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
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2202.13610 | Dominik Beese | Dominik Beese and Beg\"um Altunba\c{s} and G\"orkem G\"uzeler and
Steffen Eger | Did AI get more negative recently? | major revision, (extended) journal version with limitations section | R. Soc. open sci. 10:221159.221159 (2023) | 10.1098/rsos.221159 | null | cs.CL cs.LG cs.SI | http://creativecommons.org/licenses/by/4.0/ | In this paper, we classify scientific articles in the domain of natural
language processing (NLP) and machine learning (ML), as core subfields of
artificial intelligence (AI), into whether (i) they extend the current
state-of-the-art by the introduction of novel techniques which beat existing
models or whether (ii) they mainly criticize the existing state-of-the-art,
i.e. that it is deficient with respect to some property (e.g. wrong evaluation,
wrong datasets, misleading task specification). We refer to contributions under
(i) as having a 'positive stance' and contributions under (ii) as having a
'negative stance' (to related work). We annotate over 1.5 k papers from NLP and
ML to train a SciBERT-based model to automatically predict the stance of a
paper based on its title and abstract. We then analyse large-scale trends on
over 41 k papers from the last approximately 35 years in NLP and ML, finding
that papers have become substantially more positive over time, but negative
papers also got more negative and we observe considerably more negative papers
in recent years. Negative papers are also more influential in terms of
citations they receive.
| [
{
"version": "v1",
"created": "Mon, 28 Feb 2022 08:37:03 GMT"
},
{
"version": "v2",
"created": "Mon, 5 Sep 2022 16:48:49 GMT"
},
{
"version": "v3",
"created": "Thu, 29 Jun 2023 08:53:07 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Beese",
"Dominik",
""
],
[
"Altunbaş",
"Begüm",
""
],
[
"Güzeler",
"Görkem",
""
],
[
"Eger",
"Steffen",
""
]
] | TITLE: Did AI get more negative recently?
ABSTRACT: In this paper, we classify scientific articles in the domain of natural
language processing (NLP) and machine learning (ML), as core subfields of
artificial intelligence (AI), into whether (i) they extend the current
state-of-the-art by the introduction of novel techniques which beat existing
models or whether (ii) they mainly criticize the existing state-of-the-art,
i.e. that it is deficient with respect to some property (e.g. wrong evaluation,
wrong datasets, misleading task specification). We refer to contributions under
(i) as having a 'positive stance' and contributions under (ii) as having a
'negative stance' (to related work). We annotate over 1.5 k papers from NLP and
ML to train a SciBERT-based model to automatically predict the stance of a
paper based on its title and abstract. We then analyse large-scale trends on
over 41 k papers from the last approximately 35 years in NLP and ML, finding
that papers have become substantially more positive over time, but negative
papers also got more negative and we observe considerably more negative papers
in recent years. Negative papers are also more influential in terms of
citations they receive.
| no_new_dataset | 0.70695 |
2204.04636 | Javier Rando | Edoardo Mosca and Shreyash Agarwal and Javier Rando and Georg Groh | "That Is a Suspicious Reaction!": Interpreting Logits Variation to
Detect NLP Adversarial Attacks | ACL 2022 | null | 10.18653/v1/2022.acl-long.538 | null | cs.AI cs.CL cs.LG | http://creativecommons.org/licenses/by/4.0/ | Adversarial attacks are a major challenge faced by current machine learning
research. These purposely crafted inputs fool even the most advanced models,
precluding their deployment in safety-critical applications. Extensive research
in computer vision has been carried to develop reliable defense strategies.
However, the same issue remains less explored in natural language processing.
Our work presents a model-agnostic detector of adversarial text examples. The
approach identifies patterns in the logits of the target classifier when
perturbing the input text. The proposed detector improves the current
state-of-the-art performance in recognizing adversarial inputs and exhibits
strong generalization capabilities across different NLP models, datasets, and
word-level attacks.
| [
{
"version": "v1",
"created": "Sun, 10 Apr 2022 09:24:41 GMT"
},
{
"version": "v2",
"created": "Thu, 29 Jun 2023 13:02:28 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Mosca",
"Edoardo",
""
],
[
"Agarwal",
"Shreyash",
""
],
[
"Rando",
"Javier",
""
],
[
"Groh",
"Georg",
""
]
] | TITLE: "That Is a Suspicious Reaction!": Interpreting Logits Variation to
Detect NLP Adversarial Attacks
ABSTRACT: Adversarial attacks are a major challenge faced by current machine learning
research. These purposely crafted inputs fool even the most advanced models,
precluding their deployment in safety-critical applications. Extensive research
in computer vision has been carried to develop reliable defense strategies.
However, the same issue remains less explored in natural language processing.
Our work presents a model-agnostic detector of adversarial text examples. The
approach identifies patterns in the logits of the target classifier when
perturbing the input text. The proposed detector improves the current
state-of-the-art performance in recognizing adversarial inputs and exhibits
strong generalization capabilities across different NLP models, datasets, and
word-level attacks.
| no_new_dataset | 0.703524 |
2204.13263 | Kazuki Adachi | Kazuki Adachi, Shin'ya Yamaguchi, Atsutoshi Kumagai | Covariance-aware Feature Alignment with Pre-computed Source Statistics
for Test-time Adaptation to Multiple Image Corruptions | Extended version of the paper accepted to ICIP 2023 | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Real-world image recognition systems often face corrupted input images, which
cause distribution shifts and degrade the performance of models. These systems
often use a single prediction model in a central server and process images sent
from various environments, such as cameras distributed in cities or cars. Such
single models face images corrupted in heterogeneous ways in test time. Thus,
they require to instantly adapt to the multiple corruptions during testing
rather than being re-trained at a high cost. Test-time adaptation (TTA), which
aims to adapt models without accessing the training dataset, is one of the
settings that can address this problem. Existing TTA methods indeed work well
on a single corruption. However, the adaptation ability is limited when
multiple types of corruption occur, which is more realistic. We hypothesize
this is because the distribution shift is more complicated, and the adaptation
becomes more difficult in case of multiple corruptions. In fact, we
experimentally found that a larger distribution gap remains after TTA. To
address the distribution gap during testing, we propose a novel TTA method
named Covariance-Aware Feature alignment (CAFe). We empirically show that CAFe
outperforms prior TTA methods on image corruptions, including multiple types of
corruptions.
| [
{
"version": "v1",
"created": "Thu, 28 Apr 2022 02:50:43 GMT"
},
{
"version": "v2",
"created": "Thu, 29 Jun 2023 07:08:00 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Adachi",
"Kazuki",
""
],
[
"Yamaguchi",
"Shin'ya",
""
],
[
"Kumagai",
"Atsutoshi",
""
]
] | TITLE: Covariance-aware Feature Alignment with Pre-computed Source Statistics
for Test-time Adaptation to Multiple Image Corruptions
ABSTRACT: Real-world image recognition systems often face corrupted input images, which
cause distribution shifts and degrade the performance of models. These systems
often use a single prediction model in a central server and process images sent
from various environments, such as cameras distributed in cities or cars. Such
single models face images corrupted in heterogeneous ways in test time. Thus,
they require to instantly adapt to the multiple corruptions during testing
rather than being re-trained at a high cost. Test-time adaptation (TTA), which
aims to adapt models without accessing the training dataset, is one of the
settings that can address this problem. Existing TTA methods indeed work well
on a single corruption. However, the adaptation ability is limited when
multiple types of corruption occur, which is more realistic. We hypothesize
this is because the distribution shift is more complicated, and the adaptation
becomes more difficult in case of multiple corruptions. In fact, we
experimentally found that a larger distribution gap remains after TTA. To
address the distribution gap during testing, we propose a novel TTA method
named Covariance-Aware Feature alignment (CAFe). We empirically show that CAFe
outperforms prior TTA methods on image corruptions, including multiple types of
corruptions.
| no_new_dataset | 0.710622 |
2206.12617 | Xintao Wang | Xintao Wang, Qianyu He, Jiaqing Liang and Yanghua Xiao | Language Models as Knowledge Embeddings | This revision corrects some texts after fixing a data leakage issue | null | null | null | cs.CL cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Knowledge embeddings (KE) represent a knowledge graph (KG) by embedding
entities and relations into continuous vector spaces. Existing methods are
mainly structure-based or description-based. Structure-based methods learn
representations that preserve the inherent structure of KGs. They cannot well
represent abundant long-tail entities in real-world KGs with limited structural
information. Description-based methods leverage textual information and
language models. Prior approaches in this direction barely outperform
structure-based ones, and suffer from problems like expensive negative sampling
and restrictive description demand. In this paper, we propose LMKE, which
adopts Language Models to derive Knowledge Embeddings, aiming at both enriching
representations of long-tail entities and solving problems of prior
description-based methods. We formulate description-based KE learning with a
contrastive learning framework to improve efficiency in training and
evaluation. Experimental results show that LMKE achieves state-of-the-art
performance on KE benchmarks of link prediction and triple classification,
especially for long-tail entities.
| [
{
"version": "v1",
"created": "Sat, 25 Jun 2022 10:39:12 GMT"
},
{
"version": "v2",
"created": "Sun, 25 Jun 2023 08:23:04 GMT"
},
{
"version": "v3",
"created": "Thu, 29 Jun 2023 07:11:19 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Wang",
"Xintao",
""
],
[
"He",
"Qianyu",
""
],
[
"Liang",
"Jiaqing",
""
],
[
"Xiao",
"Yanghua",
""
]
] | TITLE: Language Models as Knowledge Embeddings
ABSTRACT: Knowledge embeddings (KE) represent a knowledge graph (KG) by embedding
entities and relations into continuous vector spaces. Existing methods are
mainly structure-based or description-based. Structure-based methods learn
representations that preserve the inherent structure of KGs. They cannot well
represent abundant long-tail entities in real-world KGs with limited structural
information. Description-based methods leverage textual information and
language models. Prior approaches in this direction barely outperform
structure-based ones, and suffer from problems like expensive negative sampling
and restrictive description demand. In this paper, we propose LMKE, which
adopts Language Models to derive Knowledge Embeddings, aiming at both enriching
representations of long-tail entities and solving problems of prior
description-based methods. We formulate description-based KE learning with a
contrastive learning framework to improve efficiency in training and
evaluation. Experimental results show that LMKE achieves state-of-the-art
performance on KE benchmarks of link prediction and triple classification,
especially for long-tail entities.
| no_new_dataset | 0.70844 |
2207.00328 | Khang Truong Giang | Khang Truong Giang, Soohwan Song, Sungho Jo | TopicFM: Robust and Interpretable Topic-Assisted Feature Matching | Accepted at AAAI-23. This version includes both main text and
supplementary materials | null | 10.1609/aaai.v37i2.25341 | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | This study addresses an image-matching problem in challenging cases, such as
large scene variations or textureless scenes. To gain robustness to such
situations, most previous studies have attempted to encode the global contexts
of a scene via graph neural networks or transformers. However, these contexts
do not explicitly represent high-level contextual information, such as
structural shapes or semantic instances; therefore, the encoded features are
still not sufficiently discriminative in challenging scenes. We propose a novel
image-matching method that applies a topic-modeling strategy to encode
high-level contexts in images. The proposed method trains latent semantic
instances called topics. It explicitly models an image as a multinomial
distribution of topics, and then performs probabilistic feature matching. This
approach improves the robustness of matching by focusing on the same semantic
areas between the images. In addition, the inferred topics provide
interpretability for matching the results, making our method explainable.
Extensive experiments on outdoor and indoor datasets show that our method
outperforms other state-of-the-art methods, particularly in challenging cases.
The code is available at https://github.com/TruongKhang/TopicFM.
| [
{
"version": "v1",
"created": "Fri, 1 Jul 2022 10:39:14 GMT"
},
{
"version": "v2",
"created": "Mon, 29 Aug 2022 06:06:09 GMT"
},
{
"version": "v3",
"created": "Tue, 29 Nov 2022 07:12:04 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Giang",
"Khang Truong",
""
],
[
"Song",
"Soohwan",
""
],
[
"Jo",
"Sungho",
""
]
] | TITLE: TopicFM: Robust and Interpretable Topic-Assisted Feature Matching
ABSTRACT: This study addresses an image-matching problem in challenging cases, such as
large scene variations or textureless scenes. To gain robustness to such
situations, most previous studies have attempted to encode the global contexts
of a scene via graph neural networks or transformers. However, these contexts
do not explicitly represent high-level contextual information, such as
structural shapes or semantic instances; therefore, the encoded features are
still not sufficiently discriminative in challenging scenes. We propose a novel
image-matching method that applies a topic-modeling strategy to encode
high-level contexts in images. The proposed method trains latent semantic
instances called topics. It explicitly models an image as a multinomial
distribution of topics, and then performs probabilistic feature matching. This
approach improves the robustness of matching by focusing on the same semantic
areas between the images. In addition, the inferred topics provide
interpretability for matching the results, making our method explainable.
Extensive experiments on outdoor and indoor datasets show that our method
outperforms other state-of-the-art methods, particularly in challenging cases.
The code is available at https://github.com/TruongKhang/TopicFM.
| no_new_dataset | 0.71103 |
2207.10774 | Bastian Wittmann | Bastian Wittmann, Fernando Navarro, Suprosanna Shit, Bjoern Menze | Focused Decoding Enables 3D Anatomical Detection by Transformers | Accepted for publication at the Journal of Machine Learning for
Biomedical Imaging (MELBA) https://melba-journal.org/2023:003 | Machine.Learning.for.Biomedical.Imaging. 2 (2023) | 10.59275/j.melba.2023-35e6 | null | cs.CV cs.AI | http://creativecommons.org/licenses/by/4.0/ | Detection Transformers represent end-to-end object detection approaches based
on a Transformer encoder-decoder architecture, exploiting the attention
mechanism for global relation modeling. Although Detection Transformers deliver
results on par with or even superior to their highly optimized CNN-based
counterparts operating on 2D natural images, their success is closely coupled
to access to a vast amount of training data. This, however, restricts the
feasibility of employing Detection Transformers in the medical domain, as
access to annotated data is typically limited. To tackle this issue and
facilitate the advent of medical Detection Transformers, we propose a novel
Detection Transformer for 3D anatomical structure detection, dubbed Focused
Decoder. Focused Decoder leverages information from an anatomical region atlas
to simultaneously deploy query anchors and restrict the cross-attention's field
of view to regions of interest, which allows for a precise focus on relevant
anatomical structures. We evaluate our proposed approach on two publicly
available CT datasets and demonstrate that Focused Decoder not only provides
strong detection results and thus alleviates the need for a vast amount of
annotated data but also exhibits exceptional and highly intuitive
explainability of results via attention weights. Our code is available at
https://github.com/bwittmann/transoar.
| [
{
"version": "v1",
"created": "Thu, 21 Jul 2022 22:17:21 GMT"
},
{
"version": "v2",
"created": "Mon, 15 Aug 2022 22:32:30 GMT"
},
{
"version": "v3",
"created": "Mon, 17 Oct 2022 09:50:32 GMT"
},
{
"version": "v4",
"created": "Sun, 26 Feb 2023 19:43:42 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Wittmann",
"Bastian",
""
],
[
"Navarro",
"Fernando",
""
],
[
"Shit",
"Suprosanna",
""
],
[
"Menze",
"Bjoern",
""
]
] | TITLE: Focused Decoding Enables 3D Anatomical Detection by Transformers
ABSTRACT: Detection Transformers represent end-to-end object detection approaches based
on a Transformer encoder-decoder architecture, exploiting the attention
mechanism for global relation modeling. Although Detection Transformers deliver
results on par with or even superior to their highly optimized CNN-based
counterparts operating on 2D natural images, their success is closely coupled
to access to a vast amount of training data. This, however, restricts the
feasibility of employing Detection Transformers in the medical domain, as
access to annotated data is typically limited. To tackle this issue and
facilitate the advent of medical Detection Transformers, we propose a novel
Detection Transformer for 3D anatomical structure detection, dubbed Focused
Decoder. Focused Decoder leverages information from an anatomical region atlas
to simultaneously deploy query anchors and restrict the cross-attention's field
of view to regions of interest, which allows for a precise focus on relevant
anatomical structures. We evaluate our proposed approach on two publicly
available CT datasets and demonstrate that Focused Decoder not only provides
strong detection results and thus alleviates the need for a vast amount of
annotated data but also exhibits exceptional and highly intuitive
explainability of results via attention weights. Our code is available at
https://github.com/bwittmann/transoar.
| no_new_dataset | 0.709447 |
2209.01999 | Fanfan Meng | F. Meng, S.-Y. Lee, O. Zografos, M. Gupta, V.D. Nguyen, G. De Micheli,
S. Cotofana, I. Asselberghs, C. Adelmann, G. Sankar Kar, S. Couet, and F.
Ciubotaru | Perspectives and Challenges of Scaled Boolean Spintronic Circuits Based
on Magnetic Tunnel Junction Transducers | This work was supported by imec Industrial Affiliation Program on
Exploratory Logic Devices. It has also received funding from the European
Union Horizon Europe research and innovation programme within the project
SPIDER under grant agreement No 101070417 | null | null | null | physics.app-ph cs.ET | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper addresses the question: Can spintronic circuits based on Magnetic
Tunnel Junction (MTJ) transducers outperform their state-of-the-art CMOS
counterparts? To this end, we use the EPFL combinational benchmark sets,
synthesize them in 7 nm CMOS and in MTJ-based spintronic technologies, and
compare the two implementation methods in terms of Energy-Delay-Product (EDP).
To fully utilize the technologies potential, CMOS and spintronic
implementations are built upon standard Boolean and Majority Gates,
respectively. For the spintronic circuits, we assumed that domain conversion
(electric/magnetic to magnetic/electric) is performed by means of MTJs and the
computation is accomplished by domain wall based majority gates, and considered
two EDP estimation scenarios: (i) Uniform Benchmarking, which ignores the
circuit's internal structure and only includes domain transducers power and
delay contributions into the calculations, and (ii) Majority-Inverter-Graph
Benchmarking, which also embeds the circuit structure, the associated critical
path delay and energy consumption by DW propagation. Our results indicate that
for the uniform case, the spintronic route is better suited for the
implementation of complex circuits with few inputs and outputs. On the other
hand, when the circuit structure is also considered via majority and inverter
synthesis, our analysis clearly indicates that in order to match and eventually
outperform CMOS performance, MTJ efficiency has to be improved by 3-4 orders of
magnitude. While it is clear that for the time being the MTJ-based-spintronic
way cannot compete with CMOS, further transducer developments may tip the
balance, which, when combined with information non-volatility, may make
spintronic implementation for certain applications that require a large number
of calculations and have a rather limited amount of interaction with the
environment.
| [
{
"version": "v1",
"created": "Mon, 5 Sep 2022 14:56:09 GMT"
},
{
"version": "v2",
"created": "Thu, 8 Sep 2022 08:24:27 GMT"
},
{
"version": "v3",
"created": "Thu, 29 Jun 2023 16:43:46 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Meng",
"F.",
""
],
[
"Lee",
"S. -Y.",
""
],
[
"Zografos",
"O.",
""
],
[
"Gupta",
"M.",
""
],
[
"Nguyen",
"V. D.",
""
],
[
"De Micheli",
"G.",
""
],
[
"Cotofana",
"S.",
""
],
[
"Asselberghs",
"I.",
""
],
[
"Adelmann",
"C.",
""
],
[
"Kar",
"G. Sankar",
""
],
[
"Couet",
"S.",
""
],
[
"Ciubotaru",
"F.",
""
]
] | TITLE: Perspectives and Challenges of Scaled Boolean Spintronic Circuits Based
on Magnetic Tunnel Junction Transducers
ABSTRACT: This paper addresses the question: Can spintronic circuits based on Magnetic
Tunnel Junction (MTJ) transducers outperform their state-of-the-art CMOS
counterparts? To this end, we use the EPFL combinational benchmark sets,
synthesize them in 7 nm CMOS and in MTJ-based spintronic technologies, and
compare the two implementation methods in terms of Energy-Delay-Product (EDP).
