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
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2208.11311 | Rui Song | Rui Song, Dai Liu, Dave Zhenyu Chen, Andreas Festag, Carsten Trinitis,
Martin Schulz, Alois Knoll | Federated Learning via Decentralized Dataset Distillation in
Resource-Constrained Edge Environments | Accepted at IJCNN 2023 | null | null | null | cs.LG cs.AI cs.CR | http://creativecommons.org/licenses/by/4.0/ | In federated learning, all networked clients contribute to the model training
cooperatively. However, with model sizes increasing, even sharing the trained
partial models often leads to severe communication bottlenecks in underlying
networks, especially when communicated iteratively. In this paper, we introduce
a federated learning framework FedD3 requiring only one-shot communication by
integrating dataset distillation instances. Instead of sharing model updates in
other federated learning approaches, FedD3 allows the connected clients to
distill the local datasets independently, and then aggregates those
decentralized distilled datasets (e.g. a few unrecognizable images) from
networks for model training. Our experimental results show that FedD3
significantly outperforms other federated learning frameworks in terms of
needed communication volumes, while it provides the additional benefit to be
able to balance the trade-off between accuracy and communication cost,
depending on usage scenario or target dataset. For instance, for training an
AlexNet model on CIFAR-10 with 10 clients under non-independent and identically
distributed (Non-IID) setting, FedD3 can either increase the accuracy by over
71% with a similar communication volume, or save 98% of communication volume,
while reaching the same accuracy, compared to other one-shot federated learning
approaches.
| [
{
"version": "v1",
"created": "Wed, 24 Aug 2022 05:36:22 GMT"
},
{
"version": "v2",
"created": "Wed, 31 Aug 2022 10:05:53 GMT"
},
{
"version": "v3",
"created": "Fri, 19 May 2023 12:30:25 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Song",
"Rui",
""
],
[
"Liu",
"Dai",
""
],
[
"Chen",
"Dave Zhenyu",
""
],
[
"Festag",
"Andreas",
""
],
[
"Trinitis",
"Carsten",
""
],
[
"Schulz",
"Martin",
""
],
[
"Knoll",
"Alois",
""
]
] | TITLE: Federated Learning via Decentralized Dataset Distillation in
Resource-Constrained Edge Environments
ABSTRACT: In federated learning, all networked clients contribute to the model training
cooperatively. However, with model sizes increasing, even sharing the trained
partial models often leads to severe communication bottlenecks in underlying
networks, especially when communicated iteratively. In this paper, we introduce
a federated learning framework FedD3 requiring only one-shot communication by
integrating dataset distillation instances. Instead of sharing model updates in
other federated learning approaches, FedD3 allows the connected clients to
distill the local datasets independently, and then aggregates those
decentralized distilled datasets (e.g. a few unrecognizable images) from
networks for model training. Our experimental results show that FedD3
significantly outperforms other federated learning frameworks in terms of
needed communication volumes, while it provides the additional benefit to be
able to balance the trade-off between accuracy and communication cost,
depending on usage scenario or target dataset. For instance, for training an
AlexNet model on CIFAR-10 with 10 clients under non-independent and identically
distributed (Non-IID) setting, FedD3 can either increase the accuracy by over
71% with a similar communication volume, or save 98% of communication volume,
while reaching the same accuracy, compared to other one-shot federated learning
approaches.
| no_new_dataset | 0.707354 |
2209.00613 | Damien Teney | Damien Teney, Yong Lin, Seong Joon Oh, Ehsan Abbasnejad | ID and OOD Performance Are Sometimes Inversely Correlated on Real-world
Datasets | null | null | null | null | cs.LG cs.CV | http://creativecommons.org/licenses/by/4.0/ | Several studies have compared the in-distribution (ID) and
out-of-distribution (OOD) performance of models in computer vision and NLP.
They report a frequent positive correlation and some surprisingly never even
observe an inverse correlation indicative of a necessary trade-off. The
possibility of inverse patterns is important to determine whether ID
performance can serve as a proxy for OOD generalization capabilities.
This paper shows with multiple datasets that inverse correlations between ID
and OOD performance do happen in real-world data - not only in theoretical
worst-case settings. We also explain theoretically how these cases can arise
even in a minimal linear setting, and why past studies could miss such cases
due to a biased selection of models.
Our observations lead to recommendations that contradict those found in much
of the current literature. - High OOD performance sometimes requires trading
off ID performance. - Focusing on ID performance alone may not lead to optimal
OOD performance. It may produce diminishing (eventually negative) returns in
OOD performance. - In these cases, studies on OOD generalization that use ID
performance for model selection (a common recommended practice) will
necessarily miss the best-performing models, making these studies blind to a
whole range of phenomena.
| [
{
"version": "v1",
"created": "Thu, 1 Sep 2022 17:27:25 GMT"
},
{
"version": "v2",
"created": "Mon, 26 Sep 2022 13:20:07 GMT"
},
{
"version": "v3",
"created": "Sun, 18 Dec 2022 09:32:03 GMT"
},
{
"version": "v4",
"created": "Fri, 19 May 2023 07:24:53 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Teney",
"Damien",
""
],
[
"Lin",
"Yong",
""
],
[
"Oh",
"Seong Joon",
""
],
[
"Abbasnejad",
"Ehsan",
""
]
] | TITLE: ID and OOD Performance Are Sometimes Inversely Correlated on Real-world
Datasets
ABSTRACT: Several studies have compared the in-distribution (ID) and
out-of-distribution (OOD) performance of models in computer vision and NLP.
They report a frequent positive correlation and some surprisingly never even
observe an inverse correlation indicative of a necessary trade-off. The
possibility of inverse patterns is important to determine whether ID
performance can serve as a proxy for OOD generalization capabilities.
This paper shows with multiple datasets that inverse correlations between ID
and OOD performance do happen in real-world data - not only in theoretical
worst-case settings. We also explain theoretically how these cases can arise
even in a minimal linear setting, and why past studies could miss such cases
due to a biased selection of models.
Our observations lead to recommendations that contradict those found in much
of the current literature. - High OOD performance sometimes requires trading
off ID performance. - Focusing on ID performance alone may not lead to optimal
OOD performance. It may produce diminishing (eventually negative) returns in
OOD performance. - In these cases, studies on OOD generalization that use ID
performance for model selection (a common recommended practice) will
necessarily miss the best-performing models, making these studies blind to a
whole range of phenomena.
| no_new_dataset | 0.709573 |
2210.03945 | Izzeddin Gur | Izzeddin Gur, Ofir Nachum, Yingjie Miao, Mustafa Safdari, Austin
Huang, Aakanksha Chowdhery, Sharan Narang, Noah Fiedel, Aleksandra Faust | Understanding HTML with Large Language Models | null | null | null | null | cs.LG cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Large language models (LLMs) have shown exceptional performance on a variety
of natural language tasks. Yet, their capabilities for HTML understanding --
i.e., parsing the raw HTML of a webpage, with applications to automation of
web-based tasks, crawling, and browser-assisted retrieval -- have not been
fully explored. We contribute HTML understanding models (fine-tuned LLMs) and
an in-depth analysis of their capabilities under three tasks: (i) Semantic
Classification of HTML elements, (ii) Description Generation for HTML inputs,
and (iii) Autonomous Web Navigation of HTML pages. While previous work has
developed dedicated architectures and training procedures for HTML
understanding, we show that LLMs pretrained on standard natural language
corpora transfer remarkably well to HTML understanding tasks. For instance,
fine-tuned LLMs are 12% more accurate at semantic classification compared to
models trained exclusively on the task dataset. Moreover, when fine-tuned on
data from the MiniWoB benchmark, LLMs successfully complete 50% more tasks
using 192x less data compared to the previous best supervised model. Out of the
LLMs we evaluate, we show evidence that T5-based models are ideal due to their
bidirectional encoder-decoder architecture. To promote further research on LLMs
for HTML understanding, we create and open-source a large-scale HTML dataset
distilled and auto-labeled from CommonCrawl.
| [
{
"version": "v1",
"created": "Sat, 8 Oct 2022 07:27:17 GMT"
},
{
"version": "v2",
"created": "Fri, 19 May 2023 17:42:56 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Gur",
"Izzeddin",
""
],
[
"Nachum",
"Ofir",
""
],
[
"Miao",
"Yingjie",
""
],
[
"Safdari",
"Mustafa",
""
],
[
"Huang",
"Austin",
""
],
[
"Chowdhery",
"Aakanksha",
""
],
[
"Narang",
"Sharan",
""
],
[
"Fiedel",
"Noah",
""
],
[
"Faust",
"Aleksandra",
""
]
] | TITLE: Understanding HTML with Large Language Models
ABSTRACT: Large language models (LLMs) have shown exceptional performance on a variety
of natural language tasks. Yet, their capabilities for HTML understanding --
i.e., parsing the raw HTML of a webpage, with applications to automation of
web-based tasks, crawling, and browser-assisted retrieval -- have not been
fully explored. We contribute HTML understanding models (fine-tuned LLMs) and
an in-depth analysis of their capabilities under three tasks: (i) Semantic
Classification of HTML elements, (ii) Description Generation for HTML inputs,
and (iii) Autonomous Web Navigation of HTML pages. While previous work has
developed dedicated architectures and training procedures for HTML
understanding, we show that LLMs pretrained on standard natural language
corpora transfer remarkably well to HTML understanding tasks. For instance,
fine-tuned LLMs are 12% more accurate at semantic classification compared to
models trained exclusively on the task dataset. Moreover, when fine-tuned on
data from the MiniWoB benchmark, LLMs successfully complete 50% more tasks
using 192x less data compared to the previous best supervised model. Out of the
LLMs we evaluate, we show evidence that T5-based models are ideal due to their
bidirectional encoder-decoder architecture. To promote further research on LLMs
for HTML understanding, we create and open-source a large-scale HTML dataset
distilled and auto-labeled from CommonCrawl.
| no_new_dataset | 0.710459 |
2211.06651 | Nils Dycke | Nils Dycke, Ilia Kuznetsov, Iryna Gurevych | NLPeer: A Unified Resource for the Computational Study of Peer Review | Accepted at the ACL 2023 main conference | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Peer review constitutes a core component of scholarly publishing; yet it
demands substantial expertise and training, and is susceptible to errors and
biases. Various applications of NLP for peer reviewing assistance aim to
support reviewers in this complex process, but the lack of clearly licensed
datasets and multi-domain corpora prevent the systematic study of NLP for peer
review. To remedy this, we introduce NLPeer -- the first ethically sourced
multidomain corpus of more than 5k papers and 11k review reports from five
different venues. In addition to the new datasets of paper drafts, camera-ready
versions and peer reviews from the NLP community, we establish a unified data
representation and augment previous peer review datasets to include parsed and
structured paper representations, rich metadata and versioning information. We
complement our resource with implementations and analysis of three reviewing
assistance tasks, including a novel guided skimming task. Our work paves the
path towards systematic, multi-faceted, evidence-based study of peer review in
NLP and beyond. The data and code are publicly available.
| [
{
"version": "v1",
"created": "Sat, 12 Nov 2022 12:29:38 GMT"
},
{
"version": "v2",
"created": "Fri, 19 May 2023 07:03:27 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Dycke",
"Nils",
""
],
[
"Kuznetsov",
"Ilia",
""
],
[
"Gurevych",
"Iryna",
""
]
] | TITLE: NLPeer: A Unified Resource for the Computational Study of Peer Review
ABSTRACT: Peer review constitutes a core component of scholarly publishing; yet it
demands substantial expertise and training, and is susceptible to errors and
biases. Various applications of NLP for peer reviewing assistance aim to
support reviewers in this complex process, but the lack of clearly licensed
datasets and multi-domain corpora prevent the systematic study of NLP for peer
review. To remedy this, we introduce NLPeer -- the first ethically sourced
multidomain corpus of more than 5k papers and 11k review reports from five
different venues. In addition to the new datasets of paper drafts, camera-ready
versions and peer reviews from the NLP community, we establish a unified data
representation and augment previous peer review datasets to include parsed and
structured paper representations, rich metadata and versioning information. We
complement our resource with implementations and analysis of three reviewing
assistance tasks, including a novel guided skimming task. Our work paves the
path towards systematic, multi-faceted, evidence-based study of peer review in
NLP and beyond. The data and code are publicly available.
| new_dataset | 0.714864 |
2212.07530 | Uri Shaham | Uri Shaham and Maha Elbayad and Vedanuj Goswami and Omer Levy and
Shruti Bhosale | Causes and Cures for Interference in Multilingual Translation | ACL 2023 | null | null | null | cs.CL cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Multilingual machine translation models can benefit from synergy between
different language pairs, but also suffer from interference. While there is a
growing number of sophisticated methods that aim to eliminate interference, our
understanding of interference as a phenomenon is still limited. This work
identifies the main factors that contribute to interference in multilingual
machine translation. Through systematic experimentation, we find that
interference (or synergy) are primarily determined by model size, data size,
and the proportion of each language pair within the total dataset. We observe
that substantial interference occurs mainly when the model is very small with
respect to the available training data, and that using standard transformer
configurations with less than one billion parameters largely alleviates
interference and promotes synergy. Moreover, we show that tuning the sampling
temperature to control the proportion of each language pair in the data is key
to balancing the amount of interference between low and high resource language
pairs effectively, and can lead to superior performance overall.
| [
{
"version": "v1",
"created": "Wed, 14 Dec 2022 22:30:55 GMT"
},
{
"version": "v2",
"created": "Mon, 15 May 2023 06:14:38 GMT"
},
{
"version": "v3",
"created": "Fri, 19 May 2023 12:26:50 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Shaham",
"Uri",
""
],
[
"Elbayad",
"Maha",
""
],
[
"Goswami",
"Vedanuj",
""
],
[
"Levy",
"Omer",
""
],
[
"Bhosale",
"Shruti",
""
]
] | TITLE: Causes and Cures for Interference in Multilingual Translation
ABSTRACT: Multilingual machine translation models can benefit from synergy between
different language pairs, but also suffer from interference. While there is a
growing number of sophisticated methods that aim to eliminate interference, our
understanding of interference as a phenomenon is still limited. This work
identifies the main factors that contribute to interference in multilingual
machine translation. Through systematic experimentation, we find that
interference (or synergy) are primarily determined by model size, data size,
and the proportion of each language pair within the total dataset. We observe
that substantial interference occurs mainly when the model is very small with
respect to the available training data, and that using standard transformer
configurations with less than one billion parameters largely alleviates
interference and promotes synergy. Moreover, we show that tuning the sampling
temperature to control the proportion of each language pair in the data is key
to balancing the amount of interference between low and high resource language
pairs effectively, and can lead to superior performance overall.
| no_new_dataset | 0.71081 |
2212.09667 | Alex Mei | Alex Mei, Sharon Levy, William Yang Wang | Foveate, Attribute, and Rationalize: Towards Physically Safe and
Trustworthy AI | In Findings of the 2023 Conference of the Association for
Computational Linguistics | null | null | null | cs.CL cs.AI cs.LG | http://creativecommons.org/licenses/by/4.0/ | Users' physical safety is an increasing concern as the market for intelligent
systems continues to grow, where unconstrained systems may recommend users
dangerous actions that can lead to serious injury. Covertly unsafe text is an
area of particular interest, as such text may arise from everyday scenarios and
are challenging to detect as harmful. We propose FARM, a novel framework
leveraging external knowledge for trustworthy rationale generation in the
context of safety. In particular, FARM foveates on missing knowledge to qualify
the information required to reason in specific scenarios and retrieves this
information with attribution to trustworthy sources. This knowledge is used to
both classify the safety of the original text and generate human-interpretable
rationales, shedding light on the risk of systems to specific user groups and
helping both stakeholders manage the risks of their systems and policymakers to
provide concrete safeguards for consumer safety. Our experiments show that FARM
obtains state-of-the-art results on the SafeText dataset, showing absolute
improvement in safety classification accuracy by 5.9%.
| [
{
"version": "v1",
"created": "Mon, 19 Dec 2022 17:51:47 GMT"
},
{
"version": "v2",
"created": "Fri, 19 May 2023 05:19:18 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Mei",
"Alex",
""
],
[
"Levy",
"Sharon",
""
],
[
"Wang",
"William Yang",
""
]
] | TITLE: Foveate, Attribute, and Rationalize: Towards Physically Safe and
Trustworthy AI
ABSTRACT: Users' physical safety is an increasing concern as the market for intelligent
systems continues to grow, where unconstrained systems may recommend users
dangerous actions that can lead to serious injury. Covertly unsafe text is an
area of particular interest, as such text may arise from everyday scenarios and
are challenging to detect as harmful. We propose FARM, a novel framework
leveraging external knowledge for trustworthy rationale generation in the
context of safety. In particular, FARM foveates on missing knowledge to qualify
the information required to reason in specific scenarios and retrieves this
information with attribution to trustworthy sources. This knowledge is used to
both classify the safety of the original text and generate human-interpretable
rationales, shedding light on the risk of systems to specific user groups and
helping both stakeholders manage the risks of their systems and policymakers to
provide concrete safeguards for consumer safety. Our experiments show that FARM
obtains state-of-the-art results on the SafeText dataset, showing absolute
improvement in safety classification accuracy by 5.9%.
| no_new_dataset | 0.710258 |
2212.10391 | Jungsoo Park | Jimin Hong, Jungsoo Park, Daeyoung Kim, Seongjae Choi, Bokyung Son,
and Jaewook Kang | Empowering Sentence Encoders with Prompting and Label Retrieval for
Zero-shot Text Classification | 9 pages, 4 figures | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | With contrastive pre-training, sentence encoders are generally optimized to
locate semantically similar samples closer to each other in their embedding
spaces. In this work, we focus on the potential of their embedding spaces to be
readily adapted to zero-shot text classification, as semantically distinct
samples are already well-separated. Our framework, RaLP (Retrieval augmented
Label Prompts for sentence encoder), encodes prompted label candidates with a
sentence encoder, then assigns the label whose prompt embedding has the highest
similarity with the input text embedding. In order to compensate for the
potentially poorly descriptive labels in their original format, RaLP retrieves
sentences that are semantically similar to the original label prompt from
external corpora and use them as additional pseudo-label prompts. RaLP achieves
competitive or stronger performance than much larger baselines on various
closed-set classification and multiple-choice QA datasets under zero-shot
settings. We show that the retrieval component plays a pivotal role in RaLP's
success, and its results are robustly attained regardless of verbalizer
variations.
| [
{
"version": "v1",
"created": "Tue, 20 Dec 2022 16:18:03 GMT"
},
{
"version": "v2",
"created": "Fri, 19 May 2023 08:37:33 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Hong",
"Jimin",
""
],
[
"Park",
"Jungsoo",
""
],
[
"Kim",
"Daeyoung",
""
],
[
"Choi",
"Seongjae",
""
],
[
"Son",
"Bokyung",
""
],
[
"Kang",
"Jaewook",
""
]
] | TITLE: Empowering Sentence Encoders with Prompting and Label Retrieval for
Zero-shot Text Classification
ABSTRACT: With contrastive pre-training, sentence encoders are generally optimized to
locate semantically similar samples closer to each other in their embedding
spaces. In this work, we focus on the potential of their embedding spaces to be
readily adapted to zero-shot text classification, as semantically distinct
samples are already well-separated. Our framework, RaLP (Retrieval augmented
Label Prompts for sentence encoder), encodes prompted label candidates with a
sentence encoder, then assigns the label whose prompt embedding has the highest
similarity with the input text embedding. In order to compensate for the
potentially poorly descriptive labels in their original format, RaLP retrieves
sentences that are semantically similar to the original label prompt from
external corpora and use them as additional pseudo-label prompts. RaLP achieves
competitive or stronger performance than much larger baselines on various
closed-set classification and multiple-choice QA datasets under zero-shot
settings. We show that the retrieval component plays a pivotal role in RaLP's
success, and its results are robustly attained regardless of verbalizer
variations.
| no_new_dataset | 0.711224 |
2212.10481 | Zhiruo Wang | Zhiruo Wang, Shuyan Zhou, Daniel Fried, Graham Neubig | Execution-Based Evaluation for Open-Domain Code Generation | null | null | null | null | cs.SE cs.AI cs.CL | http://creativecommons.org/licenses/by-sa/4.0/ | To extend the scope of coding queries to more realistic settings, we propose
ODEX, the first Open-Domain EXecution-based natural language (NL) to Python
code generation dataset. ODEX has 945 NL-Code pairs spanning 79 diverse
libraries, along with 1,707 human-written test cases for execution. Our NL-Code
pairs are harvested from StackOverflow forums to encourage natural and
practical coding queries. Moreover, ODEX supports four natural languages as
intents, in English, Spanish, Japanese, and Russian. ODEX unveils intriguing
behavioral differences among top-performing code language models (LM). While
CODEX achieves better overall results, CODEGEN improves effectively via scaling
-- CODEGEN 6.1B performs comparably with CODEX 12B. Both models show
substantial gaps between open and closed domains, but CODEGEN gaps tend to
decrease with model size while CODEX gaps increase. We release ODEX to
facilitate research into open-domain problems for the code generation
community.
| [
{
"version": "v1",
"created": "Tue, 20 Dec 2022 17:54:37 GMT"
},
{
"version": "v2",
"created": "Fri, 19 May 2023 14:27:46 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Wang",
"Zhiruo",
""
],
[
"Zhou",
"Shuyan",
""
],
[
"Fried",
"Daniel",
""
],
[
"Neubig",
"Graham",
""
]
] | TITLE: Execution-Based Evaluation for Open-Domain Code Generation
ABSTRACT: To extend the scope of coding queries to more realistic settings, we propose
ODEX, the first Open-Domain EXecution-based natural language (NL) to Python
code generation dataset. ODEX has 945 NL-Code pairs spanning 79 diverse
libraries, along with 1,707 human-written test cases for execution. Our NL-Code
pairs are harvested from StackOverflow forums to encourage natural and
practical coding queries. Moreover, ODEX supports four natural languages as
intents, in English, Spanish, Japanese, and Russian. ODEX unveils intriguing
behavioral differences among top-performing code language models (LM). While
CODEX achieves better overall results, CODEGEN improves effectively via scaling
-- CODEGEN 6.1B performs comparably with CODEX 12B. Both models show
substantial gaps between open and closed domains, but CODEGEN gaps tend to
decrease with model size while CODEX gaps increase. We release ODEX to
facilitate research into open-domain problems for the code generation
community.
| new_dataset | 0.712432 |
2301.05575 | Jo\~ao Mendes Lopes | Carolina Gon\c{c}alves, Jo\~ao M. Lopes, Sara Moccia, Daniele
Berardini, Lucia Migliorelli, and Cristina P. Santos | Deep learning-based approaches for human motion decoding in smart
walkers for rehabilitation | null | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Gait disabilities are among the most frequent worldwide. Their treatment
relies on rehabilitation therapies, in which smart walkers are being introduced
to empower the user's recovery and autonomy, while reducing the clinicians
effort. For that, these should be able to decode human motion and needs, as
early as possible. Current walkers decode motion intention using information of
wearable or embedded sensors, namely inertial units, force and hall sensors,
and lasers, whose main limitations imply an expensive solution or hinder the
perception of human movement. Smart walkers commonly lack a seamless
human-robot interaction, which intuitively understands human motions. A
contactless approach is proposed in this work, addressing human motion decoding
as an early action recognition/detection problematic, using RGB-D cameras. We
studied different deep learning-based algorithms, organised in three different
approaches, to process lower body RGB-D video sequences, recorded from an
embedded camera of a smart walker, and classify them into 4 classes (stop,
walk, turn right/left). A custom dataset involving 15 healthy participants
walking with the device was acquired and prepared, resulting in 28800 balanced
RGB-D frames, to train and evaluate the deep networks. The best results were
attained by a convolutional neural network with a channel attention mechanism,
reaching accuracy values of 99.61% and above 93%, for offline early
detection/recognition and trial simulations, respectively. Following the
hypothesis that human lower body features encode prominent information,
fostering a more robust prediction towards real-time applications, the
algorithm focus was also evaluated using Dice metric, leading to values
slightly higher than 30%. Promising results were attained for early action
detection as a human motion decoding strategy, with enhancements in the focus
of the proposed architectures.
| [
{
"version": "v1",
"created": "Fri, 13 Jan 2023 14:29:44 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Gonçalves",
"Carolina",
""
],
[
"Lopes",
"João M.",
""
],
[
"Moccia",
"Sara",
""
],
[
"Berardini",
"Daniele",
""
],
[
"Migliorelli",
"Lucia",
""
],
[
"Santos",
"Cristina P.",
""
]
] | TITLE: Deep learning-based approaches for human motion decoding in smart
walkers for rehabilitation
ABSTRACT: Gait disabilities are among the most frequent worldwide. Their treatment
relies on rehabilitation therapies, in which smart walkers are being introduced
to empower the user's recovery and autonomy, while reducing the clinicians
effort. For that, these should be able to decode human motion and needs, as
early as possible. Current walkers decode motion intention using information of
wearable or embedded sensors, namely inertial units, force and hall sensors,
and lasers, whose main limitations imply an expensive solution or hinder the
perception of human movement. Smart walkers commonly lack a seamless
human-robot interaction, which intuitively understands human motions. A
contactless approach is proposed in this work, addressing human motion decoding
as an early action recognition/detection problematic, using RGB-D cameras. We
studied different deep learning-based algorithms, organised in three different
approaches, to process lower body RGB-D video sequences, recorded from an
embedded camera of a smart walker, and classify them into 4 classes (stop,
walk, turn right/left). A custom dataset involving 15 healthy participants
walking with the device was acquired and prepared, resulting in 28800 balanced
RGB-D frames, to train and evaluate the deep networks. The best results were
attained by a convolutional neural network with a channel attention mechanism,
reaching accuracy values of 99.61% and above 93%, for offline early
detection/recognition and trial simulations, respectively. Following the
hypothesis that human lower body features encode prominent information,
fostering a more robust prediction towards real-time applications, the
algorithm focus was also evaluated using Dice metric, leading to values
slightly higher than 30%. Promising results were attained for early action
detection as a human motion decoding strategy, with enhancements in the focus
of the proposed architectures.
| new_dataset | 0.718588 |
2301.07939 | Feng Dang | Feng Dang, Qi Hu, Pengyuan Zhang | THLNet: two-stage heterogeneous lightweight network for monaural speech
enhancement | null | null | null | null | cs.SD eess.AS | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper, we propose a two-stage heterogeneous lightweight network for
monaural speech enhancement. Specifically, we design a novel two-stage
framework consisting of a coarse-grained full-band mask estimation stage and a
fine-grained low-frequency refinement stage. Instead of using a hand-designed
real-valued filter, we use a novel learnable complex-valued rectangular
bandwidth (LCRB) filter bank as an extractor of compact features. Furthermore,
considering the respective characteristics of the proposed two-stage task, we
used a heterogeneous structure, i.e., a U-shaped subnetwork as the backbone of
CoarseNet and a single-scale subnetwork as the backbone of FineNet. We
conducted experiments on the VoiceBank + DEMAND and DNS datasets to evaluate
the proposed approach. The experimental results show that the proposed method
outperforms the current state-of-the-art methods, while maintaining relatively
small model size and low computational complexity.
| [
{
"version": "v1",
"created": "Thu, 19 Jan 2023 08:17:22 GMT"
},
{
"version": "v2",
"created": "Fri, 19 May 2023 07:11:08 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Dang",
"Feng",
""
],
[
"Hu",
"Qi",
""
],
[
"Zhang",
"Pengyuan",
""
]
] | TITLE: THLNet: two-stage heterogeneous lightweight network for monaural speech
enhancement
ABSTRACT: In this paper, we propose a two-stage heterogeneous lightweight network for
monaural speech enhancement. Specifically, we design a novel two-stage
framework consisting of a coarse-grained full-band mask estimation stage and a
fine-grained low-frequency refinement stage. Instead of using a hand-designed
real-valued filter, we use a novel learnable complex-valued rectangular
bandwidth (LCRB) filter bank as an extractor of compact features. Furthermore,
considering the respective characteristics of the proposed two-stage task, we
used a heterogeneous structure, i.e., a U-shaped subnetwork as the backbone of
CoarseNet and a single-scale subnetwork as the backbone of FineNet. We
conducted experiments on the VoiceBank + DEMAND and DNS datasets to evaluate
the proposed approach. The experimental results show that the proposed method
outperforms the current state-of-the-art methods, while maintaining relatively
small model size and low computational complexity.
| no_new_dataset | 0.71039 |
2302.02231 | Kian Ahrabian | Kian Ahrabian, Xinwei Du, Richard Delwin Myloth, Arun Baalaaji Sankar
Ananthan, Jay Pujara | PubGraph: A Large-Scale Scientific Knowledge Graph | 17 Pages, 6 Figures, 9 Tables | null | null | null | cs.AI cs.LG | http://creativecommons.org/licenses/by-sa/4.0/ | Research publications are the primary vehicle for sharing scientific progress
in the form of new discoveries, methods, techniques, and insights.
Unfortunately, the lack of a large-scale, comprehensive, and easy-to-use
resource capturing the myriad relationships between publications, their
authors, and venues presents a barrier to applications for gaining a deeper
understanding of science. In this paper, we present PubGraph, a new resource
for studying scientific progress that takes the form of a large-scale knowledge
graph (KG) with more than 385M entities, 13B main edges, and 1.5B qualifier
edges. PubGraph is comprehensive and unifies data from various sources,
including Wikidata, OpenAlex, and Semantic Scholar, using the Wikidata
ontology. Beyond the metadata available from these sources, PubGraph includes
outputs from auxiliary community detection algorithms and large language
models. To further support studies on reasoning over scientific networks, we
create several large-scale benchmarks extracted from PubGraph for the core task
of knowledge graph completion (KGC). These benchmarks present many challenges
for knowledge graph embedding models, including an adversarial community-based
KGC evaluation setting, zero-shot inductive learning, and large-scale learning.
All of the aforementioned resources are accessible at https://pubgraph.isi.edu/
and released under the CC-BY-SA license. We plan to update PubGraph quarterly
to accommodate the release of new publications.
| [
{
"version": "v1",
"created": "Sat, 4 Feb 2023 20:03:55 GMT"
},
{
"version": "v2",
"created": "Fri, 19 May 2023 04:56:47 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Ahrabian",
"Kian",
""
],
[
"Du",
"Xinwei",
""
],
[
"Myloth",
"Richard Delwin",
""
],
[
"Ananthan",
"Arun Baalaaji Sankar",
""
],
[
"Pujara",
"Jay",
""
]
] | TITLE: PubGraph: A Large-Scale Scientific Knowledge Graph
ABSTRACT: Research publications are the primary vehicle for sharing scientific progress
in the form of new discoveries, methods, techniques, and insights.
Unfortunately, the lack of a large-scale, comprehensive, and easy-to-use
resource capturing the myriad relationships between publications, their
authors, and venues presents a barrier to applications for gaining a deeper
understanding of science. In this paper, we present PubGraph, a new resource
for studying scientific progress that takes the form of a large-scale knowledge
graph (KG) with more than 385M entities, 13B main edges, and 1.5B qualifier
edges. PubGraph is comprehensive and unifies data from various sources,
including Wikidata, OpenAlex, and Semantic Scholar, using the Wikidata
ontology. Beyond the metadata available from these sources, PubGraph includes
outputs from auxiliary community detection algorithms and large language
models. To further support studies on reasoning over scientific networks, we
create several large-scale benchmarks extracted from PubGraph for the core task
of knowledge graph completion (KGC). These benchmarks present many challenges
for knowledge graph embedding models, including an adversarial community-based
KGC evaluation setting, zero-shot inductive learning, and large-scale learning.
All of the aforementioned resources are accessible at https://pubgraph.isi.edu/
and released under the CC-BY-SA license. We plan to update PubGraph quarterly
to accommodate the release of new publications.
| no_new_dataset | 0.705943 |
2302.05134 | Jonas Falkner | Jonas K. Falkner and Lars Schmidt-Thieme | Neural Capacitated Clustering | Accepted at the 32nd International Joint Conference on Artificial
Intelligence (IJCAI) 2023 | null | null | null | cs.LG | http://creativecommons.org/licenses/by-sa/4.0/ | Recent work on deep clustering has found new promising methods also for
constrained clustering problems. Their typically pairwise constraints often can
be used to guide the partitioning of the data. Many problems however, feature
cluster-level constraints, e.g. the Capacitated Clustering Problem (CCP), where
each point has a weight and the total weight sum of all points in each cluster
is bounded by a prescribed capacity. In this paper we propose a new method for
the CCP, Neural Capacited Clustering, that learns a neural network to predict
the assignment probabilities of points to cluster centers from a data set of
optimal or near optimal past solutions of other problem instances. During
inference, the resulting scores are then used in an iterative k-means like
procedure to refine the assignment under capacity constraints. In our
experiments on artificial data and two real world datasets our approach
outperforms several state-of-the-art mathematical and heuristic solvers from
the literature. Moreover, we apply our method in the context of a
cluster-first-route-second approach to the Capacitated Vehicle Routing Problem
(CVRP) and show competitive results on the well-known Uchoa benchmark.
| [
{
"version": "v1",
"created": "Fri, 10 Feb 2023 09:33:44 GMT"
},
{
"version": "v2",
"created": "Fri, 19 May 2023 08:36:55 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Falkner",
"Jonas K.",
""
],
[
"Schmidt-Thieme",
"Lars",
""
]
] | TITLE: Neural Capacitated Clustering
ABSTRACT: Recent work on deep clustering has found new promising methods also for
constrained clustering problems. Their typically pairwise constraints often can
be used to guide the partitioning of the data. Many problems however, feature
cluster-level constraints, e.g. the Capacitated Clustering Problem (CCP), where
each point has a weight and the total weight sum of all points in each cluster
is bounded by a prescribed capacity. In this paper we propose a new method for
the CCP, Neural Capacited Clustering, that learns a neural network to predict
the assignment probabilities of points to cluster centers from a data set of
optimal or near optimal past solutions of other problem instances. During
inference, the resulting scores are then used in an iterative k-means like
procedure to refine the assignment under capacity constraints. In our
experiments on artificial data and two real world datasets our approach
outperforms several state-of-the-art mathematical and heuristic solvers from
the literature. Moreover, we apply our method in the context of a
cluster-first-route-second approach to the Capacitated Vehicle Routing Problem
(CVRP) and show competitive results on the well-known Uchoa benchmark.
| no_new_dataset | 0.709655 |
2302.05552 | Yiyun He | Yiyun He, Roman Vershynin, Yizhe Zhu | Algorithmically Effective Differentially Private Synthetic Data | 23 pages. to appear in COLT 2023 | null | null | null | cs.DS cs.CR math.PR math.ST stat.TH | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present a highly effective algorithmic approach for generating
$\varepsilon$-differentially private synthetic data in a bounded metric space
with near-optimal utility guarantees under the 1-Wasserstein distance. In
particular, for a dataset $X$ in the hypercube $[0,1]^d$, our algorithm
generates synthetic dataset $Y$ such that the expected 1-Wasserstein distance
between the empirical measure of $X$ and $Y$ is $O((\varepsilon n)^{-1/d})$ for
$d\geq 2$, and is $O(\log^2(\varepsilon n)(\varepsilon n)^{-1})$ for $d=1$. The
accuracy guarantee is optimal up to a constant factor for $d\geq 2$, and up to
a logarithmic factor for $d=1$. Our algorithm has a fast running time of
$O(\varepsilon dn)$ for all $d\geq 1$ and demonstrates improved accuracy
compared to the method in (Boedihardjo et al., 2022) for $d\geq 2$.
| [
{
"version": "v1",
"created": "Sat, 11 Feb 2023 00:10:35 GMT"
},
{
"version": "v2",
"created": "Tue, 14 Feb 2023 19:42:46 GMT"
},
{
"version": "v3",
"created": "Thu, 18 May 2023 21:31:46 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"He",
"Yiyun",
""
],
[
"Vershynin",
"Roman",
""
],
[
"Zhu",
"Yizhe",
""
]
] | TITLE: Algorithmically Effective Differentially Private Synthetic Data
ABSTRACT: We present a highly effective algorithmic approach for generating
$\varepsilon$-differentially private synthetic data in a bounded metric space
with near-optimal utility guarantees under the 1-Wasserstein distance. In
particular, for a dataset $X$ in the hypercube $[0,1]^d$, our algorithm
generates synthetic dataset $Y$ such that the expected 1-Wasserstein distance
between the empirical measure of $X$ and $Y$ is $O((\varepsilon n)^{-1/d})$ for
$d\geq 2$, and is $O(\log^2(\varepsilon n)(\varepsilon n)^{-1})$ for $d=1$. The
accuracy guarantee is optimal up to a constant factor for $d\geq 2$, and up to
a logarithmic factor for $d=1$. Our algorithm has a fast running time of
$O(\varepsilon dn)$ for all $d\geq 1$ and demonstrates improved accuracy
compared to the method in (Boedihardjo et al., 2022) for $d\geq 2$.
| no_new_dataset | 0.707815 |
2302.10182 | Stefan Gaugel | Stefan Gaugel, Manfred Reichert | PrecTime: A Deep Learning Architecture for Precise Time Series
Segmentation in Industrial Manufacturing Operations | Preprint | Engineering Applications of Artificial Intelligence, Volume 122,
2023 | 10.1016/j.engappai.2023.106078 | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | The fourth industrial revolution creates ubiquitous sensor data in production
plants. To generate maximum value out of these data, reliable and precise time
series-based machine learning methods like temporal neural networks are needed.
