CL cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Generative commonsense question answering (GenCQA) is a task of automatically
generating a list of answers given a question. The answer list is required to
cover all reasonable answers. This presents the considerable challenges of
producing diverse answers and ranking them properly. Incorporating a variety of
closely-related background knowledge into the encoding of questions enables the
generation of different answers. Meanwhile, learning to distinguish positive
answers from negative ones potentially enhances the probabilistic estimation of
plausibility, and accordingly, the plausibility-based ranking. Therefore, we
propose a Knowledge Enhancement and Plausibility Ranking (KEPR) approach
grounded on the Generate-Then-Rank pipeline architecture. Specifically, we
expand questions in terms of Wiktionary commonsense knowledge of keywords, and
reformulate them with normalized patterns. Dense passage retrieval is utilized
for capturing relevant knowledge, and different PLM-based (BART, GPT2 and T5)
networks are used for generating answers. On the other hand, we develop an
ELECTRA-based answer ranking model, where logistic regression is conducted
during training, with the aim of approximating different levels of plausibility
in a polar classification scenario. Extensive experiments on the benchmark
ProtoQA show that KEPR obtains substantial improvements, compared to the strong
baselines. Within the experimental models, the T5-based GenCQA with KEPR
obtains the best performance, which is up to 60.91% at the primary canonical
metric Inc@3. It outperforms the existing GenCQA models on the current
leaderboard of ProtoQA.
| [
{
"version": "v1",
"created": "Mon, 15 May 2023 04:58:37 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Li",
"Zhifeng",
""
],
[
"Zou",
"Bowei",
""
],
[
"Fan",
"Yifan",
""
],
[
"Hong",
"Yu",
""
]
] | TITLE: KEPR: Knowledge Enhancement and Plausibility Ranking for Generative
Commonsense Question Answering
ABSTRACT: Generative commonsense question answering (GenCQA) is a task of automatically
generating a list of answers given a question. The answer list is required to
cover all reasonable answers. This presents the considerable challenges of
producing diverse answers and ranking them properly. Incorporating a variety of
closely-related background knowledge into the encoding of questions enables the
generation of different answers. Meanwhile, learning to distinguish positive
answers from negative ones potentially enhances the probabilistic estimation of
plausibility, and accordingly, the plausibility-based ranking. Therefore, we
propose a Knowledge Enhancement and Plausibility Ranking (KEPR) approach
grounded on the Generate-Then-Rank pipeline architecture. Specifically, we
expand questions in terms of Wiktionary commonsense knowledge of keywords, and
reformulate them with normalized patterns. Dense passage retrieval is utilized
for capturing relevant knowledge, and different PLM-based (BART, GPT2 and T5)
networks are used for generating answers. On the other hand, we develop an
ELECTRA-based answer ranking model, where logistic regression is conducted
during training, with the aim of approximating different levels of plausibility
in a polar classification scenario. Extensive experiments on the benchmark
ProtoQA show that KEPR obtains substantial improvements, compared to the strong
baselines. Within the experimental models, the T5-based GenCQA with KEPR
obtains the best performance, which is up to 60.91% at the primary canonical
metric Inc@3. It outperforms the existing GenCQA models on the current
leaderboard of ProtoQA.
| no_new_dataset | 0.71086 |
2305.08360 | Chao Liu Dr. | Chao Liu, Xuanlin Bao, Hongyu Zhang, Neng Zhang, Haibo Hu, Xiaohong
Zhang, Meng Yan | Improving ChatGPT Prompt for Code Generation | 12 pages, 1 figure | null | null | null | cs.SE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Automated code generation can be a powerful technique for software
development, significantly reducing developers' efforts and time required to
create new code by generating it automatically based on requirements. Recently,
OpenAI's language model ChatGPT has emerged as a powerful tool for generating
human-like responses to a wide range of textual inputs (i.e., prompts),
including those related to code generation. However, the effectiveness of
ChatGPT for code generation is not well understood, and the generation
performance could be heavily influenced by the choice of prompt. To answer
these questions, we conducted experiments using the CodeXGlue dataset to
evaluate ChatGPT's capabilities for two code generation tasks, including
text-to-code and code-to-code generation. We designed prompts by leveraging the
chain-of-thought strategy with multi-step optimizations. Our results showed
that by carefully designing prompts to guide ChatGPT, the generation
performance can be improved substantially. We also analyzed the factors that
influenced the prompt design and provided insights that could guide future
research.
| [
{
"version": "v1",
"created": "Mon, 15 May 2023 05:37:33 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Liu",
"Chao",
""
],
[
"Bao",
"Xuanlin",
""
],
[
"Zhang",
"Hongyu",
""
],
[
"Zhang",
"Neng",
""
],
[
"Hu",
"Haibo",
""
],
[
"Zhang",
"Xiaohong",
""
],
[
"Yan",
"Meng",
""
]
] | TITLE: Improving ChatGPT Prompt for Code Generation
ABSTRACT: Automated code generation can be a powerful technique for software
development, significantly reducing developers' efforts and time required to
create new code by generating it automatically based on requirements. Recently,
OpenAI's language model ChatGPT has emerged as a powerful tool for generating
human-like responses to a wide range of textual inputs (i.e., prompts),
including those related to code generation. However, the effectiveness of
ChatGPT for code generation is not well understood, and the generation
performance could be heavily influenced by the choice of prompt. To answer
these questions, we conducted experiments using the CodeXGlue dataset to
evaluate ChatGPT's capabilities for two code generation tasks, including
text-to-code and code-to-code generation. We designed prompts by leveraging the
chain-of-thought strategy with multi-step optimizations. Our results showed
that by carefully designing prompts to guide ChatGPT, the generation
performance can be improved substantially. We also analyzed the factors that
influenced the prompt design and provided insights that could guide future
research.
| no_new_dataset | 0.707582 |
2305.08366 | Hiroto Honda | Hiroto Honda, Yusuke Uchida | CLRerNet: Improving Confidence of Lane Detection with LaneIoU | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Lane marker detection is a crucial component of the autonomous driving and
driver assistance systems. Modern deep lane detection methods with row-based
lane representation exhibit excellent performance on lane detection benchmarks.