To fully utilize the technologies potential, CMOS and spintronic
implementations are built upon standard Boolean and Majority Gates,
respectively. For the spintronic circuits, we assumed that domain conversion
(electric/magnetic to magnetic/electric) is performed by means of MTJs and the
computation is accomplished by domain wall based majority gates, and considered
two EDP estimation scenarios: (i) Uniform Benchmarking, which ignores the
circuit's internal structure and only includes domain transducers power and
delay contributions into the calculations, and (ii) Majority-Inverter-Graph
Benchmarking, which also embeds the circuit structure, the associated critical
path delay and energy consumption by DW propagation. Our results indicate that
for the uniform case, the spintronic route is better suited for the
implementation of complex circuits with few inputs and outputs. On the other
hand, when the circuit structure is also considered via majority and inverter
synthesis, our analysis clearly indicates that in order to match and eventually
outperform CMOS performance, MTJ efficiency has to be improved by 3-4 orders of
magnitude. While it is clear that for the time being the MTJ-based-spintronic
way cannot compete with CMOS, further transducer developments may tip the
balance, which, when combined with information non-volatility, may make
spintronic implementation for certain applications that require a large number
of calculations and have a rather limited amount of interaction with the
environment.
| no_new_dataset | 0.710465 |
2209.04477 | Lekan Molu | Leilei Cui and Lekan Molu | Robust Policy Optimization in Continuous-time Mixed
$\mathcal{H}_2/\mathcal{H}_\infty$ Stochastic Control | null | null | null | null | eess.SY cs.SY | http://creativecommons.org/licenses/by/4.0/ | Following the recent resurgence in establishing linear control theoretic
benchmarks for reinforcement leaning (RL)-based policy optimization (PO) for
complex dynamical systems with continuous state and action spaces, an optimal
control problem for a continuous-time infinite-dimensional linear stochastic
system possessing additive Brownian motion is optimized on a cost that is an
exponent of the quadratic form of the state, input, and disturbance terms. We
lay out a model-based and model-free algorithm for RL-based stochastic PO. For
the model-based algorithm, we establish rigorous convergence guarantees. For
the sampling-based algorithm, over trajectory arcs that emanate from the phase
space, we find that the Hamilton-Jacobi Bellman equation parameterizes
trajectory costs -- resulting in a discrete-time (input and state-based)
sampling scheme accompanied by unknown nonlinear dynamics with continuous-time
policy iterates. The need for known dynamics operators is circumvented and we
arrive at a reinforced PO algorithm (via policy iteration) where an upper bound
on the $\mathcal{H}_2$ norm is minimized (to guarantee stability) and a
robustness metric is enforced by maximizing the cost with respect to a
controller that includes the level of noise attenuation specified by the
system's $H_\infty$ norm. Rigorous robustness analyses is prescribed in an
input-to-state stability formalism. Our analyses and contributions are
distinguished by many natural systems characterized by additive Wiener process,
amenable to \^Ito's stochastic differential calculus in dynamic game settings.
| [
{
"version": "v1",
"created": "Fri, 9 Sep 2022 18:13:20 GMT"
},
{
"version": "v2",
"created": "Wed, 19 Oct 2022 14:50:22 GMT"
},
{
"version": "v3",
"created": "Thu, 29 Jun 2023 15:11:33 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Cui",
"Leilei",
""
],
[
"Molu",
"Lekan",
""
]
] | TITLE: Robust Policy Optimization in Continuous-time Mixed
$\mathcal{H}_2/\mathcal{H}_\infty$ Stochastic Control
ABSTRACT: Following the recent resurgence in establishing linear control theoretic
benchmarks for reinforcement leaning (RL)-based policy optimization (PO) for
complex dynamical systems with continuous state and action spaces, an optimal
control problem for a continuous-time infinite-dimensional linear stochastic
system possessing additive Brownian motion is optimized on a cost that is an
exponent of the quadratic form of the state, input, and disturbance terms. We
lay out a model-based and model-free algorithm for RL-based stochastic PO. For
the model-based algorithm, we establish rigorous convergence guarantees. For
the sampling-based algorithm, over trajectory arcs that emanate from the phase
space, we find that the Hamilton-Jacobi Bellman equation parameterizes
trajectory costs -- resulting in a discrete-time (input and state-based)
sampling scheme accompanied by unknown nonlinear dynamics with continuous-time
policy iterates. The need for known dynamics operators is circumvented and we
arrive at a reinforced PO algorithm (via policy iteration) where an upper bound
on the $\mathcal{H}_2$ norm is minimized (to guarantee stability) and a
robustness metric is enforced by maximizing the cost with respect to a
controller that includes the level of noise attenuation specified by the
system's $H_\infty$ norm. Rigorous robustness analyses is prescribed in an
input-to-state stability formalism. Our analyses and contributions are
distinguished by many natural systems characterized by additive Wiener process,
amenable to \^Ito's stochastic differential calculus in dynamic game settings.
| no_new_dataset | 0.709849 |
2211.06277 | Laura State | Laura State, Hadrien Salat, Stefania Rubrichi and Zbigniew Smoreda | Explainability in Practice: Estimating Electrification Rates from Mobile
Phone Data in Senegal | The 1st World Conference on eXplainable Artificial Intelligence (xAI
2023) | null | null | null | cs.CY cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Explainable artificial intelligence (XAI) provides explanations for not
interpretable machine learning (ML) models. While many technical approaches
exist, there is a lack of validation of these techniques on real-world
datasets. In this work, we present a use-case of XAI: an ML model which is
trained to estimate electrification rates based on mobile phone data in
Senegal. The data originate from the Data for Development challenge by Orange
in 2014/15. We apply two model-agnostic, local explanation techniques and find
that while the model can be verified, it is biased with respect to the
population density. We conclude our paper by pointing to the two main
challenges we encountered during our work: data processing and model design
that might be restricted by currently available XAI methods, and the importance
of domain knowledge to interpret explanations.
| [
{
"version": "v1",
"created": "Fri, 11 Nov 2022 15:34:12 GMT"
},
{
"version": "v2",
"created": "Thu, 29 Jun 2023 09:43:20 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"State",
"Laura",
""
],
[
"Salat",
"Hadrien",
""
],
[
"Rubrichi",
"Stefania",
""
],
[
"Smoreda",
"Zbigniew",
""
]
] | TITLE: Explainability in Practice: Estimating Electrification Rates from Mobile
Phone Data in Senegal
ABSTRACT: Explainable artificial intelligence (XAI) provides explanations for not
interpretable machine learning (ML) models. While many technical approaches
exist, there is a lack of validation of these techniques on real-world
datasets. In this work, we present a use-case of XAI: an ML model which is
trained to estimate electrification rates based on mobile phone data in
Senegal. The data originate from the Data for Development challenge by Orange
in 2014/15. We apply two model-agnostic, local explanation techniques and find
that while the model can be verified, it is biased with respect to the
population density. We conclude our paper by pointing to the two main
challenges we encountered during our work: data processing and model design
that might be restricted by currently available XAI methods, and the importance
of domain knowledge to interpret explanations.
| no_new_dataset | 0.710069 |
2211.12751 | Qiang Huang | Qiang Huang, Yanhao Wang, Anthony K. H. Tung | SAH: Shifting-aware Asymmetric Hashing for Reverse $k$-Maximum Inner
Product Search | Accepted by AAAI 2023 | null | 10.1609/aaai.v37i4.25550 | null | cs.IR cs.DB cs.DS cs.LG | http://creativecommons.org/licenses/by-nc-sa/4.0/ | This paper investigates a new yet challenging problem called Reverse
$k$-Maximum Inner Product Search (R$k$MIPS). Given a query (item) vector, a set
of item vectors, and a set of user vectors, the problem of R$k$MIPS aims to
find a set of user vectors whose inner products with the query vector are one
of the $k$ largest among the query and item vectors. We propose the first
subquadratic-time algorithm, i.e., Shifting-aware Asymmetric Hashing (SAH), to
tackle the R$k$MIPS problem. To speed up the Maximum Inner Product Search
(MIPS) on item vectors, we design a shifting-invariant asymmetric
transformation and develop a novel sublinear-time Shifting-Aware Asymmetric
Locality Sensitive Hashing (SA-ALSH) scheme. Furthermore, we devise a new
blocking strategy based on the Cone-Tree to effectively prune user vectors (in
a batch). We prove that SAH achieves a theoretical guarantee for solving the
RMIPS problem. Experimental results on five real-world datasets show that SAH
runs 4$\sim$8$\times$ faster than the state-of-the-art methods for R$k$MIPS
while achieving F1-scores of over 90\%. The code is available at
\url{https://github.com/HuangQiang/SAH}.
| [
{
"version": "v1",
"created": "Wed, 23 Nov 2022 07:26:41 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Huang",
"Qiang",
""
],
[
"Wang",
"Yanhao",
""
],
[
"Tung",
"Anthony K. H.",
""
]
] | TITLE: SAH: Shifting-aware Asymmetric Hashing for Reverse $k$-Maximum Inner
Product Search
ABSTRACT: This paper investigates a new yet challenging problem called Reverse
$k$-Maximum Inner Product Search (R$k$MIPS). Given a query (item) vector, a set
of item vectors, and a set of user vectors, the problem of R$k$MIPS aims to
find a set of user vectors whose inner products with the query vector are one
of the $k$ largest among the query and item vectors. We propose the first
subquadratic-time algorithm, i.e., Shifting-aware Asymmetric Hashing (SAH), to
tackle the R$k$MIPS problem. To speed up the Maximum Inner Product Search
(MIPS) on item vectors, we design a shifting-invariant asymmetric
transformation and develop a novel sublinear-time Shifting-Aware Asymmetric
Locality Sensitive Hashing (SA-ALSH) scheme. Furthermore, we devise a new
blocking strategy based on the Cone-Tree to effectively prune user vectors (in
a batch). We prove that SAH achieves a theoretical guarantee for solving the
RMIPS problem. Experimental results on five real-world datasets show that SAH
runs 4$\sim$8$\times$ faster than the state-of-the-art methods for R$k$MIPS
while achieving F1-scores of over 90\%. The code is available at
\url{https://github.com/HuangQiang/SAH}.
| no_new_dataset | 0.708458 |
2212.00423 | Kim Bjerge | Kim Bjerge, Carsten Eie Frigaard and Henrik Karstoft | Motion Informed Object Detection of Small Insects in Time-lapse Camera
Recordings | 10 pages, 6 figures | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Insects as pollinators play a crucial role in ecosystem management and world
food production. However, insect populations are declining, calling for
efficient methods of insect monitoring. Existing methods analyze video or
time-lapse images of insects in nature, but the analysis is challenging since
insects are small objects in complex and dynamic scenes of natural vegetation.
In this work, we provide a dataset of primary honeybees visiting three
different plant species during two months of the summer period. The dataset
consists of 107,387 annotated time-lapse images from multiple cameras,
including 9,423 annotated insects. We present a method pipeline for detecting
insects in time-lapse RGB images. The pipeline consists of a two-step process.
Firstly, the time-lapse RGB images are preprocessed to enhance insects in the
images. This Motion-Informed-Enhancement technique uses motion and colors to
enhance insects in images. Secondly, the enhanced images are subsequently fed
into a Convolutional Neural network (CNN) object detector. The method improves
the deep learning object detectors You Only Look Once (YOLO) and Faster
Region-based CNN (Faster R-CNN). Using Motion-Informed-Enhancement, the
YOLO-detector improves the average micro F1-score from 0.49 to 0.71, and the
Faster R-CNN-detector improves the average micro F1-score from 0.32 to 0.56 on
the dataset. Our dataset and proposed method provide a step forward to automate
the time-lapse camera monitoring of flying insects. The dataset is published
on: https://vision.eng.au.dk/mie/
| [
{
"version": "v1",
"created": "Thu, 1 Dec 2022 10:54:06 GMT"
},
{
"version": "v2",
"created": "Thu, 29 Jun 2023 15:01:00 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Bjerge",
"Kim",
""
],
[
"Frigaard",
"Carsten Eie",
""
],
[
"Karstoft",
"Henrik",
""
]
] | TITLE: Motion Informed Object Detection of Small Insects in Time-lapse Camera
Recordings
ABSTRACT: Insects as pollinators play a crucial role in ecosystem management and world
food production. However, insect populations are declining, calling for
efficient methods of insect monitoring. Existing methods analyze video or
time-lapse images of insects in nature, but the analysis is challenging since
insects are small objects in complex and dynamic scenes of natural vegetation.
In this work, we provide a dataset of primary honeybees visiting three
different plant species during two months of the summer period. The dataset
consists of 107,387 annotated time-lapse images from multiple cameras,
including 9,423 annotated insects. We present a method pipeline for detecting
insects in time-lapse RGB images. The pipeline consists of a two-step process.
Firstly, the time-lapse RGB images are preprocessed to enhance insects in the
images. This Motion-Informed-Enhancement technique uses motion and colors to
enhance insects in images. Secondly, the enhanced images are subsequently fed
into a Convolutional Neural network (CNN) object detector. The method improves
the deep learning object detectors You Only Look Once (YOLO) and Faster
Region-based CNN (Faster R-CNN). Using Motion-Informed-Enhancement, the
YOLO-detector improves the average micro F1-score from 0.49 to 0.71, and the
Faster R-CNN-detector improves the average micro F1-score from 0.32 to 0.56 on
the dataset. Our dataset and proposed method provide a step forward to automate
the time-lapse camera monitoring of flying insects. The dataset is published
on: https://vision.eng.au.dk/mie/
| new_dataset | 0.71882 |
2212.08208 | Mohamad Hakam Shams Eddin | Mohamad Hakam Shams Eddin, Ribana Roscher, Juergen Gall | Location-aware Adaptive Normalization: A Deep Learning Approach For
Wildfire Danger Forecasting | null | in IEEE Transactions on Geoscience and Remote Sensing, vol. 61,
pp. 1-18, 2023, Art no. 4703018 | 10.1109/TGRS.2023.3285401 | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Climate change is expected to intensify and increase extreme events in the
weather cycle. Since this has a significant impact on various sectors of our
life, recent works are concerned with identifying and predicting such extreme
events from Earth observations. With respect to wildfire danger forecasting,
previous deep learning approaches duplicate static variables along the time
dimension and neglect the intrinsic differences between static and dynamic
variables. Furthermore, most existing multi-branch architectures lose the
interconnections between the branches during the feature learning stage. To
address these issues, this paper proposes a 2D/3D two-branch convolutional
neural network (CNN) with a Location-aware Adaptive Normalization layer (LOAN).
Using LOAN as a building block, we can modulate the dynamic features
conditional on their geographical locations. Thus, our approach considers
feature properties as a unified yet compound 2D/3D model. Besides, we propose
using the sinusoidal-based encoding of the day of the year to provide the model
with explicit temporal information about the target day within the year. Our
experimental results show a better performance of our approach than other
baselines on the challenging FireCube dataset. The results show that
location-aware adaptive feature normalization is a promising technique to learn
the relation between dynamic variables and their geographic locations, which is
highly relevant for areas where remote sensing data builds the basis for
analysis. The source code is available at https://github.com/HakamShams/LOAN.
| [
{
"version": "v1",
"created": "Fri, 16 Dec 2022 00:32:38 GMT"
},
{
"version": "v2",
"created": "Fri, 7 Apr 2023 23:35:18 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Eddin",
"Mohamad Hakam Shams",
""
],
[
"Roscher",
"Ribana",
""
],
[
"Gall",
"Juergen",
""
]
] | TITLE: Location-aware Adaptive Normalization: A Deep Learning Approach For
Wildfire Danger Forecasting
ABSTRACT: Climate change is expected to intensify and increase extreme events in the
weather cycle. Since this has a significant impact on various sectors of our
life, recent works are concerned with identifying and predicting such extreme
events from Earth observations. With respect to wildfire danger forecasting,
previous deep learning approaches duplicate static variables along the time
dimension and neglect the intrinsic differences between static and dynamic
variables. Furthermore, most existing multi-branch architectures lose the
interconnections between the branches during the feature learning stage. To
address these issues, this paper proposes a 2D/3D two-branch convolutional
neural network (CNN) with a Location-aware Adaptive Normalization layer (LOAN).
Using LOAN as a building block, we can modulate the dynamic features
conditional on their geographical locations. Thus, our approach considers
feature properties as a unified yet compound 2D/3D model. Besides, we propose
using the sinusoidal-based encoding of the day of the year to provide the model
with explicit temporal information about the target day within the year. Our
experimental results show a better performance of our approach than other
baselines on the challenging FireCube dataset. The results show that
location-aware adaptive feature normalization is a promising technique to learn
the relation between dynamic variables and their geographic locations, which is
highly relevant for areas where remote sensing data builds the basis for
analysis. The source code is available at https://github.com/HakamShams/LOAN.
| no_new_dataset | 0.710653 |
2301.01404 | Xiao Shen Dr. | Xiao Shen, Dewang Sun, Shirui Pan, Xi Zhou, Laurence T. Yang | Neighbor Contrastive Learning on Learnable Graph Augmentation | null | Proceedings of AAAI Conference on Artificial Intelligence (AAAI),
pp. 9782-9791, 2023 | null | null | cs.SI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recent years, graph contrastive learning (GCL), which aims to learn
representations from unlabeled graphs, has made great progress. However, the
existing GCL methods mostly adopt human-designed graph augmentations, which are
sensitive to various graph datasets. In addition, the contrastive losses
originally developed in computer vision have been directly applied to graph
data, where the neighboring nodes are regarded as negatives and consequently
pushed far apart from the anchor. However, this is contradictory with the
homophily assumption of networks that connected nodes often belong to the same
class and should be close to each other. In this work, we propose an end-to-end
automatic GCL method, named NCLA to apply neighbor contrastive learning on
learnable graph augmentation. Several graph augmented views with adaptive
topology are automatically learned by the multi-head graph attention mechanism,
which can be compatible with various graph datasets without prior domain
knowledge. In addition, a neighbor contrastive loss is devised to allow
multiple positives per anchor by taking network topology as the supervised
signals. Both augmentations and embeddings are learned end-to-end in the
proposed NCLA. Extensive experiments on the benchmark datasets demonstrate that
NCLA yields the state-of-the-art node classification performance on
self-supervised GCL and even exceeds the supervised ones, when the labels are
extremely limited. Our code is released at https://github.com/shenxiaocam/NCLA.
| [
{
"version": "v1",
"created": "Wed, 4 Jan 2023 01:54:18 GMT"
},
{
"version": "v2",
"created": "Fri, 2 Jun 2023 07:22:36 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Shen",
"Xiao",
""
],
[
"Sun",
"Dewang",
""
],
[
"Pan",
"Shirui",
""
],
[
"Zhou",
"Xi",
""
],
[
"Yang",
"Laurence T.",
""
]
] | TITLE: Neighbor Contrastive Learning on Learnable Graph Augmentation
ABSTRACT: Recent years, graph contrastive learning (GCL), which aims to learn
representations from unlabeled graphs, has made great progress. However, the
existing GCL methods mostly adopt human-designed graph augmentations, which are
sensitive to various graph datasets. In addition, the contrastive losses
originally developed in computer vision have been directly applied to graph
data, where the neighboring nodes are regarded as negatives and consequently
pushed far apart from the anchor. However, this is contradictory with the
homophily assumption of networks that connected nodes often belong to the same
class and should be close to each other. In this work, we propose an end-to-end
automatic GCL method, named NCLA to apply neighbor contrastive learning on
learnable graph augmentation. Several graph augmented views with adaptive
topology are automatically learned by the multi-head graph attention mechanism,
which can be compatible with various graph datasets without prior domain
knowledge. In addition, a neighbor contrastive loss is devised to allow
multiple positives per anchor by taking network topology as the supervised
signals. Both augmentations and embeddings are learned end-to-end in the
proposed NCLA. Extensive experiments on the benchmark datasets demonstrate that
NCLA yields the state-of-the-art node classification performance on
self-supervised GCL and even exceeds the supervised ones, when the labels are
extremely limited. Our code is released at https://github.com/shenxiaocam/NCLA.
| no_new_dataset | 0.711268 |
2301.12251 | Luyu Jiang | Luyu Jiang, Dantong Ouyang, Qi Zhang, and Liming Zhang | DeciLS-PBO: an Effective Local Search Method for Pseudo-Boolean
Optimization | null | null | null | null | cs.AI cs.LO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Local search is an effective method for solving large-scale combinatorial
optimization problems, and it has made remarkable progress in recent years
through several subtle mechanisms. In this paper, we found two ways to improve
the local search algorithms in solving Pseudo-Boolean Optimization (PBO):
Firstly, some of those mechanisms such as unit propagation are merely used in
solving MaxSAT before, which can be generalized to solve PBO as well; Secondly,
the existing local search algorithms utilize the heuristic on variables,
so-called score, to mainly guide the search. We attempt to gain more insights
into the clause, as it plays the role of a middleman who builds a bridge
between variables and the given formula. Hence, we first extended the
combination of unit propagation-based decimation algorithm to PBO problem,
giving a further generalized definition of unit clause for PBO problem, and
apply it to the existing solver LS-PBO for constructing an initial assignment;
then, we introduced a new heuristic on clauses, dubbed care, to set a higher
priority for the clauses that are less satisfied in current iterations.
Experiments on benchmarks from the most recent PB Competition, as well as three
real-world application benchmarks including minimum-width confidence band,
wireless sensor network optimization, and seating arrangement problems show
that our algorithm DeciLS-PBO has a promising performance compared to the
state-of-the-art algorithms.
| [
{
"version": "v1",
"created": "Sat, 28 Jan 2023 17:03:56 GMT"
},
{
"version": "v2",
"created": "Thu, 29 Jun 2023 09:03:32 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Jiang",
"Luyu",
""
],
[
"Ouyang",
"Dantong",
""
],
[
"Zhang",
"Qi",
""
],
[
"Zhang",
"Liming",
""
]
] | TITLE: DeciLS-PBO: an Effective Local Search Method for Pseudo-Boolean
Optimization
ABSTRACT: Local search is an effective method for solving large-scale combinatorial
optimization problems, and it has made remarkable progress in recent years
through several subtle mechanisms. In this paper, we found two ways to improve
the local search algorithms in solving Pseudo-Boolean Optimization (PBO):
Firstly, some of those mechanisms such as unit propagation are merely used in
solving MaxSAT before, which can be generalized to solve PBO as well; Secondly,
the existing local search algorithms utilize the heuristic on variables,
so-called score, to mainly guide the search. We attempt to gain more insights
into the clause, as it plays the role of a middleman who builds a bridge
between variables and the given formula. Hence, we first extended the
combination of unit propagation-based decimation algorithm to PBO problem,
giving a further generalized definition of unit clause for PBO problem, and
apply it to the existing solver LS-PBO for constructing an initial assignment;
then, we introduced a new heuristic on clauses, dubbed care, to set a higher
priority for the clauses that are less satisfied in current iterations.
Experiments on benchmarks from the most recent PB Competition, as well as three
real-world application benchmarks including minimum-width confidence band,
wireless sensor network optimization, and seating arrangement problems show
that our algorithm DeciLS-PBO has a promising performance compared to the
state-of-the-art algorithms.
| no_new_dataset | 0.709202 |
2301.12467 | Julio Hurtado | Julio Hurtado and Dario Salvati and Rudy Semola and Mattia Bosio and
Vincenzo Lomonaco | Continual Learning for Predictive Maintenance: Overview and Challenges | null | null | 10.1016/j.iswa.2023.200251 | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | Deep learning techniques have become one of the main propellers for solving
engineering problems effectively and efficiently. For instance, Predictive
Maintenance methods have been used to improve predictions of when maintenance
is needed on different machines and operative contexts. However, deep learning
methods are not without limitations, as these models are normally trained on a
fixed distribution that only reflects the current state of the problem. Due to
internal or external factors, the state of the problem can change, and the
performance decreases due to the lack of generalization and adaptation.
Contrary to this stationary training set, real-world applications change their
environments constantly, creating the need to constantly adapt the model to
evolving scenarios. To aid in this endeavor, Continual Learning methods propose
ways to constantly adapt prediction models and incorporate new knowledge after
deployment. Despite the advantages of these techniques, there are still
challenges to applying them to real-world problems. In this work, we present a
brief introduction to predictive maintenance, non-stationary environments, and
continual learning, together with an extensive review of the current state of
applying continual learning in real-world applications and specifically in
predictive maintenance. We then discuss the current challenges of both
predictive maintenance and continual learning, proposing future directions at
the intersection of both areas. Finally, we propose a novel way to create
benchmarks that favor the application of continuous learning methods in more
realistic environments, giving specific examples of predictive maintenance.
| [
{
"version": "v1",
"created": "Sun, 29 Jan 2023 15:32:53 GMT"
},
{
"version": "v2",
"created": "Thu, 29 Jun 2023 08:55:12 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Hurtado",
"Julio",
""
],
[
"Salvati",
"Dario",
""
],
[
"Semola",
"Rudy",
""
],
[
"Bosio",
"Mattia",
""
],
[
"Lomonaco",
"Vincenzo",
""
]
] | TITLE: Continual Learning for Predictive Maintenance: Overview and Challenges
ABSTRACT: Deep learning techniques have become one of the main propellers for solving
engineering problems effectively and efficiently. For instance, Predictive
Maintenance methods have been used to improve predictions of when maintenance
is needed on different machines and operative contexts. However, deep learning
methods are not without limitations, as these models are normally trained on a
fixed distribution that only reflects the current state of the problem. Due to
internal or external factors, the state of the problem can change, and the
performance decreases due to the lack of generalization and adaptation.