This paper proposes a novel sequence-to-sequence deep learning architecture for
time series segmentation called PrecTime which tries to combine the concepts
and advantages of sliding window and dense labeling approaches. The
general-purpose architecture is evaluated on a real-world industry dataset
containing the End-of-Line testing sensor data of hydraulic pumps. We are able
to show that PrecTime outperforms five implemented state-of-the-art baseline
networks based on multiple metrics. The achieved segmentation accuracy of
around 96% shows that PrecTime can achieve results close to human intelligence
in operational state segmentation within a testing cycle.
| [
{
"version": "v1",
"created": "Fri, 27 Jan 2023 12:47:27 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Gaugel",
"Stefan",
""
],
[
"Reichert",
"Manfred",
""
]
] | TITLE: PrecTime: A Deep Learning Architecture for Precise Time Series
Segmentation in Industrial Manufacturing Operations
ABSTRACT: The fourth industrial revolution creates ubiquitous sensor data in production
plants. To generate maximum value out of these data, reliable and precise time
series-based machine learning methods like temporal neural networks are needed.
This paper proposes a novel sequence-to-sequence deep learning architecture for
time series segmentation called PrecTime which tries to combine the concepts
and advantages of sliding window and dense labeling approaches. The
general-purpose architecture is evaluated on a real-world industry dataset
containing the End-of-Line testing sensor data of hydraulic pumps. We are able
to show that PrecTime outperforms five implemented state-of-the-art baseline
networks based on multiple metrics. The achieved segmentation accuracy of
around 96% shows that PrecTime can achieve results close to human intelligence
in operational state segmentation within a testing cycle.
| no_new_dataset | 0.702083 |
2303.01777 | Satoshi Tsutsui | Satoshi Tsutsui, Zhengyang Su, Bihan Wen | Benchmarking White Blood Cell Classification Under Domain Shift | Accepted to the International Conference on Acoustics, Speech, and
Signal Processing (ICASSP) 2023. More datasets are cited | null | null | null | eess.IV cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recognizing the types of white blood cells (WBCs) in microscopic images of
human blood smears is a fundamental task in the fields of pathology and
hematology. Although previous studies have made significant contributions to
the development of methods and datasets, few papers have investigated
benchmarks or baselines that others can easily refer to. For instance, we
observed notable variations in the reported accuracies of the same
Convolutional Neural Network (CNN) model across different studies, yet no
public implementation exists to reproduce these results. In this paper, we
establish a benchmark for WBC recognition. Our results indicate that CNN-based
models achieve high accuracy when trained and tested under similar imaging
conditions. However, their performance drops significantly when tested under
different conditions. Moreover, the ResNet classifier, which has been widely
employed in previous work, exhibits an unreasonably poor generalization ability
under domain shifts due to batch normalization. We investigate this issue and
suggest some alternative normalization techniques that can mitigate it. We make
fully-reproducible code publicly
available\footnote{\url{https://github.com/apple2373/wbc-benchmark}}.
| [
{
"version": "v1",
"created": "Fri, 3 Mar 2023 08:36:19 GMT"
},
{
"version": "v2",
"created": "Fri, 19 May 2023 17:52:34 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Tsutsui",
"Satoshi",
""
],
[
"Su",
"Zhengyang",
""
],
[
"Wen",
"Bihan",
""
]
] | TITLE: Benchmarking White Blood Cell Classification Under Domain Shift
ABSTRACT: Recognizing the types of white blood cells (WBCs) in microscopic images of
human blood smears is a fundamental task in the fields of pathology and
hematology. Although previous studies have made significant contributions to
the development of methods and datasets, few papers have investigated
benchmarks or baselines that others can easily refer to. For instance, we
observed notable variations in the reported accuracies of the same
Convolutional Neural Network (CNN) model across different studies, yet no
public implementation exists to reproduce these results. In this paper, we
establish a benchmark for WBC recognition. Our results indicate that CNN-based
models achieve high accuracy when trained and tested under similar imaging
conditions. However, their performance drops significantly when tested under
different conditions. Moreover, the ResNet classifier, which has been widely
employed in previous work, exhibits an unreasonably poor generalization ability
under domain shifts due to batch normalization. We investigate this issue and
suggest some alternative normalization techniques that can mitigate it. We make
fully-reproducible code publicly
available\footnote{\url{https://github.com/apple2373/wbc-benchmark}}.
| no_new_dataset | 0.697802 |
2304.11954 | Chenlin Zhou | Chenlin Zhou, Liutao Yu, Zhaokun Zhou, Zhengyu Ma, Han Zhang, Huihui
Zhou, Yonghong Tian | Spikingformer: Spike-driven Residual Learning for Transformer-based
Spiking Neural Network | 16 pages | null | null | null | cs.NE cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Spiking neural networks (SNNs) offer a promising energy-efficient alternative
to artificial neural networks, due to their event-driven spiking computation.
However, state-of-the-art deep SNNs (including Spikformer and SEW ResNet)
suffer from non-spike computations (integer-float multiplications) caused by
the structure of their residual connection. These non-spike computations
increase SNNs' power consumption and make them unsuitable for deployment on
mainstream neuromorphic hardware, which only supports spike operations. In this
paper, we propose a hardware-friendly spike-driven residual learning
architecture for SNNs to avoid non-spike computations. Based on this residual
design, we develop Spikingformer, a pure transformer-based spiking neural
network. We evaluate Spikingformer on ImageNet, CIFAR10, CIFAR100, CIFAR10-DVS
and DVS128 Gesture datasets, and demonstrate that Spikingformer outperforms the
state-of-the-art in directly trained pure SNNs as a novel advanced backbone
(75.85$\%$ top-1 accuracy on ImageNet, + 1.04$\%$ compared with Spikformer).
Furthermore, our experiments verify that Spikingformer effectively avoids
non-spike computations and significantly reduces energy consumption by
57.34$\%$ compared with Spikformer on ImageNet. To our best knowledge, this is
the first time that a pure event-driven transformer-based SNN has been
developed.
| [
{
"version": "v1",
"created": "Mon, 24 Apr 2023 09:44:24 GMT"
},
{
"version": "v2",
"created": "Tue, 9 May 2023 06:34:40 GMT"
},
{
"version": "v3",
"created": "Fri, 19 May 2023 07:37:37 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Zhou",
"Chenlin",
""
],
[
"Yu",
"Liutao",
""
],
[
"Zhou",
"Zhaokun",
""
],
[
"Ma",
"Zhengyu",
""
],
[
"Zhang",
"Han",
""
],
[
"Zhou",
"Huihui",
""
],
[
"Tian",
"Yonghong",
""
]
] | TITLE: Spikingformer: Spike-driven Residual Learning for Transformer-based
Spiking Neural Network
ABSTRACT: Spiking neural networks (SNNs) offer a promising energy-efficient alternative
to artificial neural networks, due to their event-driven spiking computation.
However, state-of-the-art deep SNNs (including Spikformer and SEW ResNet)
suffer from non-spike computations (integer-float multiplications) caused by
the structure of their residual connection. These non-spike computations
increase SNNs' power consumption and make them unsuitable for deployment on
mainstream neuromorphic hardware, which only supports spike operations. In this
paper, we propose a hardware-friendly spike-driven residual learning
architecture for SNNs to avoid non-spike computations. Based on this residual
design, we develop Spikingformer, a pure transformer-based spiking neural
network. We evaluate Spikingformer on ImageNet, CIFAR10, CIFAR100, CIFAR10-DVS
and DVS128 Gesture datasets, and demonstrate that Spikingformer outperforms the
state-of-the-art in directly trained pure SNNs as a novel advanced backbone
(75.85$\%$ top-1 accuracy on ImageNet, + 1.04$\%$ compared with Spikformer).
Furthermore, our experiments verify that Spikingformer effectively avoids
non-spike computations and significantly reduces energy consumption by
57.34$\%$ compared with Spikformer on ImageNet. To our best knowledge, this is
the first time that a pure event-driven transformer-based SNN has been
developed.
| no_new_dataset | 0.70883 |
2305.05954 | Chenlin Zhou | Chenlin Zhou, Han Zhang, Zhaokun Zhou, Liutao Yu, Zhengyu Ma, Huihui
Zhou, Xiaopeng Fan, Yonghong Tian | Enhancing the Performance of Transformer-based Spiking Neural Networks
by SNN-optimized Downsampling with Precise Gradient Backpropagation | 12 pages | null | null | null | cs.NE cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Deep spiking neural networks (SNNs) have drawn much attention in recent years
because of their low power consumption, biological rationality and event-driven
property. However, state-of-the-art deep SNNs (including Spikformer and
Spikingformer) suffer from a critical challenge related to the imprecise
gradient backpropagation. This problem arises from the improper design of
downsampling modules in these networks, and greatly hampering the overall model
performance. In this paper, we propose ConvBN-MaxPooling-LIF (CML), an
SNN-optimized downsampling with precise gradient backpropagation. We prove that
CML can effectively overcome the imprecision of gradient backpropagation from a
theoretical perspective. In addition, we evaluate CML on ImageNet, CIFAR10,
CIFAR100, CIFAR10-DVS, DVS128-Gesture datasets, and show state-of-the-art
performance on all these datasets with significantly enhanced performances
compared with Spikingformer. For instance, our model achieves 77.64 $\%$ on
ImageNet, 96.04 $\%$ on CIFAR10, 81.4$\%$ on CIFAR10-DVS, with + 1.79$\%$ on
ImageNet, +1.16$\%$ on CIFAR100 compared with Spikingformer.
| [
{
"version": "v1",
"created": "Wed, 10 May 2023 07:48:08 GMT"
},
{
"version": "v2",
"created": "Tue, 16 May 2023 07:13:50 GMT"
},
{
"version": "v3",
"created": "Fri, 19 May 2023 07:50:16 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Zhou",
"Chenlin",
""
],
[
"Zhang",
"Han",
""
],
[
"Zhou",
"Zhaokun",
""
],
[
"Yu",
"Liutao",
""
],
[
"Ma",
"Zhengyu",
""
],
[
"Zhou",
"Huihui",
""
],
[
"Fan",
"Xiaopeng",
""
],
[
"Tian",
"Yonghong",
""
]
] | TITLE: Enhancing the Performance of Transformer-based Spiking Neural Networks
by SNN-optimized Downsampling with Precise Gradient Backpropagation
ABSTRACT: Deep spiking neural networks (SNNs) have drawn much attention in recent years
because of their low power consumption, biological rationality and event-driven
property. However, state-of-the-art deep SNNs (including Spikformer and
Spikingformer) suffer from a critical challenge related to the imprecise
gradient backpropagation. This problem arises from the improper design of
downsampling modules in these networks, and greatly hampering the overall model
performance. In this paper, we propose ConvBN-MaxPooling-LIF (CML), an
SNN-optimized downsampling with precise gradient backpropagation. We prove that
CML can effectively overcome the imprecision of gradient backpropagation from a
theoretical perspective. In addition, we evaluate CML on ImageNet, CIFAR10,
CIFAR100, CIFAR10-DVS, DVS128-Gesture datasets, and show state-of-the-art
performance on all these datasets with significantly enhanced performances
compared with Spikingformer. For instance, our model achieves 77.64 $\%$ on
ImageNet, 96.04 $\%$ on CIFAR10, 81.4$\%$ on CIFAR10-DVS, with + 1.79$\%$ on
ImageNet, +1.16$\%$ on CIFAR100 compared with Spikingformer.
| no_new_dataset | 0.710051 |
2305.07239 | Ye Deng | Ye Deng, Siqi Hui, Sanping Zhou, Deyu Meng, Jinjun Wang | T-former: An Efficient Transformer for Image Inpainting | null | ACM Multimedia 2022 | 10.1145/3503161.3548446 | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Benefiting from powerful convolutional neural networks (CNNs), learning-based
image inpainting methods have made significant breakthroughs over the years.
However, some nature of CNNs (e.g. local prior, spatially shared parameters)
limit the performance in the face of broken images with diverse and complex
forms. Recently, a class of attention-based network architectures, called
transformer, has shown significant performance on natural language processing
fields and high-level vision tasks. Compared with CNNs, attention operators are
better at long-range modeling and have dynamic weights, but their computational
complexity is quadratic in spatial resolution, and thus less suitable for
applications involving higher resolution images, such as image inpainting. In
this paper, we design a novel attention linearly related to the resolution
according to Taylor expansion. And based on this attention, a network called
$T$-former is designed for image inpainting. Experiments on several benchmark
datasets demonstrate that our proposed method achieves state-of-the-art
accuracy while maintaining a relatively low number of parameters and
computational complexity. The code can be found at
\href{https://github.com/dengyecode/T-former_image_inpainting}{github.com/dengyecode/T-former\_image\_inpainting}
| [
{
"version": "v1",
"created": "Fri, 12 May 2023 04:10:42 GMT"
},
{
"version": "v2",
"created": "Fri, 19 May 2023 02:11:54 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Deng",
"Ye",
""
],
[
"Hui",
"Siqi",
""
],
[
"Zhou",
"Sanping",
""
],
[
"Meng",
"Deyu",
""
],
[
"Wang",
"Jinjun",
""
]
] | TITLE: T-former: An Efficient Transformer for Image Inpainting
ABSTRACT: Benefiting from powerful convolutional neural networks (CNNs), learning-based
image inpainting methods have made significant breakthroughs over the years.
However, some nature of CNNs (e.g. local prior, spatially shared parameters)
limit the performance in the face of broken images with diverse and complex
forms. Recently, a class of attention-based network architectures, called
transformer, has shown significant performance on natural language processing
fields and high-level vision tasks. Compared with CNNs, attention operators are
better at long-range modeling and have dynamic weights, but their computational
complexity is quadratic in spatial resolution, and thus less suitable for
applications involving higher resolution images, such as image inpainting. In
this paper, we design a novel attention linearly related to the resolution
according to Taylor expansion. And based on this attention, a network called
$T$-former is designed for image inpainting. Experiments on several benchmark
datasets demonstrate that our proposed method achieves state-of-the-art
accuracy while maintaining a relatively low number of parameters and
computational complexity. The code can be found at
\href{https://github.com/dengyecode/T-former_image_inpainting}{github.com/dengyecode/T-former\_image\_inpainting}
| no_new_dataset | 0.710434 |
2305.07889 | Jiaji Wang | Chawit Kaewnuratchadasorn, Jiaji Wang, Chul-Woo Kim | Neural operator for structural simulation and bridge health monitoring | 20 pages, 10 figures, uses NJDapacite.sty | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Infusing deep learning with structural engineering has received widespread
attention for both forward problems (structural simulation) and inverse
problems (structural health monitoring). Based on Fourier Neural Operator, this
study proposes VINO (Vehicle-bridge Interaction Neural Operator) to serve as
the digital twin of bridge structures. VINO learns mappings between structural
response fields and damage fields. In this study, VBI-FE dataset was
established by running parametric finite element (FE) simulations considering a
random distribution of structural initial damage field. Subsequently, VBI-EXP
dataset was produced by conducting an experimental study under four damage
scenarios. After VINO was pre-trained by VBI-FE and fine-tuned by VBI-EXP from
the bridge at the healthy state, the model achieved the following two
improvements. First, forward VINO can predict structural responses from damage
field inputs more accurately than the FE model. Second, inverse VINO can
determine, localize, and quantify damages in all scenarios, suggesting the
practicality of data-driven approaches.
| [
{
"version": "v1",
"created": "Sat, 13 May 2023 10:27:08 GMT"
},
{
"version": "v2",
"created": "Fri, 19 May 2023 05:16:02 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Kaewnuratchadasorn",
"Chawit",
""
],
[
"Wang",
"Jiaji",
""
],
[
"Kim",
"Chul-Woo",
""
]
] | TITLE: Neural operator for structural simulation and bridge health monitoring
ABSTRACT: Infusing deep learning with structural engineering has received widespread
attention for both forward problems (structural simulation) and inverse
problems (structural health monitoring). Based on Fourier Neural Operator, this
study proposes VINO (Vehicle-bridge Interaction Neural Operator) to serve as
the digital twin of bridge structures. VINO learns mappings between structural
response fields and damage fields. In this study, VBI-FE dataset was
established by running parametric finite element (FE) simulations considering a
random distribution of structural initial damage field. Subsequently, VBI-EXP
dataset was produced by conducting an experimental study under four damage
scenarios. After VINO was pre-trained by VBI-FE and fine-tuned by VBI-EXP from
the bridge at the healthy state, the model achieved the following two
improvements. First, forward VINO can predict structural responses from damage
field inputs more accurately than the FE model. Second, inverse VINO can
determine, localize, and quantify damages in all scenarios, suggesting the
practicality of data-driven approaches.
| no_new_dataset | 0.708603 |
2305.08285 | Yunqi Zhu | Yunqi Zhu and Xuebing Yang and Yuanyuan Wu and Wensheng Zhang | Parameter-Efficient Fine-Tuning with Layer Pruning on Free-Text
Sequence-to-Sequence Modeling | null | null | null | null | cs.CL cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The increasing size of language models raises great research interests in
parameter-efficient fine-tuning such as LoRA that freezes the pre-trained
model, and injects small-scale trainable parameters for multiple downstream
tasks (e.g., summarization, question answering and translation). To further
enhance the efficiency of fine-tuning, we propose a framework that integrates
LoRA and structured layer pruning. The integrated framework is validated on two
created deidentified medical report summarization datasets based on
MIMIC-IV-Note and two public medical dialogue datasets. By tuning 0.6%
parameters of the original model and pruning over 30% Transformer-layers, our
framework can reduce 50% of GPU memory usage and speed up 100% of the training
phase, while preserving over 92% generation qualities on free-text
sequence-to-sequence tasks.
| [
{
"version": "v1",
"created": "Mon, 15 May 2023 00:21:08 GMT"
},
{
"version": "v2",
"created": "Thu, 18 May 2023 13:45:01 GMT"
},
{
"version": "v3",
"created": "Fri, 19 May 2023 01:29:08 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Zhu",
"Yunqi",
""
],
[
"Yang",
"Xuebing",
""
],
[
"Wu",
"Yuanyuan",
""
],
[
"Zhang",
"Wensheng",
""
]
] | TITLE: Parameter-Efficient Fine-Tuning with Layer Pruning on Free-Text
Sequence-to-Sequence Modeling
ABSTRACT: The increasing size of language models raises great research interests in
parameter-efficient fine-tuning such as LoRA that freezes the pre-trained
model, and injects small-scale trainable parameters for multiple downstream
tasks (e.g., summarization, question answering and translation). To further
enhance the efficiency of fine-tuning, we propose a framework that integrates
LoRA and structured layer pruning. The integrated framework is validated on two
created deidentified medical report summarization datasets based on
MIMIC-IV-Note and two public medical dialogue datasets. By tuning 0.6%
parameters of the original model and pruning over 30% Transformer-layers, our
framework can reduce 50% of GPU memory usage and speed up 100% of the training
phase, while preserving over 92% generation qualities on free-text
sequence-to-sequence tasks.
| no_new_dataset | 0.707165 |
2305.08592 | Yu Pei | Yu Pei (1), Jeongju Sohn (1), Sarra Habchi (2), Mike Papadakis (1)
((1) University of Luxembourg, (2) Ubisoft) | Time-based Repair for Asynchronous Wait Flaky Tests in Web Testing | null | null | null | null | cs.SE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Asynchronous waits are one of the most prevalent root causes of flaky tests
and a major time-influential factor of web application testing. To investigate
the characteristics of asynchronous wait flaky tests and their fixes in web
testing, we build a dataset of 49 reproducible flaky tests, from 26 open-source
projects, caused by asynchronous waits, along with their corresponding
developer-written fixes. Our study of these flaky tests reveals that in
approximately 63% of them (31 out of 49), developers addressed Asynchronous
Wait flaky tests by adapting the wait time, even for cases where the root
causes lie elsewhere. Based on this finding, we propose TRaf, an automated
time-based repair method for asynchronous wait flaky tests in web applications.
TRaf tackles the flakiness issues by suggesting a proper waiting time for each
asynchronous call in a web application, using code similarity and past change
history. The core insight is that as developers often make similar mistakes
more than once, hints for the efficient wait time exist in the current or past
codebase. Our analysis shows that TRaf can suggest a shorter wait time to
resolve the test flakiness compared to developer-written fixes, reducing the
test execution time by 11.1%. With additional dynamic tuning of the new wait
time, TRaf further reduces the execution time by 20.2%.
| [
{
"version": "v1",
"created": "Mon, 15 May 2023 12:17:30 GMT"
},
{
"version": "v2",
"created": "Fri, 19 May 2023 17:04:51 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Pei",
"Yu",
"",
"University of Luxembourg"
],
[
"Sohn",
"Jeongju",
"",
"University of Luxembourg"
],
[
"Habchi",
"Sarra",
"",
"Ubisoft"
],
[
"Papadakis",
"Mike",
"",
"University of Luxembourg"
]
] | TITLE: Time-based Repair for Asynchronous Wait Flaky Tests in Web Testing
ABSTRACT: Asynchronous waits are one of the most prevalent root causes of flaky tests
and a major time-influential factor of web application testing. To investigate
the characteristics of asynchronous wait flaky tests and their fixes in web
testing, we build a dataset of 49 reproducible flaky tests, from 26 open-source
projects, caused by asynchronous waits, along with their corresponding
developer-written fixes. Our study of these flaky tests reveals that in
approximately 63% of them (31 out of 49), developers addressed Asynchronous
Wait flaky tests by adapting the wait time, even for cases where the root
causes lie elsewhere. Based on this finding, we propose TRaf, an automated
time-based repair method for asynchronous wait flaky tests in web applications.
TRaf tackles the flakiness issues by suggesting a proper waiting time for each
asynchronous call in a web application, using code similarity and past change
history. The core insight is that as developers often make similar mistakes
more than once, hints for the efficient wait time exist in the current or past
codebase. Our analysis shows that TRaf can suggest a shorter wait time to
resolve the test flakiness compared to developer-written fixes, reducing the
test execution time by 11.1%. With additional dynamic tuning of the new wait
time, TRaf further reduces the execution time by 20.2%.
| new_dataset | 0.710672 |
2305.09022 | Eric Yuan | Arjun Subramonian, Xingdi Yuan, Hal Daum\'e III, Su Lin Blodgett | It Takes Two to Tango: Navigating Conceptualizations of NLP Tasks and
Measurements of Performance | null | Findings of the Association for Computational Linguistics: ACL
2023 | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Progress in NLP is increasingly measured through benchmarks; hence,
contextualizing progress requires understanding when and why practitioners may
disagree about the validity of benchmarks. We develop a taxonomy of
disagreement, drawing on tools from measurement modeling, and distinguish
between two types of disagreement: 1) how tasks are conceptualized and 2) how
measurements of model performance are operationalized. To provide evidence for
our taxonomy, we conduct a meta-analysis of relevant literature to understand
how NLP tasks are conceptualized, as well as a survey of practitioners about
their impressions of different factors that affect benchmark validity. Our
meta-analysis and survey across eight tasks, ranging from coreference
resolution to question answering, uncover that tasks are generally not clearly
and consistently conceptualized and benchmarks suffer from operationalization
disagreements. These findings support our proposed taxonomy of disagreement.
Finally, based on our taxonomy, we present a framework for constructing
benchmarks and documenting their limitations.
| [
{
"version": "v1",
"created": "Mon, 15 May 2023 21:12:07 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Subramonian",
"Arjun",
""
],
[
"Yuan",
"Xingdi",
""
],
[
"Daumé",
"Hal",
"III"
],
[
"Blodgett",
"Su Lin",
""
]
] | TITLE: It Takes Two to Tango: Navigating Conceptualizations of NLP Tasks and
Measurements of Performance
ABSTRACT: Progress in NLP is increasingly measured through benchmarks; hence,
contextualizing progress requires understanding when and why practitioners may
disagree about the validity of benchmarks. We develop a taxonomy of
disagreement, drawing on tools from measurement modeling, and distinguish
between two types of disagreement: 1) how tasks are conceptualized and 2) how
measurements of model performance are operationalized. To provide evidence for
our taxonomy, we conduct a meta-analysis of relevant literature to understand
how NLP tasks are conceptualized, as well as a survey of practitioners about
their impressions of different factors that affect benchmark validity. Our
meta-analysis and survey across eight tasks, ranging from coreference
resolution to question answering, uncover that tasks are generally not clearly
and consistently conceptualized and benchmarks suffer from operationalization
disagreements. These findings support our proposed taxonomy of disagreement.
Finally, based on our taxonomy, we present a framework for constructing
benchmarks and documenting their limitations.
| no_new_dataset | 0.70978 |
2305.09062 | Gilberto Ochoa-Ruiz | Mauricio Mendez-Ruiz, Jorge Gonzalez-Zapata, Ivan Reyes-Amezcua,
Daniel Flores-Araiza, Francisco Lopez-Tiro, Andres Mendez-Vazquez, Gilberto
Ochoa-Ruiz | SuSana Distancia is all you need: Enforcing class separability in metric
learning via two novel distance-based loss functions for few-shot image
classification | Paper submitted to a journal for publication | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by/4.0/ | Few-shot learning is a challenging area of research that aims to learn new
concepts with only a few labeled samples of data. Recent works based on
metric-learning approaches leverage the meta-learning approach, which is
encompassed by episodic tasks that make use a support (training) and query set
(test) with the objective of learning a similarity comparison metric between
those sets. Due to the lack of data, the learning process of the embedding
network becomes an important part of the few-shot task. Previous works have
addressed this problem using metric learning approaches, but the properties of
the underlying latent space and the separability of the difference classes on
it was not entirely enforced. In this work, we propose two different loss
functions which consider the importance of the embedding vectors by looking at
the intra-class and inter-class distance between the few data. The first loss
function is the Proto-Triplet Loss, which is based on the original triplet loss
with the modifications needed to better work on few-shot scenarios. The second
loss function, which we dub ICNN loss is based on an inter and intra class
nearest neighbors score, which help us to assess the quality of embeddings
obtained from the trained network. Our results, obtained from a extensive
experimental setup show a significant improvement in accuracy in the
miniImagenNet benchmark compared to other metric-based few-shot learning
methods by a margin of 2%, demonstrating the capability of these loss functions
to allow the network to generalize better to previously unseen classes. In our
experiments, we demonstrate competitive generalization capabilities to other
domains, such as the Caltech CUB, Dogs and Cars datasets compared with the
state of the art.
| [
{
"version": "v1",
"created": "Mon, 15 May 2023 23:12:09 GMT"
},
{
"version": "v2",
"created": "Wed, 17 May 2023 00:58:41 GMT"
},
{
"version": "v3",
"created": "Thu, 18 May 2023 20:41:34 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Mendez-Ruiz",
"Mauricio",
""
],
[
"Gonzalez-Zapata",
"Jorge",
""
],
[
"Reyes-Amezcua",
"Ivan",
""
],
[
"Flores-Araiza",
"Daniel",
""
],
[
"Lopez-Tiro",
"Francisco",
""
],
[
"Mendez-Vazquez",
"Andres",
""
],
[
"Ochoa-Ruiz",
"Gilberto",
""
]
] | TITLE: SuSana Distancia is all you need: Enforcing class separability in metric
learning via two novel distance-based loss functions for few-shot image
classification
ABSTRACT: Few-shot learning is a challenging area of research that aims to learn new
concepts with only a few labeled samples of data. Recent works based on
metric-learning approaches leverage the meta-learning approach, which is
encompassed by episodic tasks that make use a support (training) and query set
(test) with the objective of learning a similarity comparison metric between
those sets. Due to the lack of data, the learning process of the embedding
network becomes an important part of the few-shot task. Previous works have
addressed this problem using metric learning approaches, but the properties of
the underlying latent space and the separability of the difference classes on
it was not entirely enforced. In this work, we propose two different loss
functions which consider the importance of the embedding vectors by looking at
the intra-class and inter-class distance between the few data. The first loss
function is the Proto-Triplet Loss, which is based on the original triplet loss
with the modifications needed to better work on few-shot scenarios. The second
loss function, which we dub ICNN loss is based on an inter and intra class
nearest neighbors score, which help us to assess the quality of embeddings
obtained from the trained network. Our results, obtained from a extensive
experimental setup show a significant improvement in accuracy in the
miniImagenNet benchmark compared to other metric-based few-shot learning
methods by a margin of 2%, demonstrating the capability of these loss functions
to allow the network to generalize better to previously unseen classes. In our
experiments, we demonstrate competitive generalization capabilities to other
domains, such as the Caltech CUB, Dogs and Cars datasets compared with the
state of the art.
| no_new_dataset | 0.709868 |
2305.09193 | Chang Gao | Chang Gao, Wenxuan Zhang, Wai Lam, Lidong Bing | Easy-to-Hard Learning for Information Extraction | Findings of ACL 2023 | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Information extraction (IE) systems aim to automatically extract structured
information, such as named entities, relations between entities, and events,
from unstructured texts. While most existing work addresses a particular IE
task, universally modeling various IE tasks with one model has achieved great
success recently. Despite their success, they employ a one-stage learning
strategy, i.e., directly learning to extract the target structure given the
input text, which contradicts the human learning process. In this paper, we
propose a unified easy-to-hard learning framework consisting of three stages,
i.e., the easy stage, the hard stage, and the main stage, for IE by mimicking
the human learning process. By breaking down the learning process into multiple
stages, our framework facilitates the model to acquire general IE task
knowledge and improve its generalization ability. Extensive experiments across
four IE tasks demonstrate the effectiveness of our framework. We achieve new
state-of-the-art results on 13 out of 17 datasets. Our code is available at
\url{https://github.com/DAMO-NLP-SG/IE-E2H}.
| [
{
"version": "v1",
"created": "Tue, 16 May 2023 06:04:14 GMT"
},
{
"version": "v2",
"created": "Fri, 19 May 2023 11:27:02 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Gao",
"Chang",
""
],
[
"Zhang",
"Wenxuan",
""
],
[
"Lam",
"Wai",
""
],
[
"Bing",
"Lidong",
""
]
] | TITLE: Easy-to-Hard Learning for Information Extraction
ABSTRACT: Information extraction (IE) systems aim to automatically extract structured
information, such as named entities, relations between entities, and events,
from unstructured texts. While most existing work addresses a particular IE
task, universally modeling various IE tasks with one model has achieved great
success recently. Despite their success, they employ a one-stage learning
strategy, i.e., directly learning to extract the target structure given the
input text, which contradicts the human learning process. In this paper, we
propose a unified easy-to-hard learning framework consisting of three stages,
i.e., the easy stage, the hard stage, and the main stage, for IE by mimicking
the human learning process. By breaking down the learning process into multiple
stages, our framework facilitates the model to acquire general IE task
knowledge and improve its generalization ability. Extensive experiments across
four IE tasks demonstrate the effectiveness of our framework. We achieve new
state-of-the-art results on 13 out of 17 datasets. Our code is available at
\url{https://github.com/DAMO-NLP-SG/IE-E2H}.
| no_new_dataset | 0.707569 |
2305.09527 | Dominik Muhle | Dominik Muhle, Lukas Koestler, Krishna Murthy Jatavallabhula, Daniel
Cremers | Learning Correspondence Uncertainty via Differentiable Nonlinear Least
Squares | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We propose a differentiable nonlinear least squares framework to account for
uncertainty in relative pose estimation from feature correspondences.
Specifically, we introduce a symmetric version of the probabilistic normal
epipolar constraint, and an approach to estimate the covariance of feature
positions by differentiating through the camera pose estimation procedure. We
evaluate our approach on synthetic, as well as the KITTI and EuRoC real-world
datasets. On the synthetic dataset, we confirm that our learned covariances
accurately approximate the true noise distribution. In real world experiments,
we find that our approach consistently outperforms state-of-the-art
non-probabilistic and probabilistic approaches, regardless of the feature
extraction algorithm of choice.
| [
{
"version": "v1",
"created": "Tue, 16 May 2023 15:21:09 GMT"
},
{
"version": "v2",
"created": "Thu, 18 May 2023 18:35:23 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Muhle",
"Dominik",
""
],
[
"Koestler",
"Lukas",
""
],
[
"Jatavallabhula",
"Krishna Murthy",
""
],
[
"Cremers",
"Daniel",
""
]
] | TITLE: Learning Correspondence Uncertainty via Differentiable Nonlinear Least
Squares
ABSTRACT: We propose a differentiable nonlinear least squares framework to account for
uncertainty in relative pose estimation from feature correspondences.