Through preliminary oracle experiments, we firstly disentangle the lane
representation components to determine the direction of our approach. We show
that correct lane positions are already among the predictions of an existing
row-based detector, and the confidence scores that accurately represent
intersection-over-union (IoU) with ground truths are the most beneficial. Based
on the finding, we propose LaneIoU that better correlates with the metric, by
taking the local lane angles into consideration. We develop a novel detector
coined CLRerNet featuring LaneIoU for the target assignment cost and loss
functions aiming at the improved quality of confidence scores. Through careful
and fair benchmark including cross validation, we demonstrate that CLRerNet
outperforms the state-of-the-art by a large margin - enjoying F1 score of
81.43% compared with 80.47% of the existing method on CULane, and 86.47%
compared with 86.10% on CurveLanes.
| [
{
"version": "v1",
"created": "Mon, 15 May 2023 05:59:35 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Honda",
"Hiroto",
""
],
[
"Uchida",
"Yusuke",
""
]
] | TITLE: CLRerNet: Improving Confidence of Lane Detection with LaneIoU
ABSTRACT: Lane marker detection is a crucial component of the autonomous driving and
driver assistance systems. Modern deep lane detection methods with row-based
lane representation exhibit excellent performance on lane detection benchmarks.
Through preliminary oracle experiments, we firstly disentangle the lane
representation components to determine the direction of our approach. We show
that correct lane positions are already among the predictions of an existing
row-based detector, and the confidence scores that accurately represent
intersection-over-union (IoU) with ground truths are the most beneficial. Based
on the finding, we propose LaneIoU that better correlates with the metric, by
taking the local lane angles into consideration. We develop a novel detector
coined CLRerNet featuring LaneIoU for the target assignment cost and loss
functions aiming at the improved quality of confidence scores. Through careful
and fair benchmark including cross validation, we demonstrate that CLRerNet
outperforms the state-of-the-art by a large margin - enjoying F1 score of
81.43% compared with 80.47% of the existing method on CULane, and 86.47%
compared with 86.10% on CurveLanes.
| no_new_dataset | 0.70676 |
2305.08394 | Qiyue Yin | Meng Yao, Xueou Feng, Qiyue Yin | More Like Real World Game Challenge for Partially Observable Multi-Agent
Cooperation | null | null | null | null | cs.MA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Some standardized environments have been designed for partially observable
multi-agent cooperation, but we find most current environments are synchronous,
whereas real-world agents often have their own action spaces leading to
asynchrony. Furthermore, fixed agents number limits the scalability of action
space, whereas in reality agents number can change resulting in a flexible
action space. In addition, current environments are balanced, which is not
always the case in the real world where there may be an ability gap between
different parties leading to asymmetry. Finally, current environments tend to
have less stochasticity with simple state transitions, whereas real-world
environments can be highly stochastic and result in extremely risky. To address
this gap, we propose WarGame Challenge (WGC) inspired by the Wargame. WGC is a
lightweight, flexible, and easy-to-use environment with a clear framework that
can be easily configured by users. Along with the benchmark, we provide MARL
baseline algorithms such as QMIX and a toolkit to help algorithms complete
performance tests on WGC. Finally, we present baseline experiment results,
which demonstrate the challenges of WGC. We think WGC enrichs the partially
observable multi-agent cooperation domain and introduces more challenges that
better reflect the real-world characteristics. Code is release in
http://turingai.ia.ac.cn/data\_center/show/10.
| [
{
"version": "v1",
"created": "Mon, 15 May 2023 07:19:42 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Yao",
"Meng",
""
],
[
"Feng",
"Xueou",
""
],
[
"Yin",
"Qiyue",
""
]
] | TITLE: More Like Real World Game Challenge for Partially Observable Multi-Agent
Cooperation
ABSTRACT: Some standardized environments have been designed for partially observable
multi-agent cooperation, but we find most current environments are synchronous,
whereas real-world agents often have their own action spaces leading to
asynchrony. Furthermore, fixed agents number limits the scalability of action
space, whereas in reality agents number can change resulting in a flexible
action space. In addition, current environments are balanced, which is not
always the case in the real world where there may be an ability gap between
different parties leading to asymmetry. Finally, current environments tend to
have less stochasticity with simple state transitions, whereas real-world
environments can be highly stochastic and result in extremely risky. To address
this gap, we propose WarGame Challenge (WGC) inspired by the Wargame. WGC is a
lightweight, flexible, and easy-to-use environment with a clear framework that
can be easily configured by users. Along with the benchmark, we provide MARL
baseline algorithms such as QMIX and a toolkit to help algorithms complete
performance tests on WGC. Finally, we present baseline experiment results,
which demonstrate the challenges of WGC. We think WGC enrichs the partially
observable multi-agent cooperation domain and introduces more challenges that
better reflect the real-world characteristics. Code is release in
http://turingai.ia.ac.cn/data\_center/show/10.
| no_new_dataset | 0.708213 |
2305.08408 | Ding Jiun Huang | Ding-Jiun Huang, Yu-Ting Kao, Tieh-Hung Chuang, Ya-Chun Tsai, Jing-Kai
Lou, Shuen-Huei Guan | SB-VQA: A Stack-Based Video Quality Assessment Framework for Video
Enhancement | CVPR NTIRE 2023 | null | null | null | cs.CV eess.IV | http://creativecommons.org/licenses/by/4.0/ | In recent years, several video quality assessment (VQA) methods have been
developed, achieving high performance. However, these methods were not
specifically trained for enhanced videos, which limits their ability to predict
video quality accurately based on human subjective perception. To address this
issue, we propose a stack-based framework for VQA that outperforms existing
state-of-the-art methods on VDPVE, a dataset consisting of enhanced videos. In
addition to proposing the VQA framework for enhanced videos, we also
investigate its application on professionally generated content (PGC). To
address copyright issues with premium content, we create the PGCVQ dataset,
which consists of videos from YouTube. We evaluate our proposed approach and
state-of-the-art methods on PGCVQ, and provide new insights on the results. Our
experiments demonstrate that existing VQA algorithms can be applied to PGC
videos, and we find that VQA performance for PGC videos can be improved by
considering the plot of a play, which highlights the importance of video
semantic understanding.
| [
{
"version": "v1",
"created": "Mon, 15 May 2023 07:44:10 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Huang",
"Ding-Jiun",
""
],
[
"Kao",
"Yu-Ting",
""
],
[
"Chuang",
"Tieh-Hung",
""
],
[
"Tsai",
"Ya-Chun",
""
],
[
"Lou",
"Jing-Kai",
""
],
[
"Guan",
"Shuen-Huei",
""
]
] | TITLE: SB-VQA: A Stack-Based Video Quality Assessment Framework for Video
Enhancement
ABSTRACT: In recent years, several video quality assessment (VQA) methods have been
developed, achieving high performance. However, these methods were not
specifically trained for enhanced videos, which limits their ability to predict
video quality accurately based on human subjective perception. To address this
issue, we propose a stack-based framework for VQA that outperforms existing
state-of-the-art methods on VDPVE, a dataset consisting of enhanced videos. In
addition to proposing the VQA framework for enhanced videos, we also
investigate its application on professionally generated content (PGC). To
address copyright issues with premium content, we create the PGCVQ dataset,
which consists of videos from YouTube. We evaluate our proposed approach and
state-of-the-art methods on PGCVQ, and provide new insights on the results. Our
experiments demonstrate that existing VQA algorithms can be applied to PGC
videos, and we find that VQA performance for PGC videos can be improved by
considering the plot of a play, which highlights the importance of video
semantic understanding.