Contrary to this stationary training set, real-world applications change their
environments constantly, creating the need to constantly adapt the model to
evolving scenarios. To aid in this endeavor, Continual Learning methods propose
ways to constantly adapt prediction models and incorporate new knowledge after
deployment. Despite the advantages of these techniques, there are still
challenges to applying them to real-world problems. In this work, we present a
brief introduction to predictive maintenance, non-stationary environments, and
continual learning, together with an extensive review of the current state of
applying continual learning in real-world applications and specifically in
predictive maintenance. We then discuss the current challenges of both
predictive maintenance and continual learning, proposing future directions at
the intersection of both areas. Finally, we propose a novel way to create
benchmarks that favor the application of continuous learning methods in more
realistic environments, giving specific examples of predictive maintenance.
| no_new_dataset | 0.708843 |
2302.10017 | Rodrigo P\'erez Dattari | Rodrigo P\'erez-Dattari and Jens Kober | Stable Motion Primitives via Imitation and Contrastive Learning | null | null | null | null | cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Learning from humans allows non-experts to program robots with ease, lowering
the resources required to build complex robotic solutions. Nevertheless, such
data-driven approaches often lack the ability to provide guarantees regarding
their learned behaviors, which is critical for avoiding failures and/or
accidents. In this work, we focus on reaching/point-to-point motions, where
robots must always reach their goal, independently of their initial state. This
can be achieved by modeling motions as dynamical systems and ensuring that they
are globally asymptotically stable. Hence, we introduce a novel Contrastive
Learning loss for training Deep Neural Networks (DNN) that, when used together
with an Imitation Learning loss, enforces the aforementioned stability in the
learned motions. Differently from previous work, our method does not restrict
the structure of its function approximator, enabling its use with arbitrary
DNNs and allowing it to learn complex motions with high accuracy. We validate
it using datasets and a real robot. In the former case, motions are 2 and 4
dimensional, modeled as first- and second-order dynamical systems. In the
latter, motions are 3, 4, and 6 dimensional, of first and second order, and are
used to control a 7DoF robot manipulator in its end effector space and joint
space. More details regarding the real-world experiments are presented in:
\url{https://youtu.be/OM-2edHBRfc}.
| [
{
"version": "v1",
"created": "Mon, 20 Feb 2023 14:49:07 GMT"
},
{
"version": "v2",
"created": "Mon, 5 Jun 2023 10:18:57 GMT"
},
{
"version": "v3",
"created": "Thu, 29 Jun 2023 08:01:15 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Pérez-Dattari",
"Rodrigo",
""
],
[
"Kober",
"Jens",
""
]
] | TITLE: Stable Motion Primitives via Imitation and Contrastive Learning
ABSTRACT: Learning from humans allows non-experts to program robots with ease, lowering
the resources required to build complex robotic solutions. Nevertheless, such
data-driven approaches often lack the ability to provide guarantees regarding
their learned behaviors, which is critical for avoiding failures and/or
accidents. In this work, we focus on reaching/point-to-point motions, where
robots must always reach their goal, independently of their initial state. This
can be achieved by modeling motions as dynamical systems and ensuring that they
are globally asymptotically stable. Hence, we introduce a novel Contrastive
Learning loss for training Deep Neural Networks (DNN) that, when used together
with an Imitation Learning loss, enforces the aforementioned stability in the
learned motions. Differently from previous work, our method does not restrict
the structure of its function approximator, enabling its use with arbitrary
DNNs and allowing it to learn complex motions with high accuracy. We validate
it using datasets and a real robot. In the former case, motions are 2 and 4
dimensional, modeled as first- and second-order dynamical systems. In the
latter, motions are 3, 4, and 6 dimensional, of first and second order, and are
used to control a 7DoF robot manipulator in its end effector space and joint
space. More details regarding the real-world experiments are presented in:
\url{https://youtu.be/OM-2edHBRfc}.
| no_new_dataset | 0.710189 |
2302.12909 | Michael Menart | Raef Bassily and Crist\'obal Guzm\'an and Michael Menart | Differentially Private Algorithms for the Stochastic Saddle Point
Problem with Optimal Rates for the Strong Gap | null | null | null | null | cs.LG cs.CR math.OC stat.ML | http://creativecommons.org/licenses/by/4.0/ | We show that convex-concave Lipschitz stochastic saddle point problems (also
known as stochastic minimax optimization) can be solved under the constraint of
$(\epsilon,\delta)$-differential privacy with \emph{strong (primal-dual) gap}
rate of $\tilde O\big(\frac{1}{\sqrt{n}} + \frac{\sqrt{d}}{n\epsilon}\big)$,
where $n$ is the dataset size and $d$ is the dimension of the problem. This
rate is nearly optimal, based on existing lower bounds in differentially
private stochastic optimization. Specifically, we prove a tight upper bound on
the strong gap via novel implementation and analysis of the recursive
regularization technique repurposed for saddle point problems. We show that
this rate can be attained with
$O\big(\min\big\{\frac{n^2\epsilon^{1.5}}{\sqrt{d}}, n^{3/2}\big\}\big)$
gradient complexity, and $\tilde{O}(n)$ gradient complexity if the loss
function is smooth. As a byproduct of our method, we develop a general
algorithm that, given a black-box access to a subroutine satisfying a certain
$\alpha$ primal-dual accuracy guarantee with respect to the empirical
objective, gives a solution to the stochastic saddle point problem with a
strong gap of $\tilde{O}(\alpha+\frac{1}{\sqrt{n}})$. We show that this
$\alpha$-accuracy condition is satisfied by standard algorithms for the
empirical saddle point problem such as the proximal point method and the
stochastic gradient descent ascent algorithm. Further, we show that even for
simple problems it is possible for an algorithm to have zero weak gap and
suffer from $\Omega(1)$ strong gap. We also show that there exists a
fundamental tradeoff between stability and accuracy. Specifically, we show that
any $\Delta$-stable algorithm has empirical gap $\Omega\big(\frac{1}{\Delta
n}\big)$, and that this bound is tight. This result also holds also more
specifically for empirical risk minimization problems and may be of independent
interest.
| [
{
"version": "v1",
"created": "Fri, 24 Feb 2023 21:50:02 GMT"
},
{
"version": "v2",
"created": "Thu, 29 Jun 2023 16:22:34 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Bassily",
"Raef",
""
],
[
"Guzmán",
"Cristóbal",
""
],
[
"Menart",
"Michael",
""
]
] | TITLE: Differentially Private Algorithms for the Stochastic Saddle Point
Problem with Optimal Rates for the Strong Gap
ABSTRACT: We show that convex-concave Lipschitz stochastic saddle point problems (also
known as stochastic minimax optimization) can be solved under the constraint of
$(\epsilon,\delta)$-differential privacy with \emph{strong (primal-dual) gap}
rate of $\tilde O\big(\frac{1}{\sqrt{n}} + \frac{\sqrt{d}}{n\epsilon}\big)$,
where $n$ is the dataset size and $d$ is the dimension of the problem. This
rate is nearly optimal, based on existing lower bounds in differentially
private stochastic optimization. Specifically, we prove a tight upper bound on
the strong gap via novel implementation and analysis of the recursive
regularization technique repurposed for saddle point problems. We show that
this rate can be attained with
$O\big(\min\big\{\frac{n^2\epsilon^{1.5}}{\sqrt{d}}, n^{3/2}\big\}\big)$
gradient complexity, and $\tilde{O}(n)$ gradient complexity if the loss
function is smooth. As a byproduct of our method, we develop a general
algorithm that, given a black-box access to a subroutine satisfying a certain
$\alpha$ primal-dual accuracy guarantee with respect to the empirical
objective, gives a solution to the stochastic saddle point problem with a
strong gap of $\tilde{O}(\alpha+\frac{1}{\sqrt{n}})$. We show that this
$\alpha$-accuracy condition is satisfied by standard algorithms for the
empirical saddle point problem such as the proximal point method and the
stochastic gradient descent ascent algorithm. Further, we show that even for
simple problems it is possible for an algorithm to have zero weak gap and
suffer from $\Omega(1)$ strong gap. We also show that there exists a
fundamental tradeoff between stability and accuracy. Specifically, we show that
any $\Delta$-stable algorithm has empirical gap $\Omega\big(\frac{1}{\Delta
n}\big)$, and that this bound is tight. This result also holds also more
specifically for empirical risk minimization problems and may be of independent
interest.
| no_new_dataset | 0.710038 |
2303.02245 | Soumitri Chattopadhyay | Soumitri Chattopadhyay, Soham Ganguly, Sreejit Chaudhury, Sayan Nag,
Samiran Chattopadhyay | Exploring Self-Supervised Representation Learning For Low-Resource
Medical Image Analysis | Accepted at IEEE ICIP 2023 | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | The success of self-supervised learning (SSL) has mostly been attributed to
the availability of unlabeled yet large-scale datasets. However, in a
specialized domain such as medical imaging which is a lot different from
natural images, the assumption of data availability is unrealistic and
impractical, as the data itself is scanty and found in small databases,
collected for specific prognosis tasks. To this end, we seek to investigate the
applicability of self-supervised learning algorithms on small-scale medical
imaging datasets. In particular, we evaluate $4$ state-of-the-art SSL methods
on three publicly accessible \emph{small} medical imaging datasets. Our
investigation reveals that in-domain low-resource SSL pre-training can yield
competitive performance to transfer learning from large-scale datasets (such as
ImageNet). Furthermore, we extensively analyse our empirical findings to
provide valuable insights that can motivate for further research towards
circumventing the need for pre-training on a large image corpus. To the best of
our knowledge, this is the first attempt to holistically explore
self-supervision on low-resource medical datasets.
| [
{
"version": "v1",
"created": "Fri, 3 Mar 2023 22:26:17 GMT"
},
{
"version": "v2",
"created": "Thu, 29 Jun 2023 03:22:52 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Chattopadhyay",
"Soumitri",
""
],
[
"Ganguly",
"Soham",
""
],
[
"Chaudhury",
"Sreejit",
""
],
[
"Nag",
"Sayan",
""
],
[
"Chattopadhyay",
"Samiran",
""
]
] | TITLE: Exploring Self-Supervised Representation Learning For Low-Resource
Medical Image Analysis
ABSTRACT: The success of self-supervised learning (SSL) has mostly been attributed to
the availability of unlabeled yet large-scale datasets. However, in a
specialized domain such as medical imaging which is a lot different from
natural images, the assumption of data availability is unrealistic and
impractical, as the data itself is scanty and found in small databases,
collected for specific prognosis tasks. To this end, we seek to investigate the
applicability of self-supervised learning algorithms on small-scale medical
imaging datasets. In particular, we evaluate $4$ state-of-the-art SSL methods
on three publicly accessible \emph{small} medical imaging datasets. Our
investigation reveals that in-domain low-resource SSL pre-training can yield
competitive performance to transfer learning from large-scale datasets (such as
ImageNet). Furthermore, we extensively analyse our empirical findings to
provide valuable insights that can motivate for further research towards
circumventing the need for pre-training on a large image corpus. To the best of
our knowledge, this is the first attempt to holistically explore
self-supervision on low-resource medical datasets.
| no_new_dataset | 0.709856 |
2304.03838 | Qi Qi | Qi Qi, Shervin Ardeshir | Improving Identity-Robustness for Face Models | null | null | null | null | cs.CV cs.AI cs.DS cs.IR cs.LG | http://creativecommons.org/licenses/by/4.0/ | Despite the success of deep-learning models in many tasks, there have been
concerns about such models learning shortcuts, and their lack of robustness to
irrelevant confounders. When it comes to models directly trained on human
faces, a sensitive confounder is that of human identities. Many face-related
tasks should ideally be identity-independent, and perform uniformly across
different individuals (i.e. be fair). One way to measure and enforce such
robustness and performance uniformity is through enforcing it during training,
assuming identity-related information is available at scale. However, due to
privacy concerns and also the cost of collecting such information, this is
often not the case, and most face datasets simply contain input images and
their corresponding task-related labels. Thus, improving identity-related
robustness without the need for such annotations is of great importance. Here,
we explore using face-recognition embedding vectors, as proxies for identities,
to enforce such robustness. We propose to use the structure in the
face-recognition embedding space, to implicitly emphasize rare samples within
each class. We do so by weighting samples according to their conditional
inverse density (CID) in the proxy embedding space. Our experiments suggest
that such a simple sample weighting scheme, not only improves the training
robustness, it often improves the overall performance as a result of such
robustness. We also show that employing such constraints during training
results in models that are significantly less sensitive to different levels of
bias in the dataset.
| [
{
"version": "v1",
"created": "Fri, 7 Apr 2023 20:41:10 GMT"
},
{
"version": "v2",
"created": "Thu, 29 Jun 2023 07:05:37 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Qi",
"Qi",
""
],
[
"Ardeshir",
"Shervin",
""
]
] | TITLE: Improving Identity-Robustness for Face Models
ABSTRACT: Despite the success of deep-learning models in many tasks, there have been
concerns about such models learning shortcuts, and their lack of robustness to
irrelevant confounders. When it comes to models directly trained on human
faces, a sensitive confounder is that of human identities. Many face-related
tasks should ideally be identity-independent, and perform uniformly across
different individuals (i.e. be fair). One way to measure and enforce such
robustness and performance uniformity is through enforcing it during training,
assuming identity-related information is available at scale. However, due to
privacy concerns and also the cost of collecting such information, this is
often not the case, and most face datasets simply contain input images and
their corresponding task-related labels. Thus, improving identity-related
robustness without the need for such annotations is of great importance. Here,
we explore using face-recognition embedding vectors, as proxies for identities,
to enforce such robustness. We propose to use the structure in the
face-recognition embedding space, to implicitly emphasize rare samples within
each class. We do so by weighting samples according to their conditional
inverse density (CID) in the proxy embedding space. Our experiments suggest
that such a simple sample weighting scheme, not only improves the training
robustness, it often improves the overall performance as a result of such
robustness. We also show that employing such constraints during training
results in models that are significantly less sensitive to different levels of
bias in the dataset.
| no_new_dataset | 0.709799 |
2304.06408 | Davide Cozzolino | Riccardo Corvi, Davide Cozzolino, Giovanni Poggi, Koki Nagano, Luisa
Verdoliva | Intriguing properties of synthetic images: from generative adversarial
networks to diffusion models | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Detecting fake images is becoming a major goal of computer vision. This need
is becoming more and more pressing with the continuous improvement of synthesis
methods based on Generative Adversarial Networks (GAN), and even more with the
appearance of powerful methods based on Diffusion Models (DM). Towards this
end, it is important to gain insight into which image features better
discriminate fake images from real ones. In this paper we report on our
systematic study of a large number of image generators of different families,
aimed at discovering the most forensically relevant characteristics of real and
generated images. Our experiments provide a number of interesting observations
and shed light on some intriguing properties of synthetic images: (1) not only
the GAN models but also the DM and VQ-GAN (Vector Quantized Generative
Adversarial Networks) models give rise to visible artifacts in the Fourier
domain and exhibit anomalous regular patterns in the autocorrelation; (2) when
the dataset used to train the model lacks sufficient variety, its biases can be
transferred to the generated images; (3) synthetic and real images exhibit
significant differences in the mid-high frequency signal content, observable in
their radial and angular spectral power distributions.
| [
{
"version": "v1",
"created": "Thu, 13 Apr 2023 11:13:19 GMT"
},
{
"version": "v2",
"created": "Thu, 29 Jun 2023 15:33:42 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Corvi",
"Riccardo",
""
],
[
"Cozzolino",
"Davide",
""
],
[
"Poggi",
"Giovanni",
""
],
[
"Nagano",
"Koki",
""
],
[
"Verdoliva",
"Luisa",
""
]
] | TITLE: Intriguing properties of synthetic images: from generative adversarial
networks to diffusion models
ABSTRACT: Detecting fake images is becoming a major goal of computer vision. This need
is becoming more and more pressing with the continuous improvement of synthesis
methods based on Generative Adversarial Networks (GAN), and even more with the
appearance of powerful methods based on Diffusion Models (DM). Towards this
end, it is important to gain insight into which image features better
discriminate fake images from real ones. In this paper we report on our
systematic study of a large number of image generators of different families,
aimed at discovering the most forensically relevant characteristics of real and
generated images. Our experiments provide a number of interesting observations
and shed light on some intriguing properties of synthetic images: (1) not only
the GAN models but also the DM and VQ-GAN (Vector Quantized Generative
Adversarial Networks) models give rise to visible artifacts in the Fourier
domain and exhibit anomalous regular patterns in the autocorrelation; (2) when
the dataset used to train the model lacks sufficient variety, its biases can be
transferred to the generated images; (3) synthetic and real images exhibit
significant differences in the mid-high frequency signal content, observable in
their radial and angular spectral power distributions.
| no_new_dataset | 0.709189 |
2304.06912 | Ha Thanh Nguyen | Ha-Thanh Nguyen, Randy Goebel, Francesca Toni, Kostas Stathis, Ken
Satoh | How well do SOTA legal reasoning models support abductive reasoning? | Workshop on Logic Programming and Legal Reasoning, @ICLP 2023 | null | null | null | cs.LO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We examine how well the state-of-the-art (SOTA) models used in legal
reasoning support abductive reasoning tasks. Abductive reasoning is a form of
logical inference in which a hypothesis is formulated from a set of
observations, and that hypothesis is used to explain the observations. The
ability to formulate such hypotheses is important for lawyers and legal
scholars as it helps them articulate logical arguments, interpret laws, and
develop legal theories. Our motivation is to consider the belief that deep
learning models, especially large language models (LLMs), will soon replace
lawyers because they perform well on tasks related to legal text processing.
But to do so, we believe, requires some form of abductive hypothesis formation.
In other words, while LLMs become more popular and powerful, we want to
investigate their capacity for abductive reasoning. To pursue this goal, we
start by building a logic-augmented dataset for abductive reasoning with
498,697 samples and then use it to evaluate the performance of a SOTA model in
the legal field. Our experimental results show that although these models can
perform well on tasks related to some aspects of legal text processing, they
still fall short in supporting abductive reasoning tasks.
| [
{
"version": "v1",
"created": "Fri, 14 Apr 2023 03:36:24 GMT"
},
{
"version": "v2",
"created": "Thu, 29 Jun 2023 01:27:02 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Nguyen",
"Ha-Thanh",
""
],
[
"Goebel",
"Randy",
""
],
[
"Toni",
"Francesca",
""
],
[
"Stathis",
"Kostas",
""
],
[
"Satoh",
"Ken",
""
]
] | TITLE: How well do SOTA legal reasoning models support abductive reasoning?
ABSTRACT: We examine how well the state-of-the-art (SOTA) models used in legal
reasoning support abductive reasoning tasks. Abductive reasoning is a form of
logical inference in which a hypothesis is formulated from a set of
observations, and that hypothesis is used to explain the observations. The
ability to formulate such hypotheses is important for lawyers and legal
scholars as it helps them articulate logical arguments, interpret laws, and
develop legal theories. Our motivation is to consider the belief that deep
learning models, especially large language models (LLMs), will soon replace
lawyers because they perform well on tasks related to legal text processing.
But to do so, we believe, requires some form of abductive hypothesis formation.
In other words, while LLMs become more popular and powerful, we want to
investigate their capacity for abductive reasoning. To pursue this goal, we
start by building a logic-augmented dataset for abductive reasoning with
498,697 samples and then use it to evaluate the performance of a SOTA model in
the legal field. Our experimental results show that although these models can
perform well on tasks related to some aspects of legal text processing, they
still fall short in supporting abductive reasoning tasks.
| new_dataset | 0.713063 |
2304.08743 | Kazumi Kasaura | Kazumi Kasaura, Shuwa Miura, Tadashi Kozuno, Ryo Yonetani, Kenta
Hoshino, Yohei Hosoe | Benchmarking Actor-Critic Deep Reinforcement Learning Algorithms for
Robotics Control with Action Constraints | 8 pages, 7 figures, accepted to Robotics and Automation Letters | IEEE Robotics and Automation Letters 8(8) (2023) 4449-4456 | 10.1109/LRA.2023.3284378 | null | cs.LG cs.RO | http://creativecommons.org/licenses/by/4.0/ | This study presents a benchmark for evaluating action-constrained
reinforcement learning (RL) algorithms. In action-constrained RL, each action
taken by the learning system must comply with certain constraints. These
constraints are crucial for ensuring the feasibility and safety of actions in
real-world systems. We evaluate existing algorithms and their novel variants
across multiple robotics control environments, encompassing multiple action
constraint types. Our evaluation provides the first in-depth perspective of the
field, revealing surprising insights, including the effectiveness of a
straightforward baseline approach. The benchmark problems and associated code
utilized in our experiments are made available online at
github.com/omron-sinicx/action-constrained-RL-benchmark for further research
and development.
| [
{
"version": "v1",
"created": "Tue, 18 Apr 2023 05:45:09 GMT"
},
{
"version": "v2",
"created": "Mon, 29 May 2023 07:39:12 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Kasaura",
"Kazumi",
""
],
[
"Miura",
"Shuwa",
""
],
[
"Kozuno",
"Tadashi",
""
],
[
"Yonetani",
"Ryo",
""
],
[
"Hoshino",
"Kenta",
""
],
[
"Hosoe",
"Yohei",
""
]
] | TITLE: Benchmarking Actor-Critic Deep Reinforcement Learning Algorithms for
Robotics Control with Action Constraints
ABSTRACT: This study presents a benchmark for evaluating action-constrained
reinforcement learning (RL) algorithms. In action-constrained RL, each action
taken by the learning system must comply with certain constraints. These
constraints are crucial for ensuring the feasibility and safety of actions in
real-world systems. We evaluate existing algorithms and their novel variants
across multiple robotics control environments, encompassing multiple action
constraint types. Our evaluation provides the first in-depth perspective of the
field, revealing surprising insights, including the effectiveness of a
straightforward baseline approach. The benchmark problems and associated code
utilized in our experiments are made available online at
github.com/omron-sinicx/action-constrained-RL-benchmark for further research
and development.
| no_new_dataset | 0.710258 |
2304.13390 | Liu Hongwei | Hongwei Liu, Jian Yang, Jianfeng Zhang, Dongheng Shao, Jielong Guo,
Shaobo Li, Xuan Tang, Xian Wei | Group Equivariant BEV for 3D Object Detection | 8 pages,3 figures | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recently, 3D object detection has attracted significant attention and
achieved continuous improvement in real road scenarios. The environmental
information is collected from a single sensor or multi-sensor fusion to detect
interested objects. However, most of the current 3D object detection approaches
focus on developing advanced network architectures to improve the detection
precision of the object rather than considering the dynamic driving scenes,
where data collected from sensors equipped in the vehicle contain various
perturbation features. As a result, existing work cannot still tackle the
perturbation issue. In order to solve this problem, we propose a group
equivariant bird's eye view network (GeqBevNet) based on the group equivariant
theory, which introduces the concept of group equivariant into the BEV fusion
object detection network. The group equivariant network is embedded into the
fused BEV feature map to facilitate the BEV-level rotational equivariant
feature extraction, thus leading to lower average orientation error. In order
to demonstrate the effectiveness of the GeqBevNet, the network is verified on
the nuScenes validation dataset in which mAOE can be decreased to 0.325.
Experimental results demonstrate that GeqBevNet can extract more rotational
equivariant features in the 3D object detection of the actual road scene and
improve the performance of object orientation prediction.
| [
{
"version": "v1",
"created": "Wed, 26 Apr 2023 09:00:31 GMT"
},
{
"version": "v2",
"created": "Thu, 29 Jun 2023 03:22:08 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Liu",
"Hongwei",
""
],
[
"Yang",
"Jian",
""
],
[
"Zhang",
"Jianfeng",
""
],
[
"Shao",
"Dongheng",
""
],
[
"Guo",
"Jielong",
""
],
[
"Li",
"Shaobo",
""
],
[
"Tang",
"Xuan",
""
],
[
"Wei",
"Xian",
""
]
] | TITLE: Group Equivariant BEV for 3D Object Detection
ABSTRACT: Recently, 3D object detection has attracted significant attention and
achieved continuous improvement in real road scenarios. The environmental
information is collected from a single sensor or multi-sensor fusion to detect
interested objects. However, most of the current 3D object detection approaches
focus on developing advanced network architectures to improve the detection
precision of the object rather than considering the dynamic driving scenes,
where data collected from sensors equipped in the vehicle contain various
perturbation features. As a result, existing work cannot still tackle the
perturbation issue. In order to solve this problem, we propose a group
equivariant bird's eye view network (GeqBevNet) based on the group equivariant
theory, which introduces the concept of group equivariant into the BEV fusion
object detection network. The group equivariant network is embedded into the
fused BEV feature map to facilitate the BEV-level rotational equivariant
feature extraction, thus leading to lower average orientation error. In order
to demonstrate the effectiveness of the GeqBevNet, the network is verified on
the nuScenes validation dataset in which mAOE can be decreased to 0.325.