Specifically, we introduce a symmetric version of the probabilistic normal
epipolar constraint, and an approach to estimate the covariance of feature
positions by differentiating through the camera pose estimation procedure. We
evaluate our approach on synthetic, as well as the KITTI and EuRoC real-world
datasets. On the synthetic dataset, we confirm that our learned covariances
accurately approximate the true noise distribution. In real world experiments,
we find that our approach consistently outperforms state-of-the-art
non-probabilistic and probabilistic approaches, regardless of the feature
extraction algorithm of choice.
| no_new_dataset | 0.710465 |
2305.10246 | Linghao Feng | Linghao Feng, Dongcheng Zhao, Yi Zeng | Spiking Generative Adversarial Network with Attention Scoring Decoding | null | null | null | null | cs.NE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Generative models based on neural networks present a substantial challenge
within deep learning. As it stands, such models are primarily limited to the
domain of artificial neural networks. Spiking neural networks, as the third
generation of neural networks, offer a closer approximation to brain-like
processing due to their rich spatiotemporal dynamics. However, generative
models based on spiking neural networks are not well studied. In this work, we
pioneer constructing a spiking generative adversarial network capable of
handling complex images. Our first task was to identify the problems of
out-of-domain inconsistency and temporal inconsistency inherent in spiking
generative adversarial networks. We addressed these issues by incorporating the
Earth-Mover distance and an attention-based weighted decoding method,
significantly enhancing the performance of our algorithm across several
datasets. Experimental results reveal that our approach outperforms existing
methods on the MNIST, FashionMNIST, CIFAR10, and CelebA datasets. Moreover,
compared with hybrid spiking generative adversarial networks, where the
discriminator is an artificial analog neural network, our methodology
demonstrates closer alignment with the information processing patterns observed
in the mouse.
| [
{
"version": "v1",
"created": "Wed, 17 May 2023 14:35:45 GMT"
},
{
"version": "v2",
"created": "Thu, 18 May 2023 12:01:44 GMT"
},
{
"version": "v3",
"created": "Fri, 19 May 2023 08:08:42 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Feng",
"Linghao",
""
],
[
"Zhao",
"Dongcheng",
""
],
[
"Zeng",
"Yi",
""
]
] | TITLE: Spiking Generative Adversarial Network with Attention Scoring Decoding
ABSTRACT: Generative models based on neural networks present a substantial challenge
within deep learning. As it stands, such models are primarily limited to the
domain of artificial neural networks. Spiking neural networks, as the third
generation of neural networks, offer a closer approximation to brain-like
processing due to their rich spatiotemporal dynamics. However, generative
models based on spiking neural networks are not well studied. In this work, we
pioneer constructing a spiking generative adversarial network capable of
handling complex images. Our first task was to identify the problems of
out-of-domain inconsistency and temporal inconsistency inherent in spiking
generative adversarial networks. We addressed these issues by incorporating the
Earth-Mover distance and an attention-based weighted decoding method,
significantly enhancing the performance of our algorithm across several
datasets. Experimental results reveal that our approach outperforms existing
methods on the MNIST, FashionMNIST, CIFAR10, and CelebA datasets. Moreover,
compared with hybrid spiking generative adversarial networks, where the
discriminator is an artificial analog neural network, our methodology
demonstrates closer alignment with the information processing patterns observed
in the mouse.
| no_new_dataset | 0.707815 |
2305.10869 | Nimesh Agrawal | Nimesh Agrawal, Nikita Malik, Sandeep Kumar | Free Lunch for Privacy Preserving Distributed Graph Learning | null | null | null | null | cs.LG cs.CR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Learning on graphs is becoming prevalent in a wide range of applications
including social networks, robotics, communication, medicine, etc. These
datasets belonging to entities often contain critical private information. The
utilization of data for graph learning applications is hampered by the growing
privacy concerns from users on data sharing. Existing privacy-preserving
methods pre-process the data to extract user-side features, and only these
features are used for subsequent learning. Unfortunately, these methods are
vulnerable to adversarial attacks to infer private attributes. We present a
novel privacy-respecting framework for distributed graph learning and
graph-based machine learning. In order to perform graph learning and other
downstream tasks on the server side, this framework aims to learn features as
well as distances without requiring actual features while preserving the
original structural properties of the raw data. The proposed framework is quite
generic and highly adaptable. We demonstrate the utility of the Euclidean
space, but it can be applied with any existing method of distance approximation
and graph learning for the relevant spaces. Through extensive experimentation
on both synthetic and real datasets, we demonstrate the efficacy of the
framework in terms of comparing the results obtained without data sharing to
those obtained with data sharing as a benchmark. This is, to our knowledge, the
first privacy-preserving distributed graph learning framework.
| [
{
"version": "v1",
"created": "Thu, 18 May 2023 10:41:21 GMT"
},
{
"version": "v2",
"created": "Fri, 19 May 2023 05:06:44 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Agrawal",
"Nimesh",
""
],
[
"Malik",
"Nikita",
""
],
[
"Kumar",
"Sandeep",
""
]
] | TITLE: Free Lunch for Privacy Preserving Distributed Graph Learning
ABSTRACT: Learning on graphs is becoming prevalent in a wide range of applications
including social networks, robotics, communication, medicine, etc. These
datasets belonging to entities often contain critical private information. The
utilization of data for graph learning applications is hampered by the growing
privacy concerns from users on data sharing. Existing privacy-preserving
methods pre-process the data to extract user-side features, and only these
features are used for subsequent learning. Unfortunately, these methods are
vulnerable to adversarial attacks to infer private attributes. We present a
novel privacy-respecting framework for distributed graph learning and
graph-based machine learning. In order to perform graph learning and other
downstream tasks on the server side, this framework aims to learn features as
well as distances without requiring actual features while preserving the
original structural properties of the raw data. The proposed framework is quite
generic and highly adaptable. We demonstrate the utility of the Euclidean
space, but it can be applied with any existing method of distance approximation
and graph learning for the relevant spaces. Through extensive experimentation
on both synthetic and real datasets, we demonstrate the efficacy of the
framework in terms of comparing the results obtained without data sharing to
those obtained with data sharing as a benchmark. This is, to our knowledge, the
first privacy-preserving distributed graph learning framework.
| no_new_dataset | 0.708648 |
2305.11000 | Dong Zhang Zhang | Dong Zhang, Shimin Li, Xin Zhang, Jun Zhan, Pengyu Wang, Yaqian Zhou,
Xipeng Qiu | SpeechGPT: Empowering Large Language Models with Intrinsic Cross-Modal
Conversational Abilities | work in progress | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Multi-modal large language models are regarded as a crucial step towards
Artificial General Intelligence (AGI) and have garnered significant interest
with the emergence of ChatGPT. However, current speech-language models
typically adopt the cascade paradigm, preventing inter-modal knowledge
transfer. In this paper, we propose SpeechGPT, a large language model with
intrinsic cross-modal conversational abilities, capable of perceiving and
generating multi-model content. With discrete speech representations, we first
construct SpeechInstruct, a large-scale cross-modal speech instruction dataset.
Additionally, we employ a three-stage training strategy that includes
modality-adaptation pre-training, cross-modal instruction fine-tuning, and
chain-of-modality instruction fine-tuning. The experimental results demonstrate
that SpeechGPT has an impressive capacity to follow multi-modal human
instructions and highlight the potential of handling multiple modalities with
one model. Demos are shown in https://0nutation.github.io/SpeechGPT.github.io/.
| [
{
"version": "v1",
"created": "Thu, 18 May 2023 14:23:25 GMT"
},
{
"version": "v2",
"created": "Fri, 19 May 2023 14:41:16 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Zhang",
"Dong",
""
],
[
"Li",
"Shimin",
""
],
[
"Zhang",
"Xin",
""
],
[
"Zhan",
"Jun",
""
],
[
"Wang",
"Pengyu",
""
],
[
"Zhou",
"Yaqian",
""
],
[
"Qiu",
"Xipeng",
""
]
] | TITLE: SpeechGPT: Empowering Large Language Models with Intrinsic Cross-Modal
Conversational Abilities
ABSTRACT: Multi-modal large language models are regarded as a crucial step towards
Artificial General Intelligence (AGI) and have garnered significant interest
with the emergence of ChatGPT. However, current speech-language models
typically adopt the cascade paradigm, preventing inter-modal knowledge
transfer. In this paper, we propose SpeechGPT, a large language model with
intrinsic cross-modal conversational abilities, capable of perceiving and
generating multi-model content. With discrete speech representations, we first
construct SpeechInstruct, a large-scale cross-modal speech instruction dataset.
Additionally, we employ a three-stage training strategy that includes
modality-adaptation pre-training, cross-modal instruction fine-tuning, and
chain-of-modality instruction fine-tuning. The experimental results demonstrate
that SpeechGPT has an impressive capacity to follow multi-modal human
instructions and highlight the potential of handling multiple modalities with
one model. Demos are shown in https://0nutation.github.io/SpeechGPT.github.io/.
| new_dataset | 0.712988 |
2305.11104 | Chenhao Shuai | Chenhao Shuai, Chaohua Shi, Lu Gan and Hongqing Liu | mdctGAN: Taming transformer-based GAN for speech super-resolution with
Modified DCT spectra | 5 pages, 4 figures, INTERSPEECH 2023 | null | null | null | eess.AS cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Speech super-resolution (SSR) aims to recover a high resolution (HR) speech
from its corresponding low resolution (LR) counterpart. Recent SSR methods
focus more on the reconstruction of the magnitude spectrogram, ignoring the
importance of phase reconstruction, thereby limiting the recovery quality. To
address this issue, we propose mdctGAN, a novel SSR framework based on modified
discrete cosine transform (MDCT). By adversarial learning in the MDCT domain,
our method reconstructs HR speeches in a phase-aware manner without vocoders or
additional post-processing. Furthermore, by learning frequency consistent
features with self-attentive mechanism, mdctGAN guarantees a high quality
speech reconstruction. For VCTK corpus dataset, the experiment results show
that our model produces natural auditory quality with high MOS and PESQ scores.
It also achieves the state-of-the-art log-spectral-distance (LSD) performance
on 48 kHz target resolution from various input rates. Code is available from
https://github.com/neoncloud/mdctGAN
| [
{
"version": "v1",
"created": "Thu, 18 May 2023 16:49:46 GMT"
},
{
"version": "v2",
"created": "Fri, 19 May 2023 07:26:43 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Shuai",
"Chenhao",
""
],
[
"Shi",
"Chaohua",
""
],
[
"Gan",
"Lu",
""
],
[
"Liu",
"Hongqing",
""
]
] | TITLE: mdctGAN: Taming transformer-based GAN for speech super-resolution with
Modified DCT spectra
ABSTRACT: Speech super-resolution (SSR) aims to recover a high resolution (HR) speech
from its corresponding low resolution (LR) counterpart. Recent SSR methods
focus more on the reconstruction of the magnitude spectrogram, ignoring the
importance of phase reconstruction, thereby limiting the recovery quality. To
address this issue, we propose mdctGAN, a novel SSR framework based on modified
discrete cosine transform (MDCT). By adversarial learning in the MDCT domain,
our method reconstructs HR speeches in a phase-aware manner without vocoders or
additional post-processing. Furthermore, by learning frequency consistent
features with self-attentive mechanism, mdctGAN guarantees a high quality
speech reconstruction. For VCTK corpus dataset, the experiment results show
that our model produces natural auditory quality with high MOS and PESQ scores.
It also achieves the state-of-the-art log-spectral-distance (LSD) performance
on 48 kHz target resolution from various input rates. Code is available from
https://github.com/neoncloud/mdctGAN
| no_new_dataset | 0.709655 |
2305.11178 | Georgios Leontidis | Miles Everett, Mingjun Zhong and Georgios Leontidis | Vanishing Activations: A Symptom of Deep Capsule Networks | 9 pages, 7 figures | null | null | null | cs.CV cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Capsule Networks, an extension to Neural Networks utilizing vector or matrix
representations instead of scalars, were initially developed to create a
dynamic parse tree where visual concepts evolve from parts to complete objects.
Early implementations of Capsule Networks achieved and maintain
state-of-the-art results on various datasets. However, recent studies have
revealed shortcomings in the original Capsule Network architecture, notably its
failure to construct a parse tree and its susceptibility to vanishing gradients
when deployed in deeper networks. This paper extends the investigation to a
range of leading Capsule Network architectures, demonstrating that these issues
are not confined to the original design. We argue that the majority of Capsule
Network research has produced architectures that, while modestly divergent from
the original Capsule Network, still retain a fundamentally similar structure.
We posit that this inherent design similarity might be impeding the scalability
of Capsule Networks. Our study contributes to the broader discussion on
improving the robustness and scalability of Capsule Networks.
| [
{
"version": "v1",
"created": "Sat, 13 May 2023 15:42:26 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Everett",
"Miles",
""
],
[
"Zhong",
"Mingjun",
""
],
[
"Leontidis",
"Georgios",
""
]
] | TITLE: Vanishing Activations: A Symptom of Deep Capsule Networks
ABSTRACT: Capsule Networks, an extension to Neural Networks utilizing vector or matrix
representations instead of scalars, were initially developed to create a
dynamic parse tree where visual concepts evolve from parts to complete objects.
Early implementations of Capsule Networks achieved and maintain
state-of-the-art results on various datasets. However, recent studies have
revealed shortcomings in the original Capsule Network architecture, notably its
failure to construct a parse tree and its susceptibility to vanishing gradients
when deployed in deeper networks. This paper extends the investigation to a
range of leading Capsule Network architectures, demonstrating that these issues
are not confined to the original design. We argue that the majority of Capsule
Network research has produced architectures that, while modestly divergent from
the original Capsule Network, still retain a fundamentally similar structure.
We posit that this inherent design similarity might be impeding the scalability
of Capsule Networks. Our study contributes to the broader discussion on
improving the robustness and scalability of Capsule Networks.
| no_new_dataset | 0.710641 |
2305.11181 | Yifan Tang | Yifan Tang, M. Rahmani Dehaghani, G. Gary Wang | Comparison of Transfer Learning based Additive Manufacturing Models via
A Case Study | 16 pages, 8 figures | null | null | null | cs.LG cs.CE | http://creativecommons.org/licenses/by/4.0/ | Transfer learning (TL) based additive manufacturing (AM) modeling is an
emerging field to reuse the data from historical products and mitigate the data
insufficiency in modeling new products. Although some trials have been
conducted recently, the inherent challenges of applying TL in AM modeling are
seldom discussed, e.g., which source domain to use, how much target data is
needed, and whether to apply data preprocessing techniques. This paper aims to
answer those questions through a case study defined based on an open-source
dataset about metal AM products. In the case study, five TL methods are
integrated with decision tree regression (DTR) and artificial neural network
(ANN) to construct six TL-based models, whose performances are then compared
with the baseline DTR and ANN in a proposed validation framework. The
comparisons are used to quantify the performance of applied TL methods and are
discussed from the perspective of similarity, training data size, and data
preprocessing. Finally, the source AM domain with larger qualitative similarity
and a certain range of target-to-source training data size ratio are
recommended. Besides, the data preprocessing should be performed carefully to
balance the modeling performance and the performance improvement due to TL.
| [
{
"version": "v1",
"created": "Wed, 17 May 2023 00:29:25 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Tang",
"Yifan",
""
],
[
"Dehaghani",
"M. Rahmani",
""
],
[
"Wang",
"G. Gary",
""
]
] | TITLE: Comparison of Transfer Learning based Additive Manufacturing Models via
A Case Study
ABSTRACT: Transfer learning (TL) based additive manufacturing (AM) modeling is an
emerging field to reuse the data from historical products and mitigate the data
insufficiency in modeling new products. Although some trials have been
conducted recently, the inherent challenges of applying TL in AM modeling are
seldom discussed, e.g., which source domain to use, how much target data is
needed, and whether to apply data preprocessing techniques. This paper aims to
answer those questions through a case study defined based on an open-source
dataset about metal AM products. In the case study, five TL methods are
integrated with decision tree regression (DTR) and artificial neural network
(ANN) to construct six TL-based models, whose performances are then compared
with the baseline DTR and ANN in a proposed validation framework. The
comparisons are used to quantify the performance of applied TL methods and are
discussed from the perspective of similarity, training data size, and data
preprocessing. Finally, the source AM domain with larger qualitative similarity
and a certain range of target-to-source training data size ratio are
recommended. Besides, the data preprocessing should be performed carefully to
balance the modeling performance and the performance improvement due to TL.
| no_new_dataset | 0.708799 |
2305.11197 | Yichen Zhu | Yichen Zhu, Jian Yuan, Bo Jiang, Tao Lin, Haiming Jin, Xinbing Wang,
Chenghu Zhou | Prediction with Incomplete Data under Agnostic Mask Distribution Shift | null | null | null | null | cs.LG cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Data with missing values is ubiquitous in many applications. Recent years
have witnessed increasing attention on prediction with only incomplete data
consisting of observed features and a mask that indicates the missing pattern.
Existing methods assume that the training and testing distributions are the
same, which may be violated in real-world scenarios. In this paper, we consider
prediction with incomplete data in the presence of distribution shift. We focus
on the case where the underlying joint distribution of complete features and
label is invariant, but the missing pattern, i.e., mask distribution may shift
agnostically between training and testing. To achieve generalization, we
leverage the observation that for each mask, there is an invariant optimal
predictor. To avoid the exponential explosion when learning them separately, we
approximate the optimal predictors jointly using a double parameterization
technique. This has the undesirable side effect of allowing the learned
predictors to rely on the intra-mask correlation and that between features and
mask. We perform decorrelation to minimize this effect. Combining the
techniques above, we propose a novel prediction method called StableMiss.
Extensive experiments on both synthetic and real-world datasets show that
StableMiss is robust and outperforms state-of-the-art methods under agnostic
mask distribution shift.
| [
{
"version": "v1",
"created": "Thu, 18 May 2023 14:06:06 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Zhu",
"Yichen",
""
],
[
"Yuan",
"Jian",
""
],
[
"Jiang",
"Bo",
""
],
[
"Lin",
"Tao",
""
],
[
"Jin",
"Haiming",
""
],
[
"Wang",
"Xinbing",
""
],
[
"Zhou",
"Chenghu",
""
]
] | TITLE: Prediction with Incomplete Data under Agnostic Mask Distribution Shift
ABSTRACT: Data with missing values is ubiquitous in many applications. Recent years
have witnessed increasing attention on prediction with only incomplete data
consisting of observed features and a mask that indicates the missing pattern.
Existing methods assume that the training and testing distributions are the
same, which may be violated in real-world scenarios. In this paper, we consider
prediction with incomplete data in the presence of distribution shift. We focus
on the case where the underlying joint distribution of complete features and
label is invariant, but the missing pattern, i.e., mask distribution may shift
agnostically between training and testing. To achieve generalization, we
leverage the observation that for each mask, there is an invariant optimal
predictor. To avoid the exponential explosion when learning them separately, we
approximate the optimal predictors jointly using a double parameterization
technique. This has the undesirable side effect of allowing the learned
predictors to rely on the intra-mask correlation and that between features and
mask. We perform decorrelation to minimize this effect. Combining the
techniques above, we propose a novel prediction method called StableMiss.
Extensive experiments on both synthetic and real-world datasets show that
StableMiss is robust and outperforms state-of-the-art methods under agnostic
mask distribution shift.
| no_new_dataset | 0.709007 |
2305.11199 | Munib Mesinovic | Munib Mesinovic, Xin Ci Wong, Giri Shan Rajahram, Barbara Wanjiru
Citarella, Kalaiarasu M. Peariasamy, Frank van Someren Greve, Piero Olliaro,
Laura Merson, Lei Clifton, Christiana Kartsonaki, ISARIC Characterisation
Group | At-Admission Prediction of Mortality and Pulmonary Embolism in COVID-19
Patients Using Statistical and Machine Learning Methods: An International
Cohort Study | null | null | null | null | q-bio.QM cs.LG | http://creativecommons.org/licenses/by/4.0/ | By September, 2022, more than 600 million cases of SARS-CoV-2 infection have
been reported globally, resulting in over 6.5 million deaths. COVID-19
mortality risk estimators are often, however, developed with small
unrepresentative samples and with methodological limitations. It is highly
important to develop predictive tools for pulmonary embolism (PE) in COVID-19
patients as one of the most severe preventable complications of COVID-19. Using
a dataset of more than 800,000 COVID-19 patients from an international cohort,
we propose a cost-sensitive gradient-boosted machine learning model that
predicts occurrence of PE and death at admission. Logistic regression, Cox
proportional hazards models, and Shapley values were used to identify key
predictors for PE and death. Our prediction model had a test AUROC of 75.9% and
74.2%, and sensitivities of 67.5% and 72.7% for PE and all-cause mortality
respectively on a highly diverse and held-out test set. The PE prediction model
was also evaluated on patients in UK and Spain separately with test results of
74.5% AUROC, 63.5% sensitivity and 78.9% AUROC, 95.7% sensitivity. Age, sex,
region of admission, comorbidities (chronic cardiac and pulmonary disease,
dementia, diabetes, hypertension, cancer, obesity, smoking), and symptoms (any,
confusion, chest pain, fatigue, headache, fever, muscle or joint pain,
shortness of breath) were the most important clinical predictors at admission.
Our machine learning model developed from an international cohort can serve to
better regulate hospital risk prioritisation of at-risk patients.
| [
{
"version": "v1",
"created": "Thu, 18 May 2023 14:55:27 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Mesinovic",
"Munib",
""
],
[
"Wong",
"Xin Ci",
""
],
[
"Rajahram",
"Giri Shan",
""
],
[
"Citarella",
"Barbara Wanjiru",
""
],
[
"Peariasamy",
"Kalaiarasu M.",
""
],
[
"Greve",
"Frank van Someren",
""
],
[
"Olliaro",
"Piero",
""
],
[
"Merson",
"Laura",
""
],
[
"Clifton",
"Lei",
""
],
[
"Kartsonaki",
"Christiana",
""
],
[
"Group",
"ISARIC Characterisation",
""
]
] | TITLE: At-Admission Prediction of Mortality and Pulmonary Embolism in COVID-19
Patients Using Statistical and Machine Learning Methods: An International
Cohort Study
ABSTRACT: By September, 2022, more than 600 million cases of SARS-CoV-2 infection have
been reported globally, resulting in over 6.5 million deaths. COVID-19
mortality risk estimators are often, however, developed with small
unrepresentative samples and with methodological limitations. It is highly
important to develop predictive tools for pulmonary embolism (PE) in COVID-19
patients as one of the most severe preventable complications of COVID-19. Using
a dataset of more than 800,000 COVID-19 patients from an international cohort,
we propose a cost-sensitive gradient-boosted machine learning model that
predicts occurrence of PE and death at admission. Logistic regression, Cox
proportional hazards models, and Shapley values were used to identify key
predictors for PE and death. Our prediction model had a test AUROC of 75.9% and
74.2%, and sensitivities of 67.5% and 72.7% for PE and all-cause mortality
respectively on a highly diverse and held-out test set. The PE prediction model
was also evaluated on patients in UK and Spain separately with test results of
74.5% AUROC, 63.5% sensitivity and 78.9% AUROC, 95.7% sensitivity. Age, sex,
region of admission, comorbidities (chronic cardiac and pulmonary disease,
dementia, diabetes, hypertension, cancer, obesity, smoking), and symptoms (any,
confusion, chest pain, fatigue, headache, fever, muscle or joint pain,
shortness of breath) were the most important clinical predictors at admission.
Our machine learning model developed from an international cohort can serve to
better regulate hospital risk prioritisation of at-risk patients.
| no_new_dataset | 0.705367 |
2305.11213 | Matthew Ho | Matthew Ho, Xiaosheng Zhao, Benjamin Wandelt | Information-Ordered Bottlenecks for Adaptive Semantic Compression | 14 pages, 6 figures, 1 table, Submitted to NeurIPS 2023 | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | We present the information-ordered bottleneck (IOB), a neural layer designed
to adaptively compress data into latent variables ordered by likelihood
maximization. Without retraining, IOB nodes can be truncated at any bottleneck
width, capturing the most crucial information in the first latent variables.
Unifying several previous approaches, we show that IOBs achieve near-optimal
compression for a given encoding architecture and can assign ordering to latent
signals in a manner that is semantically meaningful. IOBs demonstrate a
remarkable ability to compress embeddings of image and text data, leveraging
the performance of SOTA architectures such as CNNs, transformers, and diffusion
models. Moreover, we introduce a novel theory for estimating global intrinsic
dimensionality with IOBs and show that they recover SOTA dimensionality
estimates for complex synthetic data. Furthermore, we showcase the utility of
these models for exploratory analysis through applications on heterogeneous
datasets, enabling computer-aided discovery of dataset complexity.
| [
{
"version": "v1",
"created": "Thu, 18 May 2023 18:00:00 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Ho",
"Matthew",
""
],
[
"Zhao",
"Xiaosheng",
""
],
[
"Wandelt",
"Benjamin",
""
]
] | TITLE: Information-Ordered Bottlenecks for Adaptive Semantic Compression
ABSTRACT: We present the information-ordered bottleneck (IOB), a neural layer designed
to adaptively compress data into latent variables ordered by likelihood
maximization. Without retraining, IOB nodes can be truncated at any bottleneck
width, capturing the most crucial information in the first latent variables.
Unifying several previous approaches, we show that IOBs achieve near-optimal
compression for a given encoding architecture and can assign ordering to latent
signals in a manner that is semantically meaningful. IOBs demonstrate a
remarkable ability to compress embeddings of image and text data, leveraging
the performance of SOTA architectures such as CNNs, transformers, and diffusion
models. Moreover, we introduce a novel theory for estimating global intrinsic
dimensionality with IOBs and show that they recover SOTA dimensionality
estimates for complex synthetic data. Furthermore, we showcase the utility of
these models for exploratory analysis through applications on heterogeneous
datasets, enabling computer-aided discovery of dataset complexity.
| no_new_dataset | 0.709982 |
2305.11229 | Tiantian Feng | Tiantian Feng and Rajat Hebbar and Shrikanth Narayanan | TrustSER: On the Trustworthiness of Fine-tuning Pre-trained Speech
Embeddings For Speech Emotion Recognition | null | null | null | null | cs.SD eess.AS | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recent studies have explored the use of pre-trained embeddings for speech
emotion recognition (SER), achieving comparable performance to conventional
methods that rely on low-level knowledge-inspired acoustic features. These
embeddings are often generated from models trained on large-scale speech
datasets using self-supervised or weakly-supervised learning objectives.
Despite the significant advancements made in SER through the use of pre-trained
embeddings, there is a limited understanding of the trustworthiness of these
methods, including privacy breaches, unfair performance, vulnerability to
adversarial attacks, and computational cost, all of which may hinder the
real-world deployment of these systems. In response, we introduce TrustSER, a
general framework designed to evaluate the trustworthiness of SER systems using
deep learning methods, with a focus on privacy, safety, fairness, and
sustainability, offering unique insights into future research in the field of
SER. Our code is publicly available under:
https://github.com/usc-sail/trust-ser.
| [
{
"version": "v1",
"created": "Thu, 18 May 2023 18:00:36 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Feng",
"Tiantian",
""
],
[
"Hebbar",
"Rajat",
""
],
[
"Narayanan",
"Shrikanth",
""
]
] | TITLE: TrustSER: On the Trustworthiness of Fine-tuning Pre-trained Speech
Embeddings For Speech Emotion Recognition
ABSTRACT: Recent studies have explored the use of pre-trained embeddings for speech
emotion recognition (SER), achieving comparable performance to conventional
methods that rely on low-level knowledge-inspired acoustic features. These
embeddings are often generated from models trained on large-scale speech
datasets using self-supervised or weakly-supervised learning objectives.
Despite the significant advancements made in SER through the use of pre-trained
embeddings, there is a limited understanding of the trustworthiness of these
methods, including privacy breaches, unfair performance, vulnerability to
adversarial attacks, and computational cost, all of which may hinder the
real-world deployment of these systems. In response, we introduce TrustSER, a
general framework designed to evaluate the trustworthiness of SER systems using
deep learning methods, with a focus on privacy, safety, fairness, and
sustainability, offering unique insights into future research in the field of
SER. Our code is publicly available under:
https://github.com/usc-sail/trust-ser.
| no_new_dataset | 0.709636 |
2305.11236 | Xinchi Qiu | Xinchi Qiu, Heng Pan, Wanru Zhao, Chenyang Ma, Pedro Porto Buarque de
Gusm\~ao, Nicholas D. Lane | Efficient Vertical Federated Learning with Secure Aggregation | Federated Learning Systems (FLSys) Workshop @ MLSys 2023 | null | null | null | cs.LG cs.AI cs.CR | http://creativecommons.org/licenses/by/4.0/ | The majority of work in privacy-preserving federated learning (FL) has been
focusing on horizontally partitioned datasets where clients share the same sets
of features and can train complete models independently. However, in many
interesting problems, such as financial fraud detection and disease detection,
individual data points are scattered across different clients/organizations in
vertical federated learning. Solutions for this type of FL require the exchange
of gradients between participants and rarely consider privacy and security
concerns, posing a potential risk of privacy leakage. In this work, we present
a novel design for training vertical FL securely and efficiently using
state-of-the-art security modules for secure aggregation. We demonstrate
empirically that our method does not impact training performance whilst
obtaining 9.1e2 ~3.8e4 speedup compared to homomorphic encryption (HE).
| [
{
"version": "v1",
"created": "Thu, 18 May 2023 18:08:36 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Qiu",
"Xinchi",
""
],
[
"Pan",
"Heng",
""
],
[
"Zhao",
"Wanru",
""
],
[
"Ma",
"Chenyang",
""
],
[
"de Gusmão",
"Pedro Porto Buarque",
""
],
[
"Lane",
"Nicholas D.",
""
]
] | TITLE: Efficient Vertical Federated Learning with Secure Aggregation
ABSTRACT: The majority of work in privacy-preserving federated learning (FL) has been
focusing on horizontally partitioned datasets where clients share the same sets
of features and can train complete models independently. However, in many
interesting problems, such as financial fraud detection and disease detection,
individual data points are scattered across different clients/organizations in
vertical federated learning. Solutions for this type of FL require the exchange
of gradients between participants and rarely consider privacy and security
concerns, posing a potential risk of privacy leakage. In this work, we present
a novel design for training vertical FL securely and efficiently using
state-of-the-art security modules for secure aggregation. We demonstrate
empirically that our method does not impact training performance whilst
obtaining 9.1e2 ~3.8e4 speedup compared to homomorphic encryption (HE).
| no_new_dataset | 0.709988 |
2305.11262 | Jiaxu Zhao | Jiaxu Zhao, Meng Fang, Zijing Shi, Yitong Li, Ling Chen, Mykola
Pechenizkiy | CHBias: Bias Evaluation and Mitigation of Chinese Conversational
Language Models | Accepted by ACL 2023 | null | null | null | cs.CL | http://creativecommons.org/licenses/by-nc-sa/4.0/ | \textit{\textbf{\textcolor{red}{Warning}:} This paper contains content that
may be offensive or upsetting.} Pretrained conversational agents have been
exposed to safety issues, exhibiting a range of stereotypical human biases such
as gender bias. However, there are still limited bias categories in current
research, and most of them only focus on English. In this paper, we introduce a
new Chinese dataset, CHBias, for bias evaluation and mitigation of Chinese
conversational language models. Apart from those previous well-explored bias
categories, CHBias includes under-explored bias categories, such as ageism and
appearance biases, which received less attention. We evaluate two popular
pretrained Chinese conversational models, CDial-GPT and EVA2.0, using CHBias.
Furthermore, to mitigate different biases, we apply several debiasing methods
to the Chinese pretrained models. Experimental results show that these Chinese
pretrained models are potentially risky for generating texts that contain
social biases, and debiasing methods using the proposed dataset can make
response generation less biased while preserving the models' conversational
capabilities.
| [
{
"version": "v1",
"created": "Thu, 18 May 2023 18:58:30 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Zhao",
"Jiaxu",
""
],
[
"Fang",
"Meng",
""
],
[
"Shi",
"Zijing",
""
],
[
"Li",
"Yitong",
""
],
[
"Chen",
"Ling",
""
],
[
"Pechenizkiy",
"Mykola",
""
]
] | TITLE: CHBias: Bias Evaluation and Mitigation of Chinese Conversational
Language Models
ABSTRACT: \textit{\textbf{\textcolor{red}{Warning}:} This paper contains content that
may be offensive or upsetting.} Pretrained conversational agents have been
exposed to safety issues, exhibiting a range of stereotypical human biases such
as gender bias. However, there are still limited bias categories in current
research, and most of them only focus on English. In this paper, we introduce a
new Chinese dataset, CHBias, for bias evaluation and mitigation of Chinese
conversational language models. Apart from those previous well-explored bias
categories, CHBias includes under-explored bias categories, such as ageism and
appearance biases, which received less attention. We evaluate two popular
pretrained Chinese conversational models, CDial-GPT and EVA2.0, using CHBias.
Furthermore, to mitigate different biases, we apply several debiasing methods
to the Chinese pretrained models. Experimental results show that these Chinese
pretrained models are potentially risky for generating texts that contain
social biases, and debiasing methods using the proposed dataset can make
response generation less biased while preserving the models' conversational
capabilities.
| new_dataset | 0.716262 |
2305.11301 | Navdeep Kaur | Ishaan Singh and Navdeep Kaur and Garima Gaur and Mausam | NeuSTIP: A Novel Neuro-Symbolic Model for Link and Time Prediction in
Temporal Knowledge Graphs | 13 pages, 2 Figures | null | null | null | cs.AI | http://creativecommons.org/licenses/by/4.0/ | While Knowledge Graph Completion (KGC) on static facts is a matured field,
Temporal Knowledge Graph Completion (TKGC), that incorporates validity time
into static facts is still in its nascent stage. The KGC methods fall into
multiple categories including embedding-based, rule-based, GNN-based,
pretrained Language Model based approaches. However, such dimensions have not
been explored in TKG. To that end, we propose a novel temporal neuro-symbolic
model, NeuSTIP, that performs link prediction and time interval prediction in a
TKG. NeuSTIP learns temporal rules in the presence of the Allen predicates that
ensure the temporal consistency between neighboring predicates in a given rule.
We further design a unique scoring function that evaluates the confidence of
the candidate answers while performing link prediction and time interval
prediction by utilizing the learned rules. Our empirical evaluation on two time
interval based TKGC datasets suggests that our model outperforms
state-of-the-art models for both link prediction and the time interval
prediction task.
| [
{
"version": "v1",
"created": "Mon, 15 May 2023 13:46:34 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Singh",
"Ishaan",
""
],
[
"Kaur",
"Navdeep",
""
],
[
"Gaur",
"Garima",
""
],
[
"Mausam",
"",
""
]
] | TITLE: NeuSTIP: A Novel Neuro-Symbolic Model for Link and Time Prediction in
Temporal Knowledge Graphs
ABSTRACT: While Knowledge Graph Completion (KGC) on static facts is a matured field,
Temporal Knowledge Graph Completion (TKGC), that incorporates validity time
into static facts is still in its nascent stage. The KGC methods fall into
multiple categories including embedding-based, rule-based, GNN-based,
pretrained Language Model based approaches. However, such dimensions have not
been explored in TKG. To that end, we propose a novel temporal neuro-symbolic
model, NeuSTIP, that performs link prediction and time interval prediction in a
TKG. NeuSTIP learns temporal rules in the presence of the Allen predicates that
ensure the temporal consistency between neighboring predicates in a given rule.