| new_dataset | 0.71638 |
2305.08409 | Florian Schintke | Florian Schintke, Ninon De Mecquenem, David Frantz, Vanessa Emanuela
Guarino, Marcus Hilbrich, Fabian Lehmann, Rebecca Sattler, Jan Arne Sparka,
Daniel Speckhard, Hermann Stolte, Anh Duc Vu, Ulf Leser | Validity Constraints for Data Analysis Workflows | null | null | null | null | cs.DC | http://creativecommons.org/licenses/by/4.0/ | Porting a scientific data analysis workflow (DAW) to a cluster
infrastructure, a new software stack, or even only a new dataset with some
notably different properties is often challenging. Despite the structured
definition of the steps (tasks) and their interdependencies during a complex
data analysis in the DAW specification, relevant assumptions may remain
unspecified and implicit. Such hidden assumptions often lead to crashing tasks
without a reasonable error message, poor performance in general,
non-terminating executions, or silent wrong results of the DAW, to name only a
few possible consequences. Searching for the causes of such errors and
drawbacks in a distributed compute cluster managed by a complex infrastructure
stack, where DAWs for large datasets typically are executed, can be tedious and
time-consuming.
We propose validity constraints (VCs) as a new concept for DAW languages to
alleviate this situation. A VC is a constraint specifying some logical
conditions that must be fulfilled at certain times for DAW executions to be
valid. When defined together with a DAW, VCs help to improve the portability,
adaptability, and reusability of DAWs by making implicit assumptions explicit.
Once specified, VC can be controlled automatically by the DAW infrastructure,
and violations can lead to meaningful error messages and graceful behaviour
(e.g., termination or invocation of repair mechanisms). We provide a broad list
of possible VCs, classify them along multiple dimensions, and compare them to
similar concepts one can find in related fields. We also provide a first sketch
for VCs' implementation into existing DAW infrastructures.
| [
{
"version": "v1",
"created": "Mon, 15 May 2023 07:44:22 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Schintke",
"Florian",
""
],
[
"De Mecquenem",
"Ninon",
""
],
[
"Frantz",
"David",
""
],
[
"Guarino",
"Vanessa Emanuela",
""
],
[
"Hilbrich",
"Marcus",
""
],
[
"Lehmann",
"Fabian",
""
],
[
"Sattler",
"Rebecca",
""
],
[
"Sparka",
"Jan Arne",
""
],
[
"Speckhard",
"Daniel",
""
],
[
"Stolte",
"Hermann",
""
],
[
"Vu",
"Anh Duc",
""
],
[
"Leser",
"Ulf",
""
]
] | TITLE: Validity Constraints for Data Analysis Workflows
ABSTRACT: Porting a scientific data analysis workflow (DAW) to a cluster
infrastructure, a new software stack, or even only a new dataset with some
notably different properties is often challenging. Despite the structured
definition of the steps (tasks) and their interdependencies during a complex
data analysis in the DAW specification, relevant assumptions may remain
unspecified and implicit. Such hidden assumptions often lead to crashing tasks
without a reasonable error message, poor performance in general,
non-terminating executions, or silent wrong results of the DAW, to name only a
few possible consequences. Searching for the causes of such errors and
drawbacks in a distributed compute cluster managed by a complex infrastructure
stack, where DAWs for large datasets typically are executed, can be tedious and
time-consuming.
We propose validity constraints (VCs) as a new concept for DAW languages to
alleviate this situation. A VC is a constraint specifying some logical
conditions that must be fulfilled at certain times for DAW executions to be
valid. When defined together with a DAW, VCs help to improve the portability,
adaptability, and reusability of DAWs by making implicit assumptions explicit.
Once specified, VC can be controlled automatically by the DAW infrastructure,
and violations can lead to meaningful error messages and graceful behaviour
(e.g., termination or invocation of repair mechanisms). We provide a broad list
of possible VCs, classify them along multiple dimensions, and compare them to
similar concepts one can find in related fields. We also provide a first sketch
for VCs' implementation into existing DAW infrastructures.
| no_new_dataset | 0.707979 |
2305.08414 | Simone Tedeschi | Simone Tedeschi, Johan Bos, Thierry Declerck, Jan Hajic, Daniel
Hershcovich, Eduard H. Hovy, Alexander Koller, Simon Krek, Steven Schockaert,
Rico Sennrich, Ekaterina Shutova, Roberto Navigli | What's the Meaning of Superhuman Performance in Today's NLU? | 9 pages, long paper at ACL 2023 proceedings | null | null | null | cs.CL cs.AI | http://creativecommons.org/licenses/by-sa/4.0/ | In the last five years, there has been a significant focus in Natural
Language Processing (NLP) on developing larger Pretrained Language Models
(PLMs) and introducing benchmarks such as SuperGLUE and SQuAD to measure their
abilities in language understanding, reasoning, and reading comprehension.
These PLMs have achieved impressive results on these benchmarks, even
surpassing human performance in some cases. This has led to claims of
superhuman capabilities and the provocative idea that certain tasks have been
solved. In this position paper, we take a critical look at these claims and ask
whether PLMs truly have superhuman abilities and what the current benchmarks
are really evaluating. We show that these benchmarks have serious limitations
affecting the comparison between humans and PLMs and provide recommendations
for fairer and more transparent benchmarks.
| [
{
"version": "v1",
"created": "Mon, 15 May 2023 07:48:31 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Tedeschi",
"Simone",
""
],
[
"Bos",
"Johan",
""
],
[
"Declerck",
"Thierry",
""
],
[
"Hajic",
"Jan",
""
],
[
"Hershcovich",
"Daniel",
""
],
[
"Hovy",
"Eduard H.",
""
],
[
"Koller",
"Alexander",
""
],
[
"Krek",
"Simon",
""
],
[
"Schockaert",
"Steven",
""
],
[
"Sennrich",
"Rico",
""
],
[
"Shutova",
"Ekaterina",
""
],
[
"Navigli",
"Roberto",
""
]
] | TITLE: What's the Meaning of Superhuman Performance in Today's NLU?