Experimental results demonstrate that GeqBevNet can extract more rotational
equivariant features in the 3D object detection of the actual road scene and
improve the performance of object orientation prediction.
| no_new_dataset | 0.708616 |
2304.14177 | Roberto Martinez Cruz | Roberto Mart\'inez-Cruz, Alvaro J. L\'opez-L\'opez, Jos\'e Portela | ChatGPT vs State-of-the-Art Models: A Benchmarking Study in Keyphrase
Generation Task | null | null | null | null | cs.CL cs.AI | http://creativecommons.org/licenses/by/4.0/ | Transformer-based language models, including ChatGPT, have demonstrated
exceptional performance in various natural language generation tasks. However,
there has been limited research evaluating ChatGPT's keyphrase generation
ability, which involves identifying informative phrases that accurately reflect
a document's content. This study seeks to address this gap by comparing
ChatGPT's keyphrase generation performance with state-of-the-art models, while
also testing its potential as a solution for two significant challenges in the
field: domain adaptation and keyphrase generation from long documents. We
conducted experiments on six publicly available datasets from scientific
articles and news domains, analyzing performance on both short and long
documents. Our results show that ChatGPT outperforms current state-of-the-art
models in all tested datasets and environments, generating high-quality
keyphrases that adapt well to diverse domains and document lengths.
| [
{
"version": "v1",
"created": "Thu, 27 Apr 2023 13:25:43 GMT"
},
{
"version": "v2",
"created": "Thu, 29 Jun 2023 13:40:42 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Martínez-Cruz",
"Roberto",
""
],
[
"López-López",
"Alvaro J.",
""
],
[
"Portela",
"José",
""
]
] | TITLE: ChatGPT vs State-of-the-Art Models: A Benchmarking Study in Keyphrase
Generation Task
ABSTRACT: Transformer-based language models, including ChatGPT, have demonstrated
exceptional performance in various natural language generation tasks. However,
there has been limited research evaluating ChatGPT's keyphrase generation
ability, which involves identifying informative phrases that accurately reflect
a document's content. This study seeks to address this gap by comparing
ChatGPT's keyphrase generation performance with state-of-the-art models, while
also testing its potential as a solution for two significant challenges in the
field: domain adaptation and keyphrase generation from long documents. We
conducted experiments on six publicly available datasets from scientific
articles and news domains, analyzing performance on both short and long
documents. Our results show that ChatGPT outperforms current state-of-the-art
models in all tested datasets and environments, generating high-quality
keyphrases that adapt well to diverse domains and document lengths.
| no_new_dataset | 0.709466 |
2304.14712 | Eneko Osaba | Eneko Osaba, Esther Villar-Rodriguez and Sebasti\'an V. Romero | Benchmark dataset and instance generator for Real-World
Three-Dimensional Bin Packing Problems | 11 pages, 4 figures | Data in Brief, 109309 (2023) | 10.1016/j.dib.2023.109309 | null | cs.AI | http://creativecommons.org/licenses/by/4.0/ | In this article, a benchmark for real-world bin packing problems is proposed.
This dataset consists of 12 instances of varying levels of complexity regarding
size (with the number of packages ranging from 38 to 53) and user-defined
requirements. In fact, several real-world-oriented restrictions were taken into
account to build these instances: i) item and bin dimensions, ii) weight
restrictions, iii) affinities among package categories iv) preferences for
package ordering and v) load balancing. Besides the data, we also offer an own
developed Python script for the dataset generation, coined Q4RealBPP-DataGen.
The benchmark was initially proposed to evaluate the performance of quantum
solvers. Therefore, the characteristics of this set of instances were designed
according to the current limitations of quantum devices. Additionally, the
dataset generator is included to allow the construction of general-purpose
benchmarks. The data introduced in this article provides a baseline that will
encourage quantum computing researchers to work on real-world bin packing
problems.
| [
{
"version": "v1",
"created": "Fri, 28 Apr 2023 09:29:43 GMT"
},
{
"version": "v2",
"created": "Tue, 9 May 2023 14:08:52 GMT"
},
{
"version": "v3",
"created": "Fri, 2 Jun 2023 08:11:15 GMT"
},
{
"version": "v4",
"created": "Thu, 29 Jun 2023 09:31:14 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Osaba",
"Eneko",
""
],
[
"Villar-Rodriguez",
"Esther",
""
],
[
"Romero",
"Sebastián V.",
""
]
] | TITLE: Benchmark dataset and instance generator for Real-World
Three-Dimensional Bin Packing Problems
ABSTRACT: In this article, a benchmark for real-world bin packing problems is proposed.
This dataset consists of 12 instances of varying levels of complexity regarding
size (with the number of packages ranging from 38 to 53) and user-defined
requirements. In fact, several real-world-oriented restrictions were taken into
account to build these instances: i) item and bin dimensions, ii) weight
restrictions, iii) affinities among package categories iv) preferences for
package ordering and v) load balancing. Besides the data, we also offer an own
developed Python script for the dataset generation, coined Q4RealBPP-DataGen.
The benchmark was initially proposed to evaluate the performance of quantum
solvers. Therefore, the characteristics of this set of instances were designed
according to the current limitations of quantum devices. Additionally, the
dataset generator is included to allow the construction of general-purpose
benchmarks. The data introduced in this article provides a baseline that will
encourage quantum computing researchers to work on real-world bin packing
problems.
| new_dataset | 0.718786 |
2305.01133 | Subrata Das | Subrata Das, Swaroop Ghosh | Randomized Reversible Gate-Based Obfuscation for Secured Compilation of
Quantum Circuit | 11 pages, 12 figures, conference | null | null | null | quant-ph cs.CR | http://creativecommons.org/licenses/by-nc-nd/4.0/ | The success of quantum circuits in providing reliable outcomes for a given
problem depends on the gate count and depth in near-term noisy quantum
computers. Quantum circuit compilers that decompose high-level gates to native
gates of the hardware and optimize the circuit play a key role in quantum
computing. However, the quality and time complexity of the optimization process
can vary significantly especially for practically relevant large-scale quantum
circuits. As a result, third-party (often less-trusted/untrusted) compilers
have emerged, claiming to provide better and faster optimization of complex
quantum circuits than so-called trusted compilers. However, untrusted compilers
can pose severe security risks, such as the theft of sensitive intellectual
property (IP) embedded within the quantum circuit. We propose an obfuscation
technique for quantum circuits using randomized reversible gates to protect
them from such attacks during compilation. The idea is to insert a small random
circuit into the original circuit and send it to the untrusted compiler. Since
the circuit function is corrupted, the adversary may get incorrect IP. However,
the user may also get incorrect output post-compilation. To circumvent this
issue, we concatenate the inverse of the random circuit in the compiled circuit
to recover the original functionality. We demonstrate the practicality of our
method by conducting exhaustive experiments on a set of benchmark circuits and
measuring the quality of obfuscation by calculating the Total Variation
Distance (TVD) metric. Our method achieves TVD of up to 1.92 and performs at
least 2X better than a previously reported obfuscation method. We also propose
a novel adversarial reverse engineering (RE) approach and show that the
proposed obfuscation is resilient against RE attacks. The proposed technique
introduces minimal degradation in fidelity (~1% to ~3% on average).
| [
{
"version": "v1",
"created": "Tue, 2 May 2023 00:24:34 GMT"
},
{
"version": "v2",
"created": "Thu, 29 Jun 2023 05:39:58 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Das",
"Subrata",
""
],
[
"Ghosh",
"Swaroop",
""
]
] | TITLE: Randomized Reversible Gate-Based Obfuscation for Secured Compilation of
Quantum Circuit
ABSTRACT: The success of quantum circuits in providing reliable outcomes for a given
problem depends on the gate count and depth in near-term noisy quantum
computers. Quantum circuit compilers that decompose high-level gates to native
gates of the hardware and optimize the circuit play a key role in quantum
computing. However, the quality and time complexity of the optimization process
can vary significantly especially for practically relevant large-scale quantum
circuits. As a result, third-party (often less-trusted/untrusted) compilers
have emerged, claiming to provide better and faster optimization of complex
quantum circuits than so-called trusted compilers. However, untrusted compilers
can pose severe security risks, such as the theft of sensitive intellectual
property (IP) embedded within the quantum circuit. We propose an obfuscation
technique for quantum circuits using randomized reversible gates to protect
them from such attacks during compilation. The idea is to insert a small random
circuit into the original circuit and send it to the untrusted compiler. Since
the circuit function is corrupted, the adversary may get incorrect IP. However,
the user may also get incorrect output post-compilation. To circumvent this
issue, we concatenate the inverse of the random circuit in the compiled circuit
to recover the original functionality. We demonstrate the practicality of our
method by conducting exhaustive experiments on a set of benchmark circuits and
measuring the quality of obfuscation by calculating the Total Variation
Distance (TVD) metric. Our method achieves TVD of up to 1.92 and performs at
least 2X better than a previously reported obfuscation method. We also propose
a novel adversarial reverse engineering (RE) approach and show that the
proposed obfuscation is resilient against RE attacks. The proposed technique
introduces minimal degradation in fidelity (~1% to ~3% on average).
| no_new_dataset | 0.709239 |
2305.02279 | Qiufeng Wang | Qiufeng Wang, Xu Yang, Shuxia Lin, Jing Wang, Xin Geng | Learngene: Inheriting Condensed Knowledge from the Ancestry Model to
Descendant Models | null | null | null | null | cs.LG cs.AI cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | During the continuous evolution of one organism's ancestry, its genes
accumulate extensive experiences and knowledge, enabling newborn descendants to
rapidly adapt to their specific environments. Motivated by this observation, we
propose a novel machine learning paradigm Learngene to enable learning models
to incorporate three key characteristics of genes. (i) Accumulating: the
knowledge is accumulated during the continuous learning of an ancestry model.
(ii) Condensing: the extensive accumulated knowledge is condensed into a much
more compact information piece, i.e., learngene. (iii) Inheriting: the
condensed learngene is inherited to make it easier for descendant models to
adapt to new environments. Since accumulating has been studied in
well-established paradigms like large-scale pre-training and lifelong learning,
we focus on condensing and inheriting, which induces three key issues and we
provide the preliminary solutions to these issues in this paper: (i) Learngene
Form: the learngene is set to a few integral layers that can preserve
significance. (ii) Learngene Condensing: we identify which layers among the
ancestry model have the most similarity as one pseudo descendant model. (iii)
Learngene Inheriting: to construct distinct descendant models for the specific
downstream tasks, we stack some randomly initialized layers to the learngene
layers. Extensive experiments across various settings, including using
different network architectures like Vision Transformer (ViT) and Convolutional
Neural Networks (CNNs) on different datasets, are carried out to confirm four
advantages of Learngene: it makes the descendant models 1) converge more
quickly, 2) exhibit less sensitivity to hyperparameters, 3) perform better, and
4) require fewer training samples to converge.
| [
{
"version": "v1",
"created": "Wed, 3 May 2023 17:15:58 GMT"
},
{
"version": "v2",
"created": "Sun, 25 Jun 2023 13:38:36 GMT"
},
{
"version": "v3",
"created": "Thu, 29 Jun 2023 14:04:26 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Wang",
"Qiufeng",
""
],
[
"Yang",
"Xu",
""
],
[
"Lin",
"Shuxia",
""
],
[
"Wang",
"Jing",
""
],
[
"Geng",
"Xin",
""
]
] | TITLE: Learngene: Inheriting Condensed Knowledge from the Ancestry Model to
Descendant Models
ABSTRACT: During the continuous evolution of one organism's ancestry, its genes
accumulate extensive experiences and knowledge, enabling newborn descendants to
rapidly adapt to their specific environments. Motivated by this observation, we
propose a novel machine learning paradigm Learngene to enable learning models
to incorporate three key characteristics of genes. (i) Accumulating: the
knowledge is accumulated during the continuous learning of an ancestry model.
(ii) Condensing: the extensive accumulated knowledge is condensed into a much
more compact information piece, i.e., learngene. (iii) Inheriting: the
condensed learngene is inherited to make it easier for descendant models to
adapt to new environments. Since accumulating has been studied in
well-established paradigms like large-scale pre-training and lifelong learning,
we focus on condensing and inheriting, which induces three key issues and we
provide the preliminary solutions to these issues in this paper: (i) Learngene
Form: the learngene is set to a few integral layers that can preserve
significance. (ii) Learngene Condensing: we identify which layers among the
ancestry model have the most similarity as one pseudo descendant model. (iii)
Learngene Inheriting: to construct distinct descendant models for the specific
downstream tasks, we stack some randomly initialized layers to the learngene
layers. Extensive experiments across various settings, including using
different network architectures like Vision Transformer (ViT) and Convolutional
Neural Networks (CNNs) on different datasets, are carried out to confirm four
advantages of Learngene: it makes the descendant models 1) converge more
quickly, 2) exhibit less sensitivity to hyperparameters, 3) perform better, and
4) require fewer training samples to converge.
| no_new_dataset | 0.7114 |
2305.16129 | Aldi Piroli | Aldi Piroli, Vinzenz Dallabetta, Johannes Kopp, Marc Walessa, Daniel
Meissner, Klaus Dietmayer | Energy-based Detection of Adverse Weather Effects in LiDAR Data | Accepted for publication in IEEE Robotics and Automation Letters
(RA-L) | IEEE Robotics and Automation Letters (RA-L) 2023 | 10.1109/LRA.2023.3282382 | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Autonomous vehicles rely on LiDAR sensors to perceive the environment.
Adverse weather conditions like rain, snow, and fog negatively affect these
sensors, reducing their reliability by introducing unwanted noise in the
measurements. In this work, we tackle this problem by proposing a novel
approach for detecting adverse weather effects in LiDAR data. We reformulate
this problem as an outlier detection task and use an energy-based framework to
detect outliers in point clouds. More specifically, our method learns to
associate low energy scores with inlier points and high energy scores with
outliers allowing for robust detection of adverse weather effects. In extensive
experiments, we show that our method performs better in adverse weather
detection and has higher robustness to unseen weather effects than previous
state-of-the-art methods. Furthermore, we show how our method can be used to
perform simultaneous outlier detection and semantic segmentation. Finally, to
help expand the research field of LiDAR perception in adverse weather, we
release the SemanticSpray dataset, which contains labeled vehicle spray data in
highway-like scenarios. The dataset is available at
https://semantic-spray-dataset.github.io .
| [
{
"version": "v1",
"created": "Thu, 25 May 2023 15:03:36 GMT"
},
{
"version": "v2",
"created": "Fri, 26 May 2023 14:19:49 GMT"
},
{
"version": "v3",
"created": "Thu, 29 Jun 2023 10:14:03 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Piroli",
"Aldi",
""
],
[
"Dallabetta",
"Vinzenz",
""
],
[
"Kopp",
"Johannes",
""
],
[
"Walessa",
"Marc",
""
],
[
"Meissner",
"Daniel",
""
],
[
"Dietmayer",
"Klaus",
""
]
] | TITLE: Energy-based Detection of Adverse Weather Effects in LiDAR Data
ABSTRACT: Autonomous vehicles rely on LiDAR sensors to perceive the environment.
Adverse weather conditions like rain, snow, and fog negatively affect these
sensors, reducing their reliability by introducing unwanted noise in the
measurements. In this work, we tackle this problem by proposing a novel
approach for detecting adverse weather effects in LiDAR data. We reformulate
this problem as an outlier detection task and use an energy-based framework to
detect outliers in point clouds. More specifically, our method learns to
associate low energy scores with inlier points and high energy scores with
outliers allowing for robust detection of adverse weather effects. In extensive
experiments, we show that our method performs better in adverse weather
detection and has higher robustness to unseen weather effects than previous
state-of-the-art methods. Furthermore, we show how our method can be used to
perform simultaneous outlier detection and semantic segmentation. Finally, to
help expand the research field of LiDAR perception in adverse weather, we
release the SemanticSpray dataset, which contains labeled vehicle spray data in
highway-like scenarios. The dataset is available at
https://semantic-spray-dataset.github.io .
| new_dataset | 0.715579 |
2305.16286 | Wangyou Zhang | Wangyou Zhang, Yanmin Qian | Weakly-Supervised Speech Pre-training: A Case Study on Target Speech
Recognition | Accepted by Interspeech; 5 pages, 1 figure, 3 tables | null | null | null | eess.AS cs.SD | http://creativecommons.org/licenses/by/4.0/ | Self-supervised learning (SSL) based speech pre-training has attracted much
attention for its capability of extracting rich representations learned from
massive unlabeled data. On the other hand, the use of weakly-supervised data is
less explored for speech pre-training. To fill this gap, we propose a
weakly-supervised speech pre-training method based on speaker-aware speech
data. It adopts a similar training procedure to the widely-used masked speech
prediction based SSL framework, while incorporating additional target-speaker
enrollment information as an auxiliary input. In this way, the learned
representation is steered towards the target speaker even in the presence of
highly overlapping interference, allowing potential applications to tasks such
as target speech recognition. Our experiments on Libri2Mix and WSJ0-2mix
datasets show that the proposed model achieves significantly better ASR
performance compared to WavLM, the state-of-the-art SSL model with denoising
capability.
| [
{
"version": "v1",
"created": "Thu, 25 May 2023 17:42:27 GMT"
},
{
"version": "v2",
"created": "Thu, 29 Jun 2023 14:42:51 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Zhang",
"Wangyou",
""
],
[
"Qian",
"Yanmin",
""
]
] | TITLE: Weakly-Supervised Speech Pre-training: A Case Study on Target Speech
Recognition
ABSTRACT: Self-supervised learning (SSL) based speech pre-training has attracted much
attention for its capability of extracting rich representations learned from
massive unlabeled data. On the other hand, the use of weakly-supervised data is
less explored for speech pre-training. To fill this gap, we propose a
weakly-supervised speech pre-training method based on speaker-aware speech
data. It adopts a similar training procedure to the widely-used masked speech
prediction based SSL framework, while incorporating additional target-speaker
enrollment information as an auxiliary input. In this way, the learned
representation is steered towards the target speaker even in the presence of
highly overlapping interference, allowing potential applications to tasks such
as target speech recognition. Our experiments on Libri2Mix and WSJ0-2mix
datasets show that the proposed model achieves significantly better ASR
performance compared to WavLM, the state-of-the-art SSL model with denoising
capability.
| no_new_dataset | 0.708862 |
2305.17595 | Shuhei Watanabe | Shuhei Watanabe | Python Wrapper for Simulating Multi-Fidelity Optimization on HPO
Benchmarks without Any Wait | null | null | null | null | cs.LG cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Hyperparameter (HP) optimization of deep learning (DL) is essential for high
performance. As DL often requires several hours to days for its training, HP
optimization (HPO) of DL is often prohibitively expensive. This boosted the
emergence of tabular or surrogate benchmarks, which enable querying the
(predictive) performance of DL with a specific HP configuration in a fraction.
However, since the actual runtime of a DL training is significantly different
from its query response time, simulators of an asynchronous HPO, e.g.
multi-fidelity optimization, must wait for the actual runtime at each iteration
in a na\"ive implementation; otherwise, the evaluation order during simulation
does not match with the real experiment. To ease this issue, we developed a
Python wrapper and describe its usage. This wrapper forces each worker to wait
so that we yield exactly the same evaluation order as in the real experiment
with only $10^{-2}$ seconds of waiting instead of waiting several hours. Our
implementation is available at
https://github.com/nabenabe0928/mfhpo-simulator/.
| [
{
"version": "v1",
"created": "Sat, 27 May 2023 23:28:54 GMT"
},
{
"version": "v2",
"created": "Thu, 29 Jun 2023 16:27:23 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Watanabe",
"Shuhei",
""
]
] | TITLE: Python Wrapper for Simulating Multi-Fidelity Optimization on HPO
Benchmarks without Any Wait
ABSTRACT: Hyperparameter (HP) optimization of deep learning (DL) is essential for high
performance. As DL often requires several hours to days for its training, HP
optimization (HPO) of DL is often prohibitively expensive. This boosted the
emergence of tabular or surrogate benchmarks, which enable querying the
(predictive) performance of DL with a specific HP configuration in a fraction.
However, since the actual runtime of a DL training is significantly different
from its query response time, simulators of an asynchronous HPO, e.g.
multi-fidelity optimization, must wait for the actual runtime at each iteration
in a na\"ive implementation; otherwise, the evaluation order during simulation
does not match with the real experiment. To ease this issue, we developed a
Python wrapper and describe its usage. This wrapper forces each worker to wait
so that we yield exactly the same evaluation order as in the real experiment
with only $10^{-2}$ seconds of waiting instead of waiting several hours. Our
implementation is available at
https://github.com/nabenabe0928/mfhpo-simulator/.
| no_new_dataset | 0.709837 |
2305.19067 | Yifu Zhang | Yifu Zhang, Hongru Li, Shimeng Shi, Youqi Li, Jiansong Zhang | Multi-source adversarial transfer learning based on similar source
domains with local features | Submitted to Information Fusion | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by/4.0/ | Transfer learning leverages knowledge from other domains and has been
successful in many applications. Transfer learning methods rely on the overall
similarity of the source and target domains. However, in some cases, it is
impossible to provide an overall similar source domain, and only some source
domains with similar local features can be provided. Can transfer learning be
achieved? In this regard, we propose a multi-source adversarial transfer
learning method based on local feature similarity to the source domain to
handle transfer scenarios where the source and target domains have only local
similarities. This method extracts transferable local features between a single
source domain and the target domain through a sub-network. Specifically, the
feature extractor of the sub-network is induced by the domain discriminator to
learn transferable knowledge between the source domain and the target domain.
The extracted features are then weighted by an attention module to suppress
non-transferable local features while enhancing transferable local features. In
order to ensure that the data from the target domain in different sub-networks
in the same batch is exactly the same, we designed a multi-source domain
independent strategy to provide the possibility for later local feature fusion
to complete the key features required. In order to verify the effectiveness of
the method, we made the dataset "Local Carvana Image Masking Dataset". Applying
the proposed method to the image segmentation task of the proposed dataset
achieves better transfer performance than other multi-source transfer learning
methods. It is shown that the designed transfer learning method is feasible for
transfer scenarios where the source and target domains have only local
similarities.
| [
{
"version": "v1",
"created": "Tue, 30 May 2023 14:29:07 GMT"
},
{
"version": "v2",
"created": "Thu, 29 Jun 2023 09:35:07 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Zhang",
"Yifu",
""
],
[
"Li",
"Hongru",
""
],
[
"Shi",
"Shimeng",
""
],
[
"Li",
"Youqi",
""
],
[
"Zhang",
"Jiansong",
""
]
] | TITLE: Multi-source adversarial transfer learning based on similar source
domains with local features
ABSTRACT: Transfer learning leverages knowledge from other domains and has been
successful in many applications. Transfer learning methods rely on the overall
similarity of the source and target domains. However, in some cases, it is
impossible to provide an overall similar source domain, and only some source
domains with similar local features can be provided. Can transfer learning be
achieved? In this regard, we propose a multi-source adversarial transfer
learning method based on local feature similarity to the source domain to
handle transfer scenarios where the source and target domains have only local
similarities. This method extracts transferable local features between a single
source domain and the target domain through a sub-network. Specifically, the
feature extractor of the sub-network is induced by the domain discriminator to
learn transferable knowledge between the source domain and the target domain.