We further design a unique scoring function that evaluates the confidence of
the candidate answers while performing link prediction and time interval
prediction by utilizing the learned rules. Our empirical evaluation on two time
interval based TKGC datasets suggests that our model outperforms
state-of-the-art models for both link prediction and the time interval
prediction task.
| no_new_dataset | 0.70844 |
2305.11315 | Zeyu Zhang | Zeyu Zhang and Steven Bethard | Improving Toponym Resolution with Better Candidate Generation,
Transformer-based Reranking, and Two-Stage Resolution | null | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Geocoding is the task of converting location mentions in text into structured
data that encodes the geospatial semantics. We propose a new architecture for
geocoding, GeoNorm. GeoNorm first uses information retrieval techniques to
generate a list of candidate entries from the geospatial ontology. Then it
reranks the candidate entries using a transformer-based neural network that
incorporates information from the ontology such as the entry's population. This
generate-and-rerank process is applied twice: first to resolve the less
ambiguous countries, states, and counties, and second to resolve the remaining
location mentions, using the identified countries, states, and counties as
context. Our proposed toponym resolution framework achieves state-of-the-art
performance on multiple datasets. Code and models are available at
\url{https://github.com/clulab/geonorm}.
| [
{
"version": "v1",
"created": "Thu, 18 May 2023 21:52:48 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Zhang",
"Zeyu",
""
],
[
"Bethard",
"Steven",
""
]
] | TITLE: Improving Toponym Resolution with Better Candidate Generation,
Transformer-based Reranking, and Two-Stage Resolution
ABSTRACT: Geocoding is the task of converting location mentions in text into structured
data that encodes the geospatial semantics. We propose a new architecture for
geocoding, GeoNorm. GeoNorm first uses information retrieval techniques to
generate a list of candidate entries from the geospatial ontology. Then it
reranks the candidate entries using a transformer-based neural network that
incorporates information from the ontology such as the entry's population. This
generate-and-rerank process is applied twice: first to resolve the less
ambiguous countries, states, and counties, and second to resolve the remaining
location mentions, using the identified countries, states, and counties as
context. Our proposed toponym resolution framework achieves state-of-the-art
performance on multiple datasets. Code and models are available at
\url{https://github.com/clulab/geonorm}.
| no_new_dataset | 0.713625 |
2305.11327 | Bhanu Prakash Voutharoja | Bhanu Prakash Voutharoja and Peng Wang and Lei Wang and Vivienne Guan | MALM: Mask Augmentation based Local Matching for Food-Recipe Retrieval | Under review. Link to the dataset repo -
https://github.com/torralba-lab/im2recipe-Pytorch#recipe1m-dataset | null | null | null | cs.CV cs.LG cs.MM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Image-to-recipe retrieval is a challenging vision-to-language task of
significant practical value. The main challenge of the task lies in the
ultra-high redundancy in the long recipe and the large variation reflected in
both food item combination and food item appearance. A de-facto idea to address
this task is to learn a shared feature embedding space in which a food image is
aligned better to its paired recipe than other recipes. However, such
supervised global matching is prone to supervision collapse, i.e., only partial
information that is necessary for distinguishing training pairs can be
identified, while other information that is potentially useful in
generalization could be lost. To mitigate such a problem, we propose a
mask-augmentation-based local matching network (MALM), where an image-text
matching module and a masked self-distillation module benefit each other
mutually to learn generalizable cross-modality representations. On one hand, we
perform local matching between the tokenized representations of image and text
to locate fine-grained cross-modality correspondence explicitly. We involve
representations of masked image patches in this process to alleviate
overfitting resulting from local matching especially when some food items are
underrepresented. On the other hand, predicting the hidden representations of
the masked patches through self-distillation helps to learn general-purpose
image representations that are expected to generalize better. And the
multi-task nature of the model enables the representations of masked patches to
be text-aware and thus facilitates the lost information reconstruction.
Experimental results on Recipe1M dataset show our method can clearly outperform
state-of-the-art (SOTA) methods. Our code will be available at
https://github.com/MyFoodChoice/MALM_Mask_Augmentation_based_Local_Matching-_for-_Food_Recipe_Retrieval
| [
{
"version": "v1",
"created": "Thu, 18 May 2023 22:25:50 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Voutharoja",
"Bhanu Prakash",
""
],
[
"Wang",
"Peng",
""
],
[
"Wang",
"Lei",
""
],
[
"Guan",
"Vivienne",
""
]
] | TITLE: MALM: Mask Augmentation based Local Matching for Food-Recipe Retrieval
ABSTRACT: Image-to-recipe retrieval is a challenging vision-to-language task of
significant practical value. The main challenge of the task lies in the
ultra-high redundancy in the long recipe and the large variation reflected in
both food item combination and food item appearance. A de-facto idea to address
this task is to learn a shared feature embedding space in which a food image is
aligned better to its paired recipe than other recipes. However, such
supervised global matching is prone to supervision collapse, i.e., only partial
information that is necessary for distinguishing training pairs can be
identified, while other information that is potentially useful in
generalization could be lost. To mitigate such a problem, we propose a
mask-augmentation-based local matching network (MALM), where an image-text
matching module and a masked self-distillation module benefit each other
mutually to learn generalizable cross-modality representations. On one hand, we
perform local matching between the tokenized representations of image and text
to locate fine-grained cross-modality correspondence explicitly. We involve
representations of masked image patches in this process to alleviate
overfitting resulting from local matching especially when some food items are
underrepresented. On the other hand, predicting the hidden representations of
the masked patches through self-distillation helps to learn general-purpose
image representations that are expected to generalize better. And the
multi-task nature of the model enables the representations of masked patches to
be text-aware and thus facilitates the lost information reconstruction.
Experimental results on Recipe1M dataset show our method can clearly outperform
state-of-the-art (SOTA) methods. Our code will be available at
https://github.com/MyFoodChoice/MALM_Mask_Augmentation_based_Local_Matching-_for-_Food_Recipe_Retrieval
| no_new_dataset | 0.711469 |
2305.11340 | Wenhao Ding | Wenhao Ding, Tong Che, Ding Zhao, Marco Pavone | Bayesian Reparameterization of Reward-Conditioned Reinforcement Learning
with Energy-based Models | Accepted to ICML 2023 | null | null | null | cs.LG cs.RO | http://creativecommons.org/licenses/by/4.0/ | Recently, reward-conditioned reinforcement learning (RCRL) has gained
popularity due to its simplicity, flexibility, and off-policy nature. However,
we will show that current RCRL approaches are fundamentally limited and fail to
address two critical challenges of RCRL -- improving generalization on high
reward-to-go (RTG) inputs, and avoiding out-of-distribution (OOD) RTG queries
during testing time. To address these challenges when training vanilla RCRL
architectures, we propose Bayesian Reparameterized RCRL (BR-RCRL), a novel set
of inductive biases for RCRL inspired by Bayes' theorem. BR-RCRL removes a core
obstacle preventing vanilla RCRL from generalizing on high RTG inputs -- a
tendency that the model treats different RTG inputs as independent values,
which we term ``RTG Independence". BR-RCRL also allows us to design an
accompanying adaptive inference method, which maximizes total returns while
avoiding OOD queries that yield unpredictable behaviors in vanilla RCRL
methods. We show that BR-RCRL achieves state-of-the-art performance on the
Gym-Mujoco and Atari offline RL benchmarks, improving upon vanilla RCRL by up
to 11%.
| [
{
"version": "v1",
"created": "Thu, 18 May 2023 23:23:08 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Ding",
"Wenhao",
""
],
[
"Che",
"Tong",
""
],
[
"Zhao",
"Ding",
""
],
[
"Pavone",
"Marco",
""
]
] | TITLE: Bayesian Reparameterization of Reward-Conditioned Reinforcement Learning
with Energy-based Models
ABSTRACT: Recently, reward-conditioned reinforcement learning (RCRL) has gained
popularity due to its simplicity, flexibility, and off-policy nature. However,
we will show that current RCRL approaches are fundamentally limited and fail to
address two critical challenges of RCRL -- improving generalization on high
reward-to-go (RTG) inputs, and avoiding out-of-distribution (OOD) RTG queries
during testing time. To address these challenges when training vanilla RCRL
architectures, we propose Bayesian Reparameterized RCRL (BR-RCRL), a novel set
of inductive biases for RCRL inspired by Bayes' theorem. BR-RCRL removes a core
obstacle preventing vanilla RCRL from generalizing on high RTG inputs -- a
tendency that the model treats different RTG inputs as independent values,
which we term ``RTG Independence". BR-RCRL also allows us to design an
accompanying adaptive inference method, which maximizes total returns while
avoiding OOD queries that yield unpredictable behaviors in vanilla RCRL
methods. We show that BR-RCRL achieves state-of-the-art performance on the
Gym-Mujoco and Atari offline RL benchmarks, improving upon vanilla RCRL by up
to 11%.
| no_new_dataset | 0.709435 |
2305.11349 | Amila Silva | Amila Silva, Ling Luo, Shanika Karunasekera, Christopher Leckie | Unsupervised Domain-agnostic Fake News Detection using Multi-modal Weak
Signals | 15 pages | null | null | null | cs.LG cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The emergence of social media as one of the main platforms for people to
access news has enabled the wide dissemination of fake news. This has motivated
numerous studies on automating fake news detection. Although there have been
limited attempts at unsupervised fake news detection, their performance suffers
due to not exploiting the knowledge from various modalities related to news
records and due to the presence of various latent biases in the existing news
datasets. To address these limitations, this work proposes an effective
framework for unsupervised fake news detection, which first embeds the
knowledge available in four modalities in news records and then proposes a
novel noise-robust self-supervised learning technique to identify the veracity
of news records from the multi-modal embeddings. Also, we propose a novel
technique to construct news datasets minimizing the latent biases in existing
news datasets. Following the proposed approach for dataset construction, we
produce a Large-scale Unlabelled News Dataset consisting 419,351 news articles
related to COVID-19, acronymed as LUND-COVID. We trained the proposed
unsupervised framework using LUND-COVID to exploit the potential of large
datasets, and evaluate it using a set of existing labelled datasets. Our
results show that the proposed unsupervised framework largely outperforms
existing unsupervised baselines for different tasks such as multi-modal fake
news detection, fake news early detection and few-shot fake news detection,
while yielding notable improvements for unseen domains during training.
| [
{
"version": "v1",
"created": "Thu, 18 May 2023 23:49:31 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Silva",
"Amila",
""
],
[
"Luo",
"Ling",
""
],
[
"Karunasekera",
"Shanika",
""
],
[
"Leckie",
"Christopher",
""
]
] | TITLE: Unsupervised Domain-agnostic Fake News Detection using Multi-modal Weak
Signals
ABSTRACT: The emergence of social media as one of the main platforms for people to
access news has enabled the wide dissemination of fake news. This has motivated
numerous studies on automating fake news detection. Although there have been
limited attempts at unsupervised fake news detection, their performance suffers
due to not exploiting the knowledge from various modalities related to news
records and due to the presence of various latent biases in the existing news
datasets. To address these limitations, this work proposes an effective
framework for unsupervised fake news detection, which first embeds the
knowledge available in four modalities in news records and then proposes a
novel noise-robust self-supervised learning technique to identify the veracity
of news records from the multi-modal embeddings. Also, we propose a novel
technique to construct news datasets minimizing the latent biases in existing
news datasets. Following the proposed approach for dataset construction, we
produce a Large-scale Unlabelled News Dataset consisting 419,351 news articles
related to COVID-19, acronymed as LUND-COVID. We trained the proposed
unsupervised framework using LUND-COVID to exploit the potential of large
datasets, and evaluate it using a set of existing labelled datasets. Our
results show that the proposed unsupervised framework largely outperforms
existing unsupervised baselines for different tasks such as multi-modal fake
news detection, fake news early detection and few-shot fake news detection,
while yielding notable improvements for unseen domains during training.
| no_new_dataset | 0.687866 |
2305.11355 | Jacob Eisenstein | Jacob Eisenstein, Vinodkumar Prabhakaran, Clara Rivera, Dorottya
Demszky, Devyani Sharma | MD3: The Multi-Dialect Dataset of Dialogues | InterSpeech 2023 | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | We introduce a new dataset of conversational speech representing English from
India, Nigeria, and the United States. The Multi-Dialect Dataset of Dialogues
(MD3) strikes a new balance between open-ended conversational speech and
task-oriented dialogue by prompting participants to perform a series of short
information-sharing tasks. This facilitates quantitative cross-dialectal
comparison, while avoiding the imposition of a restrictive task structure that
might inhibit the expression of dialect features. Preliminary analysis of the
dataset reveals significant differences in syntax and in the use of discourse
markers. The dataset, which will be made publicly available with the
publication of this paper, includes more than 20 hours of audio and more than
200,000 orthographically-transcribed tokens.
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 00:14:10 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Eisenstein",
"Jacob",
""
],
[
"Prabhakaran",
"Vinodkumar",
""
],
[
"Rivera",
"Clara",
""
],
[
"Demszky",
"Dorottya",
""
],
[
"Sharma",
"Devyani",
""
]
] | TITLE: MD3: The Multi-Dialect Dataset of Dialogues
ABSTRACT: We introduce a new dataset of conversational speech representing English from
India, Nigeria, and the United States. The Multi-Dialect Dataset of Dialogues
(MD3) strikes a new balance between open-ended conversational speech and
task-oriented dialogue by prompting participants to perform a series of short
information-sharing tasks. This facilitates quantitative cross-dialectal
comparison, while avoiding the imposition of a restrictive task structure that
might inhibit the expression of dialect features. Preliminary analysis of the
dataset reveals significant differences in syntax and in the use of discourse
markers. The dataset, which will be made publicly available with the
publication of this paper, includes more than 20 hours of audio and more than
200,000 orthographically-transcribed tokens.
| new_dataset | 0.715412 |
2305.11392 | Xiameng Qin | Mingliang Zhai, Yulin Li, Xiameng Qin, Chen Yi, Qunyi Xie, Chengquan
Zhang, Kun Yao, Yuwei Wu, Yunde Jia | Fast-StrucTexT: An Efficient Hourglass Transformer with Modality-guided
Dynamic Token Merge for Document Understanding | IJCAI 2023 | null | null | null | cs.CV cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Transformers achieve promising performance in document understanding because
of their high effectiveness and still suffer from quadratic computational
complexity dependency on the sequence length. General efficient transformers
are challenging to be directly adapted to model document. They are unable to
handle the layout representation in documents, e.g. word, line and paragraph,
on different granularity levels and seem hard to achieve a good trade-off
between efficiency and performance. To tackle the concerns, we propose
Fast-StrucTexT, an efficient multi-modal framework based on the StrucTexT
algorithm with an hourglass transformer architecture, for visual document
understanding. Specifically, we design a modality-guided dynamic token merging
block to make the model learn multi-granularity representation and prunes
redundant tokens. Additionally, we present a multi-modal interaction module
called Symmetry Cross Attention (SCA) to consider multi-modal fusion and
efficiently guide the token mergence. The SCA allows one modality input as
query to calculate cross attention with another modality in a dual phase.
Extensive experiments on FUNSD, SROIE, and CORD datasets demonstrate that our
model achieves the state-of-the-art performance and almost 1.9X faster
inference time than the state-of-the-art methods.
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 02:42:35 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Zhai",
"Mingliang",
""
],
[
"Li",
"Yulin",
""
],
[
"Qin",
"Xiameng",
""
],
[
"Yi",
"Chen",
""
],
[
"Xie",
"Qunyi",
""
],
[
"Zhang",
"Chengquan",
""
],
[
"Yao",
"Kun",
""
],
[
"Wu",
"Yuwei",
""
],
[
"Jia",
"Yunde",
""
]
] | TITLE: Fast-StrucTexT: An Efficient Hourglass Transformer with Modality-guided
Dynamic Token Merge for Document Understanding
ABSTRACT: Transformers achieve promising performance in document understanding because
of their high effectiveness and still suffer from quadratic computational
complexity dependency on the sequence length. General efficient transformers
are challenging to be directly adapted to model document. They are unable to
handle the layout representation in documents, e.g. word, line and paragraph,
on different granularity levels and seem hard to achieve a good trade-off
between efficiency and performance. To tackle the concerns, we propose
Fast-StrucTexT, an efficient multi-modal framework based on the StrucTexT
algorithm with an hourglass transformer architecture, for visual document
understanding. Specifically, we design a modality-guided dynamic token merging
block to make the model learn multi-granularity representation and prunes
redundant tokens. Additionally, we present a multi-modal interaction module
called Symmetry Cross Attention (SCA) to consider multi-modal fusion and
efficiently guide the token mergence. The SCA allows one modality input as
query to calculate cross attention with another modality in a dual phase.
Extensive experiments on FUNSD, SROIE, and CORD datasets demonstrate that our
model achieves the state-of-the-art performance and almost 1.9X faster
inference time than the state-of-the-art methods.
| no_new_dataset | 0.710848 |
2305.11394 | Tharindu Fernando | Tharindu Fernando and Harshala Gammulle and Sridha Sridharan and Simon
Denman and Clinton Fookes | Remembering What Is Important: A Factorised Multi-Head Retrieval and
Auxiliary Memory Stabilisation Scheme for Human Motion Prediction | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Humans exhibit complex motions that vary depending on the task that they are
performing, the interactions they engage in, as well as subject-specific
preferences. Therefore, forecasting future poses based on the history of the
previous motions is a challenging task. This paper presents an innovative
auxiliary-memory-powered deep neural network framework for the improved
modelling of historical knowledge. Specifically, we disentangle
subject-specific, task-specific, and other auxiliary information from the
observed pose sequences and utilise these factorised features to query the
memory. A novel Multi-Head knowledge retrieval scheme leverages these
factorised feature embeddings to perform multiple querying operations over the
historical observations captured within the auxiliary memory. Moreover, our
proposed dynamic masking strategy makes this feature disentanglement process
dynamic. Two novel loss functions are introduced to encourage diversity within
the auxiliary memory while ensuring the stability of the memory contents, such
that it can locate and store salient information that can aid the long-term
prediction of future motion, irrespective of data imbalances or the diversity
of the input data distribution. With extensive experiments conducted on two
public benchmarks, Human3.6M and CMU-Mocap, we demonstrate that these design
choices collectively allow the proposed approach to outperform the current
state-of-the-art methods by significant margins: $>$ 17\% on the Human3.6M
dataset and $>$ 9\% on the CMU-Mocap dataset.
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 02:44:58 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Fernando",
"Tharindu",
""
],
[
"Gammulle",
"Harshala",
""
],
[
"Sridharan",
"Sridha",
""
],
[
"Denman",
"Simon",
""
],
[
"Fookes",
"Clinton",
""
]
] | TITLE: Remembering What Is Important: A Factorised Multi-Head Retrieval and
Auxiliary Memory Stabilisation Scheme for Human Motion Prediction
ABSTRACT: Humans exhibit complex motions that vary depending on the task that they are
performing, the interactions they engage in, as well as subject-specific
preferences. Therefore, forecasting future poses based on the history of the
previous motions is a challenging task. This paper presents an innovative
auxiliary-memory-powered deep neural network framework for the improved
modelling of historical knowledge. Specifically, we disentangle
subject-specific, task-specific, and other auxiliary information from the
observed pose sequences and utilise these factorised features to query the
memory. A novel Multi-Head knowledge retrieval scheme leverages these
factorised feature embeddings to perform multiple querying operations over the
historical observations captured within the auxiliary memory. Moreover, our
proposed dynamic masking strategy makes this feature disentanglement process
dynamic. Two novel loss functions are introduced to encourage diversity within
the auxiliary memory while ensuring the stability of the memory contents, such
that it can locate and store salient information that can aid the long-term
prediction of future motion, irrespective of data imbalances or the diversity
of the input data distribution. With extensive experiments conducted on two
public benchmarks, Human3.6M and CMU-Mocap, we demonstrate that these design
choices collectively allow the proposed approach to outperform the current
state-of-the-art methods by significant margins: $>$ 17\% on the Human3.6M
dataset and $>$ 9\% on the CMU-Mocap dataset.
| no_new_dataset | 0.70825 |
2305.11407 | Jun Wen | Jun Wen, Jue Hou, Clara-Lea Bonzel, Yihan Zhao, Victor M. Castro,
Vivian S. Gainer, Dana Weisenfeld, Tianrun Cai, Yuk-Lam Ho, Vidul A.
Panickan, Lauren Costa, Chuan Hong, J. Michael Gaziano, Katherine P. Liao,
Junwei Lu, Kelly Cho, Tianxi Cai | LATTE: Label-efficient Incident Phenotyping from Longitudinal Electronic
Health Records | ERHs data | null | null | null | cs.AI | http://creativecommons.org/licenses/by/4.0/ | Electronic health record (EHR) data are increasingly used to support
real-world evidence (RWE) studies. Yet its ability to generate reliable RWE is
limited by the lack of readily available precise information on the timing of
clinical events such as the onset time of heart failure. We propose a
LAbel-efficienT incidenT phEnotyping (LATTE) algorithm to accurately annotate
the timing of clinical events from longitudinal EHR data. By leveraging the
pre-trained semantic embedding vectors from large-scale EHR data as prior
knowledge, LATTE selects predictive EHR features in a concept re-weighting
module by mining their relationship to the target event and compresses their
information into longitudinal visit embeddings through a visit attention
learning network. LATTE employs a recurrent neural network to capture the
sequential dependency between the target event and visit embeddings
before/after it. To improve label efficiency, LATTE constructs highly
informative longitudinal silver-standard labels from large-scale unlabeled
patients to perform unsupervised pre-training and semi-supervised joint
training. Finally, LATTE enhances cross-site portability via contrastive
representation learning. LATTE is evaluated on three analyses: the onset of
type-2 diabetes, heart failure, and the onset and relapses of multiple
sclerosis. We use various evaluation metrics present in the literature
including the $ABC_{gain}$, the proportion of reduction in the area between the
observed event indicator and the predicted cumulative incidences in reference
to the prediction per incident prevalence. LATTE consistently achieves
substantial improvement over benchmark methods such as SAMGEP and RETAIN in all
settings.
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 03:28:51 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Wen",
"Jun",
""
],
[
"Hou",
"Jue",
""
],
[
"Bonzel",
"Clara-Lea",
""
],
[
"Zhao",
"Yihan",
""
],
[
"Castro",
"Victor M.",
""
],
[
"Gainer",
"Vivian S.",
""
],
[
"Weisenfeld",
"Dana",
""
],
[
"Cai",
"Tianrun",
""
],
[
"Ho",
"Yuk-Lam",
""
],
[
"Panickan",
"Vidul A.",
""
],
[
"Costa",
"Lauren",
""
],
[
"Hong",
"Chuan",
""
],
[
"Gaziano",
"J. Michael",
""
],
[
"Liao",
"Katherine P.",
""
],
[
"Lu",
"Junwei",
""
],
[
"Cho",
"Kelly",
""
],
[
"Cai",
"Tianxi",
""
]
] | TITLE: LATTE: Label-efficient Incident Phenotyping from Longitudinal Electronic
Health Records
ABSTRACT: Electronic health record (EHR) data are increasingly used to support
real-world evidence (RWE) studies. Yet its ability to generate reliable RWE is
limited by the lack of readily available precise information on the timing of
clinical events such as the onset time of heart failure. We propose a
LAbel-efficienT incidenT phEnotyping (LATTE) algorithm to accurately annotate
the timing of clinical events from longitudinal EHR data. By leveraging the
pre-trained semantic embedding vectors from large-scale EHR data as prior
knowledge, LATTE selects predictive EHR features in a concept re-weighting
module by mining their relationship to the target event and compresses their
information into longitudinal visit embeddings through a visit attention
learning network. LATTE employs a recurrent neural network to capture the
sequential dependency between the target event and visit embeddings
before/after it. To improve label efficiency, LATTE constructs highly
informative longitudinal silver-standard labels from large-scale unlabeled
patients to perform unsupervised pre-training and semi-supervised joint
training. Finally, LATTE enhances cross-site portability via contrastive
representation learning. LATTE is evaluated on three analyses: the onset of
type-2 diabetes, heart failure, and the onset and relapses of multiple
sclerosis. We use various evaluation metrics present in the literature
including the $ABC_{gain}$, the proportion of reduction in the area between the
observed event indicator and the predicted cumulative incidences in reference
to the prediction per incident prevalence. LATTE consistently achieves
substantial improvement over benchmark methods such as SAMGEP and RETAIN in all
settings.
| no_new_dataset | 0.708648 |
2305.11437 | Achintha Wijesinghe | Achintha Wijesinghe, Songyang Zhang, Zhi Ding | PS-FedGAN: An Efficient Federated Learning Framework Based on Partially
Shared Generative Adversarial Networks For Data Privacy | null | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by/4.0/ | Federated Learning (FL) has emerged as an effective learning paradigm for
distributed computation owing to its strong potential in capturing underlying
data statistics while preserving data privacy. However, in cases of practical
data heterogeneity among FL clients, existing FL frameworks still exhibit
deficiency in capturing the overall feature properties of local client data
that exhibit disparate distributions. In response, generative adversarial
networks (GANs) have recently been exploited in FL to address data
heterogeneity since GANs can be integrated for data regeneration without
exposing original raw data. Despite some successes, existing GAN-related FL
frameworks often incur heavy communication cost and also elicit other privacy
concerns, which limit their applications in real scenarios. To this end, this
work proposes a novel FL framework that requires only partial GAN model
sharing. Named as PS-FedGAN, this new framework enhances the GAN releasing and
training mechanism to address heterogeneous data distributions across clients
and to strengthen privacy preservation at reduced communication cost,
especially over wireless networks. Our analysis demonstrates the convergence
and privacy benefits of the proposed PS-FEdGAN framework. Through experimental
results based on several well-known benchmark datasets, our proposed PS-FedGAN
shows great promise to tackle FL under non-IID client data distributions, while
securing data privacy and lowering communication overhead.
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 05:39:40 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Wijesinghe",
"Achintha",
""
],
[
"Zhang",
"Songyang",
""
],
[
"Ding",
"Zhi",
""
]
] | TITLE: PS-FedGAN: An Efficient Federated Learning Framework Based on Partially
Shared Generative Adversarial Networks For Data Privacy
ABSTRACT: Federated Learning (FL) has emerged as an effective learning paradigm for
distributed computation owing to its strong potential in capturing underlying
data statistics while preserving data privacy. However, in cases of practical
data heterogeneity among FL clients, existing FL frameworks still exhibit
deficiency in capturing the overall feature properties of local client data
that exhibit disparate distributions. In response, generative adversarial
networks (GANs) have recently been exploited in FL to address data
heterogeneity since GANs can be integrated for data regeneration without
exposing original raw data. Despite some successes, existing GAN-related FL
frameworks often incur heavy communication cost and also elicit other privacy
concerns, which limit their applications in real scenarios. To this end, this
work proposes a novel FL framework that requires only partial GAN model
sharing. Named as PS-FedGAN, this new framework enhances the GAN releasing and
training mechanism to address heterogeneous data distributions across clients
and to strengthen privacy preservation at reduced communication cost,
especially over wireless networks. Our analysis demonstrates the convergence
and privacy benefits of the proposed PS-FEdGAN framework. Through experimental
results based on several well-known benchmark datasets, our proposed PS-FedGAN
shows great promise to tackle FL under non-IID client data distributions, while
securing data privacy and lowering communication overhead.
| no_new_dataset | 0.710176 |
2305.11438 | Kaiqi Fu | Kaiqi Fu, Shaojun Gao, Shuju Shi, Xiaohai Tian, Wei Li, Zejun Ma | Phonetic and Prosody-aware Self-supervised Learning Approach for
Non-native Fluency Scoring | null | null | null | null | cs.CL eess.AS | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Speech fluency/disfluency can be evaluated by analyzing a range of phonetic
and prosodic features. Deep neural networks are commonly trained to map
fluency-related features into the human scores. However, the effectiveness of
deep learning-based models is constrained by the limited amount of labeled
training samples. To address this, we introduce a self-supervised learning
(SSL) approach that takes into account phonetic and prosody awareness for
fluency scoring. Specifically, we first pre-train the model using a
reconstruction loss function, by masking phones and their durations jointly on
a large amount of unlabeled speech and text prompts. We then fine-tune the
pre-trained model using human-annotated scoring data. Our experimental results,
conducted on datasets such as Speechocean762 and our non-native datasets, show
that our proposed method outperforms the baseline systems in terms of Pearson
correlation coefficients (PCC). Moreover, we also conduct an ablation study to
better understand the contribution of phonetic and prosody factors during the
pre-training stage.
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 05:39:41 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Fu",
"Kaiqi",
""
],
[
"Gao",
"Shaojun",
""
],
[
"Shi",
"Shuju",
""
],
[
"Tian",
"Xiaohai",
""
],
[
"Li",
"Wei",
""
],
[
"Ma",
"Zejun",
""
]
] | TITLE: Phonetic and Prosody-aware Self-supervised Learning Approach for
Non-native Fluency Scoring
ABSTRACT: Speech fluency/disfluency can be evaluated by analyzing a range of phonetic
and prosodic features. Deep neural networks are commonly trained to map
fluency-related features into the human scores. However, the effectiveness of
deep learning-based models is constrained by the limited amount of labeled
training samples. To address this, we introduce a self-supervised learning
(SSL) approach that takes into account phonetic and prosody awareness for
fluency scoring. Specifically, we first pre-train the model using a
reconstruction loss function, by masking phones and their durations jointly on
a large amount of unlabeled speech and text prompts. We then fine-tune the
pre-trained model using human-annotated scoring data. Our experimental results,
conducted on datasets such as Speechocean762 and our non-native datasets, show
that our proposed method outperforms the baseline systems in terms of Pearson
correlation coefficients (PCC). Moreover, we also conduct an ablation study to
better understand the contribution of phonetic and prosody factors during the
pre-training stage.
| no_new_dataset | 0.7082 |
2305.11439 | Runqi Wang | Runqi Wang, Hao Zheng, Xiaoyue Duan, Jianzhuang Liu, Yuning Lu, Tian
Wang, Songcen Xu, Baochang Zhang | Few-Shot Learning with Visual Distribution Calibration and Cross-Modal
Distribution Alignment | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Pre-trained vision-language models have inspired much research on few-shot
learning. However, with only a few training images, there exist two crucial
problems: (1) the visual feature distributions are easily distracted by
class-irrelevant information in images, and (2) the alignment between the
visual and language feature distributions is difficult. To deal with the
distraction problem, we propose a Selective Attack module, which consists of
trainable adapters that generate spatial attention maps of images to guide the
attacks on class-irrelevant image areas. By messing up these areas, the
critical features are captured and the visual distributions of image features
are calibrated. To better align the visual and language feature distributions
that describe the same object class, we propose a cross-modal distribution
alignment module, in which we introduce a vision-language prototype for each
class to align the distributions, and adopt the Earth Mover's Distance (EMD) to
optimize the prototypes. For efficient computation, the upper bound of EMD is
derived. In addition, we propose an augmentation strategy to increase the
diversity of the images and the text prompts, which can reduce overfitting to
the few-shot training images. Extensive experiments on 11 datasets demonstrate
that our method consistently outperforms prior arts in few-shot learning. The
implementation code will be available at https://github.com/bhrqw/SADA.
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 05:45:17 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Wang",
"Runqi",
""
],
[
"Zheng",
"Hao",
""
],
[
"Duan",
"Xiaoyue",
""
],
[
"Liu",
"Jianzhuang",
""
],
[
"Lu",
"Yuning",
""
],
[
"Wang",
"Tian",
""
],
[
"Xu",
"Songcen",
""
],
[
"Zhang",
"Baochang",
""
]
] | TITLE: Few-Shot Learning with Visual Distribution Calibration and Cross-Modal
Distribution Alignment
ABSTRACT: Pre-trained vision-language models have inspired much research on few-shot
learning. However, with only a few training images, there exist two crucial
problems: (1) the visual feature distributions are easily distracted by
class-irrelevant information in images, and (2) the alignment between the
visual and language feature distributions is difficult. To deal with the
distraction problem, we propose a Selective Attack module, which consists of
trainable adapters that generate spatial attention maps of images to guide the
attacks on class-irrelevant image areas. By messing up these areas, the
critical features are captured and the visual distributions of image features
are calibrated. To better align the visual and language feature distributions
that describe the same object class, we propose a cross-modal distribution
alignment module, in which we introduce a vision-language prototype for each
class to align the distributions, and adopt the Earth Mover's Distance (EMD) to
optimize the prototypes. For efficient computation, the upper bound of EMD is
derived. In addition, we propose an augmentation strategy to increase the
diversity of the images and the text prompts, which can reduce overfitting to
the few-shot training images. Extensive experiments on 11 datasets demonstrate
that our method consistently outperforms prior arts in few-shot learning. The
implementation code will be available at https://github.com/bhrqw/SADA.
| no_new_dataset | 0.709598 |
2305.11444 | Hiroki Ouchi | Hiroki Ouchi, Hiroyuki Shindo, Shoko Wakamiya, Yuki Matsuda, Naoya
Inoue, Shohei Higashiyama, Satoshi Nakamura, Taro Watanabe | Arukikata Travelogue Dataset | The application website for Arukikata Travelogue Dataset:
https://www.nii.ac.jp/dsc/idr/arukikata/ | null | null | null | cs.CL cs.AI cs.DL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We have constructed Arukikata Travelogue Dataset and released it free of
charge for academic research. This dataset is a Japanese text dataset with a
total of over 31 million words, comprising 4,672 Japanese domestic travelogues
and 9,607 overseas travelogues. Before providing our dataset, there was a
scarcity of widely available travelogue data for research purposes, and each
researcher had to prepare their own data. This hinders the replication of
existing studies and fair comparative analysis of experimental results. Our
dataset enables any researchers to conduct investigation on the same data and
to ensure transparency and reproducibility in research. In this paper, we
describe the academic significance, characteristics, and prospects of our
dataset.
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 05:53:49 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Ouchi",
"Hiroki",
""
],
[
"Shindo",
"Hiroyuki",
""
],
[
"Wakamiya",
"Shoko",
""
],
[
"Matsuda",
"Yuki",
""
],
[
"Inoue",
"Naoya",
""
],
[
"Higashiyama",
"Shohei",
""
],
[
"Nakamura",
"Satoshi",
""
],
[
"Watanabe",
"Taro",
""
]
] | TITLE: Arukikata Travelogue Dataset
ABSTRACT: We have constructed Arukikata Travelogue Dataset and released it free of
charge for academic research. This dataset is a Japanese text dataset with a
total of over 31 million words, comprising 4,672 Japanese domestic travelogues
and 9,607 overseas travelogues. Before providing our dataset, there was a
scarcity of widely available travelogue data for research purposes, and each
researcher had to prepare their own data. This hinders the replication of
existing studies and fair comparative analysis of experimental results. Our
dataset enables any researchers to conduct investigation on the same data and
to ensure transparency and reproducibility in research. In this paper, we
describe the academic significance, characteristics, and prospects of our
dataset.
| new_dataset | 0.714205 |
2305.11451 | Muhammad Abdullah Jamal | Muhammad Abdullah Jamal, Omid Mohareri | SurgMAE: Masked Autoencoders for Long Surgical Video Analysis | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | There has been a growing interest in using deep learning models for
processing long surgical videos, in order to automatically detect
clinical/operational activities and extract metrics that can enable workflow
efficiency tools and applications. However, training such models require vast
amounts of labeled data which is costly and not scalable. Recently,
self-supervised learning has been explored in computer vision community to
reduce the burden of the annotation cost. Masked autoencoders (MAE) got the
attention in self-supervised paradigm for Vision Transformers (ViTs) by
predicting the randomly masked regions given the visible patches of an image or
a video clip, and have shown superior performance on benchmark datasets.
However, the application of MAE in surgical data remains unexplored. In this
paper, we first investigate whether MAE can learn transferrable representations
in surgical video domain. We propose SurgMAE, which is a novel architecture
with a masking strategy based on sampling high spatio-temporal tokens for MAE.
We provide an empirical study of SurgMAE on two large scale long surgical video
datasets, and find that our method outperforms several baselines in low data
regime. We conduct extensive ablation studies to show the efficacy of our
approach and also demonstrate it's superior performance on UCF-101 to prove
it's generalizability in non-surgical datasets as well.
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 06:12:50 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Jamal",
"Muhammad Abdullah",
""
],
[
"Mohareri",
"Omid",
""
]
] | TITLE: SurgMAE: Masked Autoencoders for Long Surgical Video Analysis
ABSTRACT: There has been a growing interest in using deep learning models for
processing long surgical videos, in order to automatically detect
clinical/operational activities and extract metrics that can enable workflow
efficiency tools and applications. However, training such models require vast
amounts of labeled data which is costly and not scalable. Recently,
self-supervised learning has been explored in computer vision community to
reduce the burden of the annotation cost. Masked autoencoders (MAE) got the
attention in self-supervised paradigm for Vision Transformers (ViTs) by
predicting the randomly masked regions given the visible patches of an image or
a video clip, and have shown superior performance on benchmark datasets.
However, the application of MAE in surgical data remains unexplored. In this
paper, we first investigate whether MAE can learn transferrable representations
in surgical video domain. We propose SurgMAE, which is a novel architecture
with a masking strategy based on sampling high spatio-temporal tokens for MAE.