ABSTRACT: In the last five years, there has been a significant focus in Natural
Language Processing (NLP) on developing larger Pretrained Language Models
(PLMs) and introducing benchmarks such as SuperGLUE and SQuAD to measure their
abilities in language understanding, reasoning, and reading comprehension.
These PLMs have achieved impressive results on these benchmarks, even
surpassing human performance in some cases. This has led to claims of
superhuman capabilities and the provocative idea that certain tasks have been
solved. In this position paper, we take a critical look at these claims and ask
whether PLMs truly have superhuman abilities and what the current benchmarks
are really evaluating. We show that these benchmarks have serious limitations
affecting the comparison between humans and PLMs and provide recommendations
for fairer and more transparent benchmarks.
| no_new_dataset | 0.711281 |
2305.08428 | Dominik Stammbach | Emmanuel Bauer, Dominik Stammbach, Nianlong Gu, Elliott Ash | Legal Extractive Summarization of U.S. Court Opinions | null | null | null | null | cs.CL | http://creativecommons.org/licenses/by-nc-sa/4.0/ | This paper tackles the task of legal extractive summarization using a dataset
of 430K U.S. court opinions with key passages annotated. According to automated
summary quality metrics, the reinforcement-learning-based MemSum model is best
and even out-performs transformer-based models. In turn, expert human
evaluation shows that MemSum summaries effectively capture the key points of
lengthy court opinions. Motivated by these results, we open-source our models
to the general public. This represents progress towards democratizing law and
making U.S. court opinions more accessible to the general public.
| [
{
"version": "v1",
"created": "Mon, 15 May 2023 08:13:00 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Bauer",
"Emmanuel",
""
],
[
"Stammbach",
"Dominik",
""
],
[
"Gu",
"Nianlong",
""
],
[
"Ash",
"Elliott",
""
]
] | TITLE: Legal Extractive Summarization of U.S. Court Opinions
ABSTRACT: This paper tackles the task of legal extractive summarization using a dataset
of 430K U.S. court opinions with key passages annotated. According to automated
summary quality metrics, the reinforcement-learning-based MemSum model is best
and even out-performs transformer-based models. In turn, expert human
evaluation shows that MemSum summaries effectively capture the key points of
lengthy court opinions. Motivated by these results, we open-source our models
to the general public. This represents progress towards democratizing law and
making U.S. court opinions more accessible to the general public.
| new_dataset | 0.542863 |
2305.08433 | Mariia Zyrianova | Mariia Zyrianova, Dmytro Kalpakchi, Johan Boye | EMBRACE: Evaluation and Modifications for Boosting RACE | null | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | When training and evaluating machine reading comprehension models, it is very
important to work with high-quality datasets that are also representative of
real-world reading comprehension tasks. This requirement includes, for
instance, having questions that are based on texts of different genres and
require generating inferences or reflecting on the reading material.
In this article we turn our attention to RACE, a dataset of English texts and
corresponding multiple-choice questions (MCQs). Each MCQ consists of a question
and four alternatives (of which one is the correct answer). RACE was
constructed by Chinese teachers of English for human reading comprehension and
is widely used as training material for machine reading comprehension models.
By construction, RACE should satisfy the aforementioned quality requirements
and the purpose of this article is to check whether they are indeed satisfied.
We provide a detailed analysis of the test set of RACE for high-school
students (1045 texts and 3498 corresponding MCQs) including (1) an evaluation
of the difficulty of each MCQ and (2) annotations for the relevant pieces of
the texts (called "bases") that are used to justify the plausibility of each
alternative. A considerable number of MCQs appear not to fulfill basic
requirements for this type of reading comprehension tasks, so we additionally
identify the high-quality subset of the evaluated RACE corpus. We also
demonstrate that the distribution of the positions of the bases for the
alternatives is biased towards certain parts of texts, which is not necessarily
desirable when evaluating MCQ answering and generation models.
| [
{
"version": "v1",
"created": "Mon, 15 May 2023 08:21:32 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Zyrianova",
"Mariia",
""
],
[
"Kalpakchi",
"Dmytro",
""
],
[
"Boye",
"Johan",
""
]
] | TITLE: EMBRACE: Evaluation and Modifications for Boosting RACE
ABSTRACT: When training and evaluating machine reading comprehension models, it is very
important to work with high-quality datasets that are also representative of
real-world reading comprehension tasks. This requirement includes, for
instance, having questions that are based on texts of different genres and
require generating inferences or reflecting on the reading material.
In this article we turn our attention to RACE, a dataset of English texts and
corresponding multiple-choice questions (MCQs). Each MCQ consists of a question
and four alternatives (of which one is the correct answer). RACE was
constructed by Chinese teachers of English for human reading comprehension and
is widely used as training material for machine reading comprehension models.
By construction, RACE should satisfy the aforementioned quality requirements
and the purpose of this article is to check whether they are indeed satisfied.
We provide a detailed analysis of the test set of RACE for high-school
students (1045 texts and 3498 corresponding MCQs) including (1) an evaluation
of the difficulty of each MCQ and (2) annotations for the relevant pieces of
the texts (called "bases") that are used to justify the plausibility of each
alternative. A considerable number of MCQs appear not to fulfill basic
requirements for this type of reading comprehension tasks, so we additionally
identify the high-quality subset of the evaluated RACE corpus. We also
demonstrate that the distribution of the positions of the bases for the
alternatives is biased towards certain parts of texts, which is not necessarily
desirable when evaluating MCQ answering and generation models.
| new_dataset | 0.719208 |
2305.08435 | Antoine Kaufmann | Artem Ageev, Antoine Kaufmann | Kugelblitz: Streamlining Reconfigurable Packet Processing Pipeline
Design and Evaluation | Under submission for conference peer review | null | null | null | cs.NI cs.AR | http://creativecommons.org/licenses/by/4.0/ | Reconfigurable packet processing pipelines have emerged as a common building
block for offloading fast and efficient packet processing to programmable
switches and SmartNICs. However, designing or extending such architectures is
complex, laborious, and resource intensive, due to the large design space
coupled with the need for building architecture-specific compilers and RTL
implementations before any evaluation is feasible.