The extracted features are then weighted by an attention module to suppress
non-transferable local features while enhancing transferable local features. In
order to ensure that the data from the target domain in different sub-networks
in the same batch is exactly the same, we designed a multi-source domain
independent strategy to provide the possibility for later local feature fusion
to complete the key features required. In order to verify the effectiveness of
the method, we made the dataset "Local Carvana Image Masking Dataset". Applying
the proposed method to the image segmentation task of the proposed dataset
achieves better transfer performance than other multi-source transfer learning
methods. It is shown that the designed transfer learning method is feasible for
transfer scenarios where the source and target domains have only local
similarities.
| no_new_dataset | 0.708824 |
2306.08842 | Yaodong Yu | Yaodong Yu and Maziar Sanjabi and Yi Ma and Kamalika Chaudhuri and
Chuan Guo | ViP: A Differentially Private Foundation Model for Computer Vision | Code: https://github.com/facebookresearch/ViP-MAE. V2 adds a GitHub
link to the code | null | null | null | cs.CV cs.CR cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Artificial intelligence (AI) has seen a tremendous surge in capabilities
thanks to the use of foundation models trained on internet-scale data. On the
flip side, the uncurated nature of internet-scale data also poses significant
privacy and legal risks, as they often contain personal information or
copyrighted material that should not be trained on without permission. In this
work, we propose as a mitigation measure a recipe to train foundation vision
models with differential privacy (DP) guarantee. We identify masked
autoencoders as a suitable learning algorithm that aligns well with DP-SGD, and
train ViP -- a Vision transformer with differential Privacy -- under a strict
privacy budget of $\epsilon=8$ on the LAION400M dataset. We evaluate the
quality of representation learned by ViP using standard downstream vision
tasks; in particular, ViP achieves a (non-private) linear probing accuracy of
$55.7\%$ on ImageNet, comparable to that of end-to-end trained AlexNet (trained
and evaluated on ImageNet). Our result suggests that scaling to internet-scale
data can be practical for private learning. Code is available at
\url{https://github.com/facebookresearch/ViP-MAE}.
| [
{
"version": "v1",
"created": "Thu, 15 Jun 2023 04:06:24 GMT"
},
{
"version": "v2",
"created": "Wed, 28 Jun 2023 22:24:33 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Yu",
"Yaodong",
""
],
[
"Sanjabi",
"Maziar",
""
],
[
"Ma",
"Yi",
""
],
[
"Chaudhuri",
"Kamalika",
""
],
[
"Guo",
"Chuan",
""
]
] | TITLE: ViP: A Differentially Private Foundation Model for Computer Vision
ABSTRACT: Artificial intelligence (AI) has seen a tremendous surge in capabilities
thanks to the use of foundation models trained on internet-scale data. On the
flip side, the uncurated nature of internet-scale data also poses significant
privacy and legal risks, as they often contain personal information or
copyrighted material that should not be trained on without permission. In this
work, we propose as a mitigation measure a recipe to train foundation vision
models with differential privacy (DP) guarantee. We identify masked
autoencoders as a suitable learning algorithm that aligns well with DP-SGD, and
train ViP -- a Vision transformer with differential Privacy -- under a strict
privacy budget of $\epsilon=8$ on the LAION400M dataset. We evaluate the
quality of representation learned by ViP using standard downstream vision
tasks; in particular, ViP achieves a (non-private) linear probing accuracy of
$55.7\%$ on ImageNet, comparable to that of end-to-end trained AlexNet (trained
and evaluated on ImageNet). Our result suggests that scaling to internet-scale
data can be practical for private learning. Code is available at
\url{https://github.com/facebookresearch/ViP-MAE}.
| no_new_dataset | 0.710226 |
2306.12088 | Cheng Yang | Cheng Yang, Xue Yang, Dongxian Wu, Xiaohu Tang | An Efficient Virtual Data Generation Method for Reducing Communication
in Federated Learning | There are errors in the experimental settings in our paper | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Communication overhead is one of the major challenges in Federated
Learning(FL). A few classical schemes assume the server can extract the
auxiliary information about training data of the participants from the local
models to construct a central dummy dataset. The server uses the dummy dataset
to finetune aggregated global model to achieve the target test accuracy in
fewer communication rounds. In this paper, we summarize the above solutions
into a data-based communication-efficient FL framework. The key of the proposed
framework is to design an efficient extraction module(EM) which ensures the
dummy dataset has a positive effect on finetuning aggregated global model.
Different from the existing methods that use generator to design EM, our
proposed method, FedINIBoost borrows the idea of gradient match to construct
EM. Specifically, FedINIBoost builds a proxy dataset of the real dataset in two
steps for each participant at each communication round. Then the server
aggregates all the proxy datasets to form a central dummy dataset, which is
used to finetune aggregated global model. Extensive experiments verify the
superiority of our method compared with the existing classical method, FedAVG,
FedProx, Moon and FedFTG. Moreover, FedINIBoost plays a significant role in
finetuning the performance of aggregated global model at the initial stage of
FL.
| [
{
"version": "v1",
"created": "Wed, 21 Jun 2023 08:07:07 GMT"
},
{
"version": "v2",
"created": "Wed, 28 Jun 2023 09:22:57 GMT"
},
{
"version": "v3",
"created": "Thu, 29 Jun 2023 08:44:48 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Yang",
"Cheng",
""
],
[
"Yang",
"Xue",
""
],
[
"Wu",
"Dongxian",
""
],
[
"Tang",
"Xiaohu",
""
]
] | TITLE: An Efficient Virtual Data Generation Method for Reducing Communication
in Federated Learning
ABSTRACT: Communication overhead is one of the major challenges in Federated
Learning(FL). A few classical schemes assume the server can extract the
auxiliary information about training data of the participants from the local
models to construct a central dummy dataset. The server uses the dummy dataset
to finetune aggregated global model to achieve the target test accuracy in
fewer communication rounds. In this paper, we summarize the above solutions
into a data-based communication-efficient FL framework. The key of the proposed
framework is to design an efficient extraction module(EM) which ensures the
dummy dataset has a positive effect on finetuning aggregated global model.
Different from the existing methods that use generator to design EM, our
proposed method, FedINIBoost borrows the idea of gradient match to construct
EM. Specifically, FedINIBoost builds a proxy dataset of the real dataset in two
steps for each participant at each communication round. Then the server
aggregates all the proxy datasets to form a central dummy dataset, which is
used to finetune aggregated global model. Extensive experiments verify the
superiority of our method compared with the existing classical method, FedAVG,
FedProx, Moon and FedFTG. Moreover, FedINIBoost plays a significant role in
finetuning the performance of aggregated global model at the initial stage of
FL.
| no_new_dataset | 0.690963 |
2306.12243 | Chengchao Shen | Chengchao Shen, Dawei Liu, Hao Tang, Zhe Qu, Jianxin Wang | Inter-Instance Similarity Modeling for Contrastive Learning | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The existing contrastive learning methods widely adopt one-hot instance
discrimination as pretext task for self-supervised learning, which inevitably
neglects rich inter-instance similarities among natural images, then leading to
potential representation degeneration. In this paper, we propose a novel image
mix method, PatchMix, for contrastive learning in Vision Transformer (ViT), to
model inter-instance similarities among images. Following the nature of ViT, we
randomly mix multiple images from mini-batch in patch level to construct mixed
image patch sequences for ViT. Compared to the existing sample mix methods, our
PatchMix can flexibly and efficiently mix more than two images and simulate
more complicated similarity relations among natural images. In this manner, our
contrastive framework can significantly reduce the gap between contrastive
objective and ground truth in reality. Experimental results demonstrate that
our proposed method significantly outperforms the previous state-of-the-art on
both ImageNet-1K and CIFAR datasets, e.g., 3.0% linear accuracy improvement on
ImageNet-1K and 8.7% kNN accuracy improvement on CIFAR100. Moreover, our method
achieves the leading transfer performance on downstream tasks, object detection
and instance segmentation on COCO dataset. The code is available at
https://github.com/visresearch/patchmix
| [
{
"version": "v1",
"created": "Wed, 21 Jun 2023 13:03:47 GMT"
},
{
"version": "v2",
"created": "Sat, 24 Jun 2023 14:47:33 GMT"
},
{
"version": "v3",
"created": "Thu, 29 Jun 2023 12:14:59 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Shen",
"Chengchao",
""
],
[
"Liu",
"Dawei",
""
],
[
"Tang",
"Hao",
""
],
[
"Qu",
"Zhe",
""
],
[
"Wang",
"Jianxin",
""
]
] | TITLE: Inter-Instance Similarity Modeling for Contrastive Learning
ABSTRACT: The existing contrastive learning methods widely adopt one-hot instance
discrimination as pretext task for self-supervised learning, which inevitably
neglects rich inter-instance similarities among natural images, then leading to
potential representation degeneration. In this paper, we propose a novel image
mix method, PatchMix, for contrastive learning in Vision Transformer (ViT), to
model inter-instance similarities among images. Following the nature of ViT, we
randomly mix multiple images from mini-batch in patch level to construct mixed
image patch sequences for ViT. Compared to the existing sample mix methods, our
PatchMix can flexibly and efficiently mix more than two images and simulate
more complicated similarity relations among natural images. In this manner, our
contrastive framework can significantly reduce the gap between contrastive
objective and ground truth in reality. Experimental results demonstrate that
our proposed method significantly outperforms the previous state-of-the-art on
both ImageNet-1K and CIFAR datasets, e.g., 3.0% linear accuracy improvement on
ImageNet-1K and 8.7% kNN accuracy improvement on CIFAR100. Moreover, our method
achieves the leading transfer performance on downstream tasks, object detection
and instance segmentation on COCO dataset. The code is available at
https://github.com/visresearch/patchmix
| no_new_dataset | 0.71145 |
2306.13141 | Abeba Birhane | Abeba Birhane, Vinay Prabhu, Sang Han, Vishnu Naresh Boddeti | On Hate Scaling Laws For Data-Swamps | null | null | null | null | cs.CY | http://creativecommons.org/licenses/by/4.0/ | `Scale the model, scale the data, scale the GPU-farms' is the reigning
sentiment in the world of generative AI today. While model scaling has been
extensively studied, data scaling and its downstream impacts remain under
explored. This is especially of critical importance in the context of
visio-linguistic datasets whose main source is the World Wide Web, condensed
and packaged as the CommonCrawl dump. This large scale data-dump, which is
known to have numerous drawbacks, is repeatedly mined and serves as the
data-motherlode for large generative models. In this paper, we: 1) investigate
the effect of scaling datasets on hateful content through a comparative audit
of the LAION-400M and LAION-2B-en, containing 400 million and 2 billion samples
respectively, and 2) evaluate the downstream impact of scale on
visio-linguistic models trained on these dataset variants by measuring racial
bias of the models trained on them using the Chicago Face Dataset (CFD) as a
probe. Our results show that 1) the presence of hateful content in datasets,
when measured with a Hate Content Rate (HCR) metric on the inferences of the
Pysentimiento hate-detection Natural Language Processing (NLP) model, increased
by nearly $12\%$ and 2) societal biases and negative stereotypes were also
exacerbated with scale on the models we evaluated. As scale increased, the
tendency of the model to associate images of human faces with the `human being'
class over 7 other offensive classes reduced by half. Furthermore, for the
Black female category, the tendency of the model to associate their faces with
the `criminal' class doubled, while quintupling for Black male faces. We
present a qualitative and historical analysis of the model audit results,
reflect on our findings and its implications for dataset curation practice, and
close with a summary of our findings and potential future work to be done in
this area.
| [
{
"version": "v1",
"created": "Thu, 22 Jun 2023 18:00:17 GMT"
},
{
"version": "v2",
"created": "Wed, 28 Jun 2023 18:00:50 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Birhane",
"Abeba",
""
],
[
"Prabhu",
"Vinay",
""
],
[
"Han",
"Sang",
""
],
[
"Boddeti",
"Vishnu Naresh",
""
]
] | TITLE: On Hate Scaling Laws For Data-Swamps
ABSTRACT: `Scale the model, scale the data, scale the GPU-farms' is the reigning
sentiment in the world of generative AI today. While model scaling has been
extensively studied, data scaling and its downstream impacts remain under
explored. This is especially of critical importance in the context of
visio-linguistic datasets whose main source is the World Wide Web, condensed
and packaged as the CommonCrawl dump. This large scale data-dump, which is
known to have numerous drawbacks, is repeatedly mined and serves as the
data-motherlode for large generative models. In this paper, we: 1) investigate
the effect of scaling datasets on hateful content through a comparative audit
of the LAION-400M and LAION-2B-en, containing 400 million and 2 billion samples
respectively, and 2) evaluate the downstream impact of scale on
visio-linguistic models trained on these dataset variants by measuring racial
bias of the models trained on them using the Chicago Face Dataset (CFD) as a
probe. Our results show that 1) the presence of hateful content in datasets,
when measured with a Hate Content Rate (HCR) metric on the inferences of the
Pysentimiento hate-detection Natural Language Processing (NLP) model, increased
by nearly $12\%$ and 2) societal biases and negative stereotypes were also
exacerbated with scale on the models we evaluated. As scale increased, the
tendency of the model to associate images of human faces with the `human being'
class over 7 other offensive classes reduced by half. Furthermore, for the
Black female category, the tendency of the model to associate their faces with
the `criminal' class doubled, while quintupling for Black male faces. We
present a qualitative and historical analysis of the model audit results,
reflect on our findings and its implications for dataset curation practice, and
close with a summary of our findings and potential future work to be done in
this area.
| no_new_dataset | 0.709384 |
2306.13651 | Jonas Geiping | Neel Jain, Khalid Saifullah, Yuxin Wen, John Kirchenbauer, Manli Shu,
Aniruddha Saha, Micah Goldblum, Jonas Geiping and Tom Goldstein | Bring Your Own Data! Self-Supervised Evaluation for Large Language
Models | Code is available at https://github.com/neelsjain/BYOD. First two
authors contributed equally. 21 pages, 22 figures | null | null | null | cs.CL cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | With the rise of Large Language Models (LLMs) and their ubiquitous deployment
in diverse domains, measuring language model behavior on realistic data is
imperative. For example, a company deploying a client-facing chatbot must
ensure that the model will not respond to client requests with profanity.
Current evaluations approach this problem using small, domain-specific datasets
with human-curated labels. These evaluation sets are often sampled from a
narrow and simplified distribution, and data sources can unknowingly be leaked
into the training set which can lead to misleading evaluations. To bypass these
drawbacks, we propose a framework for self-supervised evaluation of LLMs by
analyzing their sensitivity or invariance to transformations on the input text.
Self-supervised evaluation can directly monitor LLM behavior on datasets
collected in the wild or streamed during live model deployment. We demonstrate
self-supervised evaluation strategies for measuring closed-book knowledge,
toxicity, and long-range context dependence, in addition to sensitivity to
grammatical structure and tokenization errors. When comparisons to similar
human-labeled benchmarks are available, we find strong correlations between
self-supervised and human-supervised evaluations. The self-supervised paradigm
complements current evaluation strategies that rely on labeled data.
| [
{
"version": "v1",
"created": "Fri, 23 Jun 2023 17:59:09 GMT"
},
{
"version": "v2",
"created": "Thu, 29 Jun 2023 17:30:45 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Jain",
"Neel",
""
],
[
"Saifullah",
"Khalid",
""
],
[
"Wen",
"Yuxin",
""
],
[
"Kirchenbauer",
"John",
""
],
[
"Shu",
"Manli",
""
],
[
"Saha",
"Aniruddha",
""
],
[
"Goldblum",
"Micah",
""
],
[
"Geiping",
"Jonas",
""
],
[
"Goldstein",
"Tom",
""
]
] | TITLE: Bring Your Own Data! Self-Supervised Evaluation for Large Language
Models
ABSTRACT: With the rise of Large Language Models (LLMs) and their ubiquitous deployment
in diverse domains, measuring language model behavior on realistic data is
imperative. For example, a company deploying a client-facing chatbot must
ensure that the model will not respond to client requests with profanity.
Current evaluations approach this problem using small, domain-specific datasets
with human-curated labels. These evaluation sets are often sampled from a
narrow and simplified distribution, and data sources can unknowingly be leaked
into the training set which can lead to misleading evaluations. To bypass these
drawbacks, we propose a framework for self-supervised evaluation of LLMs by
analyzing their sensitivity or invariance to transformations on the input text.
Self-supervised evaluation can directly monitor LLM behavior on datasets
collected in the wild or streamed during live model deployment. We demonstrate
self-supervised evaluation strategies for measuring closed-book knowledge,
toxicity, and long-range context dependence, in addition to sensitivity to
grammatical structure and tokenization errors. When comparisons to similar
human-labeled benchmarks are available, we find strong correlations between
self-supervised and human-supervised evaluations. The self-supervised paradigm
complements current evaluation strategies that rely on labeled data.
| no_new_dataset | 0.709831 |
2306.14406 | Xinquan Yang | Xinquan Yang and Jinheng Xie and Xuguang Li and Xuechen Li and Xin Li
and Linlin Shen and Yongqiang Deng | TCEIP: Text Condition Embedded Regression Network for Dental Implant
Position Prediction | MICCAI 2023 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | When deep neural network has been proposed to assist the dentist in designing
the location of dental implant, most of them are targeting simple cases where
only one missing tooth is available. As a result, literature works do not work
well when there are multiple missing teeth and easily generate false
predictions when the teeth are sparsely distributed. In this paper, we are
trying to integrate a weak supervision text, the target region, to the implant
position regression network, to address above issues. We propose a text
condition embedded implant position regression network (TCEIP), to embed the
text condition into the encoder-decoder framework for improvement of the
regression performance. A cross-modal interaction that consists of cross-modal
attention (CMA) and knowledge alignment module (KAM) is proposed to facilitate
the interaction between features of images and texts. The CMA module performs a
cross-attention between the image feature and the text condition, and the KAM
mitigates the knowledge gap between the image feature and the image encoder of
the CLIP. Extensive experiments on a dental implant dataset through five-fold
cross-validation demonstrated that the proposed TCEIP achieves superior
performance than existing methods.
| [
{
"version": "v1",
"created": "Mon, 26 Jun 2023 03:38:43 GMT"
},
{
"version": "v2",
"created": "Thu, 29 Jun 2023 12:52:56 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Yang",
"Xinquan",
""
],
[
"Xie",
"Jinheng",
""
],
[
"Li",
"Xuguang",
""
],
[
"Li",
"Xuechen",
""
],
[
"Li",
"Xin",
""
],
[
"Shen",
"Linlin",
""
],
[
"Deng",
"Yongqiang",
""
]
] | TITLE: TCEIP: Text Condition Embedded Regression Network for Dental Implant
Position Prediction
ABSTRACT: When deep neural network has been proposed to assist the dentist in designing
the location of dental implant, most of them are targeting simple cases where
only one missing tooth is available. As a result, literature works do not work
well when there are multiple missing teeth and easily generate false
predictions when the teeth are sparsely distributed. In this paper, we are
trying to integrate a weak supervision text, the target region, to the implant
position regression network, to address above issues. We propose a text
condition embedded implant position regression network (TCEIP), to embed the
text condition into the encoder-decoder framework for improvement of the
regression performance. A cross-modal interaction that consists of cross-modal
attention (CMA) and knowledge alignment module (KAM) is proposed to facilitate
the interaction between features of images and texts. The CMA module performs a
cross-attention between the image feature and the text condition, and the KAM
mitigates the knowledge gap between the image feature and the image encoder of
the CLIP. Extensive experiments on a dental implant dataset through five-fold
cross-validation demonstrated that the proposed TCEIP achieves superior
performance than existing methods.
| no_new_dataset | 0.709818 |
2306.14584 | Guillem Delgado | Guillem Delgado, Andoni Cort\'es, Sara Garc\'ia, Est\'ibaliz Loyo,
Maialen Berasategi, Nerea Aranjuelo | Methodology for generating synthetic labeled datasets for visual
container inspection | null | Transportation Research Part E: Logistics and Transportation
Review, Volume 175, 2023, 103174 | 10.1016/j.tre.2023.103174 | null | cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Nowadays, containerized freight transport is one of the most important
transportation systems that is undergoing an automation process due to the Deep
Learning success. However, it suffers from a lack of annotated data in order to
incorporate state-of-the-art neural network models to its systems. In this
paper we present an innovative methodology to generate a realistic, varied,
balanced, and labelled dataset for visual inspection task of containers in a
dock environment. In addition, we validate this methodology with multiple
visual tasks recurrently found in the state of the art. We prove that the
generated synthetic labelled dataset allows to train a deep neural network that
can be used in a real world scenario. On the other side, using this methodology
we provide the first open synthetic labelled dataset called SeaFront available
in: https://datasets.vicomtech.org/di21-seafront/readme.txt.
| [
{
"version": "v1",
"created": "Mon, 26 Jun 2023 10:51:18 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Delgado",
"Guillem",
""
],
[
"Cortés",
"Andoni",
""
],
[
"García",
"Sara",
""
],
[
"Loyo",
"Estíbaliz",
""
],
[
"Berasategi",
"Maialen",
""
],
[
"Aranjuelo",
"Nerea",
""
]
] | TITLE: Methodology for generating synthetic labeled datasets for visual
container inspection
ABSTRACT: Nowadays, containerized freight transport is one of the most important
transportation systems that is undergoing an automation process due to the Deep
Learning success. However, it suffers from a lack of annotated data in order to
incorporate state-of-the-art neural network models to its systems. In this
paper we present an innovative methodology to generate a realistic, varied,
balanced, and labelled dataset for visual inspection task of containers in a
dock environment. In addition, we validate this methodology with multiple
visual tasks recurrently found in the state of the art. We prove that the
generated synthetic labelled dataset allows to train a deep neural network that
can be used in a real world scenario. On the other side, using this methodology
we provide the first open synthetic labelled dataset called SeaFront available
in: https://datasets.vicomtech.org/di21-seafront/readme.txt.
| new_dataset | 0.714055 |
2306.15138 | Gang Wu | Yongyan Guo and Gang Wu | A Restarted Large-Scale Spectral Clustering with Self-Guiding and Block
Diagonal Representation | 36 pages | null | null | null | cs.LG cs.NA math.NA | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Spectral clustering is one of the most popular unsupervised machine learning
methods. Constructing similarity matrix is crucial to this type of method. In
most existing works, the similarity matrix is computed once for all or is
updated alternatively. However, the former is difficult to reflect
comprehensive relationships among data points, and the latter is time-consuming
and is even infeasible for large-scale problems. In this work, we propose a
restarted clustering framework with self-guiding and block diagonal
representation. An advantage of the strategy is that some useful clustering
information obtained from previous cycles could be preserved as much as
possible. To the best of our knowledge, this is the first work that applies
restarting strategy to spectral clustering. The key difference is that we
reclassify the samples in each cycle of our method, while they are classified
only once in existing methods. To further release the overhead, we introduce a
block diagonal representation with Nystr\"{o}m approximation for constructing
the similarity matrix. Theoretical results are established to show the
rationality of inexact computations in spectral clustering. Comprehensive
experiments are performed on some benchmark databases, which show the
superiority of our proposed algorithms over many state-of-the-art algorithms
for large-scale problems. Specifically, our framework has a potential boost for
clustering algorithms and works well even using an initial guess chosen
randomly.
| [
{
"version": "v1",
"created": "Tue, 27 Jun 2023 01:38:52 GMT"
},
{
"version": "v2",
"created": "Thu, 29 Jun 2023 06:51:09 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Guo",
"Yongyan",
""
],
[
"Wu",
"Gang",
""
]
] | TITLE: A Restarted Large-Scale Spectral Clustering with Self-Guiding and Block
Diagonal Representation
ABSTRACT: Spectral clustering is one of the most popular unsupervised machine learning
methods. Constructing similarity matrix is crucial to this type of method. In
most existing works, the similarity matrix is computed once for all or is
updated alternatively. However, the former is difficult to reflect
comprehensive relationships among data points, and the latter is time-consuming
and is even infeasible for large-scale problems. In this work, we propose a
restarted clustering framework with self-guiding and block diagonal
representation. An advantage of the strategy is that some useful clustering
information obtained from previous cycles could be preserved as much as
possible. To the best of our knowledge, this is the first work that applies
restarting strategy to spectral clustering. The key difference is that we
reclassify the samples in each cycle of our method, while they are classified
only once in existing methods. To further release the overhead, we introduce a
block diagonal representation with Nystr\"{o}m approximation for constructing
the similarity matrix. Theoretical results are established to show the
rationality of inexact computations in spectral clustering. Comprehensive
experiments are performed on some benchmark databases, which show the
superiority of our proposed algorithms over many state-of-the-art algorithms
for large-scale problems. Specifically, our framework has a potential boost for
clustering algorithms and works well even using an initial guess chosen
randomly.
| no_new_dataset | 0.711262 |
2306.15662 | Jiaye Wu | Jiaye Wu, Sanjoy Chowdhury, Hariharmano Shanmugaraja, David Jacobs,
and Soumyadip Sengupta | Measured Albedo in the Wild: Filling the Gap in Intrinsics Evaluation | Accepted into ICCP2023 | null | null | null | cs.CV | http://creativecommons.org/licenses/by-sa/4.0/ | Intrinsic image decomposition and inverse rendering are long-standing
problems in computer vision. To evaluate albedo recovery, most algorithms
report their quantitative performance with a mean Weighted Human Disagreement
Rate (WHDR) metric on the IIW dataset. However, WHDR focuses only on relative
albedo values and often fails to capture overall quality of the albedo. In
order to comprehensively evaluate albedo, we collect a new dataset, Measured
Albedo in the Wild (MAW), and propose three new metrics that complement WHDR:
intensity, chromaticity and texture metrics. We show that existing algorithms
often improve WHDR metric but perform poorly on other metrics. We then finetune
different algorithms on our MAW dataset to significantly improve the quality of
the reconstructed albedo both quantitatively and qualitatively. Since the
proposed intensity, chromaticity, and texture metrics and the WHDR are all
complementary we further introduce a relative performance measure that captures
average performance. By analysing existing algorithms we show that there is
significant room for improvement. Our dataset and evaluation metrics will
enable researchers to develop algorithms that improve albedo reconstruction.