We provide an empirical study of SurgMAE on two large scale long surgical video
datasets, and find that our method outperforms several baselines in low data
regime. We conduct extensive ablation studies to show the efficacy of our
approach and also demonstrate it's superior performance on UCF-101 to prove
it's generalizability in non-surgical datasets as well.
| no_new_dataset | 0.709007 |
2305.11460 | Shiyao Ding | Shiyao Ding and Takayuki Ito | Self-Agreement: A Framework for Fine-tuning Language Models to Find
Agreement among Diverse Opinions | null | null | null | null | cs.CL cs.AI cs.MA | http://creativecommons.org/licenses/by/4.0/ | Finding an agreement among diverse opinions is a challenging topic in
multiagent systems. Recently, large language models (LLMs) have shown great
potential in addressing this challenge due to their remarkable capabilities in
comprehending human opinions and generating human-like text. However, they
typically rely on extensive human-annotated data. In this paper, we propose
Self-Agreement, a novel framework for fine-tuning LLMs to autonomously find
agreement using data generated by LLM itself. Specifically, our approach
employs the generative pre-trained transformer-3 (GPT-3) to generate multiple
opinions for each question in a question dataset and create several agreement
candidates among these opinions. Then, a bidirectional encoder representations
from transformers (BERT)-based model evaluates the agreement score of each
agreement candidate and selects the one with the highest agreement score. This
process yields a dataset of question-opinion-agreements, which we use to
fine-tune a pre-trained LLM for discovering agreements among diverse opinions.
Remarkably, a pre-trained LLM fine-tuned by our Self-Agreement framework
achieves comparable performance to GPT-3 with only 1/25 of its parameters,
showcasing its ability to identify agreement among various opinions without the
need for human-annotated data.
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 06:27:16 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Ding",
"Shiyao",
""
],
[
"Ito",
"Takayuki",
""
]
] | TITLE: Self-Agreement: A Framework for Fine-tuning Language Models to Find
Agreement among Diverse Opinions
ABSTRACT: Finding an agreement among diverse opinions is a challenging topic in
multiagent systems. Recently, large language models (LLMs) have shown great
potential in addressing this challenge due to their remarkable capabilities in
comprehending human opinions and generating human-like text. However, they
typically rely on extensive human-annotated data. In this paper, we propose
Self-Agreement, a novel framework for fine-tuning LLMs to autonomously find
agreement using data generated by LLM itself. Specifically, our approach
employs the generative pre-trained transformer-3 (GPT-3) to generate multiple
opinions for each question in a question dataset and create several agreement
candidates among these opinions. Then, a bidirectional encoder representations
from transformers (BERT)-based model evaluates the agreement score of each
agreement candidate and selects the one with the highest agreement score. This
process yields a dataset of question-opinion-agreements, which we use to
fine-tune a pre-trained LLM for discovering agreements among diverse opinions.
Remarkably, a pre-trained LLM fine-tuned by our Self-Agreement framework
achieves comparable performance to GPT-3 with only 1/25 of its parameters,
showcasing its ability to identify agreement among various opinions without the
need for human-annotated data.
| new_dataset | 0.716609 |
2305.11462 | Anupiya Nugaliyadde Dr | Anupiya Nugaliyadde | Extending Memory for Language Modelling | null | null | null | null | cs.CL | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Breakthroughs in deep learning and memory networks have made major advances
in natural language understanding. Language is sequential and information
carried through the sequence can be captured through memory networks. Learning
the sequence is one of the key aspects in learning the language. However,
memory networks are not capable of holding infinitely long sequences in their
memories and are limited by various constraints such as the vanishing or
exploding gradient problem. Therefore, natural language understanding models
are affected when presented with long sequential text. We introduce Long Term
Memory network (LTM) to learn from infinitely long sequences. LTM gives
priority to the current inputs to allow it to have a high impact. Language
modeling is an important factor in natural language understanding. LTM was
tested in language modeling, which requires long term memory. LTM is tested on
Penn Tree bank dataset, Google Billion Word dataset and WikiText-2 dataset. We
compare LTM with other language models which require long term memory.
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 06:30:19 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Nugaliyadde",
"Anupiya",
""
]
] | TITLE: Extending Memory for Language Modelling
ABSTRACT: Breakthroughs in deep learning and memory networks have made major advances
in natural language understanding. Language is sequential and information
carried through the sequence can be captured through memory networks. Learning
the sequence is one of the key aspects in learning the language. However,
memory networks are not capable of holding infinitely long sequences in their
memories and are limited by various constraints such as the vanishing or
exploding gradient problem. Therefore, natural language understanding models
are affected when presented with long sequential text. We introduce Long Term
Memory network (LTM) to learn from infinitely long sequences. LTM gives
priority to the current inputs to allow it to have a high impact. Language
modeling is an important factor in natural language understanding. LTM was
tested in language modeling, which requires long term memory. LTM is tested on
Penn Tree bank dataset, Google Billion Word dataset and WikiText-2 dataset. We
compare LTM with other language models which require long term memory.
| no_new_dataset | 0.711281 |
2305.11482 | Yihong Tang | Yihong Tang, Bo Wang, Miao Fang, Dongming Zhao, Kun Huang, Ruifang He,
Yuexian Hou | Enhancing Personalized Dialogue Generation with Contrastive Latent
Variables: Combining Sparse and Dense Persona | ACL 2023 | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The personalized dialogue explores the consistent relationship between
dialogue generation and personality. Existing personalized dialogue agents
model persona profiles from three resources: sparse or dense persona
descriptions and dialogue histories. However, sparse structured persona
attributes are explicit but uninformative, dense persona texts contain rich
persona descriptions with much noise, and dialogue history query is both noisy
and uninformative for persona modeling. In this work, we combine the advantages
of the three resources to obtain a richer and more accurate persona. We design
a Contrastive Latent Variable-based model (CLV) that clusters the dense persona
descriptions into sparse categories, which are combined with the history query
to generate personalized responses. Experimental results on Chinese and English
datasets demonstrate our model's superiority in personalization.
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 07:24:27 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Tang",
"Yihong",
""
],
[
"Wang",
"Bo",
""
],
[
"Fang",
"Miao",
""
],
[
"Zhao",
"Dongming",
""
],
[
"Huang",
"Kun",
""
],
[
"He",
"Ruifang",
""
],
[
"Hou",
"Yuexian",
""
]
] | TITLE: Enhancing Personalized Dialogue Generation with Contrastive Latent
Variables: Combining Sparse and Dense Persona
ABSTRACT: The personalized dialogue explores the consistent relationship between
dialogue generation and personality. Existing personalized dialogue agents
model persona profiles from three resources: sparse or dense persona
descriptions and dialogue histories. However, sparse structured persona
attributes are explicit but uninformative, dense persona texts contain rich
persona descriptions with much noise, and dialogue history query is both noisy
and uninformative for persona modeling. In this work, we combine the advantages
of the three resources to obtain a richer and more accurate persona. We design
a Contrastive Latent Variable-based model (CLV) that clusters the dense persona
descriptions into sparse categories, which are combined with the history query
to generate personalized responses. Experimental results on Chinese and English
datasets demonstrate our model's superiority in personalization.
| no_new_dataset | 0.712589 |
2305.11497 | Chenchi Zhang | Chenchi Zhang, Jun Xiao, Lei Chen, Jian Shao, Long Chen | TreePrompt: Learning to Compose Tree Prompts for Explainable Visual
Grounding | null | null | null | null | cs.CV cs.AI cs.CL cs.MM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Prompt tuning has achieved great success in transferring the knowledge from
large pretrained vision-language models into downstream tasks, and has
dominated the performance on visual grounding (VG). However, almost all
existing prompt tuning paradigms suffer from poor interpretability. In this
paper, we argue that their poor interpretability is attributed to the holistic
prompt generation and inference process. By "holistic", we mean that they
usually directly learn a set of vectors as the prompt (i.e., prompt
generation), and use the learned global prompt to augment the textual input for
the VG model (i.e., prompt inference). To this end, we propose a new prompt
construction paradigm with explicit explainable ability, named TreePrompt.
Specifically, we first deconstruct a complex sentence into a tree, that is
consistent with human reasoning. Then, following the syntax tree, we compose a
structured prompt in a bottom-up manner. Thanks to this step-by-step prompt
construction process, each intermediate prompt (i.e., tree node) permits us to
understand the reasoning process. Extensive ablations on various backbones and
benchmarks consistently demonstrate the effectiveness and interpretability of
our TreePrompt.
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 07:52:22 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Zhang",
"Chenchi",
""
],
[
"Xiao",
"Jun",
""
],
[
"Chen",
"Lei",
""
],
[
"Shao",
"Jian",
""
],
[
"Chen",
"Long",
""
]
] | TITLE: TreePrompt: Learning to Compose Tree Prompts for Explainable Visual
Grounding
ABSTRACT: Prompt tuning has achieved great success in transferring the knowledge from
large pretrained vision-language models into downstream tasks, and has
dominated the performance on visual grounding (VG). However, almost all
existing prompt tuning paradigms suffer from poor interpretability. In this
paper, we argue that their poor interpretability is attributed to the holistic
prompt generation and inference process. By "holistic", we mean that they
usually directly learn a set of vectors as the prompt (i.e., prompt
generation), and use the learned global prompt to augment the textual input for
the VG model (i.e., prompt inference). To this end, we propose a new prompt
construction paradigm with explicit explainable ability, named TreePrompt.
Specifically, we first deconstruct a complex sentence into a tree, that is
consistent with human reasoning. Then, following the syntax tree, we compose a
structured prompt in a bottom-up manner. Thanks to this step-by-step prompt
construction process, each intermediate prompt (i.e., tree node) permits us to
understand the reasoning process. Extensive ablations on various backbones and
benchmarks consistently demonstrate the effectiveness and interpretability of
our TreePrompt.
| no_new_dataset | 0.71027 |
2305.11498 | Xingyu Bai | Xingyu Bai, Taiqiang Wu, Han Guo, Zhe Zhao, Xuefeng Yang, Jiayi Li,
Weijie Liu, Qi Ju, Weigang Guo, Yujiu Yang | Recouple Event Field via Probabilistic Bias for Event Extraction | Published in: ICASSP 2023 - 2023 IEEE International Conference on
Acoustics, Speech and Signal Processing (ICASSP) | null | null | null | cs.CL cs.AI cs.IR | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Event Extraction (EE), aiming to identify and classify event triggers and
arguments from event mentions, has benefited from pre-trained language models
(PLMs). However, existing PLM-based methods ignore the information of
trigger/argument fields, which is crucial for understanding event schemas. To
this end, we propose a Probabilistic reCoupling model enhanced Event extraction
framework (ProCE). Specifically, we first model the syntactic-related event
fields as probabilistic biases, to clarify the event fields from ambiguous
entanglement. Furthermore, considering multiple occurrences of the same
triggers/arguments in EE, we explore probabilistic interaction strategies among
multiple fields of the same triggers/arguments, to recouple the corresponding
clarified distributions and capture more latent information fields. Experiments
on EE datasets demonstrate the effectiveness and generalization of our proposed
approach.
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 07:55:37 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Bai",
"Xingyu",
""
],
[
"Wu",
"Taiqiang",
""
],
[
"Guo",
"Han",
""
],
[
"Zhao",
"Zhe",
""
],
[
"Yang",
"Xuefeng",
""
],
[
"Li",
"Jiayi",
""
],
[
"Liu",
"Weijie",
""
],
[
"Ju",
"Qi",
""
],
[
"Guo",
"Weigang",
""
],
[
"Yang",
"Yujiu",
""
]
] | TITLE: Recouple Event Field via Probabilistic Bias for Event Extraction
ABSTRACT: Event Extraction (EE), aiming to identify and classify event triggers and
arguments from event mentions, has benefited from pre-trained language models
(PLMs). However, existing PLM-based methods ignore the information of
trigger/argument fields, which is crucial for understanding event schemas. To
this end, we propose a Probabilistic reCoupling model enhanced Event extraction
framework (ProCE). Specifically, we first model the syntactic-related event
fields as probabilistic biases, to clarify the event fields from ambiguous
entanglement. Furthermore, considering multiple occurrences of the same
triggers/arguments in EE, we explore probabilistic interaction strategies among
multiple fields of the same triggers/arguments, to recouple the corresponding
clarified distributions and capture more latent information fields. Experiments
on EE datasets demonstrate the effectiveness and generalization of our proposed
approach.
| no_new_dataset | 0.710653 |
2305.11501 | Yu Zhao | Yu Zhao, Yike Wu, Xiangrui Cai, Ying Zhang, Haiwei Zhang, Xiaojie Yuan | From Alignment to Entailment: A Unified Textual Entailment Framework for
Entity Alignment | Accepted by ACL 2023 Findings | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Entity Alignment (EA) aims to find the equivalent entities between two
Knowledge Graphs (KGs). Existing methods usually encode the triples of entities
as embeddings and learn to align the embeddings, which prevents the direct
interaction between the original information of the cross-KG entities.
Moreover, they encode the relational triples and attribute triples of an entity
in heterogeneous embedding spaces, which prevents them from helping each other.
In this paper, we transform both triples into unified textual sequences, and
model the EA task as a bi-directional textual entailment task between the
sequences of cross-KG entities. Specifically, we feed the sequences of two
entities simultaneously into a pre-trained language model (PLM) and propose two
kinds of PLM-based entity aligners that model the entailment probability
between sequences as the similarity between entities. Our approach captures the
unified correlation pattern of two kinds of information between entities, and
explicitly models the fine-grained interaction between original entity
information. The experiments on five cross-lingual EA datasets show that our
approach outperforms the state-of-the-art EA methods and enables the mutual
enhancement of the heterogeneous information. Codes are available at
https://github.com/OreOZhao/TEA.
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 08:06:50 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Zhao",
"Yu",
""
],
[
"Wu",
"Yike",
""
],
[
"Cai",
"Xiangrui",
""
],
[
"Zhang",
"Ying",
""
],
[
"Zhang",
"Haiwei",
""
],
[
"Yuan",
"Xiaojie",
""
]
] | TITLE: From Alignment to Entailment: A Unified Textual Entailment Framework for
Entity Alignment
ABSTRACT: Entity Alignment (EA) aims to find the equivalent entities between two
Knowledge Graphs (KGs). Existing methods usually encode the triples of entities
as embeddings and learn to align the embeddings, which prevents the direct
interaction between the original information of the cross-KG entities.
Moreover, they encode the relational triples and attribute triples of an entity
in heterogeneous embedding spaces, which prevents them from helping each other.
In this paper, we transform both triples into unified textual sequences, and
model the EA task as a bi-directional textual entailment task between the
sequences of cross-KG entities. Specifically, we feed the sequences of two
entities simultaneously into a pre-trained language model (PLM) and propose two
kinds of PLM-based entity aligners that model the entailment probability
between sequences as the similarity between entities. Our approach captures the
unified correlation pattern of two kinds of information between entities, and
explicitly models the fine-grained interaction between original entity
information. The experiments on five cross-lingual EA datasets show that our
approach outperforms the state-of-the-art EA methods and enables the mutual
enhancement of the heterogeneous information. Codes are available at
https://github.com/OreOZhao/TEA.
| no_new_dataset | 0.709988 |
2305.11503 | Xiuying Chen | Xiuying Chen, Mingzhe Li, Shen Gao, Xin Cheng, Qiang Yang, Qishen
Zhang, Xin Gao, Xiangliang Zhang | A Topic-aware Summarization Framework with Different Modal Side
Information | SIGIR 2023, 10 pages | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Automatic summarization plays an important role in the exponential document
growth on the Web. On content websites such as CNN.com and WikiHow.com, there
often exist various kinds of side information along with the main document for
attention attraction and easier understanding, such as videos, images, and
queries. Such information can be used for better summarization, as they often
explicitly or implicitly mention the essence of the article. However, most of
the existing side-aware summarization methods are designed to incorporate
either single-modal or multi-modal side information, and cannot effectively
adapt to each other. In this paper, we propose a general summarization
framework, which can flexibly incorporate various modalities of side
information. The main challenges in designing a flexible summarization model
with side information include: (1) the side information can be in textual or
visual format, and the model needs to align and unify it with the document into
the same semantic space, (2) the side inputs can contain information from
various aspects, and the model should recognize the aspects useful for
summarization. To address these two challenges, we first propose a unified
topic encoder, which jointly discovers latent topics from the document and
various kinds of side information. The learned topics flexibly bridge and guide
the information flow between multiple inputs in a graph encoder through a
topic-aware interaction. We secondly propose a triplet contrastive learning
mechanism to align the single-modal or multi-modal information into a unified
semantic space, where the summary quality is enhanced by better understanding
the document and side information. Results show that our model significantly
surpasses strong baselines on three public single-modal or multi-modal
benchmark summarization datasets.
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 08:09:45 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Chen",
"Xiuying",
""
],
[
"Li",
"Mingzhe",
""
],
[
"Gao",
"Shen",
""
],
[
"Cheng",
"Xin",
""
],
[
"Yang",
"Qiang",
""
],
[
"Zhang",
"Qishen",
""
],
[
"Gao",
"Xin",
""
],
[
"Zhang",
"Xiangliang",
""
]
] | TITLE: A Topic-aware Summarization Framework with Different Modal Side
Information
ABSTRACT: Automatic summarization plays an important role in the exponential document
growth on the Web. On content websites such as CNN.com and WikiHow.com, there
often exist various kinds of side information along with the main document for
attention attraction and easier understanding, such as videos, images, and
queries. Such information can be used for better summarization, as they often
explicitly or implicitly mention the essence of the article. However, most of
the existing side-aware summarization methods are designed to incorporate
either single-modal or multi-modal side information, and cannot effectively
adapt to each other. In this paper, we propose a general summarization
framework, which can flexibly incorporate various modalities of side
information. The main challenges in designing a flexible summarization model
with side information include: (1) the side information can be in textual or
visual format, and the model needs to align and unify it with the document into
the same semantic space, (2) the side inputs can contain information from
various aspects, and the model should recognize the aspects useful for
summarization. To address these two challenges, we first propose a unified
topic encoder, which jointly discovers latent topics from the document and
various kinds of side information. The learned topics flexibly bridge and guide
the information flow between multiple inputs in a graph encoder through a
topic-aware interaction. We secondly propose a triplet contrastive learning
mechanism to align the single-modal or multi-modal information into a unified
semantic space, where the summary quality is enhanced by better understanding
the document and side information. Results show that our model significantly
surpasses strong baselines on three public single-modal or multi-modal
benchmark summarization datasets.
| no_new_dataset | 0.710672 |
2305.11504 | He Huaqing | Huaqing He, Li Lin, Zhiyuan Cai, Pujin Cheng, Xiaoying Tang | JOINEDTrans: Prior Guided Multi-task Transformer for Joint Optic
Disc/Cup Segmentation and Fovea Detection | 11 pages, 6 figures | null | null | null | eess.IV cs.CV cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Deep learning-based image segmentation and detection models have largely
improved the efficiency of analyzing retinal landmarks such as optic disc (OD),
optic cup (OC), and fovea. However, factors including ophthalmic
disease-related lesions and low image quality issues may severely complicate
automatic OD/OC segmentation and fovea detection. Most existing works treat the
identification of each landmark as a single task, and take into account no
prior information. To address these issues, we propose a prior guided
multi-task transformer framework for joint OD/OC segmentation and fovea
detection, named JOINEDTrans. JOINEDTrans effectively combines various spatial
features of the fundus images, relieving the structural distortions induced by
lesions and other imaging issues. It contains a segmentation branch and a
detection branch. To be noted, we employ an encoder pretrained in a vessel
segmentation task to effectively exploit the positional relationship among
vessel, OD/OC, and fovea, successfully incorporating spatial prior into the
proposed JOINEDTrans framework. There are a coarse stage and a fine stage in
JOINEDTrans. In the coarse stage, OD/OC coarse segmentation and fovea heatmap
localization are obtained through a joint segmentation and detection module. In
the fine stage, we crop regions of interest for subsequent refinement and use
predictions obtained in the coarse stage to provide additional information for
better performance and faster convergence. Experimental results demonstrate
that JOINEDTrans outperforms existing state-of-the-art methods on the publicly
available GAMMA, REFUGE, and PALM fundus image datasets. We make our code
available at https://github.com/HuaqingHe/JOINEDTrans
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 08:10:43 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"He",
"Huaqing",
""
],
[
"Lin",
"Li",
""
],
[
"Cai",
"Zhiyuan",
""
],
[
"Cheng",
"Pujin",
""
],
[
"Tang",
"Xiaoying",
""
]
] | TITLE: JOINEDTrans: Prior Guided Multi-task Transformer for Joint Optic
Disc/Cup Segmentation and Fovea Detection
ABSTRACT: Deep learning-based image segmentation and detection models have largely
improved the efficiency of analyzing retinal landmarks such as optic disc (OD),
optic cup (OC), and fovea. However, factors including ophthalmic
disease-related lesions and low image quality issues may severely complicate
automatic OD/OC segmentation and fovea detection. Most existing works treat the
identification of each landmark as a single task, and take into account no
prior information. To address these issues, we propose a prior guided
multi-task transformer framework for joint OD/OC segmentation and fovea
detection, named JOINEDTrans. JOINEDTrans effectively combines various spatial
features of the fundus images, relieving the structural distortions induced by
lesions and other imaging issues. It contains a segmentation branch and a
detection branch. To be noted, we employ an encoder pretrained in a vessel
segmentation task to effectively exploit the positional relationship among
vessel, OD/OC, and fovea, successfully incorporating spatial prior into the
proposed JOINEDTrans framework. There are a coarse stage and a fine stage in
JOINEDTrans. In the coarse stage, OD/OC coarse segmentation and fovea heatmap
localization are obtained through a joint segmentation and detection module. In
the fine stage, we crop regions of interest for subsequent refinement and use
predictions obtained in the coarse stage to provide additional information for
better performance and faster convergence. Experimental results demonstrate
that JOINEDTrans outperforms existing state-of-the-art methods on the publicly
available GAMMA, REFUGE, and PALM fundus image datasets. We make our code
available at https://github.com/HuaqingHe/JOINEDTrans
| no_new_dataset | 0.712401 |
2305.11513 | Aaron Jie | Leiping Jie, Hui Zhang | When SAM Meets Shadow Detection | Technical Report | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | As a promptable generic object segmentation model, segment anything model
(SAM) has recently attracted significant attention, and also demonstrates its
powerful performance. Nevertheless, it still meets its Waterloo when
encountering several tasks, e.g., medical image segmentation, camouflaged
object detection, etc. In this report, we try SAM on an unexplored popular
task: shadow detection. Specifically, four benchmarks were chosen and evaluated
with widely used metrics. The experimental results show that the performance
for shadow detection using SAM is not satisfactory, especially when comparing
with the elaborate models. Code is available at
https://github.com/LeipingJie/SAMSh.
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 08:26:08 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Jie",
"Leiping",
""
],
[
"Zhang",
"Hui",
""
]
] | TITLE: When SAM Meets Shadow Detection
ABSTRACT: As a promptable generic object segmentation model, segment anything model
(SAM) has recently attracted significant attention, and also demonstrates its
powerful performance. Nevertheless, it still meets its Waterloo when
encountering several tasks, e.g., medical image segmentation, camouflaged
object detection, etc. In this report, we try SAM on an unexplored popular
task: shadow detection. Specifically, four benchmarks were chosen and evaluated
with widely used metrics. The experimental results show that the performance
for shadow detection using SAM is not satisfactory, especially when comparing
with the elaborate models. Code is available at
https://github.com/LeipingJie/SAMSh.
| no_new_dataset | 0.711281 |
2305.11522 | Heyuan Li | Heyuan Li, Bo Wang, Yu Cheng, Mohan Kankanhalli, Robby T. Tan | DSFNet: Dual Space Fusion Network for Occlusion-Robust 3D Dense Face
Alignment | Accepted into CVPR'23 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Sensitivity to severe occlusion and large view angles limits the usage
scenarios of the existing monocular 3D dense face alignment methods. The
state-of-the-art 3DMM-based method, directly regresses the model's
coefficients, underutilizing the low-level 2D spatial and semantic information,
which can actually offer cues for face shape and orientation. In this work, we
demonstrate how modeling 3D facial geometry in image and model space jointly
can solve the occlusion and view angle problems. Instead of predicting the
whole face directly, we regress image space features in the visible facial
region by dense prediction first. Subsequently, we predict our model's
coefficients based on the regressed feature of the visible regions, leveraging
the prior knowledge of whole face geometry from the morphable models to
complete the invisible regions. We further propose a fusion network that
combines the advantages of both the image and model space predictions to
achieve high robustness and accuracy in unconstrained scenarios. Thanks to the
proposed fusion module, our method is robust not only to occlusion and large
pitch and roll view angles, which is the benefit of our image space approach,
but also to noise and large yaw angles, which is the benefit of our model space
method. Comprehensive evaluations demonstrate the superior performance of our
method compared with the state-of-the-art methods. On the 3D dense face
alignment task, we achieve 3.80% NME on the AFLW2000-3D dataset, which
outperforms the state-of-the-art method by 5.5%. Code is available at
https://github.com/lhyfst/DSFNet.
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 08:43:37 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Li",
"Heyuan",
""
],
[
"Wang",
"Bo",
""
],
[
"Cheng",
"Yu",
""
],
[
"Kankanhalli",
"Mohan",
""
],
[
"Tan",
"Robby T.",
""
]
] | TITLE: DSFNet: Dual Space Fusion Network for Occlusion-Robust 3D Dense Face
Alignment
ABSTRACT: Sensitivity to severe occlusion and large view angles limits the usage
scenarios of the existing monocular 3D dense face alignment methods. The
state-of-the-art 3DMM-based method, directly regresses the model's
coefficients, underutilizing the low-level 2D spatial and semantic information,
which can actually offer cues for face shape and orientation. In this work, we
demonstrate how modeling 3D facial geometry in image and model space jointly
can solve the occlusion and view angle problems. Instead of predicting the
whole face directly, we regress image space features in the visible facial
region by dense prediction first. Subsequently, we predict our model's
coefficients based on the regressed feature of the visible regions, leveraging
the prior knowledge of whole face geometry from the morphable models to
complete the invisible regions. We further propose a fusion network that
combines the advantages of both the image and model space predictions to
achieve high robustness and accuracy in unconstrained scenarios. Thanks to the
proposed fusion module, our method is robust not only to occlusion and large
pitch and roll view angles, which is the benefit of our image space approach,
but also to noise and large yaw angles, which is the benefit of our model space
method. Comprehensive evaluations demonstrate the superior performance of our
method compared with the state-of-the-art methods. On the 3D dense face
alignment task, we achieve 3.80% NME on the AFLW2000-3D dataset, which
outperforms the state-of-the-art method by 5.5%. Code is available at
https://github.com/lhyfst/DSFNet.
| no_new_dataset | 0.710478 |
2305.11529 | Hafida Benhidour | Hanan S. Murayshid, Hafida Benhidour, Said Kerrache | A Sequence-to-Sequence Approach for Arabic Pronoun Resolution | null | null | null | null | cs.CL cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper proposes a sequence-to-sequence learning approach for Arabic
pronoun resolution, which explores the effectiveness of using advanced natural
language processing (NLP) techniques, specifically Bi-LSTM and the BERT
pre-trained Language Model, in solving the pronoun resolution problem in
Arabic. The proposed approach is evaluated on the AnATAr dataset, and its
performance is compared to several baseline models, including traditional
machine learning models and handcrafted feature-based models. Our results
demonstrate that the proposed model outperforms the baseline models, which
include KNN, logistic regression, and SVM, across all metrics. In addition, we
explore the effectiveness of various modifications to the model, including
concatenating the anaphor text beside the paragraph text as input, adding a
mask to focus on candidate scores, and filtering candidates based on gender and
number agreement with the anaphor. Our results show that these modifications
significantly improve the model's performance, achieving up to 81% on MRR and
71% for F1 score while also demonstrating higher precision, recall, and
accuracy. These findings suggest that the proposed model is an effective
approach to Arabic pronoun resolution and highlights the potential benefits of
leveraging advanced NLP neural models.
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 08:53:41 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Murayshid",
"Hanan S.",
""
],
[
"Benhidour",
"Hafida",
""
],
[
"Kerrache",
"Said",
""
]
] | TITLE: A Sequence-to-Sequence Approach for Arabic Pronoun Resolution
ABSTRACT: This paper proposes a sequence-to-sequence learning approach for Arabic
pronoun resolution, which explores the effectiveness of using advanced natural
language processing (NLP) techniques, specifically Bi-LSTM and the BERT
pre-trained Language Model, in solving the pronoun resolution problem in
Arabic. The proposed approach is evaluated on the AnATAr dataset, and its
performance is compared to several baseline models, including traditional
machine learning models and handcrafted feature-based models. Our results
demonstrate that the proposed model outperforms the baseline models, which
include KNN, logistic regression, and SVM, across all metrics. In addition, we
explore the effectiveness of various modifications to the model, including
concatenating the anaphor text beside the paragraph text as input, adding a
mask to focus on candidate scores, and filtering candidates based on gender and
number agreement with the anaphor. Our results show that these modifications
significantly improve the model's performance, achieving up to 81% on MRR and
71% for F1 score while also demonstrating higher precision, recall, and
accuracy. These findings suggest that the proposed model is an effective
approach to Arabic pronoun resolution and highlights the potential benefits of
leveraging advanced NLP neural models.
| no_new_dataset | 0.711387 |
2305.11536 | Piyush Kumar Garg | Piyush Kumar Garg, Roshni Chakraborty, and Sourav Kumar Dandapat | PORTRAIT: a hybrid aPproach tO cReate extractive ground-TRuth summAry
for dIsaster evenT | null | null | null | null | cs.CL cs.SI | http://creativecommons.org/licenses/by/4.0/ | Disaster summarization approaches provide an overview of the important
information posted during disaster events on social media platforms, such as,
Twitter. However, the type of information posted significantly varies across
disasters depending on several factors like the location, type, severity, etc.
Verification of the effectiveness of disaster summarization approaches still
suffer due to the lack of availability of good spectrum of datasets along with
the ground-truth summary. Existing approaches for ground-truth summary
generation (ground-truth for extractive summarization) relies on the wisdom and
intuition of the annotators. Annotators are provided with a complete set of
input tweets from which a subset of tweets is selected by the annotators for
the summary. This process requires immense human effort and significant time.
Additionally, this intuition-based selection of the tweets might lead to a high
variance in summaries generated across annotators. Therefore, to handle these
challenges, we propose a hybrid (semi-automated) approach (PORTRAIT) where we
partly automate the ground-truth summary generation procedure. This approach
reduces the effort and time of the annotators while ensuring the quality of the
created ground-truth summary. We validate the effectiveness of PORTRAIT on 5
disaster events through quantitative and qualitative comparisons of
ground-truth summaries generated by existing intuitive approaches, a
semi-automated approach, and PORTRAIT. We prepare and release the ground-truth
summaries for 5 disaster events which consist of both natural and man-made
disaster events belonging to 4 different countries. Finally, we provide a study
about the performance of various state-of-the-art summarization approaches on
the ground-truth summaries generated by PORTRAIT using ROUGE-N F1-scores.
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 09:07:52 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Garg",
"Piyush Kumar",
""
],
[
"Chakraborty",
"Roshni",
""
],
[
"Dandapat",
"Sourav Kumar",
""
]
] | TITLE: PORTRAIT: a hybrid aPproach tO cReate extractive ground-TRuth summAry
for dIsaster evenT
ABSTRACT: Disaster summarization approaches provide an overview of the important
information posted during disaster events on social media platforms, such as,
Twitter. However, the type of information posted significantly varies across
disasters depending on several factors like the location, type, severity, etc.
Verification of the effectiveness of disaster summarization approaches still
suffer due to the lack of availability of good spectrum of datasets along with
the ground-truth summary. Existing approaches for ground-truth summary
generation (ground-truth for extractive summarization) relies on the wisdom and
intuition of the annotators. Annotators are provided with a complete set of
input tweets from which a subset of tweets is selected by the annotators for
the summary. This process requires immense human effort and significant time.
Additionally, this intuition-based selection of the tweets might lead to a high
variance in summaries generated across annotators. Therefore, to handle these
challenges, we propose a hybrid (semi-automated) approach (PORTRAIT) where we
partly automate the ground-truth summary generation procedure. This approach
reduces the effort and time of the annotators while ensuring the quality of the
created ground-truth summary. We validate the effectiveness of PORTRAIT on 5
disaster events through quantitative and qualitative comparisons of
ground-truth summaries generated by existing intuitive approaches, a
semi-automated approach, and PORTRAIT. We prepare and release the ground-truth
summaries for 5 disaster events which consist of both natural and man-made
disaster events belonging to 4 different countries. Finally, we provide a study
about the performance of various state-of-the-art summarization approaches on
the ground-truth summaries generated by PORTRAIT using ROUGE-N F1-scores.
| no_new_dataset | 0.712026 |
2305.11556 | Sira Vegas | Sira Vegas, Sebastian Elbaum | Pitfalls in Experiments with DNN4SE: An Analysis of the State of the
Practice | null | null | null | null | cs.SE | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Software engineering techniques are increasingly relying on deep learning
approaches to support many software engineering tasks, from bug triaging to
code generation. To assess the efficacy of such techniques researchers
typically perform controlled experiments. Conducting these experiments,
however, is particularly challenging given the complexity of the space of
variables involved, from specialized and intricate architectures and algorithms
to a large number of training hyper-parameters and choices of evolving
datasets, all compounded by how rapidly the machine learning technology is
advancing, and the inherent sources of randomness in the training process. In
this work we conduct a mapping study, examining 194 experiments with techniques
that rely on deep neural networks appearing in 55 papers published in premier
software engineering venues to provide a characterization of the
state-of-the-practice, pinpointing experiments common trends and pitfalls. Our
study reveals that most of the experiments, including those that have received
ACM artifact badges, have fundamental limitations that raise doubts about the
reliability of their findings. More specifically, we find: weak analyses to
determine that there is a true relationship between independent and dependent
variables (87% of the experiments); limited control over the space of DNN
relevant variables, which can render a relationship between dependent variables
and treatments that may not be causal but rather correlational (100% of the
experiments); and lack of specificity in terms of what are the DNN variables
and their values utilized in the experiments (86% of the experiments) to define
the treatments being applied, which makes it unclear whether the techniques
designed are the ones being assessed, or how the sources of extraneous
variation are controlled. We provide some practical recommendations to address
these limitations.
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 09:55:48 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Vegas",
"Sira",
""
],
[
"Elbaum",
"Sebastian",
""
]
] | TITLE: Pitfalls in Experiments with DNN4SE: An Analysis of the State of the
Practice
ABSTRACT: Software engineering techniques are increasingly relying on deep learning
approaches to support many software engineering tasks, from bug triaging to
code generation. To assess the efficacy of such techniques researchers
typically perform controlled experiments. Conducting these experiments,
however, is particularly challenging given the complexity of the space of
variables involved, from specialized and intricate architectures and algorithms
to a large number of training hyper-parameters and choices of evolving
datasets, all compounded by how rapidly the machine learning technology is
advancing, and the inherent sources of randomness in the training process. In
this work we conduct a mapping study, examining 194 experiments with techniques
that rely on deep neural networks appearing in 55 papers published in premier
software engineering venues to provide a characterization of the
state-of-the-practice, pinpointing experiments common trends and pitfalls. Our
study reveals that most of the experiments, including those that have received
ACM artifact badges, have fundamental limitations that raise doubts about the
reliability of their findings. More specifically, we find: weak analyses to
determine that there is a true relationship between independent and dependent
variables (87% of the experiments); limited control over the space of DNN
relevant variables, which can render a relationship between dependent variables
and treatments that may not be causal but rather correlational (100% of the
experiments); and lack of specificity in terms of what are the DNN variables
and their values utilized in the experiments (86% of the experiments) to define
the treatments being applied, which makes it unclear whether the techniques
designed are the ones being assessed, or how the sources of extraneous
variation are controlled. We provide some practical recommendations to address
these limitations.
| no_new_dataset | 0.705164 |
2305.11560 | Matteo Ferrante | Matteo Ferrante, Furkan Ozcelik, Tommaso Boccato, Rufin VanRullen,
Nicola Toschi | Brain Captioning: Decoding human brain activity into images and text | null | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by/4.0/ | Every day, the human brain processes an immense volume of visual information,
relying on intricate neural mechanisms to perceive and interpret these stimuli.