In this paper, we present Kugelblitz, a framework for rapidly designing,
implementing and evaluating reconfigurable packet processing pipeline
architectures. Kugelblitz comprises domain specific abstractions for separately
representing pipeline architectures and protocol processing programs, a
compiler for automatically mapping programs to specific architectures, and a
hardware generator outputting complete RTL implementations for specific
pipeline architectures. Kugelblitz enables meaningful full-system evaluation
including chip area and timing from VLSI synthesis of the generated RTL, and
end-to-end system benchmarks running real application workloads in full system
simulation. In our evaluation, we demonstrate that Kugelblitz generates
realistic and efficient RTL implementations, scales to large pipelines, enables
fast design exploration, and provides meaningful application offload
performance results.
| [
{
"version": "v1",
"created": "Mon, 15 May 2023 08:27:38 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Ageev",
"Artem",
""
],
[
"Kaufmann",
"Antoine",
""
]
] | TITLE: Kugelblitz: Streamlining Reconfigurable Packet Processing Pipeline
Design and Evaluation
ABSTRACT: Reconfigurable packet processing pipelines have emerged as a common building
block for offloading fast and efficient packet processing to programmable
switches and SmartNICs. However, designing or extending such architectures is
complex, laborious, and resource intensive, due to the large design space
coupled with the need for building architecture-specific compilers and RTL
implementations before any evaluation is feasible.
In this paper, we present Kugelblitz, a framework for rapidly designing,
implementing and evaluating reconfigurable packet processing pipeline
architectures. Kugelblitz comprises domain specific abstractions for separately
representing pipeline architectures and protocol processing programs, a
compiler for automatically mapping programs to specific architectures, and a
hardware generator outputting complete RTL implementations for specific
pipeline architectures. Kugelblitz enables meaningful full-system evaluation
including chip area and timing from VLSI synthesis of the generated RTL, and
end-to-end system benchmarks running real application workloads in full system
simulation. In our evaluation, we demonstrate that Kugelblitz generates
realistic and efficient RTL implementations, scales to large pipelines, enables
fast design exploration, and provides meaningful application offload
performance results.
| no_new_dataset | 0.707222 |
2305.08462 | Xin Xiao | Xin Xiao, Daiguo Zhou, Jiagao Hu, Yi Hu, Yongchao Xu | Not All Pixels Are Equal: Learning Pixel Hardness for Semantic
Segmentation | null | null | null | null | cs.CV cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Semantic segmentation has recently witnessed great progress. Despite the
impressive overall results, the segmentation performance in some hard areas
(e.g., small objects or thin parts) is still not promising. A straightforward
solution is hard sample mining, which is widely used in object detection. Yet,
most existing hard pixel mining strategies for semantic segmentation often rely
on pixel's loss value, which tends to decrease during training. Intuitively,
the pixel hardness for segmentation mainly depends on image structure and is
expected to be stable. In this paper, we propose to learn pixel hardness for
semantic segmentation, leveraging hardness information contained in global and
historical loss values. More precisely, we add a gradient-independent branch
for learning a hardness level (HL) map by maximizing hardness-weighted
segmentation loss, which is minimized for the segmentation head. This
encourages large hardness values in difficult areas, leading to appropriate and
stable HL map. Despite its simplicity, the proposed method can be applied to
most segmentation methods with no and marginal extra cost during inference and
training, respectively. Without bells and whistles, the proposed method
achieves consistent/significant improvement (1.37% mIoU on average) over most
popular semantic segmentation methods on Cityscapes dataset, and demonstrates
good generalization ability across domains. The source codes are available at
https://github.com/Menoly-xin/Hardness-Level-Learning .
| [
{
"version": "v1",
"created": "Mon, 15 May 2023 09:04:46 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Xiao",
"Xin",
""
],
[
"Zhou",
"Daiguo",
""
],
[
"Hu",
"Jiagao",
""
],
[
"Hu",
"Yi",
""
],
[
"Xu",
"Yongchao",
""
]
] | TITLE: Not All Pixels Are Equal: Learning Pixel Hardness for Semantic
Segmentation
ABSTRACT: Semantic segmentation has recently witnessed great progress. Despite the
impressive overall results, the segmentation performance in some hard areas
(e.g., small objects or thin parts) is still not promising. A straightforward
solution is hard sample mining, which is widely used in object detection. Yet,
most existing hard pixel mining strategies for semantic segmentation often rely
on pixel's loss value, which tends to decrease during training. Intuitively,
the pixel hardness for segmentation mainly depends on image structure and is
expected to be stable. In this paper, we propose to learn pixel hardness for
semantic segmentation, leveraging hardness information contained in global and
historical loss values. More precisely, we add a gradient-independent branch
for learning a hardness level (HL) map by maximizing hardness-weighted
segmentation loss, which is minimized for the segmentation head. This
encourages large hardness values in difficult areas, leading to appropriate and
stable HL map. Despite its simplicity, the proposed method can be applied to
most segmentation methods with no and marginal extra cost during inference and
training, respectively. Without bells and whistles, the proposed method
achieves consistent/significant improvement (1.37% mIoU on average) over most
popular semantic segmentation methods on Cityscapes dataset, and demonstrates
good generalization ability across domains. The source codes are available at
https://github.com/Menoly-xin/Hardness-Level-Learning .
| no_new_dataset | 0.712445 |
2305.08502 | Reut Apel | Reut Apel, Tom Braude, Amir Kantor, Eyal Kolman | MeeQA: Natural Questions in Meeting Transcripts | null | null | null | null | cs.CL cs.LG | http://creativecommons.org/licenses/by/4.0/ | We present MeeQA, a dataset for natural-language question answering over
meeting transcripts. It includes real questions asked during meetings by its
participants. The dataset contains 48K question-answer pairs, extracted from
422 meeting transcripts, spanning multiple domains. Questions in transcripts
pose a special challenge as they are not always clear, and considerable context
may be required in order to provide an answer. Further, many questions asked
during meetings are left unanswered. To improve baseline model performance on
this type of questions, we also propose a novel loss function, \emph{Flat
Hierarchical Loss}, designed to enhance performance over questions with no
answer in the text. Our experiments demonstrate the advantage of using our
approach over standard QA models.
| [
{
"version": "v1",
"created": "Mon, 15 May 2023 10:02:47 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Apel",
"Reut",
""
],
[
"Braude",
"Tom",
""
],
[
"Kantor",
"Amir",
""
],
[
"Kolman",
"Eyal",
""
]
] | TITLE: MeeQA: Natural Questions in Meeting Transcripts
ABSTRACT: We present MeeQA, a dataset for natural-language question answering over
meeting transcripts. It includes real questions asked during meetings by its
participants. The dataset contains 48K question-answer pairs, extracted from
422 meeting transcripts, spanning multiple domains. Questions in transcripts
pose a special challenge as they are not always clear, and considerable context
may be required in order to provide an answer. Further, many questions asked
during meetings are left unanswered. To improve baseline model performance on
this type of questions, we also propose a novel loss function, \emph{Flat
Hierarchical Loss}, designed to enhance performance over questions with no
answer in the text. Our experiments demonstrate the advantage of using our
approach over standard QA models.
| new_dataset | 0.714392 |
2305.08509 | Tongkun Liu | Tongkun Liu, Bing Li, Xiao Du, Bingke Jiang, Xiao Jin, Liuyi Jin, Zhuo
Zhao | Component-aware anomaly detection framework for adjustable and logical
industrial visual inspection | 13 pages, 15 figures | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Industrial visual inspection aims at detecting surface defects in products
during the manufacturing process. Although existing anomaly detection models
have shown great performance on many public benchmarks, their limited
adjustability and ability to detect logical anomalies hinder their broader use
in real-world settings. To this end, in this paper, we propose a novel
component-aware anomaly detection framework (ComAD) which can simultaneously
achieve adjustable and logical anomaly detection for industrial scenarios.