Code and Data available at: https://measuredalbedo.github.io/
| [
{
"version": "v1",
"created": "Tue, 27 Jun 2023 17:55:33 GMT"
},
{
"version": "v2",
"created": "Thu, 29 Jun 2023 17:42:44 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Wu",
"Jiaye",
""
],
[
"Chowdhury",
"Sanjoy",
""
],
[
"Shanmugaraja",
"Hariharmano",
""
],
[
"Jacobs",
"David",
""
],
[
"Sengupta",
"Soumyadip",
""
]
] | TITLE: Measured Albedo in the Wild: Filling the Gap in Intrinsics Evaluation
ABSTRACT: Intrinsic image decomposition and inverse rendering are long-standing
problems in computer vision. To evaluate albedo recovery, most algorithms
report their quantitative performance with a mean Weighted Human Disagreement
Rate (WHDR) metric on the IIW dataset. However, WHDR focuses only on relative
albedo values and often fails to capture overall quality of the albedo. In
order to comprehensively evaluate albedo, we collect a new dataset, Measured
Albedo in the Wild (MAW), and propose three new metrics that complement WHDR:
intensity, chromaticity and texture metrics. We show that existing algorithms
often improve WHDR metric but perform poorly on other metrics. We then finetune
different algorithms on our MAW dataset to significantly improve the quality of
the reconstructed albedo both quantitatively and qualitatively. Since the
proposed intensity, chromaticity, and texture metrics and the WHDR are all
complementary we further introduce a relative performance measure that captures
average performance. By analysing existing algorithms we show that there is
significant room for improvement. Our dataset and evaluation metrics will
enable researchers to develop algorithms that improve albedo reconstruction.
Code and Data available at: https://measuredalbedo.github.io/
| new_dataset | 0.71703 |
2306.15786 | Sebastian M\"uller | Sebastian M\"uller, Vanessa Toborek, Katharina Beckh, Matthias Jakobs,
Christian Bauckhage and Pascal Welke | An Empirical Evaluation of the Rashomon Effect in Explainable Machine
Learning | null | null | null | null | cs.LG cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The Rashomon Effect describes the following phenomenon: for a given dataset
there may exist many models with equally good performance but with different
solution strategies. The Rashomon Effect has implications for Explainable
Machine Learning, especially for the comparability of explanations. We provide
a unified view on three different comparison scenarios and conduct a
quantitative evaluation across different datasets, models, attribution methods,
and metrics. We find that hyperparameter-tuning plays a role and that metric
selection matters. Our results provide empirical support for previously
anecdotal evidence and exhibit challenges for both scientists and
practitioners.
| [
{
"version": "v1",
"created": "Tue, 27 Jun 2023 20:32:07 GMT"
},
{
"version": "v2",
"created": "Thu, 29 Jun 2023 07:55:55 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Müller",
"Sebastian",
""
],
[
"Toborek",
"Vanessa",
""
],
[
"Beckh",
"Katharina",
""
],
[
"Jakobs",
"Matthias",
""
],
[
"Bauckhage",
"Christian",
""
],
[
"Welke",
"Pascal",
""
]
] | TITLE: An Empirical Evaluation of the Rashomon Effect in Explainable Machine
Learning
ABSTRACT: The Rashomon Effect describes the following phenomenon: for a given dataset
there may exist many models with equally good performance but with different
solution strategies. The Rashomon Effect has implications for Explainable
Machine Learning, especially for the comparability of explanations. We provide
a unified view on three different comparison scenarios and conduct a
quantitative evaluation across different datasets, models, attribution methods,
and metrics. We find that hyperparameter-tuning plays a role and that metric
selection matters. Our results provide empirical support for previously
anecdotal evidence and exhibit challenges for both scientists and
practitioners.
| no_new_dataset | 0.712045 |
2306.16125 | Simon Sanchez Viloria Mr | Simon Sanchez Viloria, Daniel Peix del R\'io, Rub\'en Berm\'udez Cabo,
Guillermo Arturo Arrojo Fuentes, Isabel Segura-Bedmar | A Framework for Identifying Depression on Social Media:
MentalRiskES@IberLEF 2023 | Submitted to the Proceedings of IberLEF 2023, September 2023, Ja\'en,
Spain | null | null | null | cs.CL cs.AI | http://creativecommons.org/licenses/by/4.0/ | This paper describes our participation in the MentalRiskES task at IberLEF
2023. The task involved predicting the likelihood of an individual experiencing
depression based on their social media activity. The dataset consisted of
conversations from 175 Telegram users, each labeled according to their evidence
of suffering from the disorder. We used a combination of traditional machine
learning and deep learning techniques to solve four predictive subtasks: binary
classification, simple regression, multiclass classification, and multi-output
regression.
We approached this by training a model to solve the multi-output regression
case and then transforming the predictions to work for the other three
subtasks.
We compare the performance of two modeling approaches: fine-tuning a
BERT-based model directly for the task or using its embeddings as inputs to a
linear regressor, with the latter yielding better results. The code to
reproduce our results can be found at:
https://github.com/simonsanvil/EarlyDepression-MentalRiskES
| [
{
"version": "v1",
"created": "Wed, 28 Jun 2023 11:53:07 GMT"
},
{
"version": "v2",
"created": "Thu, 29 Jun 2023 07:02:59 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Viloria",
"Simon Sanchez",
""
],
[
"del Río",
"Daniel Peix",
""
],
[
"Cabo",
"Rubén Bermúdez",
""
],
[
"Fuentes",
"Guillermo Arturo Arrojo",
""
],
[
"Segura-Bedmar",
"Isabel",
""
]
] | TITLE: A Framework for Identifying Depression on Social Media:
MentalRiskES@IberLEF 2023
ABSTRACT: This paper describes our participation in the MentalRiskES task at IberLEF
2023. The task involved predicting the likelihood of an individual experiencing
depression based on their social media activity. The dataset consisted of
conversations from 175 Telegram users, each labeled according to their evidence
of suffering from the disorder. We used a combination of traditional machine
learning and deep learning techniques to solve four predictive subtasks: binary
classification, simple regression, multiclass classification, and multi-output
regression.
We approached this by training a model to solve the multi-output regression
case and then transforming the predictions to work for the other three
subtasks.
We compare the performance of two modeling approaches: fine-tuning a
BERT-based model directly for the task or using its embeddings as inputs to a
linear regressor, with the latter yielding better results. The code to
reproduce our results can be found at:
https://github.com/simonsanvil/EarlyDepression-MentalRiskES
| no_new_dataset | 0.709384 |
2306.16425 | Pengye Zhang | Wei Zhang, Pengye Zhang, Bo Zhang, Xingxing Wang, Dong Wang | A Collaborative Transfer Learning Framework for Cross-domain
Recommendation | KDD2023 accepted | null | null | null | cs.IR cs.LG | http://creativecommons.org/licenses/by/4.0/ | In the recommendation systems, there are multiple business domains to meet
the diverse interests and needs of users, and the click-through rate(CTR) of
each domain can be quite different, which leads to the demand for CTR
prediction modeling for different business domains. The industry solution is to
use domain-specific models or transfer learning techniques for each domain. The
disadvantage of the former is that the data from other domains is not utilized
by a single domain model, while the latter leverage all the data from different
domains, but the fine-tuned model of transfer learning may trap the model in a
local optimum of the source domain, making it difficult to fit the target
domain. Meanwhile, significant differences in data quantity and feature schemas
between different domains, known as domain shift, may lead to negative transfer
in the process of transferring. To overcome these challenges, we propose the
Collaborative Cross-Domain Transfer Learning Framework (CCTL). CCTL evaluates
the information gain of the source domain on the target domain using a
symmetric companion network and adjusts the information transfer weight of each
source domain sample using the information flow network. This approach enables
full utilization of other domain data while avoiding negative migration.
Additionally, a representation enhancement network is used as an auxiliary task
to preserve domain-specific features. Comprehensive experiments on both public
and real-world industrial datasets, CCTL achieved SOTA score on offline
metrics. At the same time, the CCTL algorithm has been deployed in Meituan,
bringing 4.37% CTR and 5.43% GMV lift, which is significant to the business.
| [
{
"version": "v1",
"created": "Mon, 26 Jun 2023 09:43:58 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Zhang",
"Wei",
""
],
[
"Zhang",
"Pengye",
""
],
[
"Zhang",
"Bo",
""
],
[
"Wang",
"Xingxing",
""
],
[
"Wang",
"Dong",
""
]
] | TITLE: A Collaborative Transfer Learning Framework for Cross-domain
Recommendation
ABSTRACT: In the recommendation systems, there are multiple business domains to meet
the diverse interests and needs of users, and the click-through rate(CTR) of
each domain can be quite different, which leads to the demand for CTR
prediction modeling for different business domains. The industry solution is to
use domain-specific models or transfer learning techniques for each domain. The
disadvantage of the former is that the data from other domains is not utilized
by a single domain model, while the latter leverage all the data from different
domains, but the fine-tuned model of transfer learning may trap the model in a
local optimum of the source domain, making it difficult to fit the target
domain. Meanwhile, significant differences in data quantity and feature schemas
between different domains, known as domain shift, may lead to negative transfer
in the process of transferring. To overcome these challenges, we propose the
Collaborative Cross-Domain Transfer Learning Framework (CCTL). CCTL evaluates
the information gain of the source domain on the target domain using a
symmetric companion network and adjusts the information transfer weight of each
source domain sample using the information flow network. This approach enables
full utilization of other domain data while avoiding negative migration.
Additionally, a representation enhancement network is used as an auxiliary task
to preserve domain-specific features. Comprehensive experiments on both public
and real-world industrial datasets, CCTL achieved SOTA score on offline
metrics. At the same time, the CCTL algorithm has been deployed in Meituan,
bringing 4.37% CTR and 5.43% GMV lift, which is significant to the business.
| no_new_dataset | 0.711825 |
2306.16431 | Joran Michiels | Joran Michiels, Maarten De Vos, Johan Suykens | Increasing Performance And Sample Efficiency With Model-agnostic
Interactive Feature Attributions | null | null | null | null | cs.LG cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Model-agnostic feature attributions can provide local insights in complex ML
models. If the explanation is correct, a domain expert can validate and trust
the model's decision. However, if it contradicts the expert's knowledge,
related work only corrects irrelevant features to improve the model. To allow
for unlimited interaction, in this paper we provide model-agnostic
implementations for two popular explanation methods (Occlusion and Shapley
values) to enforce entirely different attributions in the complex model. For a
particular set of samples, we use the corrected feature attributions to
generate extra local data, which is used to retrain the model to have the right
explanation for the samples. Through simulated and real data experiments on a
variety of models we show how our proposed approach can significantly improve
the model's performance only by augmenting its training dataset based on
corrected explanations. Adding our interactive explanations to active learning
settings increases the sample efficiency significantly and outperforms existing
explanatory interactive strategies. Additionally we explore how a domain expert
can provide feature attributions which are sufficiently correct to improve the
model.
| [
{
"version": "v1",
"created": "Wed, 28 Jun 2023 15:23:28 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Michiels",
"Joran",
""
],
[
"De Vos",
"Maarten",
""
],
[
"Suykens",
"Johan",
""
]
] | TITLE: Increasing Performance And Sample Efficiency With Model-agnostic
Interactive Feature Attributions
ABSTRACT: Model-agnostic feature attributions can provide local insights in complex ML
models. If the explanation is correct, a domain expert can validate and trust
the model's decision. However, if it contradicts the expert's knowledge,
related work only corrects irrelevant features to improve the model. To allow
for unlimited interaction, in this paper we provide model-agnostic
implementations for two popular explanation methods (Occlusion and Shapley
values) to enforce entirely different attributions in the complex model. For a
particular set of samples, we use the corrected feature attributions to
generate extra local data, which is used to retrain the model to have the right
explanation for the samples. Through simulated and real data experiments on a
variety of models we show how our proposed approach can significantly improve
the model's performance only by augmenting its training dataset based on
corrected explanations. Adding our interactive explanations to active learning
settings increases the sample efficiency significantly and outperforms existing
explanatory interactive strategies. Additionally we explore how a domain expert
can provide feature attributions which are sufficiently correct to improve the
model.
| no_new_dataset | 0.710584 |
2306.16482 | Aniket Pal | Aniket Pal, Krishna Pratap Singh | DenseBAM-GI: Attention Augmented DeneseNet with momentum aided GRU for
HMER | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | The task of recognising Handwritten Mathematical Expressions (HMER) is
crucial in the fields of digital education and scholarly research. However, it
is difficult to accurately determine the length and complex spatial
relationships among symbols in handwritten mathematical expressions. In this
study, we present a novel encoder-decoder architecture (DenseBAM-GI) for HMER,
where the encoder has a Bottleneck Attention Module (BAM) to improve feature
representation and the decoder has a Gated Input-GRU (GI-GRU) unit with an
extra gate to make decoding long and complex expressions easier. The proposed
model is an efficient and lightweight architecture with performance equivalent
to state-of-the-art models in terms of Expression Recognition Rate (exprate).
It also performs better in terms of top 1, 2, and 3 error accuracy across the
CROHME 2014, 2016, and 2019 datasets. DenseBAM-GI achieves the best exprate
among all models on the CROHME 2019 dataset. Importantly, these successes are
accomplished with a drop in the complexity of the calculation and a reduction
in the need for GPU memory.
| [
{
"version": "v1",
"created": "Wed, 28 Jun 2023 18:12:23 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Pal",
"Aniket",
""
],
[
"Singh",
"Krishna Pratap",
""
]
] | TITLE: DenseBAM-GI: Attention Augmented DeneseNet with momentum aided GRU for
HMER
ABSTRACT: The task of recognising Handwritten Mathematical Expressions (HMER) is
crucial in the fields of digital education and scholarly research. However, it
is difficult to accurately determine the length and complex spatial
relationships among symbols in handwritten mathematical expressions. In this
study, we present a novel encoder-decoder architecture (DenseBAM-GI) for HMER,
where the encoder has a Bottleneck Attention Module (BAM) to improve feature
representation and the decoder has a Gated Input-GRU (GI-GRU) unit with an
extra gate to make decoding long and complex expressions easier. The proposed
model is an efficient and lightweight architecture with performance equivalent
to state-of-the-art models in terms of Expression Recognition Rate (exprate).
It also performs better in terms of top 1, 2, and 3 error accuracy across the
CROHME 2014, 2016, and 2019 datasets. DenseBAM-GI achieves the best exprate
among all models on the CROHME 2019 dataset. Importantly, these successes are
accomplished with a drop in the complexity of the calculation and a reduction
in the need for GPU memory.
| no_new_dataset | 0.710622 |
2306.16495 | Quanzhi Li | Quanzhi Li, Yang Chao, Dong Li, Yao Lu, Chi Zhang | Event Detection from Social Media Stream: Methods, Datasets and
Opportunities | 8 pages | null | null | null | cs.SI cs.AI cs.IR | http://creativecommons.org/licenses/by/4.0/ | Social media streams contain large and diverse amount of information, ranging
from daily-life stories to the latest global and local events and news.
Twitter, especially, allows a fast spread of events happening real time, and
enables individuals and organizations to stay informed of the events happening
now. Event detection from social media data poses different challenges from
traditional text and is a research area that has attracted much attention in
recent years. In this paper, we survey a wide range of event detection methods
for Twitter data stream, helping readers understand the recent development in
this area. We present the datasets available to the public. Furthermore, a few
research opportunities
| [
{
"version": "v1",
"created": "Wed, 28 Jun 2023 18:40:03 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Li",
"Quanzhi",
""
],
[
"Chao",
"Yang",
""
],
[
"Li",
"Dong",
""
],
[
"Lu",
"Yao",
""
],
[
"Zhang",
"Chi",
""
]
] | TITLE: Event Detection from Social Media Stream: Methods, Datasets and
Opportunities
ABSTRACT: Social media streams contain large and diverse amount of information, ranging
from daily-life stories to the latest global and local events and news.
Twitter, especially, allows a fast spread of events happening real time, and
enables individuals and organizations to stay informed of the events happening
now. Event detection from social media data poses different challenges from
traditional text and is a research area that has attracted much attention in
recent years. In this paper, we survey a wide range of event detection methods
for Twitter data stream, helping readers understand the recent development in
this area. We present the datasets available to the public. Furthermore, a few
research opportunities
| no_new_dataset | 0.709177 |
2306.16503 | Sukriti Verma | Sukriti Verma, Ayush Chopra, Jayakumar Subramanian, Mausoom Sarkar,
Nikaash Puri, Piyush Gupta, Balaji Krishnamurthy | SARC: Soft Actor Retrospective Critic | Accepted at RLDM 2022 | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by/4.0/ | The two-time scale nature of SAC, which is an actor-critic algorithm, is
characterised by the fact that the critic estimate has not converged for the
actor at any given time, but since the critic learns faster than the actor, it
ensures eventual consistency between the two. Various strategies have been
introduced in literature to learn better gradient estimates to help achieve
better convergence. Since gradient estimates depend upon the critic, we posit
that improving the critic can provide a better gradient estimate for the actor
at each time. Utilizing this, we propose Soft Actor Retrospective Critic
(SARC), where we augment the SAC critic loss with another loss term -
retrospective loss - leading to faster critic convergence and consequently,
better policy gradient estimates for the actor. An existing implementation of
SAC can be easily adapted to SARC with minimal modifications. Through extensive
experimentation and analysis, we show that SARC provides consistent improvement
over SAC on benchmark environments. We plan to open-source the code and all
experiment data at: https://github.com/sukritiverma1996/SARC.
| [
{
"version": "v1",
"created": "Wed, 28 Jun 2023 18:50:18 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Verma",
"Sukriti",
""
],
[
"Chopra",
"Ayush",
""
],
[
"Subramanian",
"Jayakumar",
""
],
[
"Sarkar",
"Mausoom",
""
],
[
"Puri",
"Nikaash",
""
],
[
"Gupta",
"Piyush",
""
],
[
"Krishnamurthy",
"Balaji",
""
]
] | TITLE: SARC: Soft Actor Retrospective Critic
ABSTRACT: The two-time scale nature of SAC, which is an actor-critic algorithm, is
characterised by the fact that the critic estimate has not converged for the
actor at any given time, but since the critic learns faster than the actor, it
ensures eventual consistency between the two. Various strategies have been
introduced in literature to learn better gradient estimates to help achieve
better convergence. Since gradient estimates depend upon the critic, we posit
that improving the critic can provide a better gradient estimate for the actor
at each time. Utilizing this, we propose Soft Actor Retrospective Critic
(SARC), where we augment the SAC critic loss with another loss term -
retrospective loss - leading to faster critic convergence and consequently,
better policy gradient estimates for the actor. An existing implementation of
SAC can be easily adapted to SARC with minimal modifications. Through extensive
experimentation and analysis, we show that SARC provides consistent improvement
over SAC on benchmark environments. We plan to open-source the code and all
experiment data at: https://github.com/sukritiverma1996/SARC.
| no_new_dataset | 0.710019 |
2306.16539 | Hanxue Gu | Hanxue Gu, Haoyu Dong, Nicholas Konz, Maciej A. Mazurowski | A systematic study of the foreground-background imbalance problem in
deep learning for object detection | null | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by/4.0/ | The class imbalance problem in deep learning has been explored in several
studies, but there has yet to be a systematic analysis of this phenomenon in
object detection. Here, we present comprehensive analyses and experiments of
the foreground-background (F-B) imbalance problem in object detection, which is
very common and caused by small, infrequent objects of interest. We
experimentally study the effects of different aspects of F-B imbalance (object
size, number of objects, dataset size, object type) on detection performance.
In addition, we also compare 9 leading methods for addressing this problem,
including Faster-RCNN, SSD, OHEM, Libra-RCNN, Focal-Loss, GHM, PISA, YOLO-v3,
and GFL with a range of datasets from different imaging domains. We conclude
that (1) the F-B imbalance can indeed cause a significant drop in detection
performance, (2) The detection performance is more affected by F-B imbalance
when fewer training data are available, (3) in most cases, decreasing object
size leads to larger performance drop than decreasing number of objects, given
the same change in the ratio of object pixels to non-object pixels, (6) among
all selected methods, Libra-RCNN and PISA demonstrate the best performance in
addressing the issue of F-B imbalance. (7) When the training dataset size is
large, the choice of method is not impactful (8) Soft-sampling methods,
including focal-loss, GHM, and GFL, perform fairly well on average but are
relatively unstable.
| [
{
"version": "v1",
"created": "Wed, 28 Jun 2023 20:27:11 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Gu",
"Hanxue",
""
],
[
"Dong",
"Haoyu",
""
],
[
"Konz",
"Nicholas",
""
],
[
"Mazurowski",
"Maciej A.",
""
]
] | TITLE: A systematic study of the foreground-background imbalance problem in
deep learning for object detection
ABSTRACT: The class imbalance problem in deep learning has been explored in several
studies, but there has yet to be a systematic analysis of this phenomenon in
object detection. Here, we present comprehensive analyses and experiments of
the foreground-background (F-B) imbalance problem in object detection, which is
very common and caused by small, infrequent objects of interest. We
experimentally study the effects of different aspects of F-B imbalance (object
size, number of objects, dataset size, object type) on detection performance.