Recent breakthroughs in functional magnetic resonance imaging (fMRI) have
enabled scientists to extract visual information from human brain activity
patterns. In this study, we present an innovative method for decoding brain
activity into meaningful images and captions, with a specific focus on brain
captioning due to its enhanced flexibility as compared to brain decoding into
images. Our approach takes advantage of cutting-edge image captioning models
and incorporates a unique image reconstruction pipeline that utilizes latent
diffusion models and depth estimation. We utilized the Natural Scenes Dataset,
a comprehensive fMRI dataset from eight subjects who viewed images from the
COCO dataset. We employed the Generative Image-to-text Transformer (GIT) as our
backbone for captioning and propose a new image reconstruction pipeline based
on latent diffusion models. The method involves training regularized linear
regression models between brain activity and extracted features. Additionally,
we incorporated depth maps from the ControlNet model to further guide the
reconstruction process. We evaluate our methods using quantitative metrics for
both generated captions and images. Our brain captioning approach outperforms
existing methods, while our image reconstruction pipeline generates plausible
images with improved spatial relationships. In conclusion, we demonstrate
significant progress in brain decoding, showcasing the enormous potential of
integrating vision and language to better understand human cognition. Our
approach provides a flexible platform for future research, with potential
applications in various fields, including neural art, style transfer, and
portable devices.
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 09:57:19 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Ferrante",
"Matteo",
""
],
[
"Ozcelik",
"Furkan",
""
],
[
"Boccato",
"Tommaso",
""
],
[
"VanRullen",
"Rufin",
""
],
[
"Toschi",
"Nicola",
""
]
] | TITLE: Brain Captioning: Decoding human brain activity into images and text
ABSTRACT: Every day, the human brain processes an immense volume of visual information,
relying on intricate neural mechanisms to perceive and interpret these stimuli.
Recent breakthroughs in functional magnetic resonance imaging (fMRI) have
enabled scientists to extract visual information from human brain activity
patterns. In this study, we present an innovative method for decoding brain
activity into meaningful images and captions, with a specific focus on brain
captioning due to its enhanced flexibility as compared to brain decoding into
images. Our approach takes advantage of cutting-edge image captioning models
and incorporates a unique image reconstruction pipeline that utilizes latent
diffusion models and depth estimation. We utilized the Natural Scenes Dataset,
a comprehensive fMRI dataset from eight subjects who viewed images from the
COCO dataset. We employed the Generative Image-to-text Transformer (GIT) as our
backbone for captioning and propose a new image reconstruction pipeline based
on latent diffusion models. The method involves training regularized linear
regression models between brain activity and extracted features. Additionally,
we incorporated depth maps from the ControlNet model to further guide the
reconstruction process. We evaluate our methods using quantitative metrics for
both generated captions and images. Our brain captioning approach outperforms
existing methods, while our image reconstruction pipeline generates plausible
images with improved spatial relationships. In conclusion, we demonstrate
significant progress in brain decoding, showcasing the enormous potential of
integrating vision and language to better understand human cognition. Our
approach provides a flexible platform for future research, with potential
applications in various fields, including neural art, style transfer, and
portable devices.
| no_new_dataset | 0.708427 |
2305.11592 | Piyush Kumar Garg | Piyush Kumar Garg, Roshni Chakraborty, Srishti Gupta, and Sourav Kumar
Dandapat | IKDSumm: Incorporating Key-phrases into BERT for extractive Disaster
Tweet Summarization | null | null | null | null | cs.CL cs.SI | http://creativecommons.org/licenses/by/4.0/ | Online social media platforms, such as Twitter, are one of the most valuable
sources of information during disaster events. Therefore, humanitarian
organizations, government agencies, and volunteers rely on a summary of this
information, i.e., tweets, for effective disaster management. Although there
are several existing supervised and unsupervised approaches for automated tweet
summary approaches, these approaches either require extensive labeled
information or do not incorporate specific domain knowledge of disasters.
Additionally, the most recent approaches to disaster summarization have
proposed BERT-based models to enhance the summary quality. However, for further
improved performance, we introduce the utilization of domain-specific knowledge
without any human efforts to understand the importance (salience) of a tweet
which further aids in summary creation and improves summary quality. In this
paper, we propose a disaster-specific tweet summarization framework, IKDSumm,
which initially identifies the crucial and important information from each
tweet related to a disaster through key-phrases of that tweet. We identify
these key-phrases by utilizing the domain knowledge (using existing ontology)
of disasters without any human intervention. Further, we utilize these
key-phrases to automatically generate a summary of the tweets. Therefore, given
tweets related to a disaster, IKDSumm ensures fulfillment of the summarization
key objectives, such as information coverage, relevance, and diversity in
summary without any human intervention. We evaluate the performance of IKDSumm
with 8 state-of-the-art techniques on 12 disaster datasets. The evaluation
results show that IKDSumm outperforms existing techniques by approximately
2-79% in terms of ROUGE-N F1-score.
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 11:05:55 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Garg",
"Piyush Kumar",
""
],
[
"Chakraborty",
"Roshni",
""
],
[
"Gupta",
"Srishti",
""
],
[
"Dandapat",
"Sourav Kumar",
""
]
] | TITLE: IKDSumm: Incorporating Key-phrases into BERT for extractive Disaster
Tweet Summarization
ABSTRACT: Online social media platforms, such as Twitter, are one of the most valuable
sources of information during disaster events. Therefore, humanitarian
organizations, government agencies, and volunteers rely on a summary of this
information, i.e., tweets, for effective disaster management. Although there
are several existing supervised and unsupervised approaches for automated tweet
summary approaches, these approaches either require extensive labeled
information or do not incorporate specific domain knowledge of disasters.
Additionally, the most recent approaches to disaster summarization have
proposed BERT-based models to enhance the summary quality. However, for further
improved performance, we introduce the utilization of domain-specific knowledge
without any human efforts to understand the importance (salience) of a tweet
which further aids in summary creation and improves summary quality. In this
paper, we propose a disaster-specific tweet summarization framework, IKDSumm,
which initially identifies the crucial and important information from each
tweet related to a disaster through key-phrases of that tweet. We identify
these key-phrases by utilizing the domain knowledge (using existing ontology)
of disasters without any human intervention. Further, we utilize these
key-phrases to automatically generate a summary of the tweets. Therefore, given
tweets related to a disaster, IKDSumm ensures fulfillment of the summarization
key objectives, such as information coverage, relevance, and diversity in
summary without any human intervention. We evaluate the performance of IKDSumm
with 8 state-of-the-art techniques on 12 disaster datasets. The evaluation
results show that IKDSumm outperforms existing techniques by approximately
2-79% in terms of ROUGE-N F1-score.
| no_new_dataset | 0.709252 |
2305.11601 | Baorui Ma | Baorui Ma, Junsheng Zhou, Yu-Shen Liu, Zhizhong Han | Towards Better Gradient Consistency for Neural Signed Distance Functions
via Level Set Alignment | To appear at CVPR2023. Project page:
https://github.com/mabaorui/TowardsBetterGradient | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Neural signed distance functions (SDFs) have shown remarkable capability in
representing geometry with details. However, without signed distance
supervision, it is still a challenge to infer SDFs from point clouds or
multi-view images using neural networks. In this paper, we claim that gradient
consistency in the field, indicated by the parallelism of level sets, is the
key factor affecting the inference accuracy. Hence, we propose a level set
alignment loss to evaluate the parallelism of level sets, which can be
minimized to achieve better gradient consistency. Our novelty lies in that we
can align all level sets to the zero level set by constraining gradients at
queries and their projections on the zero level set in an adaptive way. Our
insight is to propagate the zero level set to everywhere in the field through
consistent gradients to eliminate uncertainty in the field that is caused by
the discreteness of 3D point clouds or the lack of observations from multi-view
images. Our proposed loss is a general term which can be used upon different
methods to infer SDFs from 3D point clouds and multi-view images. Our numerical
and visual comparisons demonstrate that our loss can significantly improve the
accuracy of SDFs inferred from point clouds or multi-view images under various
benchmarks. Code and data are available at
https://github.com/mabaorui/TowardsBetterGradient .
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 11:28:05 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Ma",
"Baorui",
""
],
[
"Zhou",
"Junsheng",
""
],
[
"Liu",
"Yu-Shen",
""
],
[
"Han",
"Zhizhong",
""
]
] | TITLE: Towards Better Gradient Consistency for Neural Signed Distance Functions
via Level Set Alignment
ABSTRACT: Neural signed distance functions (SDFs) have shown remarkable capability in
representing geometry with details. However, without signed distance
supervision, it is still a challenge to infer SDFs from point clouds or
multi-view images using neural networks. In this paper, we claim that gradient
consistency in the field, indicated by the parallelism of level sets, is the
key factor affecting the inference accuracy. Hence, we propose a level set
alignment loss to evaluate the parallelism of level sets, which can be
minimized to achieve better gradient consistency. Our novelty lies in that we
can align all level sets to the zero level set by constraining gradients at
queries and their projections on the zero level set in an adaptive way. Our
insight is to propagate the zero level set to everywhere in the field through
consistent gradients to eliminate uncertainty in the field that is caused by
the discreteness of 3D point clouds or the lack of observations from multi-view
images. Our proposed loss is a general term which can be used upon different
methods to infer SDFs from 3D point clouds and multi-view images. Our numerical
and visual comparisons demonstrate that our loss can significantly improve the
accuracy of SDFs inferred from point clouds or multi-view images under various
benchmarks. Code and data are available at
https://github.com/mabaorui/TowardsBetterGradient .
| no_new_dataset | 0.710622 |
2305.11603 | Tom Hosking | Tom Hosking and Hao Tang and Mirella Lapata | Attributable and Scalable Opinion Summarization | ACL 2023 | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | We propose a method for unsupervised opinion summarization that encodes
sentences from customer reviews into a hierarchical discrete latent space, then
identifies common opinions based on the frequency of their encodings. We are
able to generate both abstractive summaries by decoding these frequent
encodings, and extractive summaries by selecting the sentences assigned to the
same frequent encodings. Our method is attributable, because the model
identifies sentences used to generate the summary as part of the summarization
process. It scales easily to many hundreds of input reviews, because
aggregation is performed in the latent space rather than over long sequences of
tokens. We also demonstrate that our appraoch enables a degree of control,
generating aspect-specific summaries by restricting the model to parts of the
encoding space that correspond to desired aspects (e.g., location or food).
Automatic and human evaluation on two datasets from different domains
demonstrates that our method generates summaries that are more informative than
prior work and better grounded in the input reviews.
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 11:30:37 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Hosking",
"Tom",
""
],
[
"Tang",
"Hao",
""
],
[
"Lapata",
"Mirella",
""
]
] | TITLE: Attributable and Scalable Opinion Summarization
ABSTRACT: We propose a method for unsupervised opinion summarization that encodes
sentences from customer reviews into a hierarchical discrete latent space, then
identifies common opinions based on the frequency of their encodings. We are
able to generate both abstractive summaries by decoding these frequent
encodings, and extractive summaries by selecting the sentences assigned to the
same frequent encodings. Our method is attributable, because the model
identifies sentences used to generate the summary as part of the summarization
process. It scales easily to many hundreds of input reviews, because
aggregation is performed in the latent space rather than over long sequences of
tokens. We also demonstrate that our appraoch enables a degree of control,
generating aspect-specific summaries by restricting the model to parts of the
encoding space that correspond to desired aspects (e.g., location or food).
Automatic and human evaluation on two datasets from different domains
demonstrates that our method generates summaries that are more informative than
prior work and better grounded in the input reviews.
| no_new_dataset | 0.709837 |
2305.11654 | Rui Song | Rui Song, Lingjuan Lyu, Wei Jiang, Andreas Festag and Alois Knoll | V2X-Boosted Federated Learning for Cooperative Intelligent
Transportation Systems with Contextual Client Selection | Accepted at ICRA 2023 Workshop on Collaborative Perception and
Learning | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | Machine learning (ML) has revolutionized transportation systems, enabling
autonomous driving and smart traffic services. Federated learning (FL)
overcomes privacy constraints by training ML models in distributed systems,
exchanging model parameters instead of raw data. However, the dynamic states of
connected vehicles affect the network connection quality and influence the FL
performance. To tackle this challenge, we propose a contextual client selection
pipeline that uses Vehicle-to-Everything (V2X) messages to select clients based
on the predicted communication latency. The pipeline includes: (i) fusing V2X
messages, (ii) predicting future traffic topology, (iii) pre-clustering clients
based on local data distribution similarity, and (iv) selecting clients with
minimal latency for future model aggregation. Experiments show that our
pipeline outperforms baselines on various datasets, particularly in non-iid
settings.
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 13:09:33 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Song",
"Rui",
""
],
[
"Lyu",
"Lingjuan",
""
],
[
"Jiang",
"Wei",
""
],
[
"Festag",
"Andreas",
""
],
[
"Knoll",
"Alois",
""
]
] | TITLE: V2X-Boosted Federated Learning for Cooperative Intelligent
Transportation Systems with Contextual Client Selection
ABSTRACT: Machine learning (ML) has revolutionized transportation systems, enabling
autonomous driving and smart traffic services. Federated learning (FL)
overcomes privacy constraints by training ML models in distributed systems,
exchanging model parameters instead of raw data. However, the dynamic states of
connected vehicles affect the network connection quality and influence the FL
performance. To tackle this challenge, we propose a contextual client selection
pipeline that uses Vehicle-to-Everything (V2X) messages to select clients based
on the predicted communication latency. The pipeline includes: (i) fusing V2X
messages, (ii) predicting future traffic topology, (iii) pre-clustering clients
based on local data distribution similarity, and (iv) selecting clients with
minimal latency for future model aggregation. Experiments show that our
pipeline outperforms baselines on various datasets, particularly in non-iid
settings.
| no_new_dataset | 0.709856 |
2305.11655 | Bhaskar Biswas | Bhaskar Biswas, Dmitry Ignatyev, Argyrios Zolotas and Antonios
Tsourdos | Region of Attraction Estimation Using Union Theorem in Sum-of-Squares
Optimization | null | null | null | BB.5.2023.1 | eess.SY cs.SY math.OC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Appropriate estimation of Region of Attraction for a nonlinear dynamical
system plays a key role in system analysis and control design. Sum-of-Squares
optimization is a powerful tool enabling Region of Attraction estimation for
polynomial dynamical systems. Employment of a positive definite function called
shape function within the Sum-of-Squares procedure helps to find a richer
representation of the Lyapunov function and a larger corresponding Region of
Attraction estimation. However, existing Sum-of-Squares optimization techniques
demonstrate very conservative results. The main novelty of this paper is the
Union theorem which enables the use of multiple shape functions to create a
polynomial Lyapunov function encompassing all the areas generated by the shape
functions. The main contribution of this paper is a novel
computationally-efficient numerical method for Region of Attraction estimation,
which remarkably improves estimation performance and overcomes limitations of
existing methods, while maintaining the resultant Lyapunov function polynomial,
thus facilitating control system design and construction of control Lyapunov
function with enhanced Region of Attraction using conventional Sum-of-Squares
tools. A mathematical proof of the Union theorem along with its application to
the numerical algorithm of Region of Attraction estimation is provided. The
method yields significantly enlarged Region of Attraction estimations even for
systems with non-symmetric or unbounded Region of Attraction, which is
demonstrated via simulations of several benchmark examples.
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 13:11:05 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Biswas",
"Bhaskar",
""
],
[
"Ignatyev",
"Dmitry",
""
],
[
"Zolotas",
"Argyrios",
""
],
[
"Tsourdos",
"Antonios",
""
]
] | TITLE: Region of Attraction Estimation Using Union Theorem in Sum-of-Squares
Optimization
ABSTRACT: Appropriate estimation of Region of Attraction for a nonlinear dynamical
system plays a key role in system analysis and control design. Sum-of-Squares
optimization is a powerful tool enabling Region of Attraction estimation for
polynomial dynamical systems. Employment of a positive definite function called
shape function within the Sum-of-Squares procedure helps to find a richer
representation of the Lyapunov function and a larger corresponding Region of
Attraction estimation. However, existing Sum-of-Squares optimization techniques
demonstrate very conservative results. The main novelty of this paper is the
Union theorem which enables the use of multiple shape functions to create a
polynomial Lyapunov function encompassing all the areas generated by the shape
functions. The main contribution of this paper is a novel
computationally-efficient numerical method for Region of Attraction estimation,
which remarkably improves estimation performance and overcomes limitations of
existing methods, while maintaining the resultant Lyapunov function polynomial,
thus facilitating control system design and construction of control Lyapunov
function with enhanced Region of Attraction using conventional Sum-of-Squares
tools. A mathematical proof of the Union theorem along with its application to
the numerical algorithm of Region of Attraction estimation is provided. The
method yields significantly enlarged Region of Attraction estimations even for
systems with non-symmetric or unbounded Region of Attraction, which is
demonstrated via simulations of several benchmark examples.
| no_new_dataset | 0.710415 |
2305.11663 | Jill Walker Rettberg | Jill Walker Rettberg | Algorithmic failure as a humanities methodology: machine learning's
mispredictions identify rich cases for qualitative analysis | null | Big Data & Society 9(2) 2022 | 10.1177/2053951722113129 | null | cs.LG cs.AI cs.CL cs.CY | http://creativecommons.org/licenses/by-nc-sa/4.0/ | This commentary tests a methodology proposed by Munk et al. (2022) for using
failed predictions in machine learning as a method to identify ambiguous and
rich cases for qualitative analysis. Using a dataset describing actions
performed by fictional characters interacting with machine vision technologies
in 500 artworks, movies, novels and videogames, I trained a simple machine
learning algorithm (using the kNN algorithm in R) to predict whether or not an
action was active or passive using only information about the fictional
characters. Predictable actions were generally unemotional and unambiguous
activities where machine vision technologies were treated as simple tools.
Unpredictable actions, that is, actions that the algorithm could not correctly
predict, were more ambivalent and emotionally loaded, with more complex power
relationships between characters and technologies. The results thus support
Munk et al.'s theory that failed predictions can be productively used to
identify rich cases for qualitative analysis. This test goes beyond simply
replicating Munk et al.'s results by demonstrating that the method can be
applied to a broader humanities domain, and that it does not require complex
neural networks but can also work with a simpler machine learning algorithm.
Further research is needed to develop an understanding of what kinds of data
the method is useful for and which kinds of machine learning are most
generative. To support this, the R code required to produce the results is
included so the test can be replicated. The code can also be reused or adapted
to test the method on other datasets.
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 13:24:32 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Rettberg",
"Jill Walker",
""
]
] | TITLE: Algorithmic failure as a humanities methodology: machine learning's
mispredictions identify rich cases for qualitative analysis
ABSTRACT: This commentary tests a methodology proposed by Munk et al. (2022) for using
failed predictions in machine learning as a method to identify ambiguous and
rich cases for qualitative analysis. Using a dataset describing actions
performed by fictional characters interacting with machine vision technologies
in 500 artworks, movies, novels and videogames, I trained a simple machine
learning algorithm (using the kNN algorithm in R) to predict whether or not an
action was active or passive using only information about the fictional
characters. Predictable actions were generally unemotional and unambiguous
activities where machine vision technologies were treated as simple tools.
Unpredictable actions, that is, actions that the algorithm could not correctly
predict, were more ambivalent and emotionally loaded, with more complex power
relationships between characters and technologies. The results thus support
Munk et al.'s theory that failed predictions can be productively used to
identify rich cases for qualitative analysis. This test goes beyond simply
replicating Munk et al.'s results by demonstrating that the method can be
applied to a broader humanities domain, and that it does not require complex
neural networks but can also work with a simpler machine learning algorithm.
Further research is needed to develop an understanding of what kinds of data
the method is useful for and which kinds of machine learning are most
generative. To support this, the R code required to produce the results is
included so the test can be replicated. The code can also be reused or adapted
to test the method on other datasets.
| no_new_dataset | 0.708616 |
2305.11664 | JingYuan Zhu | Jingyuan Zhu, Huimin Ma, Jiansheng Chen, Jian Yuan | Few-shot 3D Shape Generation | 23 pages, 15 figures | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Realistic and diverse 3D shape generation is helpful for a wide variety of
applications such as virtual reality, gaming, and animation. Modern generative
models, such as GANs and diffusion models, learn from large-scale datasets and
generate new samples following similar data distributions. However, when
training data is limited, deep neural generative networks overfit and tend to
replicate training samples. Prior works focus on few-shot image generation to
produce high-quality and diverse results using a few target images.
Unfortunately, abundant 3D shape data is typically hard to obtain as well. In
this work, we make the first attempt to realize few-shot 3D shape generation by
adapting generative models pre-trained on large source domains to target
domains using limited data. To relieve overfitting and keep considerable
diversity, we propose to maintain the probability distributions of the pairwise
relative distances between adapted samples at feature-level and shape-level
during domain adaptation. Our approach only needs the silhouettes of few-shot
target samples as training data to learn target geometry distributions and
achieve generated shapes with diverse topology and textures. Moreover, we
introduce several metrics to evaluate the quality and diversity of few-shot 3D
shape generation. The effectiveness of our approach is demonstrated
qualitatively and quantitatively under a series of few-shot 3D shape adaptation
setups.
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 13:30:10 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Zhu",
"Jingyuan",
""
],
[
"Ma",
"Huimin",
""
],
[
"Chen",
"Jiansheng",
""
],
[
"Yuan",
"Jian",
""
]
] | TITLE: Few-shot 3D Shape Generation
ABSTRACT: Realistic and diverse 3D shape generation is helpful for a wide variety of
applications such as virtual reality, gaming, and animation. Modern generative
models, such as GANs and diffusion models, learn from large-scale datasets and
generate new samples following similar data distributions. However, when
training data is limited, deep neural generative networks overfit and tend to
replicate training samples. Prior works focus on few-shot image generation to
produce high-quality and diverse results using a few target images.
Unfortunately, abundant 3D shape data is typically hard to obtain as well. In
this work, we make the first attempt to realize few-shot 3D shape generation by
adapting generative models pre-trained on large source domains to target
domains using limited data. To relieve overfitting and keep considerable
diversity, we propose to maintain the probability distributions of the pairwise
relative distances between adapted samples at feature-level and shape-level
during domain adaptation. Our approach only needs the silhouettes of few-shot
target samples as training data to learn target geometry distributions and
achieve generated shapes with diverse topology and textures. Moreover, we
introduce several metrics to evaluate the quality and diversity of few-shot 3D
shape generation. The effectiveness of our approach is demonstrated
qualitatively and quantitatively under a series of few-shot 3D shape adaptation
setups.
| no_new_dataset | 0.710791 |
2305.11681 | Li Ding | Li Ding, Edward Pantridge, Lee Spector | Probabilistic Lexicase Selection | GECCO 2023 | null | 10.1145/3583131.3590375 | null | cs.NE cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Lexicase selection is a widely used parent selection algorithm in genetic
programming, known for its success in various task domains such as program
synthesis, symbolic regression, and machine learning. Due to its non-parametric
and recursive nature, calculating the probability of each individual being
selected by lexicase selection has been proven to be an NP-hard problem, which
discourages deeper theoretical understanding and practical improvements to the
algorithm. In this work, we introduce probabilistic lexicase selection
(plexicase selection), a novel parent selection algorithm that efficiently
approximates the probability distribution of lexicase selection. Our method not
only demonstrates superior problem-solving capabilities as a semantic-aware
selection method, but also benefits from having a probabilistic representation
of the selection process for enhanced efficiency and flexibility. Experiments
are conducted in two prevalent domains in genetic programming: program
synthesis and symbolic regression, using standard benchmarks including PSB and
SRBench. The empirical results show that plexicase selection achieves
state-of-the-art problem-solving performance that is competitive to the
lexicase selection, and significantly outperforms lexicase selection in
computation efficiency.
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 13:57:04 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Ding",
"Li",
""
],
[
"Pantridge",
"Edward",
""
],
[
"Spector",
"Lee",
""
]
] | TITLE: Probabilistic Lexicase Selection
ABSTRACT: Lexicase selection is a widely used parent selection algorithm in genetic
programming, known for its success in various task domains such as program
synthesis, symbolic regression, and machine learning. Due to its non-parametric
and recursive nature, calculating the probability of each individual being
selected by lexicase selection has been proven to be an NP-hard problem, which
discourages deeper theoretical understanding and practical improvements to the
algorithm. In this work, we introduce probabilistic lexicase selection
(plexicase selection), a novel parent selection algorithm that efficiently
approximates the probability distribution of lexicase selection. Our method not
only demonstrates superior problem-solving capabilities as a semantic-aware
selection method, but also benefits from having a probabilistic representation
of the selection process for enhanced efficiency and flexibility. Experiments
are conducted in two prevalent domains in genetic programming: program
synthesis and symbolic regression, using standard benchmarks including PSB and
SRBench. The empirical results show that plexicase selection achieves
state-of-the-art problem-solving performance that is competitive to the
lexicase selection, and significantly outperforms lexicase selection in
computation efficiency.
| no_new_dataset | 0.711074 |
2305.11692 | Long Bai | Long Bai, Mobarakol Islam, Lalithkumar Seenivasan, Hongliang Ren | Surgical-VQLA: Transformer with Gated Vision-Language Embedding for
Visual Question Localized-Answering in Robotic Surgery | To appear in IEEE ICRA 2023. Code and data availability:
https://github.com/longbai1006/Surgical-VQLA | null | null | null | cs.CV cs.AI cs.CL cs.LG cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Despite the availability of computer-aided simulators and recorded videos of
surgical procedures, junior residents still heavily rely on experts to answer
their queries. However, expert surgeons are often overloaded with clinical and
academic workloads and limit their time in answering. For this purpose, we
develop a surgical question-answering system to facilitate robot-assisted
surgical scene and activity understanding from recorded videos. Most of the
existing VQA methods require an object detector and regions based feature
extractor to extract visual features and fuse them with the embedded text of
the question for answer generation. However, (1) surgical object detection
model is scarce due to smaller datasets and lack of bounding box annotation;
(2) current fusion strategy of heterogeneous modalities like text and image is
naive; (3) the localized answering is missing, which is crucial in complex
surgical scenarios. In this paper, we propose Visual Question
Localized-Answering in Robotic Surgery (Surgical-VQLA) to localize the specific
surgical area during the answer prediction. To deal with the fusion of the
heterogeneous modalities, we design gated vision-language embedding (GVLE) to
build input patches for the Language Vision Transformer (LViT) to predict the
answer. To get localization, we add the detection head in parallel with the
prediction head of the LViT. We also integrate GIoU loss to boost localization
performance by preserving the accuracy of the question-answering model. We
annotate two datasets of VQLA by utilizing publicly available surgical videos
from MICCAI challenges EndoVis-17 and 18. Our validation results suggest that
Surgical-VQLA can better understand the surgical scene and localize the
specific area related to the question-answering. GVLE presents an efficient
language-vision embedding technique by showing superior performance over the
existing benchmarks.
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 14:13:47 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Bai",
"Long",
""
],
[
"Islam",
"Mobarakol",
""
],
[
"Seenivasan",
"Lalithkumar",
""
],
[
"Ren",
"Hongliang",
""
]
] | TITLE: Surgical-VQLA: Transformer with Gated Vision-Language Embedding for
Visual Question Localized-Answering in Robotic Surgery
ABSTRACT: Despite the availability of computer-aided simulators and recorded videos of
surgical procedures, junior residents still heavily rely on experts to answer
their queries. However, expert surgeons are often overloaded with clinical and
academic workloads and limit their time in answering. For this purpose, we
develop a surgical question-answering system to facilitate robot-assisted
surgical scene and activity understanding from recorded videos. Most of the
existing VQA methods require an object detector and regions based feature
extractor to extract visual features and fuse them with the embedded text of
the question for answer generation. However, (1) surgical object detection
model is scarce due to smaller datasets and lack of bounding box annotation;
(2) current fusion strategy of heterogeneous modalities like text and image is
naive; (3) the localized answering is missing, which is crucial in complex
surgical scenarios. In this paper, we propose Visual Question
Localized-Answering in Robotic Surgery (Surgical-VQLA) to localize the specific
surgical area during the answer prediction. To deal with the fusion of the
heterogeneous modalities, we design gated vision-language embedding (GVLE) to
build input patches for the Language Vision Transformer (LViT) to predict the
answer. To get localization, we add the detection head in parallel with the
prediction head of the LViT. We also integrate GIoU loss to boost localization
performance by preserving the accuracy of the question-answering model. We
annotate two datasets of VQLA by utilizing publicly available surgical videos
from MICCAI challenges EndoVis-17 and 18. Our validation results suggest that
Surgical-VQLA can better understand the surgical scene and localize the
specific area related to the question-answering. GVLE presents an efficient
language-vision embedding technique by showing superior performance over the
existing benchmarks.
| no_new_dataset | 0.711469 |
2305.11715 | Xiyao Jin | Xiyao Jin, Yao Hao, Jessica Hilliard, Zhehao Zhang, Maria A. Thomas,
Hua Li, Abhinav K. Jha, Geoffrey D. Hugo | A quality assurance framework for real-time monitoring of deep learning
segmentation models in radiotherapy | null | null | null | null | eess.IV cs.CV physics.med-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | To safely deploy deep learning models in the clinic, a quality assurance
framework is needed for routine or continuous monitoring of input-domain shift
and the models' performance without ground truth contours. In this work,
cardiac substructure segmentation was used as an example task to establish a QA
framework. A benchmark dataset consisting of Computed Tomography (CT) images
along with manual cardiac delineations of 241 patients were collected,
including one 'common' image domain and five 'uncommon' domains. Segmentation
models were tested on the benchmark dataset for an initial evaluation of model
capacity and limitations. An image domain shift detector was developed by
utilizing a trained Denoising autoencoder (DAE) and two hand-engineered
features. Another Variational Autoencoder (VAE) was also trained to estimate
the shape quality of the auto-segmentation results. Using the extracted
features from the image/segmentation pair as inputs, a regression model was
trained to predict the per-patient segmentation accuracy, measured by Dice
coefficient similarity (DSC). The framework was tested across 19 segmentation
models to evaluate the generalizability of the entire framework.
As results, the predicted DSC of regression models achieved a mean absolute
error (MAE) ranging from 0.036 to 0.046 with an averaged MAE of 0.041. When
tested on the benchmark dataset, the performances of all segmentation models
were not significantly affected by scanning parameters: FOV, slice thickness
and reconstructions kernels. For input images with Poisson noise, CNN-based
segmentation models demonstrated a decreased DSC ranging from 0.07 to 0.41,
while the transformer-based model was not significantly affected.
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 14:51:05 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Jin",
"Xiyao",
""
],
[
"Hao",
"Yao",
""
],
[
"Hilliard",
"Jessica",
""
],
[
"Zhang",
"Zhehao",
""
],
[
"Thomas",
"Maria A.",
""
],
[
"Li",
"Hua",
""
],
[
"Jha",
"Abhinav K.",
""
],
[
"Hugo",
"Geoffrey D.",
""
]
] | TITLE: A quality assurance framework for real-time monitoring of deep learning
segmentation models in radiotherapy
ABSTRACT: To safely deploy deep learning models in the clinic, a quality assurance
framework is needed for routine or continuous monitoring of input-domain shift
and the models' performance without ground truth contours. In this work,
cardiac substructure segmentation was used as an example task to establish a QA
framework. A benchmark dataset consisting of Computed Tomography (CT) images
along with manual cardiac delineations of 241 patients were collected,
including one 'common' image domain and five 'uncommon' domains. Segmentation
models were tested on the benchmark dataset for an initial evaluation of model
capacity and limitations. An image domain shift detector was developed by
utilizing a trained Denoising autoencoder (DAE) and two hand-engineered
features. Another Variational Autoencoder (VAE) was also trained to estimate
the shape quality of the auto-segmentation results. Using the extracted
features from the image/segmentation pair as inputs, a regression model was
trained to predict the per-patient segmentation accuracy, measured by Dice
coefficient similarity (DSC). The framework was tested across 19 segmentation
models to evaluate the generalizability of the entire framework.
As results, the predicted DSC of regression models achieved a mean absolute
error (MAE) ranging from 0.036 to 0.046 with an averaged MAE of 0.041. When
tested on the benchmark dataset, the performances of all segmentation models
were not significantly affected by scanning parameters: FOV, slice thickness
and reconstructions kernels. For input images with Poisson noise, CNN-based
segmentation models demonstrated a decreased DSC ranging from 0.07 to 0.41,
while the transformer-based model was not significantly affected.
| no_new_dataset | 0.709233 |
2305.11729 | Ioanna Diamanti | Ioanna Diamanti, Antigoni Tsiami, Petros Koutras and Petros Maragos | ViDaS Video Depth-aware Saliency Network | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We introduce ViDaS, a two-stream, fully convolutional Video, Depth-Aware
Saliency network to address the problem of attention modeling ``in-the-wild",
via saliency prediction in videos. Contrary to existing visual saliency
approaches using only RGB frames as input, our network employs also depth as an
additional modality. The network consists of two visual streams, one for the
RGB frames, and one for the depth frames. Both streams follow an
encoder-decoder approach and are fused to obtain a final saliency map. The
network is trained end-to-end and is evaluated in a variety of different
databases with eye-tracking data, containing a wide range of video content.
Although the publicly available datasets do not contain depth, we estimate it
using three different state-of-the-art methods, to enable comparisons and a
deeper insight. Our method outperforms in most cases state-of-the-art models
and our RGB-only variant, which indicates that depth can be beneficial to
accurately estimating saliency in videos displayed on a 2D screen. Depth has
been widely used to assist salient object detection problems, where it has been
proven to be very beneficial. Our problem though differs significantly from
salient object detection, since it is not restricted to specific salient
objects, but predicts human attention in a more general aspect. These two
problems not only have different objectives, but also different ground truth
data and evaluation metrics. To our best knowledge, this is the first
competitive deep learning video saliency estimation approach that combines both
RGB and Depth features to address the general problem of saliency estimation
``in-the-wild". The code will be publicly released.
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 15:04:49 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Diamanti",
"Ioanna",
""
],
[
"Tsiami",
"Antigoni",
""
],
[
"Koutras",
"Petros",
""
],
[
"Maragos",
"Petros",
""
]
] | TITLE: ViDaS Video Depth-aware Saliency Network
ABSTRACT: We introduce ViDaS, a two-stream, fully convolutional Video, Depth-Aware
Saliency network to address the problem of attention modeling ``in-the-wild",
via saliency prediction in videos. Contrary to existing visual saliency
approaches using only RGB frames as input, our network employs also depth as an
additional modality. The network consists of two visual streams, one for the
RGB frames, and one for the depth frames. Both streams follow an
encoder-decoder approach and are fused to obtain a final saliency map. The
network is trained end-to-end and is evaluated in a variety of different
databases with eye-tracking data, containing a wide range of video content.