Specifically, we propose to segment images into multiple components based on a
lightweight and nearly training-free unsupervised semantic segmentation model.
Then, we design an interpretable logical anomaly detection model through
modeling the metrological features of each component and their relationships.
Despite its simplicity, our framework achieves state-of-the-art performance on
image-level logical anomaly detection. Meanwhile, segmenting a product image
into multiple components provides a novel perspective for industrial visual
inspection, demonstrating great potential in model customization, noise
resistance, and anomaly classification. The code will be available at
https://github.com/liutongkun/ComAD.
| [
{
"version": "v1",
"created": "Mon, 15 May 2023 10:18:52 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Liu",
"Tongkun",
""
],
[
"Li",
"Bing",
""
],
[
"Du",
"Xiao",
""
],
[
"Jiang",
"Bingke",
""
],
[
"Jin",
"Xiao",
""
],
[
"Jin",
"Liuyi",
""
],
[
"Zhao",
"Zhuo",
""
]
] | TITLE: Component-aware anomaly detection framework for adjustable and logical
industrial visual inspection
ABSTRACT: Industrial visual inspection aims at detecting surface defects in products
during the manufacturing process. Although existing anomaly detection models
have shown great performance on many public benchmarks, their limited
adjustability and ability to detect logical anomalies hinder their broader use
in real-world settings. To this end, in this paper, we propose a novel
component-aware anomaly detection framework (ComAD) which can simultaneously
achieve adjustable and logical anomaly detection for industrial scenarios.
Specifically, we propose to segment images into multiple components based on a
lightweight and nearly training-free unsupervised semantic segmentation model.
Then, we design an interpretable logical anomaly detection model through
modeling the metrological features of each component and their relationships.
Despite its simplicity, our framework achieves state-of-the-art performance on
image-level logical anomaly detection. Meanwhile, segmenting a product image
into multiple components provides a novel perspective for industrial visual
inspection, demonstrating great potential in model customization, noise
resistance, and anomaly classification. The code will be available at
https://github.com/liutongkun/ComAD.
| no_new_dataset | 0.708824 |
2305.08522 | Jingyi Wang | Jingyi Wang, Jinfa Huang, Can Zhang, and Zhidong Deng | Cross-Modality Time-Variant Relation Learning for Generating Dynamic
Scene Graphs | Preprint. Accepted by ICRA 2023 | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Dynamic scene graphs generated from video clips could help enhance the
semantic visual understanding in a wide range of challenging tasks such as
environmental perception, autonomous navigation, and task planning of
self-driving vehicles and mobile robots. In the process of temporal and spatial
modeling during dynamic scene graph generation, it is particularly intractable
to learn time-variant relations in dynamic scene graphs among frames. In this
paper, we propose a Time-variant Relation-aware TRansformer (TR$^2$), which
aims to model the temporal change of relations in dynamic scene graphs.
Explicitly, we leverage the difference of text embeddings of prompted sentences
about relation labels as the supervision signal for relations. In this way,
cross-modality feature guidance is realized for the learning of time-variant
relations. Implicitly, we design a relation feature fusion module with a
transformer and an additional message token that describes the difference
between adjacent frames. Extensive experiments on the Action Genome dataset
prove that our TR$^2$ can effectively model the time-variant relations. TR$^2$
significantly outperforms previous state-of-the-art methods under two different
settings by 2.1% and 2.6% respectively.
| [
{
"version": "v1",
"created": "Mon, 15 May 2023 10:30:38 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Wang",
"Jingyi",
""
],
[
"Huang",
"Jinfa",
""
],
[
"Zhang",
"Can",
""
],
[
"Deng",
"Zhidong",
""
]
] | TITLE: Cross-Modality Time-Variant Relation Learning for Generating Dynamic
Scene Graphs
ABSTRACT: Dynamic scene graphs generated from video clips could help enhance the
semantic visual understanding in a wide range of challenging tasks such as
environmental perception, autonomous navigation, and task planning of
self-driving vehicles and mobile robots. In the process of temporal and spatial
modeling during dynamic scene graph generation, it is particularly intractable
to learn time-variant relations in dynamic scene graphs among frames. In this
paper, we propose a Time-variant Relation-aware TRansformer (TR$^2$), which
aims to model the temporal change of relations in dynamic scene graphs.
Explicitly, we leverage the difference of text embeddings of prompted sentences
about relation labels as the supervision signal for relations. In this way,
cross-modality feature guidance is realized for the learning of time-variant
relations. Implicitly, we design a relation feature fusion module with a
transformer and an additional message token that describes the difference
between adjacent frames. Extensive experiments on the Action Genome dataset
prove that our TR$^2$ can effectively model the time-variant relations. TR$^2$
significantly outperforms previous state-of-the-art methods under two different
settings by 2.1% and 2.6% respectively.
| no_new_dataset | 0.709057 |
2305.08532 | Paola Cecilia Torrico Mor\'on | Paola Torrico Mor\'on, Sahar Salimpour, Lei Fu, Xianjia Yu, Jorge
Pe\~na Queralta, Tomi Westerlund | Benchmarking UWB-Based Infrastructure-Free Positioning and Multi-Robot
Relative Localization: Dataset and Characterization | 6 pages, 8 figures, 3 tables | null | null | null | cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Ultra-wideband (UWB) positioning has emerged as a low-cost and dependable
localization solution for multiple use cases, from mobile robots to asset
tracking within the Industrial IoT. The technology is mature and the scientific
literature contains multiple datasets and methods for localization based on
fixed UWB nodes. At the same time, research in UWB-based relative localization
and infrastructure-free localization is gaining traction, further domains.
tools and datasets in this domain are scarce. Therefore, we introduce in this
paper a novel dataset for benchmarking infrastructure-free relative
localization targeting the domain of multi-robot systems. Compared to previous
datasets, we analyze the performance of different relative localization
approaches for a much wider variety of scenarios with varying numbers of fixed
and mobile nodes. A motion capture system provides ground truth data, are
multi-modal and include inertial or odometry measurements for benchmarking
sensor fusion methods. Additionally, the dataset contains measurements of
ranging accuracy based on the relative orientation of antennas and a
comprehensive set of measurements for ranging between a single pair of nodes.