In addition, we also compare 9 leading methods for addressing this problem,
including Faster-RCNN, SSD, OHEM, Libra-RCNN, Focal-Loss, GHM, PISA, YOLO-v3,
and GFL with a range of datasets from different imaging domains. We conclude
that (1) the F-B imbalance can indeed cause a significant drop in detection
performance, (2) The detection performance is more affected by F-B imbalance
when fewer training data are available, (3) in most cases, decreasing object
size leads to larger performance drop than decreasing number of objects, given
the same change in the ratio of object pixels to non-object pixels, (6) among
all selected methods, Libra-RCNN and PISA demonstrate the best performance in
addressing the issue of F-B imbalance. (7) When the training dataset size is
large, the choice of method is not impactful (8) Soft-sampling methods,
including focal-loss, GHM, and GFL, perform fairly well on average but are
relatively unstable.
| no_new_dataset | 0.708988 |
2306.16551 | Jinhee Yu | Jinhee Yu, Jingdao Chen, Lalitha Dabbiru, Christopher T. Goodin | Analysis of LiDAR Configurations on Off-road Semantic Segmentation
Performance | null | null | 10.1117/12.2663098 | null | cs.CV cs.RO eess.IV | http://creativecommons.org/licenses/by/4.0/ | This paper investigates the impact of LiDAR configuration shifts on the
performance of 3D LiDAR point cloud semantic segmentation models, a topic not
extensively studied before. We explore the effect of using different LiDAR
channels when training and testing a 3D LiDAR point cloud semantic segmentation
model, utilizing Cylinder3D for the experiments. A Cylinder3D model is trained
and tested on simulated 3D LiDAR point cloud datasets created using the
Mississippi State University Autonomous Vehicle Simulator (MAVS) and 32, 64
channel 3D LiDAR point clouds of the RELLIS-3D dataset collected in a
real-world off-road environment. Our experimental results demonstrate that
sensor and spatial domain shifts significantly impact the performance of
LiDAR-based semantic segmentation models. In the absence of spatial domain
changes between training and testing, models trained and tested on the same
sensor type generally exhibited better performance. Moreover, higher-resolution
sensors showed improved performance compared to those with lower-resolution
ones. However, results varied when spatial domain changes were present. In some
cases, the advantage of a sensor's higher resolution led to better performance
both with and without sensor domain shifts. In other instances, the higher
resolution resulted in overfitting within a specific domain, causing a lack of
generalization capability and decreased performance when tested on data with
different sensor configurations.
| [
{
"version": "v1",
"created": "Wed, 28 Jun 2023 20:41:45 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Yu",
"Jinhee",
""
],
[
"Chen",
"Jingdao",
""
],
[
"Dabbiru",
"Lalitha",
""
],
[
"Goodin",
"Christopher T.",
""
]
] | TITLE: Analysis of LiDAR Configurations on Off-road Semantic Segmentation
Performance
ABSTRACT: This paper investigates the impact of LiDAR configuration shifts on the
performance of 3D LiDAR point cloud semantic segmentation models, a topic not
extensively studied before. We explore the effect of using different LiDAR
channels when training and testing a 3D LiDAR point cloud semantic segmentation
model, utilizing Cylinder3D for the experiments. A Cylinder3D model is trained
and tested on simulated 3D LiDAR point cloud datasets created using the
Mississippi State University Autonomous Vehicle Simulator (MAVS) and 32, 64
channel 3D LiDAR point clouds of the RELLIS-3D dataset collected in a
real-world off-road environment. Our experimental results demonstrate that
sensor and spatial domain shifts significantly impact the performance of
LiDAR-based semantic segmentation models. In the absence of spatial domain
changes between training and testing, models trained and tested on the same
sensor type generally exhibited better performance. Moreover, higher-resolution
sensors showed improved performance compared to those with lower-resolution
ones. However, results varied when spatial domain changes were present. In some
cases, the advantage of a sensor's higher resolution led to better performance
both with and without sensor domain shifts. In other instances, the higher
resolution resulted in overfitting within a specific domain, causing a lack of
generalization capability and decreased performance when tested on data with
different sensor configurations.
| new_dataset | 0.712001 |
2306.16552 | Meiyu Zhong | Meiyu Zhong, Ravi Tandon | Learning Fair Classifiers via Min-Max F-divergence Regularization | null | null | null | null | cs.LG cs.AI cs.CY cs.IT math.IT | http://creativecommons.org/licenses/by/4.0/ | As machine learning (ML) based systems are adopted in domains such as law
enforcement, criminal justice, finance, hiring and admissions, ensuring the
fairness of ML aided decision-making is becoming increasingly important. In
this paper, we focus on the problem of fair classification, and introduce a
novel min-max F-divergence regularization framework for learning fair
classification models while preserving high accuracy. Our framework consists of
two trainable networks, namely, a classifier network and a bias/fairness
estimator network, where the fairness is measured using the statistical notion
of F-divergence. We show that F-divergence measures possess convexity and
differentiability properties, and their variational representation make them
widely applicable in practical gradient based training methods. The proposed
framework can be readily adapted to multiple sensitive attributes and for high
dimensional datasets. We study the F-divergence based training paradigm for two
types of group fairness constraints, namely, demographic parity and equalized
odds. We present a comprehensive set of experiments for several real-world data
sets arising in multiple domains (including COMPAS, Law Admissions, Adult
Income, and CelebA datasets). To quantify the fairness-accuracy tradeoff, we
introduce the notion of fairness-accuracy receiver operating characteristic
(FA-ROC) and a corresponding \textit{low-bias} FA-ROC, which we argue is an
appropriate measure to evaluate different classifiers. In comparison to several
existing approaches for learning fair classifiers (including pre-processing,
post-processing and other regularization methods), we show that the proposed
F-divergence based framework achieves state-of-the-art performance with respect
to the trade-off between accuracy and fairness.
| [
{
"version": "v1",
"created": "Wed, 28 Jun 2023 20:42:04 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Zhong",
"Meiyu",
""
],
[
"Tandon",
"Ravi",
""
]
] | TITLE: Learning Fair Classifiers via Min-Max F-divergence Regularization
ABSTRACT: As machine learning (ML) based systems are adopted in domains such as law
enforcement, criminal justice, finance, hiring and admissions, ensuring the
fairness of ML aided decision-making is becoming increasingly important. In
this paper, we focus on the problem of fair classification, and introduce a
novel min-max F-divergence regularization framework for learning fair
classification models while preserving high accuracy. Our framework consists of
two trainable networks, namely, a classifier network and a bias/fairness
estimator network, where the fairness is measured using the statistical notion
of F-divergence. We show that F-divergence measures possess convexity and
differentiability properties, and their variational representation make them
widely applicable in practical gradient based training methods. The proposed
framework can be readily adapted to multiple sensitive attributes and for high
dimensional datasets. We study the F-divergence based training paradigm for two
types of group fairness constraints, namely, demographic parity and equalized
odds. We present a comprehensive set of experiments for several real-world data
sets arising in multiple domains (including COMPAS, Law Admissions, Adult
Income, and CelebA datasets). To quantify the fairness-accuracy tradeoff, we
introduce the notion of fairness-accuracy receiver operating characteristic
(FA-ROC) and a corresponding \textit{low-bias} FA-ROC, which we argue is an
appropriate measure to evaluate different classifiers. In comparison to several
existing approaches for learning fair classifiers (including pre-processing,
post-processing and other regularization methods), we show that the proposed
F-divergence based framework achieves state-of-the-art performance with respect
to the trade-off between accuracy and fairness.
| no_new_dataset | 0.709856 |
2306.16606 | Cristhian David Forigua Diaz | Cristhian Forigua, Maria Escobar, Jordi Pont-Tuset, Kevis-Kokitsi
Maninis and Pablo Arbel\'aez | EgoCOL: Egocentric Camera pose estimation for Open-world 3D object
Localization @Ego4D challenge 2023 | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present EgoCOL, an egocentric camera pose estimation method for open-world
3D object localization. Our method leverages sparse camera pose reconstructions
in a two-fold manner, video and scan independently, to estimate the camera pose
of egocentric frames in 3D renders with high recall and precision. We
extensively evaluate our method on the Visual Query (VQ) 3D object localization
Ego4D benchmark. EgoCOL can estimate 62% and 59% more camera poses than the
Ego4D baseline in the Ego4D Visual Queries 3D Localization challenge at CVPR
2023 in the val and test sets, respectively. Our code is publicly available at
https://github.com/BCV-Uniandes/EgoCOL
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 00:17:23 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Forigua",
"Cristhian",
""
],
[
"Escobar",
"Maria",
""
],
[
"Pont-Tuset",
"Jordi",
""
],
[
"Maninis",
"Kevis-Kokitsi",
""
],
[
"Arbeláez",
"Pablo",
""
]
] | TITLE: EgoCOL: Egocentric Camera pose estimation for Open-world 3D object
Localization @Ego4D challenge 2023
ABSTRACT: We present EgoCOL, an egocentric camera pose estimation method for open-world
3D object localization. Our method leverages sparse camera pose reconstructions
in a two-fold manner, video and scan independently, to estimate the camera pose
of egocentric frames in 3D renders with high recall and precision. We
extensively evaluate our method on the Visual Query (VQ) 3D object localization
Ego4D benchmark. EgoCOL can estimate 62% and 59% more camera poses than the
Ego4D baseline in the Ego4D Visual Queries 3D Localization challenge at CVPR
2023 in the val and test sets, respectively. Our code is publicly available at
https://github.com/BCV-Uniandes/EgoCOL
| no_new_dataset | 0.711656 |
2306.16612 | Minsoo Kang | Minsoo Kang, Suhyun Kim | GuidedMixup: An Efficient Mixup Strategy Guided by Saliency Maps | Published at AAAI2023 (Oral) | Proceedings of the AAAI Conference on Artificial Intelligence,
37(1), 2023, 1096-1104 | 10.1609/aaai.v37i1.25191 | null | cs.CV cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Data augmentation is now an essential part of the image training process, as
it effectively prevents overfitting and makes the model more robust against
noisy datasets. Recent mixing augmentation strategies have advanced to generate
the mixup mask that can enrich the saliency information, which is a supervisory
signal. However, these methods incur a significant computational burden to
optimize the mixup mask. From this motivation, we propose a novel
saliency-aware mixup method, GuidedMixup, which aims to retain the salient
regions in mixup images with low computational overhead. We develop an
efficient pairing algorithm that pursues to minimize the conflict of salient
regions of paired images and achieve rich saliency in mixup images. Moreover,
GuidedMixup controls the mixup ratio for each pixel to better preserve the
salient region by interpolating two paired images smoothly. The experiments on
several datasets demonstrate that GuidedMixup provides a good trade-off between
augmentation overhead and generalization performance on classification
datasets. In addition, our method shows good performance in experiments with
corrupted or reduced datasets.
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 00:55:51 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Kang",
"Minsoo",
""
],
[
"Kim",
"Suhyun",
""
]
] | TITLE: GuidedMixup: An Efficient Mixup Strategy Guided by Saliency Maps
ABSTRACT: Data augmentation is now an essential part of the image training process, as
it effectively prevents overfitting and makes the model more robust against
noisy datasets. Recent mixing augmentation strategies have advanced to generate
the mixup mask that can enrich the saliency information, which is a supervisory
signal. However, these methods incur a significant computational burden to
optimize the mixup mask. From this motivation, we propose a novel
saliency-aware mixup method, GuidedMixup, which aims to retain the salient
regions in mixup images with low computational overhead. We develop an
efficient pairing algorithm that pursues to minimize the conflict of salient
regions of paired images and achieve rich saliency in mixup images. Moreover,
GuidedMixup controls the mixup ratio for each pixel to better preserve the
salient region by interpolating two paired images smoothly. The experiments on
several datasets demonstrate that GuidedMixup provides a good trade-off between
augmentation overhead and generalization performance on classification
datasets. In addition, our method shows good performance in experiments with
corrupted or reduced datasets.
| no_new_dataset | 0.708975 |
2306.16615 | Ju Yong Chang | Sungho Chun, Sungbum Park, Ju Yong Chang | Representation learning of vertex heatmaps for 3D human mesh
reconstruction from multi-view images | ICIP 2023 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This study addresses the problem of 3D human mesh reconstruction from
multi-view images. Recently, approaches that directly estimate the skinned
multi-person linear model (SMPL)-based human mesh vertices based on volumetric
heatmap representation from input images have shown good performance. We show
that representation learning of vertex heatmaps using an autoencoder helps
improve the performance of such approaches. Vertex heatmap autoencoder (VHA)
learns the manifold of plausible human meshes in the form of latent codes using
AMASS, which is a large-scale motion capture dataset. Body code predictor (BCP)
utilizes the learned body prior from VHA for human mesh reconstruction from
multi-view images through latent code-based supervision and transfer of
pretrained weights. According to experiments on Human3.6M and LightStage
datasets, the proposed method outperforms previous methods and achieves
state-of-the-art human mesh reconstruction performance.
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 01:16:03 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Chun",
"Sungho",
""
],
[
"Park",
"Sungbum",
""
],
[
"Chang",
"Ju Yong",
""
]
] | TITLE: Representation learning of vertex heatmaps for 3D human mesh
reconstruction from multi-view images
ABSTRACT: This study addresses the problem of 3D human mesh reconstruction from
multi-view images. Recently, approaches that directly estimate the skinned
multi-person linear model (SMPL)-based human mesh vertices based on volumetric
heatmap representation from input images have shown good performance. We show
that representation learning of vertex heatmaps using an autoencoder helps
improve the performance of such approaches. Vertex heatmap autoencoder (VHA)
learns the manifold of plausible human meshes in the form of latent codes using
AMASS, which is a large-scale motion capture dataset. Body code predictor (BCP)
utilizes the learned body prior from VHA for human mesh reconstruction from
multi-view images through latent code-based supervision and transfer of
pretrained weights. According to experiments on Human3.6M and LightStage
datasets, the proposed method outperforms previous methods and achieves
state-of-the-art human mesh reconstruction performance.
| no_new_dataset | 0.709239 |
2306.16621 | Dexter Antonio | Dexter Antonio, Hannah O'Toole, Randy Carney, Ambarish Kulkarni, Ahmet
Palazoglu | Assessing the Performance of 1D-Convolution Neural Networks to Predict
Concentration of Mixture Components from Raman Spectra | 7 pages, 7 figures | null | null | null | eess.SP cs.LG q-bio.QM stat.AP | http://creativecommons.org/licenses/by/4.0/ | An emerging application of Raman spectroscopy is monitoring the state of
chemical reactors during biologic drug production. Raman shift intensities
scale linearly with the concentrations of chemical species and thus can be used
to analytically determine real-time concentrations using non-destructive light
irradiation in a label-free manner. Chemometric algorithms are used to
interpret Raman spectra produced from complex mixtures of bioreactor contents
as a reaction evolves. Finding the optimal algorithm for a specific bioreactor
environment is challenging due to the lack of freely available Raman mixture
datasets. The RaMix Python package addresses this challenge by enabling the
generation of synthetic Raman mixture datasets with controllable noise levels
to assess the utility of different chemometric algorithm types for real-time
monitoring applications. To demonstrate the capabilities of this package and
compare the performance of different chemometric algorithms, 48 datasets of
simulated spectra were generated using the RaMix Python package. The four
tested algorithms include partial least squares regression (PLS), a simple
neural network, a simple convolutional neural network (simple CNN), and a 1D
convolutional neural network with a ResNet architecture (ResNet). The
performance of the PLS and simple CNN model was found to be comparable, with
the PLS algorithm slightly outperforming the other models on 83\% of the data
sets. The simple CNN model outperforms the other models on large, high noise
datasets, demonstrating the superior capability of convolutional neural
networks compared to PLS in analyzing noisy spectra. These results demonstrate
the promise of CNNs to automatically extract concentration information from
unprocessed, noisy spectra, allowing for better process control of industrial
drug production. Code for this project is available at
github.com/DexterAntonio/RaMix.
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 01:41:07 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Antonio",
"Dexter",
""
],
[
"O'Toole",
"Hannah",
""
],
[
"Carney",
"Randy",
""
],
[
"Kulkarni",
"Ambarish",
""
],
[
"Palazoglu",
"Ahmet",
""
]
] | TITLE: Assessing the Performance of 1D-Convolution Neural Networks to Predict
Concentration of Mixture Components from Raman Spectra
ABSTRACT: An emerging application of Raman spectroscopy is monitoring the state of
chemical reactors during biologic drug production. Raman shift intensities
scale linearly with the concentrations of chemical species and thus can be used
to analytically determine real-time concentrations using non-destructive light
irradiation in a label-free manner. Chemometric algorithms are used to
interpret Raman spectra produced from complex mixtures of bioreactor contents
as a reaction evolves. Finding the optimal algorithm for a specific bioreactor
environment is challenging due to the lack of freely available Raman mixture
datasets. The RaMix Python package addresses this challenge by enabling the
generation of synthetic Raman mixture datasets with controllable noise levels
to assess the utility of different chemometric algorithm types for real-time
monitoring applications. To demonstrate the capabilities of this package and
compare the performance of different chemometric algorithms, 48 datasets of
simulated spectra were generated using the RaMix Python package. The four
tested algorithms include partial least squares regression (PLS), a simple
neural network, a simple convolutional neural network (simple CNN), and a 1D
convolutional neural network with a ResNet architecture (ResNet). The
performance of the PLS and simple CNN model was found to be comparable, with
the PLS algorithm slightly outperforming the other models on 83\% of the data
sets. The simple CNN model outperforms the other models on large, high noise
datasets, demonstrating the superior capability of convolutional neural
networks compared to PLS in analyzing noisy spectra. These results demonstrate
the promise of CNNs to automatically extract concentration information from
unprocessed, noisy spectra, allowing for better process control of industrial
drug production. Code for this project is available at
github.com/DexterAntonio/RaMix.
| no_new_dataset | 0.710038 |
2306.16627 | Leonardo Placidi | Leonardo Placidi, Ryuichiro Hataya, Toshio Mori, Koki Aoyama, Hayata
Morisaki, Kosuke Mitarai, Keisuke Fujii | MNISQ: A Large-Scale Quantum Circuit Dataset for Machine Learning on/for
Quantum Computers in the NISQ era | Preprint. Under review | null | null | null | quant-ph cs.LG | http://creativecommons.org/licenses/by-sa/4.0/ | We introduce the first large-scale dataset, MNISQ, for both the Quantum and
the Classical Machine Learning community during the Noisy Intermediate-Scale
Quantum era. MNISQ consists of 4,950,000 data points organized in 9
subdatasets. Building our dataset from the quantum encoding of classical
information (e.g., MNIST dataset), we deliver a dataset in a dual form: in
quantum form, as circuits, and in classical form, as quantum circuit
descriptions (quantum programming language, QASM). In fact, also the Machine
Learning research related to quantum computers undertakes a dual challenge:
enhancing machine learning exploiting the power of quantum computers, while
also leveraging state-of-the-art classical machine learning methodologies to
help the advancement of quantum computing. Therefore, we perform circuit
classification on our dataset, tackling the task with both quantum and
classical models. In the quantum endeavor, we test our circuit dataset with
Quantum Kernel methods, and we show excellent results up to $97\%$ accuracy. In
the classical world, the underlying quantum mechanical structures within the
quantum circuit data are not trivial. Nevertheless, we test our dataset on
three classical models: Structured State Space sequence model (S4), Transformer
and LSTM. In particular, the S4 model applied on the tokenized QASM sequences
reaches an impressive $77\%$ accuracy. These findings illustrate that quantum
circuit-related datasets are likely to be quantum advantageous, but also that
state-of-the-art machine learning methodologies can competently classify and
recognize quantum circuits. We finally entrust the quantum and classical
machine learning community the fundamental challenge to build more
quantum-classical datasets like ours and to build future benchmarks from our
experiments. The dataset is accessible on GitHub and its circuits are easily
run in qulacs or qiskit.
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 02:04:14 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Placidi",
"Leonardo",
""
],
[
"Hataya",
"Ryuichiro",
""
],
[
"Mori",
"Toshio",
""
],
[
"Aoyama",
"Koki",
""
],
[
"Morisaki",
"Hayata",
""
],
[
"Mitarai",
"Kosuke",
""
],
[
"Fujii",
"Keisuke",
""
]
] | TITLE: MNISQ: A Large-Scale Quantum Circuit Dataset for Machine Learning on/for
Quantum Computers in the NISQ era
ABSTRACT: We introduce the first large-scale dataset, MNISQ, for both the Quantum and
the Classical Machine Learning community during the Noisy Intermediate-Scale
Quantum era. MNISQ consists of 4,950,000 data points organized in 9
subdatasets. Building our dataset from the quantum encoding of classical
information (e.g., MNIST dataset), we deliver a dataset in a dual form: in
quantum form, as circuits, and in classical form, as quantum circuit
descriptions (quantum programming language, QASM). In fact, also the Machine
Learning research related to quantum computers undertakes a dual challenge:
enhancing machine learning exploiting the power of quantum computers, while
also leveraging state-of-the-art classical machine learning methodologies to
help the advancement of quantum computing. Therefore, we perform circuit
classification on our dataset, tackling the task with both quantum and
classical models. In the quantum endeavor, we test our circuit dataset with
Quantum Kernel methods, and we show excellent results up to $97\%$ accuracy. In
the classical world, the underlying quantum mechanical structures within the
quantum circuit data are not trivial. Nevertheless, we test our dataset on
three classical models: Structured State Space sequence model (S4), Transformer
and LSTM. In particular, the S4 model applied on the tokenized QASM sequences
reaches an impressive $77\%$ accuracy. These findings illustrate that quantum
circuit-related datasets are likely to be quantum advantageous, but also that
state-of-the-art machine learning methodologies can competently classify and
recognize quantum circuits. We finally entrust the quantum and classical
machine learning community the fundamental challenge to build more
quantum-classical datasets like ours and to build future benchmarks from our
experiments. The dataset is accessible on GitHub and its circuits are easily
run in qulacs or qiskit.
| no_new_dataset | 0.581125 |
2306.16630 | Youyang Qu | Youyang Qu, Lichuan Ma, Wenjie Ye, Xuemeng Zhai, Shui Yu, Yunfeng Li,
and David Smith | Towards Blockchain-Assisted Privacy-Aware Data Sharing For Edge
Intelligence: A Smart Healthcare Perspective | null | null | null | null | cs.CR cs.AI | http://creativecommons.org/licenses/by/4.0/ | The popularization of intelligent healthcare devices and big data analytics
significantly boosts the development of smart healthcare networks (SHNs). To
enhance the precision of diagnosis, different participants in SHNs share health
data that contains sensitive information. Therefore, the data exchange process
raises privacy concerns, especially when the integration of health data from
multiple sources (linkage attack) results in further leakage. Linkage attack is
a type of dominant attack in the privacy domain, which can leverage various
data sources for private data mining. Furthermore, adversaries launch poisoning
attacks to falsify the health data, which leads to misdiagnosing or even
physical damage. To protect private health data, we propose a personalized
differential privacy model based on the trust levels among users. The trust is
evaluated by a defined community density, while the corresponding privacy
protection level is mapped to controllable randomized noise constrained by
differential privacy. To avoid linkage attacks in personalized differential
privacy, we designed a noise correlation decoupling mechanism using a Markov
stochastic process. In addition, we build the community model on a blockchain,
which can mitigate the risk of poisoning attacks during differentially private
data transmission over SHNs. To testify the effectiveness and superiority of
the proposed approach, we conduct extensive experiments on benchmark datasets.