Although the publicly available datasets do not contain depth, we estimate it
using three different state-of-the-art methods, to enable comparisons and a
deeper insight. Our method outperforms in most cases state-of-the-art models
and our RGB-only variant, which indicates that depth can be beneficial to
accurately estimating saliency in videos displayed on a 2D screen. Depth has
been widely used to assist salient object detection problems, where it has been
proven to be very beneficial. Our problem though differs significantly from
salient object detection, since it is not restricted to specific salient
objects, but predicts human attention in a more general aspect. These two
problems not only have different objectives, but also different ground truth
data and evaluation metrics. To our best knowledge, this is the first
competitive deep learning video saliency estimation approach that combines both
RGB and Depth features to address the general problem of saliency estimation
``in-the-wild". The code will be publicly released.
| no_new_dataset | 0.709667 |
2305.11733 | Mengke Li | Mengke Li, Yiu-ming Cheung, Yang Lu | Long-tailed Visual Recognition via Gaussian Clouded Logit Adjustment | Published as a conference paper at CVPR 2022. arXiv admin note: text
overlap with arXiv:2305.10648 | Proceedings of the IEEE/CVF Conference on Computer Vision and
Pattern Recognition (CVPR), 2022, pp. 6929-6938 | 10.1109/CVPR52688.2022.00680 | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Long-tailed data is still a big challenge for deep neural networks, even
though they have achieved great success on balanced data. We observe that
vanilla training on long-tailed data with cross-entropy loss makes the
instance-rich head classes severely squeeze the spatial distribution of the
tail classes, which leads to difficulty in classifying tail class samples.
Furthermore, the original cross-entropy loss can only propagate gradient
short-lively because the gradient in softmax form rapidly approaches zero as
the logit difference increases. This phenomenon is called softmax saturation.
It is unfavorable for training on balanced data, but can be utilized to adjust
the validity of the samples in long-tailed data, thereby solving the distorted
embedding space of long-tailed problems. To this end, this paper proposes the
Gaussian clouded logit adjustment by Gaussian perturbation of different class
logits with varied amplitude. We define the amplitude of perturbation as cloud
size and set relatively large cloud sizes to tail classes. The large cloud size
can reduce the softmax saturation and thereby making tail class samples more
active as well as enlarging the embedding space. To alleviate the bias in a
classifier, we therefore propose the class-based effective number sampling
strategy with classifier re-training. Extensive experiments on benchmark
datasets validate the superior performance of the proposed method. Source code
is available at https://github.com/Keke921/GCLLoss.
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 15:11:06 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Li",
"Mengke",
""
],
[
"Cheung",
"Yiu-ming",
""
],
[
"Lu",
"Yang",
""
]
] | TITLE: Long-tailed Visual Recognition via Gaussian Clouded Logit Adjustment
ABSTRACT: Long-tailed data is still a big challenge for deep neural networks, even
though they have achieved great success on balanced data. We observe that
vanilla training on long-tailed data with cross-entropy loss makes the
instance-rich head classes severely squeeze the spatial distribution of the
tail classes, which leads to difficulty in classifying tail class samples.
Furthermore, the original cross-entropy loss can only propagate gradient
short-lively because the gradient in softmax form rapidly approaches zero as
the logit difference increases. This phenomenon is called softmax saturation.
It is unfavorable for training on balanced data, but can be utilized to adjust
the validity of the samples in long-tailed data, thereby solving the distorted
embedding space of long-tailed problems. To this end, this paper proposes the
Gaussian clouded logit adjustment by Gaussian perturbation of different class
logits with varied amplitude. We define the amplitude of perturbation as cloud
size and set relatively large cloud sizes to tail classes. The large cloud size
can reduce the softmax saturation and thereby making tail class samples more
active as well as enlarging the embedding space. To alleviate the bias in a
classifier, we therefore propose the class-based effective number sampling
strategy with classifier re-training. Extensive experiments on benchmark
datasets validate the superior performance of the proposed method. Source code
is available at https://github.com/Keke921/GCLLoss.
| no_new_dataset | 0.714261 |
2305.11739 | Tuomas Eerola | Tuomas Eerola, Daniel Batrakhanov, Nastaran Vatankhah Barazandeh,
Kaisa Kraft, Lumi Haraguchi, Lasse Lensu, Sanna Suikkanen, Jukka Sepp\"al\"a,
Timo Tamminen, Heikki K\"alvi\"ainen | Survey of Automatic Plankton Image Recognition: Challenges, Existing
Solutions and Future Perspectives | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Planktonic organisms are key components of aquatic ecosystems and respond
quickly to changes in the environment, therefore their monitoring is vital to
understand the changes in the environment. Yet, monitoring plankton at
appropriate scales still remains a challenge, limiting our understanding of
functioning of aquatic systems and their response to changes. Modern plankton
imaging instruments can be utilized to sample at high frequencies, enabling
novel possibilities to study plankton populations. However, manual analysis of
the data is costly, time consuming and expert based, making such approach
unsuitable for large-scale application and urging for automatic solutions. The
key problem related to the utilization of plankton datasets through image
analysis is plankton recognition. Despite the large amount of research done,
automatic methods have not been widely adopted for operational use. In this
paper, a comprehensive survey on existing solutions for automatic plankton
recognition is presented. First, we identify the most notable challenges that
that make the development of plankton recognition systems difficult. Then, we
provide a detailed description of solutions for these challenges proposed in
plankton recognition literature. Finally, we propose a workflow to identify the
specific challenges in new datasets and the recommended approaches to address
them. For many of the challenges, applicable solutions exist. However,
important challenges remain unsolved: 1) the domain shift between the datasets
hindering the development of a general plankton recognition system that would
work across different imaging instruments, 2) the difficulty to identify and
process the images of previously unseen classes, and 3) the uncertainty in
expert annotations that affects the training of the machine learning models for
recognition. These challenges should be addressed in the future research.
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 15:20:00 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Eerola",
"Tuomas",
""
],
[
"Batrakhanov",
"Daniel",
""
],
[
"Barazandeh",
"Nastaran Vatankhah",
""
],
[
"Kraft",
"Kaisa",
""
],
[
"Haraguchi",
"Lumi",
""
],
[
"Lensu",
"Lasse",
""
],
[
"Suikkanen",
"Sanna",
""
],
[
"Seppälä",
"Jukka",
""
],
[
"Tamminen",
"Timo",
""
],
[
"Kälviäinen",
"Heikki",
""
]
] | TITLE: Survey of Automatic Plankton Image Recognition: Challenges, Existing
Solutions and Future Perspectives
ABSTRACT: Planktonic organisms are key components of aquatic ecosystems and respond
quickly to changes in the environment, therefore their monitoring is vital to
understand the changes in the environment. Yet, monitoring plankton at
appropriate scales still remains a challenge, limiting our understanding of
functioning of aquatic systems and their response to changes. Modern plankton
imaging instruments can be utilized to sample at high frequencies, enabling
novel possibilities to study plankton populations. However, manual analysis of
the data is costly, time consuming and expert based, making such approach
unsuitable for large-scale application and urging for automatic solutions. The
key problem related to the utilization of plankton datasets through image
analysis is plankton recognition. Despite the large amount of research done,
automatic methods have not been widely adopted for operational use. In this
paper, a comprehensive survey on existing solutions for automatic plankton
recognition is presented. First, we identify the most notable challenges that
that make the development of plankton recognition systems difficult. Then, we
provide a detailed description of solutions for these challenges proposed in
plankton recognition literature. Finally, we propose a workflow to identify the
specific challenges in new datasets and the recommended approaches to address
them. For many of the challenges, applicable solutions exist. However,
important challenges remain unsolved: 1) the domain shift between the datasets
hindering the development of a general plankton recognition system that would
work across different imaging instruments, 2) the difficulty to identify and
process the images of previously unseen classes, and 3) the uncertainty in
expert annotations that affects the training of the machine learning models for
recognition. These challenges should be addressed in the future research.
| no_new_dataset | 0.711268 |
2305.11759 | Charith Peris | Mustafa Safa Ozdayi and Charith Peris and Jack FitzGerald and
Christophe Dupuy and Jimit Majmudar and Haidar Khan and Rahil Parikh and
Rahul Gupta | Controlling the Extraction of Memorized Data from Large Language Models
via Prompt-Tuning | 5 pages, 3 Figures, ACL 2023 | null | null | null | cs.CL cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Large Language Models (LLMs) are known to memorize significant portions of
their training data. Parts of this memorized content have been shown to be
extractable by simply querying the model, which poses a privacy risk. We
present a novel approach which uses prompt-tuning to control the extraction
rates of memorized content in LLMs. We present two prompt training strategies
to increase and decrease extraction rates, which correspond to an attack and a
defense, respectively. We demonstrate the effectiveness of our techniques by
using models from the GPT-Neo family on a public benchmark. For the 1.3B
parameter GPT-Neo model, our attack yields a 9.3 percentage point increase in
extraction rate compared to our baseline. Our defense can be tuned to achieve
different privacy-utility trade-offs by a user-specified hyperparameter. We
achieve an extraction rate reduction of up to 97.7% relative to our baseline,
with a perplexity increase of 16.9%.
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 15:45:29 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Ozdayi",
"Mustafa Safa",
""
],
[
"Peris",
"Charith",
""
],
[
"FitzGerald",
"Jack",
""
],
[
"Dupuy",
"Christophe",
""
],
[
"Majmudar",
"Jimit",
""
],
[
"Khan",
"Haidar",
""
],
[
"Parikh",
"Rahil",
""
],
[
"Gupta",
"Rahul",
""
]
] | TITLE: Controlling the Extraction of Memorized Data from Large Language Models
via Prompt-Tuning
ABSTRACT: Large Language Models (LLMs) are known to memorize significant portions of
their training data. Parts of this memorized content have been shown to be
extractable by simply querying the model, which poses a privacy risk. We
present a novel approach which uses prompt-tuning to control the extraction
rates of memorized content in LLMs. We present two prompt training strategies
to increase and decrease extraction rates, which correspond to an attack and a
defense, respectively. We demonstrate the effectiveness of our techniques by
using models from the GPT-Neo family on a public benchmark. For the 1.3B
parameter GPT-Neo model, our attack yields a 9.3 percentage point increase in
extraction rate compared to our baseline. Our defense can be tuned to achieve
different privacy-utility trade-offs by a user-specified hyperparameter. We
achieve an extraction rate reduction of up to 97.7% relative to our baseline,
with a perplexity increase of 16.9%.
| no_new_dataset | 0.710239 |
2305.11773 | Mohsen Dastpak | Mohsen Dastpak, Fausto Errico, Ola Jabali, and Federico Malucelli | Dynamic Routing for the Electric Vehicle Shortest Path Problem with
Charging Station Occupancy Information | null | null | null | null | math.OC cs.SY eess.SY | http://creativecommons.org/licenses/by-nc-sa/4.0/ | We study EVs traveling from origin to destination in the shortest time,
focusing on long-distance settings with energy requirements exceeding EV
autonomy. The EV may charge its battery at public Charging Stations (CSs),
which are subject to uncertain waiting times. We model CSs using appropriately
defined queues, whose status is revealed upon the EV arrival. However, we
consider the availability of real-time binary Occupancy Indicator (OI)
information, signaling if a CS is busy or not. At each OI update, we determine
the sequence of CSs to visit along with associated charging quantities. We name
the resulting problem the Electric Vehicle Shortest Path Problem with charging
station Occupancy Indicator information (EVSPP-OI). In this problem, we
consider that the EV is allowed to partially charge its battery, and we model
charging times via piecewise linear charging functions that depend on the CS
technology. We propose an MDP formulation for the EVSPP-OI and develop a
reoptimization algorithm that establishes the sequence of CS visits and
charging amounts based on system updates. Specifically, we propose a
simulation-based approach to estimate the waiting time of the EV at a CS as a
function of its arrival time. As the path to a CS may consist of multiple
intermediate CS stops, estimating the arrival times at each CS is fairly
intricate. To this end, we propose an efficient heuristic that yields
approximate lower bounds on the arrival time of the EV at each CS. We use these
estimations to define a deterministic EVSPP, which we solve with an existing
algorithm. We conduct a comprehensive computational study and compare the
performance of our methodology with a benchmark that observes the status of CSs
only upon arrival. Results show that our method reduces waiting times and total
trip duration by an average of 23.7%-95.4% and 1.4%-18.5%, respectively.
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 15:59:12 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Dastpak",
"Mohsen",
""
],
[
"Errico",
"Fausto",
""
],
[
"Jabali",
"Ola",
""
],
[
"Malucelli",
"Federico",
""
]
] | TITLE: Dynamic Routing for the Electric Vehicle Shortest Path Problem with
Charging Station Occupancy Information
ABSTRACT: We study EVs traveling from origin to destination in the shortest time,
focusing on long-distance settings with energy requirements exceeding EV
autonomy. The EV may charge its battery at public Charging Stations (CSs),
which are subject to uncertain waiting times. We model CSs using appropriately
defined queues, whose status is revealed upon the EV arrival. However, we
consider the availability of real-time binary Occupancy Indicator (OI)
information, signaling if a CS is busy or not. At each OI update, we determine
the sequence of CSs to visit along with associated charging quantities. We name
the resulting problem the Electric Vehicle Shortest Path Problem with charging
station Occupancy Indicator information (EVSPP-OI). In this problem, we
consider that the EV is allowed to partially charge its battery, and we model
charging times via piecewise linear charging functions that depend on the CS
technology. We propose an MDP formulation for the EVSPP-OI and develop a
reoptimization algorithm that establishes the sequence of CS visits and
charging amounts based on system updates. Specifically, we propose a
simulation-based approach to estimate the waiting time of the EV at a CS as a
function of its arrival time. As the path to a CS may consist of multiple
intermediate CS stops, estimating the arrival times at each CS is fairly
intricate. To this end, we propose an efficient heuristic that yields
approximate lower bounds on the arrival time of the EV at each CS. We use these
estimations to define a deterministic EVSPP, which we solve with an existing
algorithm. We conduct a comprehensive computational study and compare the
performance of our methodology with a benchmark that observes the status of CSs
only upon arrival. Results show that our method reduces waiting times and total
trip duration by an average of 23.7%-95.4% and 1.4%-18.5%, respectively.
| no_new_dataset | 0.706981 |
2305.11791 | Huiming Wang | Huiming Wang, Liying Cheng, Wenxuan Zhang, De Wen Soh, Lidong Bing | Enhancing Few-shot NER with Prompt Ordering based Data Augmentation | 7 pages, 2 figures | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recently, data augmentation (DA) methods have been proven to be effective for
pre-trained language models (PLMs) in low-resource settings, including few-shot
named entity recognition (NER). However, conventional NER DA methods are mostly
aimed at sequence labeling models, i.e., token-level classification, and few
are compatible with unified autoregressive generation frameworks, which can
handle a wider range of NER tasks, such as nested NER. Furthermore, these
generation frameworks have a strong assumption that the entities will appear in
the target sequence with the same left-to-right order as the source sequence.
In this paper, we claim that there is no need to keep this strict order, and
more diversified but reasonable target entity sequences can be provided during
the training stage as a novel DA method. Nevertheless, a naive mixture of
augmented data can confuse the model since one source sequence will then be
paired with different target sequences. Therefore, we propose a simple but
effective Prompt Ordering based Data Augmentation (PODA) method to improve the
training of unified autoregressive generation frameworks under few-shot NER
scenarios. Experimental results on three public NER datasets and further
analyses demonstrate the effectiveness of our approach.
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 16:25:43 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Wang",
"Huiming",
""
],
[
"Cheng",
"Liying",
""
],
[
"Zhang",
"Wenxuan",
""
],
[
"Soh",
"De Wen",
""
],
[
"Bing",
"Lidong",
""
]
] | TITLE: Enhancing Few-shot NER with Prompt Ordering based Data Augmentation
ABSTRACT: Recently, data augmentation (DA) methods have been proven to be effective for
pre-trained language models (PLMs) in low-resource settings, including few-shot
named entity recognition (NER). However, conventional NER DA methods are mostly
aimed at sequence labeling models, i.e., token-level classification, and few
are compatible with unified autoregressive generation frameworks, which can
handle a wider range of NER tasks, such as nested NER. Furthermore, these
generation frameworks have a strong assumption that the entities will appear in
the target sequence with the same left-to-right order as the source sequence.
In this paper, we claim that there is no need to keep this strict order, and
more diversified but reasonable target entity sequences can be provided during
the training stage as a novel DA method. Nevertheless, a naive mixture of
augmented data can confuse the model since one source sequence will then be
paired with different target sequences. Therefore, we propose a simple but
effective Prompt Ordering based Data Augmentation (PODA) method to improve the
training of unified autoregressive generation frameworks under few-shot NER
scenarios. Experimental results on three public NER datasets and further
analyses demonstrate the effectiveness of our approach.
| no_new_dataset | 0.710879 |
2305.11795 | Lydia Abady | Lydia Abady, Giovanna Maria Dimitri and Mauro Barni | A One-Class Classifier for the Detection of GAN Manipulated
Multi-Spectral Satellite Images | 17 pages, 4 figures | null | null | null | cs.CV eess.IV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The highly realistic image quality achieved by current image generative
models has many academic and industrial applications. To limit the use of such
models to benign applications, though, it is necessary that tools to
conclusively detect whether an image has been generated synthetically or not
are developed. For this reason, several detectors have been developed providing
excellent performance in computer vision applications, however, they can not be
applied as they are to multispectral satellite images, and hence new models
must be trained. In general, two-class classifiers can achieve very good
detection accuracies, however they are not able to generalise to image domains
and generative models architectures different than those used during training.
For this reason, in this paper, we propose a one-class classifier based on
Vector Quantized Variational Autoencoder 2 (VQ-VAE 2) features to overcome the
limitations of two-class classifiers. First, we emphasize the generalization
problem that binary classifiers suffer from by training and testing an
EfficientNet-B4 architecture on multiple multispectral datasets. Then we show
that, since the VQ-VAE 2 based classifier is trained only on pristine images,
it is able to detect images belonging to different domains and generated by
architectures that have not been used during training. Last, we compare the two
classifiers head-to-head on the same generated datasets, highlighting the
superiori generalization capabilities of the VQ-VAE 2-based detector.
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 16:30:50 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Abady",
"Lydia",
""
],
[
"Dimitri",
"Giovanna Maria",
""
],
[
"Barni",
"Mauro",
""
]
] | TITLE: A One-Class Classifier for the Detection of GAN Manipulated
Multi-Spectral Satellite Images
ABSTRACT: The highly realistic image quality achieved by current image generative
models has many academic and industrial applications. To limit the use of such
models to benign applications, though, it is necessary that tools to
conclusively detect whether an image has been generated synthetically or not
are developed. For this reason, several detectors have been developed providing
excellent performance in computer vision applications, however, they can not be
applied as they are to multispectral satellite images, and hence new models
must be trained. In general, two-class classifiers can achieve very good
detection accuracies, however they are not able to generalise to image domains
and generative models architectures different than those used during training.
For this reason, in this paper, we propose a one-class classifier based on
Vector Quantized Variational Autoencoder 2 (VQ-VAE 2) features to overcome the
limitations of two-class classifiers. First, we emphasize the generalization
problem that binary classifiers suffer from by training and testing an
EfficientNet-B4 architecture on multiple multispectral datasets. Then we show
that, since the VQ-VAE 2 based classifier is trained only on pristine images,
it is able to detect images belonging to different domains and generated by
architectures that have not been used during training. Last, we compare the two
classifiers head-to-head on the same generated datasets, highlighting the
superiori generalization capabilities of the VQ-VAE 2-based detector.
| no_new_dataset | 0.709831 |
2305.11805 | Emine Kucukbenli | Franco Pellegrini, Ruggero Lot, Yusuf Shaidu, Emine K\"u\c{c}\"ukbenli | PANNA 2.0: Efficient neural network interatomic potentials and new
architectures | Submitted to J Chem Phys Special Topic on Software for Atomistic
Machine Learning | null | null | null | physics.comp-ph cond-mat.mtrl-sci cs.LG physics.chem-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present the latest release of PANNA 2.0 (Properties from Artificial Neural
Network Architectures), a code for the generation of neural network interatomic
potentials based on local atomic descriptors and multilayer perceptrons. Built
on a new back end, this new release of PANNA features improved tools for
customizing and monitoring network training, better GPU support including a
fast descriptor calculator, new plugins for external codes and a new
architecture for the inclusion of long-range electrostatic interactions through
a variational charge equilibration scheme. We present an overview of the main
features of the new code, and several benchmarks comparing the accuracy of
PANNA models to the state of the art, on commonly used benchmarks as well as
richer datasets.
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 16:41:59 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Pellegrini",
"Franco",
""
],
[
"Lot",
"Ruggero",
""
],
[
"Shaidu",
"Yusuf",
""
],
[
"Küçükbenli",
"Emine",
""
]
] | TITLE: PANNA 2.0: Efficient neural network interatomic potentials and new
architectures
ABSTRACT: We present the latest release of PANNA 2.0 (Properties from Artificial Neural
Network Architectures), a code for the generation of neural network interatomic
potentials based on local atomic descriptors and multilayer perceptrons. Built
on a new back end, this new release of PANNA features improved tools for
customizing and monitoring network training, better GPU support including a
fast descriptor calculator, new plugins for external codes and a new
architecture for the inclusion of long-range electrostatic interactions through
a variational charge equilibration scheme. We present an overview of the main
features of the new code, and several benchmarks comparing the accuracy of
PANNA models to the state of the art, on commonly used benchmarks as well as
richer datasets.
| no_new_dataset | 0.708036 |
2305.11811 | Yang You | Yang You, Vincent Thomas, Francis Colas, Olivier Buffet | Monte-Carlo Search for an Equilibrium in Dec-POMDPs | Accepted to UAI 2023, preliminary version | null | null | null | cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Decentralized partially observable Markov decision processes (Dec-POMDPs)
formalize the problem of designing individual controllers for a group of
collaborative agents under stochastic dynamics and partial observability.
Seeking a global optimum is difficult (NEXP complete), but seeking a Nash
equilibrium -- each agent policy being a best response to the other agents --
is more accessible, and allowed addressing infinite-horizon problems with
solutions in the form of finite state controllers. In this paper, we show that
this approach can be adapted to cases where only a generative model (a
simulator) of the Dec-POMDP is available. This requires relying on a
simulation-based POMDP solver to construct an agent's FSC node by node. A
related process is used to heuristically derive initial FSCs. Experiment with
benchmarks shows that MC-JESP is competitive with exisiting Dec-POMDP solvers,
even better than many offline methods using explicit models.
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 16:47:46 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"You",
"Yang",
""
],
[
"Thomas",
"Vincent",
""
],
[
"Colas",
"Francis",
""
],
[
"Buffet",
"Olivier",
""
]
] | TITLE: Monte-Carlo Search for an Equilibrium in Dec-POMDPs
ABSTRACT: Decentralized partially observable Markov decision processes (Dec-POMDPs)
formalize the problem of designing individual controllers for a group of
collaborative agents under stochastic dynamics and partial observability.
Seeking a global optimum is difficult (NEXP complete), but seeking a Nash
equilibrium -- each agent policy being a best response to the other agents --
is more accessible, and allowed addressing infinite-horizon problems with
solutions in the form of finite state controllers. In this paper, we show that
this approach can be adapted to cases where only a generative model (a
simulator) of the Dec-POMDP is available. This requires relying on a
simulation-based POMDP solver to construct an agent's FSC node by node. A
related process is used to heuristically derive initial FSCs. Experiment with
benchmarks shows that MC-JESP is competitive with exisiting Dec-POMDP solvers,
even better than many offline methods using explicit models.
| no_new_dataset | 0.70866 |
2305.11832 | Agathe Senellart | Agathe Senellart, Cl\'ement Chadebec, St\'ephanie Allassonni\`ere | Improving Multimodal Joint Variational Autoencoders through Normalizing
Flows and Correlation Analysis | null | null | null | null | stat.ML cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We propose a new multimodal variational autoencoder that enables to generate
from the joint distribution and conditionally to any number of complex
modalities. The unimodal posteriors are conditioned on the Deep Canonical
Correlation Analysis embeddings which preserve the shared information across
modalities leading to more coherent cross-modal generations. Furthermore, we
use Normalizing Flows to enrich the unimodal posteriors and achieve more
diverse data generation. Finally, we propose to use a Product of Experts for
inferring one modality from several others which makes the model scalable to
any number of modalities. We demonstrate that our method improves likelihood
estimates, diversity of the generations and in particular coherence metrics in
the conditional generations on several datasets.
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 17:15:34 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Senellart",
"Agathe",
""
],
[
"Chadebec",
"Clément",
""
],
[
"Allassonnière",
"Stéphanie",
""
]
] | TITLE: Improving Multimodal Joint Variational Autoencoders through Normalizing
Flows and Correlation Analysis
ABSTRACT: We propose a new multimodal variational autoencoder that enables to generate
from the joint distribution and conditionally to any number of complex
modalities. The unimodal posteriors are conditioned on the Deep Canonical
Correlation Analysis embeddings which preserve the shared information across
modalities leading to more coherent cross-modal generations. Furthermore, we
use Normalizing Flows to enrich the unimodal posteriors and achieve more
diverse data generation. Finally, we propose to use a Product of Experts for
inferring one modality from several others which makes the model scalable to
any number of modalities. We demonstrate that our method improves likelihood
estimates, diversity of the generations and in particular coherence metrics in
the conditional generations on several datasets.
| no_new_dataset | 0.711186 |
2305.11840 | Akshita Jha | Akshita Jha, Aida Davani, Chandan K. Reddy, Shachi Dave, Vinodkumar
Prabhakaran, Sunipa Dev | SeeGULL: A Stereotype Benchmark with Broad Geo-Cultural Coverage
Leveraging Generative Models | null | null | null | null | cs.CL cs.CY | http://creativecommons.org/licenses/by/4.0/ | Stereotype benchmark datasets are crucial to detect and mitigate social
stereotypes about groups of people in NLP models. However, existing datasets
are limited in size and coverage, and are largely restricted to stereotypes
prevalent in the Western society. This is especially problematic as language
technologies gain hold across the globe. To address this gap, we present
SeeGULL, a broad-coverage stereotype dataset, built by utilizing generative
capabilities of large language models such as PaLM, and GPT-3, and leveraging a
globally diverse rater pool to validate the prevalence of those stereotypes in
society. SeeGULL is in English, and contains stereotypes about identity groups
spanning 178 countries across 8 different geo-political regions across 6
continents, as well as state-level identities within the US and India. We also
include fine-grained offensiveness scores for different stereotypes and
demonstrate their global disparities. Furthermore, we include comparative
annotations about the same groups by annotators living in the region vs. those
that are based in North America, and demonstrate that within-region stereotypes
about groups differ from those prevalent in North America. CONTENT WARNING:
This paper contains stereotype examples that may be offensive.
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 17:30:19 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Jha",
"Akshita",
""
],
[
"Davani",
"Aida",
""
],
[
"Reddy",
"Chandan K.",
""
],
[
"Dave",
"Shachi",
""
],
[
"Prabhakaran",
"Vinodkumar",
""
],
[
"Dev",
"Sunipa",
""
]
] | TITLE: SeeGULL: A Stereotype Benchmark with Broad Geo-Cultural Coverage
Leveraging Generative Models
ABSTRACT: Stereotype benchmark datasets are crucial to detect and mitigate social
stereotypes about groups of people in NLP models. However, existing datasets
are limited in size and coverage, and are largely restricted to stereotypes
prevalent in the Western society. This is especially problematic as language
technologies gain hold across the globe. To address this gap, we present
SeeGULL, a broad-coverage stereotype dataset, built by utilizing generative
capabilities of large language models such as PaLM, and GPT-3, and leveraging a
globally diverse rater pool to validate the prevalence of those stereotypes in
society. SeeGULL is in English, and contains stereotypes about identity groups
spanning 178 countries across 8 different geo-political regions across 6
continents, as well as state-level identities within the US and India. We also
include fine-grained offensiveness scores for different stereotypes and
demonstrate their global disparities. Furthermore, we include comparative
annotations about the same groups by annotators living in the region vs. those
that are based in North America, and demonstrate that within-region stereotypes
about groups differ from those prevalent in North America. CONTENT WARNING:
This paper contains stereotype examples that may be offensive.
| new_dataset | 0.715579 |
2305.11845 | Yujie Qian | Yujie Qian, Jiang Guo, Zhengkai Tu, Connor W. Coley, Regina Barzilay | RxnScribe: A Sequence Generation Model for Reaction Diagram Parsing | To be published in the Journal of Chemical Information and Modeling | null | null | null | cs.CL cs.AI cs.CV | http://creativecommons.org/licenses/by/4.0/ | Reaction diagram parsing is the task of extracting reaction schemes from a
diagram in the chemistry literature. The reaction diagrams can be arbitrarily
complex, thus robustly parsing them into structured data is an open challenge.
In this paper, we present RxnScribe, a machine learning model for parsing
reaction diagrams of varying styles. We formulate this structured prediction
task with a sequence generation approach, which condenses the traditional
pipeline into an end-to-end model. We train RxnScribe on a dataset of 1,378
diagrams and evaluate it with cross validation, achieving an 80.0% soft match
F1 score, with significant improvements over previous models. Our code and data
are publicly available at https://github.com/thomas0809/RxnScribe.
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 17:37:28 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Qian",
"Yujie",
""
],
[
"Guo",
"Jiang",
""
],
[
"Tu",
"Zhengkai",
""
],
[
"Coley",
"Connor W.",
""
],
[
"Barzilay",
"Regina",
""
]
] | TITLE: RxnScribe: A Sequence Generation Model for Reaction Diagram Parsing
ABSTRACT: Reaction diagram parsing is the task of extracting reaction schemes from a
diagram in the chemistry literature. The reaction diagrams can be arbitrarily
complex, thus robustly parsing them into structured data is an open challenge.
In this paper, we present RxnScribe, a machine learning model for parsing
reaction diagrams of varying styles. We formulate this structured prediction
task with a sequence generation approach, which condenses the traditional
pipeline into an end-to-end model. We train RxnScribe on a dataset of 1,378
diagrams and evaluate it with cross validation, achieving an 80.0% soft match
F1 score, with significant improvements over previous models. Our code and data
are publicly available at https://github.com/thomas0809/RxnScribe.
| no_new_dataset | 0.706209 |
2305.11846 | Ziyi Yang | Zineng Tang, Ziyi Yang, Chenguang Zhu, Michael Zeng, Mohit Bansal | Any-to-Any Generation via Composable Diffusion | Project Page: https://codi-gen.github.io | null | null | null | cs.CV cs.CL cs.LG cs.SD eess.AS | http://creativecommons.org/licenses/by/4.0/ | We present Composable Diffusion (CoDi), a novel generative model capable of
generating any combination of output modalities, such as language, image,
video, or audio, from any combination of input modalities. Unlike existing
generative AI systems, CoDi can generate multiple modalities in parallel and
its input is not limited to a subset of modalities like text or image. Despite
the absence of training datasets for many combinations of modalities, we
propose to align modalities in both the input and output space. This allows
CoDi to freely condition on any input combination and generate any group of
modalities, even if they are not present in the training data. CoDi employs a
novel composable generation strategy which involves building a shared
multimodal space by bridging alignment in the diffusion process, enabling the
synchronized generation of intertwined modalities, such as temporally aligned
video and audio. Highly customizable and flexible, CoDi achieves strong
joint-modality generation quality, and outperforms or is on par with the
unimodal state-of-the-art for single-modality synthesis. The project page with
demonstrations and code is at https://codi-gen.github.io
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 17:38:32 GMT"
}
] | 2023-05-22T00:00:00 | [
[
"Tang",
"Zineng",
""
],
[
"Yang",
"Ziyi",
""
],
[
"Zhu",
"Chenguang",
""
],
[
"Zeng",
"Michael",
""
],
[
"Bansal",
"Mohit",
""
]
] | TITLE: Any-to-Any Generation via Composable Diffusion
ABSTRACT: We present Composable Diffusion (CoDi), a novel generative model capable of
generating any combination of output modalities, such as language, image,
video, or audio, from any combination of input modalities. Unlike existing
generative AI systems, CoDi can generate multiple modalities in parallel and
its input is not limited to a subset of modalities like text or image. Despite
the absence of training datasets for many combinations of modalities, we
propose to align modalities in both the input and output space. This allows
CoDi to freely condition on any input combination and generate any group of
modalities, even if they are not present in the training data. CoDi employs a
novel composable generation strategy which involves building a shared
multimodal space by bridging alignment in the diffusion process, enabling the
synchronized generation of intertwined modalities, such as temporally aligned
video and audio. Highly customizable and flexible, CoDi achieves strong
joint-modality generation quality, and outperforms or is on par with the
unimodal state-of-the-art for single-modality synthesis. The project page with
demonstrations and code is at https://codi-gen.github.io
| no_new_dataset | 0.709403 |
2012.01724 | Ping-Yang Chen | Ping-Yang Chen, Ming-Ching Chang, Jun-Wei Hsieh, Yong-Sheng Chen | Parallel Residual Bi-Fusion Feature Pyramid Network for Accurate
Single-Shot Object Detection | accepted by IEEE transactions on Image Processing | IEEE Transactions on Image Processing, vol. 30, pp. 9099-9111,
2021 | 10.1109/TIP.2021.3118953 | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper proposes the Parallel Residual Bi-Fusion Feature Pyramid Network
(PRB-FPN) for fast and accurate single-shot object detection. Feature Pyramid
(FP) is widely used in recent visual detection, however the top-down pathway of
FP cannot preserve accurate localization due to pooling shifting. The advantage
of FP is weakened as deeper backbones with more layers are used. In addition,
it cannot keep up accurate detection of both small and large objects at the
same time. To address these issues, we propose a new parallel FP structure with
bi-directional (top-down and bottom-up) fusion and associated improvements to
retain high-quality features for accurate localization. We provide the
following design improvements: (1) A parallel bifusion FP structure with a
bottom-up fusion module (BFM) to detect both small and large objects at once
with high accuracy. (2) A concatenation and re-organization (CORE) module
provides a bottom-up pathway for feature fusion, which leads to the
bi-directional fusion FP that can recover lost information from lower-layer
feature maps. (3) The CORE feature is further purified to retain richer
contextual information. Such CORE purification in both top-down and bottom-up
pathways can be finished in only a few iterations. (4) The adding of a residual
design to CORE leads to a new Re-CORE module that enables easy training and
integration with a wide range of deeper or lighter backbones. The proposed
network achieves state-of-the-art performance on the UAVDT17 and MS COCO
datasets. Code is available at https://github.com/pingyang1117/PRBNet_PyTorch.
| [
{
"version": "v1",
"created": "Thu, 3 Dec 2020 06:51:20 GMT"
},
{
"version": "v2",
"created": "Fri, 17 Sep 2021 19:32:03 GMT"
},
{
"version": "v3",
"created": "Wed, 7 Sep 2022 09:36:06 GMT"
},
{
"version": "v4",
"created": "Mon, 20 Feb 2023 14:30:38 GMT"
},
{
"version": "v5",
"created": "Thu, 18 May 2023 15:33:06 GMT"
}
] | 2023-05-19T00:00:00 | [
[
"Chen",
"Ping-Yang",
""
],
[
"Chang",
"Ming-Ching",
""
],
[
"Hsieh",
"Jun-Wei",
""
],
[
"Chen",
"Yong-Sheng",
""
]
] | TITLE: Parallel Residual Bi-Fusion Feature Pyramid Network for Accurate
Single-Shot Object Detection
ABSTRACT: This paper proposes the Parallel Residual Bi-Fusion Feature Pyramid Network
(PRB-FPN) for fast and accurate single-shot object detection. Feature Pyramid
(FP) is widely used in recent visual detection, however the top-down pathway of
FP cannot preserve accurate localization due to pooling shifting. The advantage
of FP is weakened as deeper backbones with more layers are used. In addition,
it cannot keep up accurate detection of both small and large objects at the
same time. To address these issues, we propose a new parallel FP structure with
bi-directional (top-down and bottom-up) fusion and associated improvements to
retain high-quality features for accurate localization. We provide the
following design improvements: (1) A parallel bifusion FP structure with a
bottom-up fusion module (BFM) to detect both small and large objects at once
with high accuracy. (2) A concatenation and re-organization (CORE) module
provides a bottom-up pathway for feature fusion, which leads to the
bi-directional fusion FP that can recover lost information from lower-layer
feature maps. (3) The CORE feature is further purified to retain richer
contextual information. Such CORE purification in both top-down and bottom-up
pathways can be finished in only a few iterations. (4) The adding of a residual
design to CORE leads to a new Re-CORE module that enables easy training and
integration with a wide range of deeper or lighter backbones. The proposed
network achieves state-of-the-art performance on the UAVDT17 and MS COCO
datasets. Code is available at https://github.com/pingyang1117/PRBNet_PyTorch.
| no_new_dataset | 0.713656 |
2105.02046 | Yuan Zhou | Yuan Zhou, Yanrong Guo, Shijie Hao, Richang Hong, Jiebo Luo | Few-shot Partial Multi-view Learning | 18 pages | null | null | null | cs.CV cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | It is often the case that data are with multiple views in real-world
applications. Fully exploring the information of each view is significant for
making data more representative. However, due to various limitations and
failures in data collection and pre-processing, it is inevitable for real data
to suffer from view missing and data scarcity. The coexistence of these two
issues makes it more challenging to achieve the pattern classification task.