Our experimental analysis shows that high accuracy can be localization, but the
variability of the ranging error is significant across different settings and
setups.
| [
{
"version": "v1",
"created": "Mon, 15 May 2023 10:43:46 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Morón",
"Paola Torrico",
""
],
[
"Salimpour",
"Sahar",
""
],
[
"Fu",
"Lei",
""
],
[
"Yu",
"Xianjia",
""
],
[
"Queralta",
"Jorge Peña",
""
],
[
"Westerlund",
"Tomi",
""
]
] | TITLE: Benchmarking UWB-Based Infrastructure-Free Positioning and Multi-Robot
Relative Localization: Dataset and Characterization
ABSTRACT: Ultra-wideband (UWB) positioning has emerged as a low-cost and dependable
localization solution for multiple use cases, from mobile robots to asset
tracking within the Industrial IoT. The technology is mature and the scientific
literature contains multiple datasets and methods for localization based on
fixed UWB nodes. At the same time, research in UWB-based relative localization
and infrastructure-free localization is gaining traction, further domains.
tools and datasets in this domain are scarce. Therefore, we introduce in this
paper a novel dataset for benchmarking infrastructure-free relative
localization targeting the domain of multi-robot systems. Compared to previous
datasets, we analyze the performance of different relative localization
approaches for a much wider variety of scenarios with varying numbers of fixed
and mobile nodes. A motion capture system provides ground truth data, are
multi-modal and include inertial or odometry measurements for benchmarking
sensor fusion methods. Additionally, the dataset contains measurements of
ranging accuracy based on the relative orientation of antennas and a
comprehensive set of measurements for ranging between a single pair of nodes.
Our experimental analysis shows that high accuracy can be localization, but the
variability of the ranging error is significant across different settings and
setups.
| new_dataset | 0.716845 |
2305.08544 | Hua-Lei Yin | Min-Gang Zhou, Zhi-Ping Liu, Hua-Lei Yin, Chen-Long Li, Tong-Kai Xu,
Zeng-Bing Chen | Quantum Neural Network for Quantum Neural Computing | 10 pages, 6 figures | Research 6, 0134 (2023) | 10.34133/research.0134 | null | quant-ph cs.AI | http://creativecommons.org/licenses/by/4.0/ | Neural networks have achieved impressive breakthroughs in both industry and
academia. How to effectively develop neural networks on quantum computing
devices is a challenging open problem. Here, we propose a new quantum neural
network model for quantum neural computing using (classically-controlled)
single-qubit operations and measurements on real-world quantum systems with
naturally occurring environment-induced decoherence, which greatly reduces the
difficulties of physical implementations. Our model circumvents the problem
that the state-space size grows exponentially with the number of neurons,
thereby greatly reducing memory requirements and allowing for fast optimization
with traditional optimization algorithms. We benchmark our model for
handwritten digit recognition and other nonlinear classification tasks. The
results show that our model has an amazing nonlinear classification ability and
robustness to noise. Furthermore, our model allows quantum computing to be
applied in a wider context and inspires the earlier development of a quantum
neural computer than standard quantum computers.
| [
{
"version": "v1",
"created": "Mon, 15 May 2023 11:16:47 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Zhou",
"Min-Gang",
""
],
[
"Liu",
"Zhi-Ping",
""
],
[
"Yin",
"Hua-Lei",
""
],
[
"Li",
"Chen-Long",
""
],
[
"Xu",
"Tong-Kai",
""
],
[
"Chen",
"Zeng-Bing",
""
]
] | TITLE: Quantum Neural Network for Quantum Neural Computing
ABSTRACT: Neural networks have achieved impressive breakthroughs in both industry and
academia. How to effectively develop neural networks on quantum computing
devices is a challenging open problem. Here, we propose a new quantum neural
network model for quantum neural computing using (classically-controlled)
single-qubit operations and measurements on real-world quantum systems with
naturally occurring environment-induced decoherence, which greatly reduces the
difficulties of physical implementations. Our model circumvents the problem
that the state-space size grows exponentially with the number of neurons,
thereby greatly reducing memory requirements and allowing for fast optimization
with traditional optimization algorithms. We benchmark our model for
handwritten digit recognition and other nonlinear classification tasks. The
results show that our model has an amazing nonlinear classification ability and
robustness to noise. Furthermore, our model allows quantum computing to be
applied in a wider context and inspires the earlier development of a quantum
neural computer than standard quantum computers.
| no_new_dataset | 0.712607 |
2305.08551 | Elahe Arani | Ibrahim Batuhan Akkaya, Senthilkumar S. Kathiresan, Elahe Arani,
Bahram Zonooz | Enhancing Performance of Vision Transformers on Small Datasets through
Local Inductive Bias Incorporation | null | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Vision transformers (ViTs) achieve remarkable performance on large datasets,
but tend to perform worse than convolutional neural networks (CNNs) when
trained from scratch on smaller datasets, possibly due to a lack of local
inductive bias in the architecture. Recent studies have therefore added
locality to the architecture and demonstrated that it can help ViTs achieve
performance comparable to CNNs in the small-size dataset regime. Existing
methods, however, are architecture-specific or have higher computational and
memory costs. Thus, we propose a module called Local InFormation Enhancer
(LIFE) that extracts patch-level local information and incorporates it into the
embeddings used in the self-attention block of ViTs. Our proposed module is
memory and computation efficient, as well as flexible enough to process
auxiliary tokens such as the classification and distillation tokens. Empirical
results show that the addition of the LIFE module improves the performance of
ViTs on small image classification datasets. We further demonstrate how the
effect can be extended to downstream tasks, such as object detection and
semantic segmentation. In addition, we introduce a new visualization method,
Dense Attention Roll-Out, specifically designed for dense prediction tasks,
allowing the generation of class-specific attention maps utilizing the
attention maps of all tokens.
| [
{
"version": "v1",
"created": "Mon, 15 May 2023 11:23:18 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Akkaya",
"Ibrahim Batuhan",
""
],
[
"Kathiresan",
"Senthilkumar S.",
""
],
[
"Arani",
"Elahe",
""
],
[
"Zonooz",
"Bahram",
""
]
] | TITLE: Enhancing Performance of Vision Transformers on Small Datasets through
Local Inductive Bias Incorporation
ABSTRACT: Vision transformers (ViTs) achieve remarkable performance on large datasets,
but tend to perform worse than convolutional neural networks (CNNs) when
trained from scratch on smaller datasets, possibly due to a lack of local
inductive bias in the architecture. Recent studies have therefore added
locality to the architecture and demonstrated that it can help ViTs achieve
performance comparable to CNNs in the small-size dataset regime. Existing
methods, however, are architecture-specific or have higher computational and
memory costs. Thus, we propose a module called Local InFormation Enhancer
(LIFE) that extracts patch-level local information and incorporates it into the
embeddings used in the self-attention block of ViTs. Our proposed module is
memory and computation efficient, as well as flexible enough to process
auxiliary tokens such as the classification and distillation tokens. Empirical
results show that the addition of the LIFE module improves the performance of
ViTs on small image classification datasets. We further demonstrate how the
effect can be extended to downstream tasks, such as object detection and
semantic segmentation. In addition, we introduce a new visualization method,
Dense Attention Roll-Out, specifically designed for dense prediction tasks,
allowing the generation of class-specific attention maps utilizing the
attention maps of all tokens.
| no_new_dataset | 0.710377 |
2305.08583 | Bruno Pinaud | Georgios Panayiotou, Matteo Magnani, Bruno Pinaud (LaBRI, UB) | Towards efficient multilayer network data management | null | French Regional Conference on Complex Systems FRCCS 2023, Roberto
Interdonato, Cyrille Bertelle, May 2023, Le Havre, France | null | null | cs.DB | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Real-world multilayer networks can be very large and there can be multiple
choices regarding what should be modeled as a layer. Therefore, there is a need
for their effective storage and manipulation. Currently, multilayer network
analysis software use different data structures and manipulation operators. We
aim to categorize operators in order to assess which structures work best for
certain operator classes and data features. In this work, we propose a
preliminary taxonomy of layer and data manipulation operators. We also design
and execute a benchmark of select software and operators to identify potential
for optimization.
| [
{
"version": "v1",
"created": "Mon, 15 May 2023 12:10:03 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Panayiotou",
"Georgios",
"",
"LaBRI, UB"
],
[
"Magnani",
"Matteo",
"",
"LaBRI, UB"
],
[
"Pinaud",
"Bruno",
"",
"LaBRI, UB"
]
] | TITLE: Towards efficient multilayer network data management
ABSTRACT: Real-world multilayer networks can be very large and there can be multiple
choices regarding what should be modeled as a layer. Therefore, there is a need
for their effective storage and manipulation. Currently, multilayer network
analysis software use different data structures and manipulation operators. We
aim to categorize operators in order to assess which structures work best for
certain operator classes and data features. In this work, we propose a
preliminary taxonomy of layer and data manipulation operators. We also design
and execute a benchmark of select software and operators to identify potential
for optimization.
| no_new_dataset | 0.709849 |
2305.08586 | Lingyuan Kong | Lingyuan Kong, Hao Ding, Guangwei Hu | GCNSLIM: Graph Convolutional Network with Sparse Linear Methods for
E-government Service Recommendation | null | null | null | null | cs.IR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Graph Convolutional Networks have made significant strides in Collabora-tive
Filtering recommendations. However, existing GCN-based CF methods are mainly
based on matrix factorization and incorporate some optimization tech-niques to
enhance performance, which are not enough to handle the complexities of diverse
real-world recommendation scenarios. E-government service recommendation is a
crucial area for recommendation re-search as it involves rigid aspects of
people's lives. However, it has not received ad-equate attention in comparison
to other recommendation scenarios like news and music recommendation. We
empirically find that when existing GCN-based CF methods are directly applied
to e-government service recommendation, they are limited by the MF framework
and showing poor performance. This is because MF's equal treatment of users and
items is not appropriate for scenarios where the number of users and items is
unbalanced. In this work, we propose a new model, GCNSLIM, which combines GCN
and sparse linear methods instead of combining GCN and MF to accommodate
e-government service recommendation. In particular, GCNSLIM explicitly injects
high-order collaborative signals obtained from multi-layer light graph
convolutions into the item similarity matrix in the SLIM frame-work,
effectively improving the recommendation accuracy. In addition, we propose two
optimization measures, removing layer 0 embedding and adding nonlinear
acti-vation, to further adapt to the characteristics of e-government service
recommenda-tion scenarios. Furthermore, we propose a joint optimization mode to
adapt to more diverse recommendation scenarios. We conduct extensive
experiments on a real e-government service dataset and a common public dataset
and demonstrate the ef-fectiveness of GCNSLIM in recommendation accuracy and
operational performance.
| [
{
"version": "v1",
"created": "Mon, 15 May 2023 12:12:39 GMT"
}
] | 2023-05-16T00:00:00 | [
[
"Kong",
"Lingyuan",
""
],
[
"Ding",
"Hao",
""
],
[
"Hu",
"Guangwei",
""
]
] | TITLE: GCNSLIM: Graph Convolutional Network with Sparse Linear Methods for
E-government Service Recommendation
ABSTRACT: Graph Convolutional Networks have made significant strides in Collabora-tive
Filtering recommendations. However, existing GCN-based CF methods are mainly
based on matrix factorization and incorporate some optimization tech-niques to
enhance performance, which are not enough to handle the complexities of diverse
real-world recommendation scenarios. E-government service recommendation is a
crucial area for recommendation re-search as it involves rigid aspects of
people's lives. However, it has not received ad-equate attention in comparison
to other recommendation scenarios like news and music recommendation. We
empirically find that when existing GCN-based CF methods are directly applied
to e-government service recommendation, they are limited by the MF framework
and showing poor performance. This is because MF's equal treatment of users and
items is not appropriate for scenarios where the number of users and items is
unbalanced. In this work, we propose a new model, GCNSLIM, which combines GCN
and sparse linear methods instead of combining GCN and MF to accommodate
e-government service recommendation. In particular, GCNSLIM explicitly injects
high-order collaborative signals obtained from multi-layer light graph
convolutions into the item similarity matrix in the SLIM frame-work,
effectively improving the recommendation accuracy. In addition, we propose two
optimization measures, removing layer 0 embedding and adding nonlinear
acti-vation, to further adapt to the characteristics of e-government service
recommenda-tion scenarios. Furthermore, we propose a joint optimization mode to
adapt to more diverse recommendation scenarios. We conduct extensive
experiments on a real e-government service dataset and a common public dataset
and demonstrate the ef-fectiveness of GCNSLIM in recommendation accuracy and
operational performance.
| no_new_dataset | 0.706152 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.