Currently, to our best knowledge, few appropriate methods can well-handle these
two issues simultaneously. Aiming to draw more attention from the community to
this challenge, we propose a new task in this paper, called few-shot partial
multi-view learning, which focuses on overcoming the negative impact of the
view-missing issue in the low-data regime. The challenges of this task are
twofold: (i) it is difficult to overcome the impact of data scarcity under the
interference of missing views; (ii) the limited number of data exacerbates
information scarcity, thus making it harder to address the view-missing issue
in turn. To address these challenges, we propose a new unified Gaussian
dense-anchoring method. The unified dense anchors are learned for the limited
partial multi-view data, thereby anchoring them into a unified dense
representation space where the influence of data scarcity and view missing can
be alleviated. We conduct extensive experiments to evaluate our method. The
results on Cub-googlenet-doc2vec, Handwritten, Caltech102, Scene15, Animal,
ORL, tieredImagenet, and Birds-200-2011 datasets validate its effectiveness.
| [
{
"version": "v1",
"created": "Wed, 5 May 2021 13:34:43 GMT"
},
{
"version": "v2",
"created": "Thu, 3 Mar 2022 08:46:44 GMT"
},
{
"version": "v3",
"created": "Thu, 23 Feb 2023 09:40:18 GMT"
},
{
"version": "v4",
"created": "Thu, 18 May 2023 13:25:25 GMT"
}
] | 2023-05-19T00:00:00 | [
[
"Zhou",
"Yuan",
""
],
[
"Guo",
"Yanrong",
""
],
[
"Hao",
"Shijie",
""
],
[
"Hong",
"Richang",
""
],
[
"Luo",
"Jiebo",
""
]
] | TITLE: Few-shot Partial Multi-view Learning
ABSTRACT: It is often the case that data are with multiple views in real-world
applications. Fully exploring the information of each view is significant for
making data more representative. However, due to various limitations and
failures in data collection and pre-processing, it is inevitable for real data
to suffer from view missing and data scarcity. The coexistence of these two
issues makes it more challenging to achieve the pattern classification task.
Currently, to our best knowledge, few appropriate methods can well-handle these
two issues simultaneously. Aiming to draw more attention from the community to
this challenge, we propose a new task in this paper, called few-shot partial
multi-view learning, which focuses on overcoming the negative impact of the
view-missing issue in the low-data regime. The challenges of this task are
twofold: (i) it is difficult to overcome the impact of data scarcity under the
interference of missing views; (ii) the limited number of data exacerbates
information scarcity, thus making it harder to address the view-missing issue
in turn. To address these challenges, we propose a new unified Gaussian
dense-anchoring method. The unified dense anchors are learned for the limited
partial multi-view data, thereby anchoring them into a unified dense
representation space where the influence of data scarcity and view missing can
be alleviated. We conduct extensive experiments to evaluate our method. The
results on Cub-googlenet-doc2vec, Handwritten, Caltech102, Scene15, Animal,
ORL, tieredImagenet, and Birds-200-2011 datasets validate its effectiveness.
| no_new_dataset | 0.710848 |
2109.04726 | Dong-Ho Lee | Dong-Ho Lee, Ravi Kiran Selvam, Sheikh Muhammad Sarwar, Bill Yuchen
Lin, Fred Morstatter, Jay Pujara, Elizabeth Boschee, James Allan and Xiang
Ren | AutoTriggER: Label-Efficient and Robust Named Entity Recognition with
Auxiliary Trigger Extraction | 15 pages, 13 figures, EACL 2023 | null | null | null | cs.CL cs.IR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Deep neural models for named entity recognition (NER) have shown impressive
results in overcoming label scarcity and generalizing to unseen entities by
leveraging distant supervision and auxiliary information such as explanations.
However, the costs of acquiring such additional information are generally
prohibitive. In this paper, we present a novel two-stage framework
(AutoTriggER) to improve NER performance by automatically generating and
leveraging ``entity triggers'' which are human-readable cues in the text that
help guide the model to make better decisions. Our framework leverages post-hoc
explanation to generate rationales and strengthens a model's prior knowledge
using an embedding interpolation technique. This approach allows models to
exploit triggers to infer entity boundaries and types instead of solely
memorizing the entity words themselves. Through experiments on three
well-studied NER datasets, AutoTriggER shows strong label-efficiency, is
capable of generalizing to unseen entities, and outperforms the RoBERTa-CRF
baseline by nearly 0.5 F1 points on average.
| [
{
"version": "v1",
"created": "Fri, 10 Sep 2021 08:11:56 GMT"
},
{
"version": "v2",
"created": "Thu, 14 Oct 2021 20:21:12 GMT"
},
{
"version": "v3",
"created": "Thu, 18 May 2023 06:04:10 GMT"
}
] | 2023-05-19T00:00:00 | [
[
"Lee",
"Dong-Ho",
""
],
[
"Selvam",
"Ravi Kiran",
""
],
[
"Sarwar",
"Sheikh Muhammad",
""
],
[
"Lin",
"Bill Yuchen",
""
],
[
"Morstatter",
"Fred",
""
],
[
"Pujara",
"Jay",
""
],
[
"Boschee",
"Elizabeth",
""
],
[
"Allan",
"James",
""
],
[
"Ren",
"Xiang",
""
]
] | TITLE: AutoTriggER: Label-Efficient and Robust Named Entity Recognition with
Auxiliary Trigger Extraction
ABSTRACT: Deep neural models for named entity recognition (NER) have shown impressive
results in overcoming label scarcity and generalizing to unseen entities by
leveraging distant supervision and auxiliary information such as explanations.
However, the costs of acquiring such additional information are generally
prohibitive. In this paper, we present a novel two-stage framework
(AutoTriggER) to improve NER performance by automatically generating and
leveraging ``entity triggers'' which are human-readable cues in the text that
help guide the model to make better decisions. Our framework leverages post-hoc
explanation to generate rationales and strengthens a model's prior knowledge
using an embedding interpolation technique. This approach allows models to
exploit triggers to infer entity boundaries and types instead of solely
memorizing the entity words themselves. Through experiments on three
well-studied NER datasets, AutoTriggER shows strong label-efficiency, is
capable of generalizing to unseen entities, and outperforms the RoBERTa-CRF
baseline by nearly 0.5 F1 points on average.
| no_new_dataset | 0.709233 |
2201.02018 | Tom\'a\v{s} Dlask | Tom\'a\v{s} Dlask, Tom\'a\v{s} Werner, Simon de Givry | Super-Reparametrizations of Weighted CSPs: Properties and Optimization
Perspective | null | null | 10.1007/s10601-023-09343-6 | null | math.OC cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The notion of reparametrizations of Weighted CSPs (WCSPs) (also known as
equivalence-preserving transformations of WCSPs) is well-known and finds its
use in many algorithms to approximate or bound the optimal WCSP value. In
contrast, the concept of super-reparametrizations (which are changes of the
weights that keep or increase the WCSP objective for every assignment) was
already proposed but never studied in detail. To fill this gap, we present a
number of theoretical properties of super-reparametrizations and compare them
to those of reparametrizations. Furthermore, we propose a framework for
computing upper bounds on the optimal value of the (maximization version of)
WCSP using super-reparametrizations. We show that it is in principle possible
to employ arbitrary (under some technical conditions) constraint propagation
rules to improve the bound. For arc consistency in particular, the method
reduces to the known Virtual AC (VAC) algorithm. We implemented the method for
singleton arc consistency (SAC) and compared it to other strong local
consistencies in WCSPs on a public benchmark. The results show that the bounds
obtained from SAC are superior for many instance groups.
| [
{
"version": "v1",
"created": "Thu, 6 Jan 2022 11:49:27 GMT"
},
{
"version": "v2",
"created": "Wed, 17 May 2023 19:20:59 GMT"
}
] | 2023-05-19T00:00:00 | [
[
"Dlask",
"Tomáš",
""
],
[
"Werner",
"Tomáš",
""
],
[
"de Givry",
"Simon",
""
]
] | TITLE: Super-Reparametrizations of Weighted CSPs: Properties and Optimization
Perspective
ABSTRACT: The notion of reparametrizations of Weighted CSPs (WCSPs) (also known as
equivalence-preserving transformations of WCSPs) is well-known and finds its
use in many algorithms to approximate or bound the optimal WCSP value. In
contrast, the concept of super-reparametrizations (which are changes of the
weights that keep or increase the WCSP objective for every assignment) was
already proposed but never studied in detail. To fill this gap, we present a
number of theoretical properties of super-reparametrizations and compare them
to those of reparametrizations. Furthermore, we propose a framework for
computing upper bounds on the optimal value of the (maximization version of)
WCSP using super-reparametrizations. We show that it is in principle possible
to employ arbitrary (under some technical conditions) constraint propagation
rules to improve the bound. For arc consistency in particular, the method
reduces to the known Virtual AC (VAC) algorithm. We implemented the method for
singleton arc consistency (SAC) and compared it to other strong local
consistencies in WCSPs on a public benchmark. The results show that the bounds
obtained from SAC are superior for many instance groups.
| no_new_dataset | 0.708629 |
2202.13047 | Chujie Zheng | Chujie Zheng, Sahand Sabour, Jiaxin Wen, Zheng Zhang, Minlie Huang | AugESC: Dialogue Augmentation with Large Language Models for Emotional
Support Conversation | Findings of ACL 2023 | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Crowdsourced dialogue corpora are usually limited in scale and topic coverage
due to the expensive cost of data curation. This would hinder the
generalization of downstream dialogue models to open-domain topics. In this
work, we leverage large language models for dialogue augmentation in the task
of emotional support conversation (ESC). By treating dialogue augmentation as a
dialogue completion task, we prompt a fine-tuned language model to complete
full dialogues from available dialogue posts of various topics, which are then
postprocessed based on heuristics. Applying this approach, we construct AugESC,
an augmented dataset for the ESC task, which largely extends the scale and
topic coverage of the crowdsourced ESConv corpus. Through comprehensive human
evaluation, we demonstrate that our approach is superior to strong baselines of
dialogue augmentation and that AugESC has comparable dialogue quality to the
crowdsourced corpus. We also conduct human interactive evaluation and prove
that post-training on AugESC improves downstream dialogue models'
generalization ability to open-domain topics. These results suggest the utility
of AugESC and highlight the potential of large language models in improving
data-scarce dialogue generation tasks.
| [
{
"version": "v1",
"created": "Sat, 26 Feb 2022 03:17:08 GMT"
},
{
"version": "v2",
"created": "Wed, 10 May 2023 08:34:56 GMT"
},
{
"version": "v3",
"created": "Thu, 18 May 2023 05:25:49 GMT"
}
] | 2023-05-19T00:00:00 | [
[
"Zheng",
"Chujie",
""
],
[
"Sabour",
"Sahand",
""
],
[
"Wen",
"Jiaxin",
""
],
[
"Zhang",
"Zheng",
""
],
[
"Huang",
"Minlie",
""
]
] | TITLE: AugESC: Dialogue Augmentation with Large Language Models for Emotional
Support Conversation
ABSTRACT: Crowdsourced dialogue corpora are usually limited in scale and topic coverage
due to the expensive cost of data curation. This would hinder the
generalization of downstream dialogue models to open-domain topics. In this
work, we leverage large language models for dialogue augmentation in the task
of emotional support conversation (ESC). By treating dialogue augmentation as a
dialogue completion task, we prompt a fine-tuned language model to complete
full dialogues from available dialogue posts of various topics, which are then
postprocessed based on heuristics. Applying this approach, we construct AugESC,
an augmented dataset for the ESC task, which largely extends the scale and
topic coverage of the crowdsourced ESConv corpus. Through comprehensive human
evaluation, we demonstrate that our approach is superior to strong baselines of
dialogue augmentation and that AugESC has comparable dialogue quality to the
crowdsourced corpus. We also conduct human interactive evaluation and prove
that post-training on AugESC improves downstream dialogue models'
generalization ability to open-domain topics. These results suggest the utility
of AugESC and highlight the potential of large language models in improving
data-scarce dialogue generation tasks.
| no_new_dataset | 0.685331 |
2203.16973 | Sreyan Ghosh | Ashish Seth and Lodagala V S V Durga Prasad and Sreyan Ghosh and S.
Umesh | Analyzing the factors affecting usefulness of Self-Supervised
Pre-trained Representations for Speech Recognition | null | null | null | null | cs.CL cs.SD eess.AS | http://creativecommons.org/licenses/by/4.0/ | Self-supervised learning (SSL) to learn high-level speech representations has
been a popular approach to building Automatic Speech Recognition (ASR) systems
in low-resource settings. However, the common assumption made in literature is
that a considerable amount of unlabeled data is available for the same domain
or language that can be leveraged for SSL pre-training, which we acknowledge is
not feasible in a real-world setting. In this paper, as part of the Interspeech
Gram Vaani ASR challenge, we try to study the effect of domain, language,
dataset size, and other aspects of our upstream pre-training SSL data on the
final performance low-resource downstream ASR task. We also build on the
continued pre-training paradigm to study the effect of prior knowledge
possessed by models trained using SSL. Extensive experiments and studies reveal
that the performance of ASR systems is susceptible to the data used for SSL
pre-training. Their performance improves with an increase in similarity and
volume of pre-training data. We believe our work will be helpful to the speech
community in building better ASR systems in low-resource settings and steer
research towards improving generalization in SSL-based pre-training for speech
systems.
| [
{
"version": "v1",
"created": "Thu, 31 Mar 2022 11:48:24 GMT"
},
{
"version": "v2",
"created": "Mon, 4 Apr 2022 12:35:46 GMT"
},
{
"version": "v3",
"created": "Thu, 18 Aug 2022 15:08:50 GMT"
},
{
"version": "v4",
"created": "Thu, 18 May 2023 01:24:47 GMT"
}
] | 2023-05-19T00:00:00 | [
[
"Seth",
"Ashish",
""
],
[
"Prasad",
"Lodagala V S V Durga",
""
],
[
"Ghosh",
"Sreyan",
""
],
[
"Umesh",
"S.",
""
]
] | TITLE: Analyzing the factors affecting usefulness of Self-Supervised
Pre-trained Representations for Speech Recognition
ABSTRACT: Self-supervised learning (SSL) to learn high-level speech representations has
been a popular approach to building Automatic Speech Recognition (ASR) systems
in low-resource settings. However, the common assumption made in literature is
that a considerable amount of unlabeled data is available for the same domain
or language that can be leveraged for SSL pre-training, which we acknowledge is
not feasible in a real-world setting. In this paper, as part of the Interspeech
Gram Vaani ASR challenge, we try to study the effect of domain, language,
dataset size, and other aspects of our upstream pre-training SSL data on the
final performance low-resource downstream ASR task. We also build on the
continued pre-training paradigm to study the effect of prior knowledge
possessed by models trained using SSL. Extensive experiments and studies reveal
that the performance of ASR systems is susceptible to the data used for SSL
pre-training. Their performance improves with an increase in similarity and
volume of pre-training data. We believe our work will be helpful to the speech
community in building better ASR systems in low-resource settings and steer
research towards improving generalization in SSL-based pre-training for speech
systems.
| no_new_dataset | 0.709636 |
2204.03083 | Davide Cozzolino | Davide Cozzolino, Alessandro Pianese, Matthias Nie{\ss}ner, Luisa
Verdoliva | Audio-Visual Person-of-Interest DeepFake Detection | null | null | null | null | cs.CV cs.SD eess.AS | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Face manipulation technology is advancing very rapidly, and new methods are
being proposed day by day. The aim of this work is to propose a deepfake
detector that can cope with the wide variety of manipulation methods and
scenarios encountered in the real world. Our key insight is that each person
has specific characteristics that a synthetic generator likely cannot
reproduce. Accordingly, we extract audio-visual features which characterize the
identity of a person, and use them to create a person-of-interest (POI)
deepfake detector. We leverage a contrastive learning paradigm to learn the
moving-face and audio segment embeddings that are most discriminative for each
identity. As a result, when the video and/or audio of a person is manipulated,
its representation in the embedding space becomes inconsistent with the real
identity, allowing reliable detection. Training is carried out exclusively on
real talking-face video; thus, the detector does not depend on any specific
manipulation method and yields the highest generalization ability. In addition,
our method can detect both single-modality (audio-only, video-only) and
multi-modality (audio-video) attacks, and is robust to low-quality or corrupted
videos. Experiments on a wide variety of datasets confirm that our method
ensures a SOTA performance, especially on low quality videos. Code is publicly
available on-line at https://github.com/grip-unina/poi-forensics.
| [
{
"version": "v1",
"created": "Wed, 6 Apr 2022 20:51:40 GMT"
},
{
"version": "v2",
"created": "Wed, 21 Dec 2022 23:07:27 GMT"
},
{
"version": "v3",
"created": "Thu, 18 May 2023 06:56:42 GMT"
}
] | 2023-05-19T00:00:00 | [
[
"Cozzolino",
"Davide",
""
],
[
"Pianese",
"Alessandro",
""
],
[
"Nießner",
"Matthias",
""
],
[
"Verdoliva",
"Luisa",
""
]
] | TITLE: Audio-Visual Person-of-Interest DeepFake Detection
ABSTRACT: Face manipulation technology is advancing very rapidly, and new methods are
being proposed day by day. The aim of this work is to propose a deepfake
detector that can cope with the wide variety of manipulation methods and
scenarios encountered in the real world. Our key insight is that each person
has specific characteristics that a synthetic generator likely cannot
reproduce. Accordingly, we extract audio-visual features which characterize the
identity of a person, and use them to create a person-of-interest (POI)
deepfake detector. We leverage a contrastive learning paradigm to learn the
moving-face and audio segment embeddings that are most discriminative for each
identity. As a result, when the video and/or audio of a person is manipulated,
its representation in the embedding space becomes inconsistent with the real
identity, allowing reliable detection. Training is carried out exclusively on
real talking-face video; thus, the detector does not depend on any specific
manipulation method and yields the highest generalization ability. In addition,
our method can detect both single-modality (audio-only, video-only) and
multi-modality (audio-video) attacks, and is robust to low-quality or corrupted
videos. Experiments on a wide variety of datasets confirm that our method
ensures a SOTA performance, especially on low quality videos. Code is publicly
available on-line at https://github.com/grip-unina/poi-forensics.
| no_new_dataset | 0.710848 |
2204.06674 | Anthony Colas | Anthony Colas, Mehrdad Alvandipour, Daisy Zhe Wang | GAP: A Graph-aware Language Model Framework for Knowledge Graph-to-Text
Generation | Accepted as a Main Conference Long paper at COLING 2022 | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Recent improvements in KG-to-text generation are due to additional auxiliary
pre-training tasks designed to give the fine-tune task a boost in performance.
These tasks require extensive computational resources while only suggesting
marginal improvements. Here, we demonstrate that by fusing graph-aware elements
into existing pre-trained language models, we are able to outperform
state-of-the-art models and close the gap imposed by additional pre-training
tasks. We do so by proposing a mask structure to capture neighborhood
information and a novel type encoder that adds a bias to the graph-attention
weights depending on the connection type. Experiments on two KG-to-text
benchmark datasets show our models are competitive while involving fewer
parameters and no additional pre-training tasks. By formulating the problem as
a framework, we can interchange the various proposed components and begin
interpreting KG-to-text generative models based on the topological and type
information found in a graph.
| [
{
"version": "v1",
"created": "Wed, 13 Apr 2022 23:53:37 GMT"
},
{
"version": "v2",
"created": "Thu, 13 Oct 2022 07:59:54 GMT"
},
{
"version": "v3",
"created": "Fri, 2 Dec 2022 03:11:11 GMT"
},
{
"version": "v4",
"created": "Thu, 18 May 2023 14:36:32 GMT"
}
] | 2023-05-19T00:00:00 | [
[
"Colas",
"Anthony",
""
],
[
"Alvandipour",
"Mehrdad",
""
],
[
"Wang",
"Daisy Zhe",
""
]
] | TITLE: GAP: A Graph-aware Language Model Framework for Knowledge Graph-to-Text
Generation
ABSTRACT: Recent improvements in KG-to-text generation are due to additional auxiliary
pre-training tasks designed to give the fine-tune task a boost in performance.
These tasks require extensive computational resources while only suggesting
marginal improvements. Here, we demonstrate that by fusing graph-aware elements
into existing pre-trained language models, we are able to outperform
state-of-the-art models and close the gap imposed by additional pre-training
tasks. We do so by proposing a mask structure to capture neighborhood
information and a novel type encoder that adds a bias to the graph-attention
weights depending on the connection type. Experiments on two KG-to-text
benchmark datasets show our models are competitive while involving fewer
parameters and no additional pre-training tasks. By formulating the problem as
a framework, we can interchange the various proposed components and begin
interpreting KG-to-text generative models based on the topological and type
information found in a graph.
| no_new_dataset | 0.708395 |
2205.10505 | Fuzhao Xue | Fuzhao Xue, Jianghai Chen, Aixin Sun, Xiaozhe Ren, Zangwei Zheng,
Xiaoxin He, Yongming Chen, Xin Jiang, Yang You | A Study on Transformer Configuration and Training Objective | Accepted at ICML 2023 | null | null | null | cs.LG cs.AI cs.CL cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Transformer-based models have delivered impressive results on many tasks,
particularly vision and language tasks. In many model training situations,
conventional configurations are typically adopted. For example, we often set
the base model with hidden dimensions (i.e. model width) to be 768 and the
number of transformer layers (i.e. model depth) to be 12. In this paper, we
revisit these conventional configurations. Through theoretical analysis and
experimental evaluation, we show that the masked autoencoder is effective in
alleviating the over-smoothing issue in deep transformer training. Based on
this finding, we propose Bamboo, an idea of using deeper and narrower
transformer configurations, for masked autoencoder training. On ImageNet, with
such a simple change in configuration, re-designed model achieves 87.1% top-1
accuracy and outperforms SoTA models like MAE and BEiT. On language tasks,
re-designed model outperforms BERT with default setting by 1.1 points on
average, on GLUE datasets.
| [
{
"version": "v1",
"created": "Sat, 21 May 2022 05:17:11 GMT"
},
{
"version": "v2",
"created": "Tue, 24 May 2022 08:03:25 GMT"
},
{
"version": "v3",
"created": "Thu, 18 May 2023 16:08:10 GMT"
}
] | 2023-05-19T00:00:00 | [
[
"Xue",
"Fuzhao",
""
],
[
"Chen",
"Jianghai",
""
],
[
"Sun",
"Aixin",
""
],
[
"Ren",
"Xiaozhe",
""
],
[
"Zheng",
"Zangwei",
""
],
[
"He",
"Xiaoxin",
""
],
[
"Chen",
"Yongming",
""
],
[
"Jiang",
"Xin",
""
],
[
"You",
"Yang",
""
]
] | TITLE: A Study on Transformer Configuration and Training Objective
ABSTRACT: Transformer-based models have delivered impressive results on many tasks,
particularly vision and language tasks. In many model training situations,
conventional configurations are typically adopted. For example, we often set
the base model with hidden dimensions (i.e. model width) to be 768 and the
number of transformer layers (i.e. model depth) to be 12. In this paper, we
revisit these conventional configurations. Through theoretical analysis and
experimental evaluation, we show that the masked autoencoder is effective in
alleviating the over-smoothing issue in deep transformer training. Based on
this finding, we propose Bamboo, an idea of using deeper and narrower
transformer configurations, for masked autoencoder training. On ImageNet, with
such a simple change in configuration, re-designed model achieves 87.1% top-1
accuracy and outperforms SoTA models like MAE and BEiT. On language tasks,
re-designed model outperforms BERT with default setting by 1.1 points on
average, on GLUE datasets.
| no_new_dataset | 0.71123 |
2205.14495 | Massimo Caccia | Massimo Caccia, Jonas Mueller, Taesup Kim, Laurent Charlin, Rasool
Fakoor | Task-Agnostic Continual Reinforcement Learning: Gaining Insights and
Overcoming Challenges | null | CoLLAs 2023 | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | Continual learning (CL) enables the development of models and agents that
learn from a sequence of tasks while addressing the limitations of standard
deep learning approaches, such as catastrophic forgetting. In this work, we
investigate the factors that contribute to the performance differences between
task-agnostic CL and multi-task (MTL) agents. We pose two hypotheses: (1)
task-agnostic methods might provide advantages in settings with limited data,
computation, or high dimensionality, and (2) faster adaptation may be
particularly beneficial in continual learning settings, helping to mitigate the
effects of catastrophic forgetting. To investigate these hypotheses, we
introduce a replay-based recurrent reinforcement learning (3RL) methodology for
task-agnostic CL agents. We assess 3RL on a synthetic task and the Meta-World
benchmark, which includes 50 unique manipulation tasks. Our results demonstrate
that 3RL outperforms baseline methods and can even surpass its multi-task
equivalent in challenging settings with high dimensionality. We also show that
the recurrent task-agnostic agent consistently outperforms or matches the
performance of its transformer-based counterpart. These findings provide
insights into the advantages of task-agnostic CL over task-aware MTL approaches
and highlight the potential of task-agnostic methods in resource-constrained,
high-dimensional, and multi-task environments.
| [
{
"version": "v1",
"created": "Sat, 28 May 2022 17:59:00 GMT"
},
{
"version": "v2",
"created": "Sun, 18 Sep 2022 14:30:50 GMT"
},
{
"version": "v3",
"created": "Wed, 17 May 2023 18:23:06 GMT"
}
] | 2023-05-19T00:00:00 | [
[
"Caccia",
"Massimo",
""
],
[
"Mueller",
"Jonas",
""
],
[
"Kim",
"Taesup",
""
],
[
"Charlin",
"Laurent",
""
],
[
"Fakoor",
"Rasool",
""
]
] | TITLE: Task-Agnostic Continual Reinforcement Learning: Gaining Insights and
Overcoming Challenges
ABSTRACT: Continual learning (CL) enables the development of models and agents that
learn from a sequence of tasks while addressing the limitations of standard
deep learning approaches, such as catastrophic forgetting. In this work, we
investigate the factors that contribute to the performance differences between
task-agnostic CL and multi-task (MTL) agents. We pose two hypotheses: (1)
task-agnostic methods might provide advantages in settings with limited data,
computation, or high dimensionality, and (2) faster adaptation may be
particularly beneficial in continual learning settings, helping to mitigate the
effects of catastrophic forgetting. To investigate these hypotheses, we
introduce a replay-based recurrent reinforcement learning (3RL) methodology for
task-agnostic CL agents. We assess 3RL on a synthetic task and the Meta-World
benchmark, which includes 50 unique manipulation tasks. Our results demonstrate
that 3RL outperforms baseline methods and can even surpass its multi-task
equivalent in challenging settings with high dimensionality. We also show that
the recurrent task-agnostic agent consistently outperforms or matches the
performance of its transformer-based counterpart. These findings provide
insights into the advantages of task-agnostic CL over task-aware MTL approaches
and highlight the potential of task-agnostic methods in resource-constrained,
high-dimensional, and multi-task environments.
| no_new_dataset | 0.707967 |
2207.00148 | Haoran Yang | Haoran Yang, Hongxu Chen, Sixiao Zhang, Xiangguo Sun, Qian Li, Xiangyu
Zhao, Guandong Xu | Generating Counterfactual Hard Negative Samples for Graph Contrastive
Learning | ACCEPTED BY WWW'2023 | null | 10.1145/3543507.3583499 | null | cs.LG cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Graph contrastive learning has emerged as a powerful tool for unsupervised
graph representation learning. The key to the success of graph contrastive
learning is to acquire high-quality positive and negative samples as
contrasting pairs for the purpose of learning underlying structural semantics
of the input graph. Recent works usually sample negative samples from the same
training batch with the positive samples, or from an external irrelevant graph.
However, a significant limitation lies in such strategies, which is the
unavoidable problem of sampling false negative samples. In this paper, we
propose a novel method to utilize \textbf{C}ounterfactual mechanism to generate
artificial hard negative samples for \textbf{G}raph \textbf{C}ontrastive
learning, namely \textbf{CGC}, which has a different perspective compared to
those sampling-based strategies. We utilize counterfactual mechanism to produce
hard negative samples, which ensures that the generated samples are similar to,
but have labels that different from the positive sample. The proposed method
achieves satisfying results on several datasets compared to some traditional
unsupervised graph learning methods and some SOTA graph contrastive learning
methods. We also conduct some supplementary experiments to give an extensive
illustration of the proposed method, including the performances of CGC with
different hard negative samples and evaluations for hard negative samples
generated with different similarity measurements.
| [
{
"version": "v1",
"created": "Fri, 1 Jul 2022 02:19:59 GMT"
},
{
"version": "v2",
"created": "Sun, 12 Feb 2023 04:38:02 GMT"
},
{
"version": "v3",
"created": "Thu, 18 May 2023 05:41:42 GMT"
}
] | 2023-05-19T00:00:00 | [
[
"Yang",
"Haoran",
""
],
[
"Chen",
"Hongxu",
""
],
[
"Zhang",
"Sixiao",
""
],
[
"Sun",
"Xiangguo",
""
],
[
"Li",
"Qian",
""
],
[
"Zhao",
"Xiangyu",
""
],
[
"Xu",
"Guandong",
""
]
] | TITLE: Generating Counterfactual Hard Negative Samples for Graph Contrastive
Learning
ABSTRACT: Graph contrastive learning has emerged as a powerful tool for unsupervised
graph representation learning. The key to the success of graph contrastive
learning is to acquire high-quality positive and negative samples as
contrasting pairs for the purpose of learning underlying structural semantics
of the input graph. Recent works usually sample negative samples from the same
training batch with the positive samples, or from an external irrelevant graph.
However, a significant limitation lies in such strategies, which is the
unavoidable problem of sampling false negative samples. In this paper, we
propose a novel method to utilize \textbf{C}ounterfactual mechanism to generate
artificial hard negative samples for \textbf{G}raph \textbf{C}ontrastive
learning, namely \textbf{CGC}, which has a different perspective compared to
those sampling-based strategies. We utilize counterfactual mechanism to produce
hard negative samples, which ensures that the generated samples are similar to,
but have labels that different from the positive sample. The proposed method
achieves satisfying results on several datasets compared to some traditional
unsupervised graph learning methods and some SOTA graph contrastive learning
methods. We also conduct some supplementary experiments to give an extensive
illustration of the proposed method, including the performances of CGC with
different hard negative samples and evaluations for hard negative samples
generated with different similarity measurements.
| no_new_dataset | 0.710051 |
2207.08597 | Hossein Hajiabolhassan | Hossein Hajiabolhassan, Zahra Taheri, Ali Hojatnia, Yavar Taheri
Yeganeh | FunQG: Molecular Representation Learning Via Quotient Graphs | null | null | 10.1021/acs.jcim.3c00445 | null | cs.LG cs.AI q-bio.BM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Learning expressive molecular representations is crucial to facilitate the
accurate prediction of molecular properties. Despite the significant
advancement of graph neural networks (GNNs) in molecular representation
learning, they generally face limitations such as neighbors-explosion,
under-reaching, over-smoothing, and over-squashing. Also, GNNs usually have
high computational costs because of the large-scale number of parameters.
Typically, such limitations emerge or increase when facing relatively
large-size graphs or using a deeper GNN model architecture. An idea to overcome
these problems is to simplify a molecular graph into a small, rich, and
informative one, which is more efficient and less challenging to train GNNs. To
this end, we propose a novel molecular graph coarsening framework named FunQG
utilizing Functional groups, as influential building blocks of a molecule to
determine its properties, based on a graph-theoretic concept called Quotient
Graph. By experiments, we show that the resulting informative graphs are much
smaller than the molecular graphs and thus are good candidates for training
GNNs. We apply the FunQG on popular molecular property prediction benchmarks
and then compare the performance of some popular baseline GNNs on the obtained
datasets with the performance of several state-of-the-art baselines on the
original datasets. By experiments, this method significantly outperforms
previous baselines on various datasets, besides its dramatic reduction in the
number of parameters and low computational costs. Therefore, the FunQG can be
used as a simple, cost-effective, and robust method for solving the molecular
representation learning problem.
| [
{
"version": "v1",
"created": "Mon, 18 Jul 2022 13:36:20 GMT"
},
{
"version": "v2",
"created": "Fri, 4 Nov 2022 08:42:25 GMT"
}
] | 2023-05-19T00:00:00 | [
[
"Hajiabolhassan",
"Hossein",
""
],
[
"Taheri",
"Zahra",
""
],
[
"Hojatnia",
"Ali",
""
],
[
"Yeganeh",
"Yavar Taheri",
""
]
] | TITLE: FunQG: Molecular Representation Learning Via Quotient Graphs
ABSTRACT: Learning expressive molecular representations is crucial to facilitate the
accurate prediction of molecular properties. Despite the significant
advancement of graph neural networks (GNNs) in molecular representation
learning, they generally face limitations such as neighbors-explosion,
under-reaching, over-smoothing, and over-squashing. Also, GNNs usually have
high computational costs because of the large-scale number of parameters.
Typically, such limitations emerge or increase when facing relatively
large-size graphs or using a deeper GNN model architecture. An idea to overcome
these problems is to simplify a molecular graph into a small, rich, and
informative one, which is more efficient and less challenging to train GNNs. To
this end, we propose a novel molecular graph coarsening framework named FunQG
utilizing Functional groups, as influential building blocks of a molecule to
determine its properties, based on a graph-theoretic concept called Quotient
Graph. By experiments, we show that the resulting informative graphs are much
smaller than the molecular graphs and thus are good candidates for training
GNNs. We apply the FunQG on popular molecular property prediction benchmarks
and then compare the performance of some popular baseline GNNs on the obtained
datasets with the performance of several state-of-the-art baselines on the
original datasets. By experiments, this method significantly outperforms
previous baselines on various datasets, besides its dramatic reduction in the
number of parameters and low computational costs. Therefore, the FunQG can be
used as a simple, cost-effective, and robust method for solving the molecular
representation learning problem.
| no_new_dataset | 0.710804 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.