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 02:06:04 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Qu",
"Youyang",
""
],
[
"Ma",
"Lichuan",
""
],
[
"Ye",
"Wenjie",
""
],
[
"Zhai",
"Xuemeng",
""
],
[
"Yu",
"Shui",
""
],
[
"Li",
"Yunfeng",
""
],
[
"Smith",
"David",
""
]
] | TITLE: Towards Blockchain-Assisted Privacy-Aware Data Sharing For Edge
Intelligence: A Smart Healthcare Perspective
ABSTRACT: The popularization of intelligent healthcare devices and big data analytics
significantly boosts the development of smart healthcare networks (SHNs). To
enhance the precision of diagnosis, different participants in SHNs share health
data that contains sensitive information. Therefore, the data exchange process
raises privacy concerns, especially when the integration of health data from
multiple sources (linkage attack) results in further leakage. Linkage attack is
a type of dominant attack in the privacy domain, which can leverage various
data sources for private data mining. Furthermore, adversaries launch poisoning
attacks to falsify the health data, which leads to misdiagnosing or even
physical damage. To protect private health data, we propose a personalized
differential privacy model based on the trust levels among users. The trust is
evaluated by a defined community density, while the corresponding privacy
protection level is mapped to controllable randomized noise constrained by
differential privacy. To avoid linkage attacks in personalized differential
privacy, we designed a noise correlation decoupling mechanism using a Markov
stochastic process. In addition, we build the community model on a blockchain,
which can mitigate the risk of poisoning attacks during differentially private
data transmission over SHNs. To testify the effectiveness and superiority of
the proposed approach, we conduct extensive experiments on benchmark datasets.
| no_new_dataset | 0.712426 |
2306.16636 | Tianwen Wei | Tianwen Wei, Jian Luan, Wei Liu, Shuang Dong, Bin Wang | CMATH: Can Your Language Model Pass Chinese Elementary School Math Test? | null | null | null | null | cs.CL cs.AI cs.LG | http://creativecommons.org/licenses/by/4.0/ | We present the Chinese Elementary School Math Word Problems (CMATH) dataset,
comprising 1.7k elementary school-level math word problems with detailed
annotations, source from actual Chinese workbooks and exams. This dataset aims
to provide a benchmark tool for assessing the following question: to what grade
level of elementary school math do the abilities of popular large language
models (LLMs) correspond? We evaluate a variety of popular LLMs, including both
commercial and open-source options, and discover that only GPT-4 achieves
success (accuracy $\geq$ 60\%) across all six elementary school grades, while
other models falter at different grade levels. Furthermore, we assess the
robustness of several top-performing LLMs by augmenting the original problems
in the CMATH dataset with distracting information. Our findings reveal that
GPT-4 is able to maintains robustness, while other model fail. We anticipate
that our study will expose limitations in LLMs' arithmetic and reasoning
capabilities, and promote their ongoing development and advancement.
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 02:19:50 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Wei",
"Tianwen",
""
],
[
"Luan",
"Jian",
""
],
[
"Liu",
"Wei",
""
],
[
"Dong",
"Shuang",
""
],
[
"Wang",
"Bin",
""
]
] | TITLE: CMATH: Can Your Language Model Pass Chinese Elementary School Math Test?
ABSTRACT: We present the Chinese Elementary School Math Word Problems (CMATH) dataset,
comprising 1.7k elementary school-level math word problems with detailed
annotations, source from actual Chinese workbooks and exams. This dataset aims
to provide a benchmark tool for assessing the following question: to what grade
level of elementary school math do the abilities of popular large language
models (LLMs) correspond? We evaluate a variety of popular LLMs, including both
commercial and open-source options, and discover that only GPT-4 achieves
success (accuracy $\geq$ 60\%) across all six elementary school grades, while
other models falter at different grade levels. Furthermore, we assess the
robustness of several top-performing LLMs by augmenting the original problems
in the CMATH dataset with distracting information. Our findings reveal that
GPT-4 is able to maintains robustness, while other model fail. We anticipate
that our study will expose limitations in LLMs' arithmetic and reasoning
capabilities, and promote their ongoing development and advancement.
| new_dataset | 0.71584 |
2306.16638 | Ha Thanh Nguyen | Ha Thanh Nguyen, Randy Goebel, Francesca Toni, Kostas Stathis, Ken
Satoh | A negation detection assessment of GPTs: analysis with the xNot360
dataset | null | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Negation is a fundamental aspect of natural language, playing a critical role
in communication and comprehension. Our study assesses the negation detection
performance of Generative Pre-trained Transformer (GPT) models, specifically
GPT-2, GPT-3, GPT-3.5, and GPT-4. We focus on the identification of negation in
natural language using a zero-shot prediction approach applied to our custom
xNot360 dataset. Our approach examines sentence pairs labeled to indicate
whether the second sentence negates the first. Our findings expose a
considerable performance disparity among the GPT models, with GPT-4 surpassing
its counterparts and GPT-3.5 displaying a marked performance reduction. The
overall proficiency of the GPT models in negation detection remains relatively
modest, indicating that this task pushes the boundaries of their natural
language understanding capabilities. We not only highlight the constraints of
GPT models in handling negation but also emphasize the importance of logical
reliability in high-stakes domains such as healthcare, science, and law.
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 02:27:48 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Nguyen",
"Ha Thanh",
""
],
[
"Goebel",
"Randy",
""
],
[
"Toni",
"Francesca",
""
],
[
"Stathis",
"Kostas",
""
],
[
"Satoh",
"Ken",
""
]
] | TITLE: A negation detection assessment of GPTs: analysis with the xNot360
dataset
ABSTRACT: Negation is a fundamental aspect of natural language, playing a critical role
in communication and comprehension. Our study assesses the negation detection
performance of Generative Pre-trained Transformer (GPT) models, specifically
GPT-2, GPT-3, GPT-3.5, and GPT-4. We focus on the identification of negation in
natural language using a zero-shot prediction approach applied to our custom
xNot360 dataset. Our approach examines sentence pairs labeled to indicate
whether the second sentence negates the first. Our findings expose a
considerable performance disparity among the GPT models, with GPT-4 surpassing
its counterparts and GPT-3.5 displaying a marked performance reduction. The
overall proficiency of the GPT models in negation detection remains relatively
modest, indicating that this task pushes the boundaries of their natural
language understanding capabilities. We not only highlight the constraints of
GPT models in handling negation but also emphasize the importance of logical
reliability in high-stakes domains such as healthcare, science, and law.
| new_dataset | 0.715617 |
2306.16645 | Jinhong Ni | Jinhong Ni, Yalong Bai, Wei Zhang, Ting Yao, Tao Mei | Deep Equilibrium Multimodal Fusion | null | null | null | null | cs.CV cs.MM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Multimodal fusion integrates the complementary information present in
multiple modalities and has gained much attention recently. Most existing
fusion approaches either learn a fixed fusion strategy during training and
inference, or are only capable of fusing the information to a certain extent.
Such solutions may fail to fully capture the dynamics of interactions across
modalities especially when there are complex intra- and inter-modality
correlations to be considered for informative multimodal fusion. In this paper,
we propose a novel deep equilibrium (DEQ) method towards multimodal fusion via
seeking a fixed point of the dynamic multimodal fusion process and modeling the
feature correlations in an adaptive and recursive manner. This new way encodes
the rich information within and across modalities thoroughly from low level to
high level for efficacious downstream multimodal learning and is readily
pluggable to various multimodal frameworks. Extensive experiments on BRCA,
MM-IMDB, CMU-MOSI, SUN RGB-D, and VQA-v2 demonstrate the superiority of our DEQ
fusion. More remarkably, DEQ fusion consistently achieves state-of-the-art
performance on multiple multimodal benchmarks. The code will be released.
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 03:02:20 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Ni",
"Jinhong",
""
],
[
"Bai",
"Yalong",
""
],
[
"Zhang",
"Wei",
""
],
[
"Yao",
"Ting",
""
],
[
"Mei",
"Tao",
""
]
] | TITLE: Deep Equilibrium Multimodal Fusion
ABSTRACT: Multimodal fusion integrates the complementary information present in
multiple modalities and has gained much attention recently. Most existing
fusion approaches either learn a fixed fusion strategy during training and
inference, or are only capable of fusing the information to a certain extent.
Such solutions may fail to fully capture the dynamics of interactions across
modalities especially when there are complex intra- and inter-modality
correlations to be considered for informative multimodal fusion. In this paper,
we propose a novel deep equilibrium (DEQ) method towards multimodal fusion via
seeking a fixed point of the dynamic multimodal fusion process and modeling the
feature correlations in an adaptive and recursive manner. This new way encodes
the rich information within and across modalities thoroughly from low level to
high level for efficacious downstream multimodal learning and is readily
pluggable to various multimodal frameworks. Extensive experiments on BRCA,
MM-IMDB, CMU-MOSI, SUN RGB-D, and VQA-v2 demonstrate the superiority of our DEQ
fusion. More remarkably, DEQ fusion consistently achieves state-of-the-art
performance on multiple multimodal benchmarks. The code will be released.
| no_new_dataset | 0.708219 |
2306.16650 | Liqiang Jing | Liqiang Jing, Xuemeng Song, Kun Ouyang, Mengzhao Jia, Liqiang Nie | Multi-source Semantic Graph-based Multimodal Sarcasm Explanation
Generation | Accepted by ACL 2023 main conference | ACL 2023 | null | null | cs.CL cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Multimodal Sarcasm Explanation (MuSE) is a new yet challenging task, which
aims to generate a natural language sentence for a multimodal social post (an
image as well as its caption) to explain why it contains sarcasm. Although the
existing pioneer study has achieved great success with the BART backbone, it
overlooks the gap between the visual feature space and the decoder semantic
space, the object-level metadata of the image, as well as the potential
external knowledge. To solve these limitations, in this work, we propose a
novel mulTi-source sEmantic grAph-based Multimodal sarcasm explanation scheme,
named TEAM. In particular, TEAM extracts the object-level semantic meta-data
instead of the traditional global visual features from the input image.
Meanwhile, TEAM resorts to ConceptNet to obtain the external related knowledge
concepts for the input text and the extracted object meta-data. Thereafter,
TEAM introduces a multi-source semantic graph that comprehensively characterize
the multi-source (i.e., caption, object meta-data, external knowledge) semantic
relations to facilitate the sarcasm reasoning. Extensive experiments on a
public released dataset MORE verify the superiority of our model over
cutting-edge methods.
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 03:26:10 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Jing",
"Liqiang",
""
],
[
"Song",
"Xuemeng",
""
],
[
"Ouyang",
"Kun",
""
],
[
"Jia",
"Mengzhao",
""
],
[
"Nie",
"Liqiang",
""
]
] | TITLE: Multi-source Semantic Graph-based Multimodal Sarcasm Explanation
Generation
ABSTRACT: Multimodal Sarcasm Explanation (MuSE) is a new yet challenging task, which
aims to generate a natural language sentence for a multimodal social post (an
image as well as its caption) to explain why it contains sarcasm. Although the
existing pioneer study has achieved great success with the BART backbone, it
overlooks the gap between the visual feature space and the decoder semantic
space, the object-level metadata of the image, as well as the potential
external knowledge. To solve these limitations, in this work, we propose a
novel mulTi-source sEmantic grAph-based Multimodal sarcasm explanation scheme,
named TEAM. In particular, TEAM extracts the object-level semantic meta-data
instead of the traditional global visual features from the input image.
Meanwhile, TEAM resorts to ConceptNet to obtain the external related knowledge
concepts for the input text and the extracted object meta-data. Thereafter,
TEAM introduces a multi-source semantic graph that comprehensively characterize
the multi-source (i.e., caption, object meta-data, external knowledge) semantic
relations to facilitate the sarcasm reasoning. Extensive experiments on a
public released dataset MORE verify the superiority of our model over
cutting-edge methods.
| no_new_dataset | 0.70576 |
2306.16665 | Jing Mai | Jing Mai, Jiarui Wang, Zhixiong Di, Guojie Luo, Yun Liang and Yibo Lin | OpenPARF: An Open-Source Placement and Routing Framework for Large-Scale
Heterogeneous FPGAs with Deep Learning Toolkit | null | null | null | null | cs.AR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper proposes OpenPARF, an open-source placement and routing framework
for large-scale FPGA designs. OpenPARF is implemented with the deep learning
toolkit PyTorch and supports massive parallelization on GPU. The framework
proposes a novel asymmetric multi-electrostatic field system to solve FPGA
placement. It considers fine-grained routing resources inside configurable
logic blocks (CLBs) for FPGA routing and supports large-scale irregular routing
resource graphs. Experimental results on ISPD 2016 and ISPD 2017 FPGA contest
benchmarks and industrial benchmarks demonstrate that OpenPARF can achieve
0.4-12.7% improvement in routed wirelength and more than $2\times$ speedup in
placement. We believe that OpenPARF can pave the road for developing FPGA
physical design engines and stimulate further research on related topics.
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 03:53:52 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Mai",
"Jing",
""
],
[
"Wang",
"Jiarui",
""
],
[
"Di",
"Zhixiong",
""
],
[
"Luo",
"Guojie",
""
],
[
"Liang",
"Yun",
""
],
[
"Lin",
"Yibo",
""
]
] | TITLE: OpenPARF: An Open-Source Placement and Routing Framework for Large-Scale
Heterogeneous FPGAs with Deep Learning Toolkit
ABSTRACT: This paper proposes OpenPARF, an open-source placement and routing framework
for large-scale FPGA designs. OpenPARF is implemented with the deep learning
toolkit PyTorch and supports massive parallelization on GPU. The framework
proposes a novel asymmetric multi-electrostatic field system to solve FPGA
placement. It considers fine-grained routing resources inside configurable
logic blocks (CLBs) for FPGA routing and supports large-scale irregular routing
resource graphs. Experimental results on ISPD 2016 and ISPD 2017 FPGA contest
benchmarks and industrial benchmarks demonstrate that OpenPARF can achieve
0.4-12.7% improvement in routed wirelength and more than $2\times$ speedup in
placement. We believe that OpenPARF can pave the road for developing FPGA
physical design engines and stimulate further research on related topics.
| no_new_dataset | 0.709069 |
2306.16682 | Antonino Furnari | Antonino Furnari, Giovanni Maria Farinella | Streaming egocentric action anticipation: An evaluation scheme and
approach | Published in Computer Vision and Image Understanding, 2023. arXiv
admin note: text overlap with arXiv:2110.05386 | null | 10.1016/j.cviu.2023.103763 | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Egocentric action anticipation aims to predict the future actions the camera
wearer will perform from the observation of the past. While predictions about
the future should be available before the predicted events take place, most
approaches do not pay attention to the computational time required to make such
predictions. As a result, current evaluation schemes assume that predictions
are available right after the input video is observed, i.e., presuming a
negligible runtime, which may lead to overly optimistic evaluations. We propose
a streaming egocentric action evaluation scheme which assumes that predictions
are performed online and made available only after the model has processed the
current input segment, which depends on its runtime. To evaluate all models
considering the same prediction horizon, we hence propose that slower models
should base their predictions on temporal segments sampled ahead of time. Based
on the observation that model runtime can affect performance in the considered
streaming evaluation scenario, we further propose a lightweight action
anticipation model based on feed-forward 3D CNNs which is optimized using
knowledge distillation techniques with a novel past-to-future distillation
loss. Experiments on the three popular datasets EPIC-KITCHENS-55,
EPIC-KITCHENS-100 and EGTEA Gaze+ show that (i) the proposed evaluation scheme
induces a different ranking on state-of-the-art methods as compared to classic
evaluations, (ii) lightweight approaches tend to outmatch more computationally
expensive ones, and (iii) the proposed model based on feed-forward 3D CNNs and
knowledge distillation outperforms current art in the streaming egocentric
action anticipation scenario.
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 04:53:29 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Furnari",
"Antonino",
""
],
[
"Farinella",
"Giovanni Maria",
""
]
] | TITLE: Streaming egocentric action anticipation: An evaluation scheme and
approach
ABSTRACT: Egocentric action anticipation aims to predict the future actions the camera
wearer will perform from the observation of the past. While predictions about
the future should be available before the predicted events take place, most
approaches do not pay attention to the computational time required to make such
predictions. As a result, current evaluation schemes assume that predictions
are available right after the input video is observed, i.e., presuming a
negligible runtime, which may lead to overly optimistic evaluations. We propose
a streaming egocentric action evaluation scheme which assumes that predictions
are performed online and made available only after the model has processed the
current input segment, which depends on its runtime. To evaluate all models
considering the same prediction horizon, we hence propose that slower models
should base their predictions on temporal segments sampled ahead of time. Based
on the observation that model runtime can affect performance in the considered
streaming evaluation scenario, we further propose a lightweight action
anticipation model based on feed-forward 3D CNNs which is optimized using
knowledge distillation techniques with a novel past-to-future distillation
loss. Experiments on the three popular datasets EPIC-KITCHENS-55,
EPIC-KITCHENS-100 and EGTEA Gaze+ show that (i) the proposed evaluation scheme
induces a different ranking on state-of-the-art methods as compared to classic
evaluations, (ii) lightweight approaches tend to outmatch more computationally
expensive ones, and (iii) the proposed model based on feed-forward 3D CNNs and
knowledge distillation outperforms current art in the streaming egocentric
action anticipation scenario.
| no_new_dataset | 0.708648 |
2306.16684 | Bradley Theilman | Bradley H. Theilman, Felix Wang, Fred Rothganger, James B. Aimone | Decomposing spiking neural networks with Graphical Neural Activity
Threads | null | null | null | SAND2023-05685O | cs.NE q-bio.NC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A satisfactory understanding of information processing in spiking neural
networks requires appropriate computational abstractions of neural activity.
Traditionally, the neural population state vector has been the most common
abstraction applied to spiking neural networks, but this requires artificially
partitioning time into bins that are not obviously relevant to the network
itself. We introduce a distinct set of techniques for analyzing spiking neural
networks that decomposes neural activity into multiple, disjoint, parallel
threads of activity. We construct these threads by estimating the degree of
causal relatedness between pairs of spikes, then use these estimates to
construct a directed acyclic graph that traces how the network activity evolves
through individual spikes. We find that this graph of spiking activity
naturally decomposes into disjoint connected components that overlap in space
and time, which we call Graphical Neural Activity Threads (GNATs). We provide
an efficient algorithm for finding analogous threads that reoccur in large
spiking datasets, revealing that seemingly distinct spike trains are composed
of similar underlying threads of activity, a hallmark of compositionality. The
picture of spiking neural networks provided by our GNAT analysis points to new
abstractions for spiking neural computation that are naturally adapted to the
spatiotemporally distributed dynamics of spiking neural networks.
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 05:10:11 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Theilman",
"Bradley H.",
""
],
[
"Wang",
"Felix",
""
],
[
"Rothganger",
"Fred",
""
],
[
"Aimone",
"James B.",
""
]
] | TITLE: Decomposing spiking neural networks with Graphical Neural Activity
Threads
ABSTRACT: A satisfactory understanding of information processing in spiking neural
networks requires appropriate computational abstractions of neural activity.
Traditionally, the neural population state vector has been the most common
abstraction applied to spiking neural networks, but this requires artificially
partitioning time into bins that are not obviously relevant to the network
itself. We introduce a distinct set of techniques for analyzing spiking neural
networks that decomposes neural activity into multiple, disjoint, parallel
threads of activity. We construct these threads by estimating the degree of
causal relatedness between pairs of spikes, then use these estimates to
construct a directed acyclic graph that traces how the network activity evolves
through individual spikes. We find that this graph of spiking activity
naturally decomposes into disjoint connected components that overlap in space
and time, which we call Graphical Neural Activity Threads (GNATs). We provide
an efficient algorithm for finding analogous threads that reoccur in large
spiking datasets, revealing that seemingly distinct spike trains are composed
of similar underlying threads of activity, a hallmark of compositionality. The
picture of spiking neural networks provided by our GNAT analysis points to new
abstractions for spiking neural computation that are naturally adapted to the
spatiotemporally distributed dynamics of spiking neural networks.
| no_new_dataset | 0.712407 |
2306.16697 | Mingli Zhu | Mingli Zhu, Shaokui Wei, Hongyuan Zha, Baoyuan Wu | Neural Polarizer: A Lightweight and Effective Backdoor Defense via
Purifying Poisoned Features | null | null | null | null | cs.AI cs.CR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recent studies have demonstrated the susceptibility of deep neural networks
to backdoor attacks. Given a backdoored model, its prediction of a poisoned
sample with trigger will be dominated by the trigger information, though
trigger information and benign information coexist. Inspired by the mechanism
of the optical polarizer that a polarizer could pass light waves with
particular polarizations while filtering light waves with other polarizations,
we propose a novel backdoor defense method by inserting a learnable neural
polarizer into the backdoored model as an intermediate layer, in order to
purify the poisoned sample via filtering trigger information while maintaining
benign information. The neural polarizer is instantiated as one lightweight
linear transformation layer, which is learned through solving a well designed
bi-level optimization problem, based on a limited clean dataset. Compared to
other fine-tuning-based defense methods which often adjust all parameters of
the backdoored model, the proposed method only needs to learn one additional
layer, such that it is more efficient and requires less clean data. Extensive
experiments demonstrate the effectiveness and efficiency of our method in
removing backdoors across various neural network architectures and datasets,
especially in the case of very limited clean data.
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 05:39:58 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Zhu",
"Mingli",
""
],
[
"Wei",
"Shaokui",
""
],
[
"Zha",
"Hongyuan",
""
],
[
"Wu",
"Baoyuan",
""
]
] | TITLE: Neural Polarizer: A Lightweight and Effective Backdoor Defense via
Purifying Poisoned Features
ABSTRACT: Recent studies have demonstrated the susceptibility of deep neural networks
to backdoor attacks. Given a backdoored model, its prediction of a poisoned
sample with trigger will be dominated by the trigger information, though
trigger information and benign information coexist. Inspired by the mechanism
of the optical polarizer that a polarizer could pass light waves with
particular polarizations while filtering light waves with other polarizations,
we propose a novel backdoor defense method by inserting a learnable neural
polarizer into the backdoored model as an intermediate layer, in order to
purify the poisoned sample via filtering trigger information while maintaining
benign information. The neural polarizer is instantiated as one lightweight
linear transformation layer, which is learned through solving a well designed
bi-level optimization problem, based on a limited clean dataset. Compared to
other fine-tuning-based defense methods which often adjust all parameters of
the backdoored model, the proposed method only needs to learn one additional
layer, such that it is more efficient and requires less clean data. Extensive
experiments demonstrate the effectiveness and efficiency of our method in
removing backdoors across various neural network architectures and datasets,
especially in the case of very limited clean data.
| no_new_dataset | 0.708635 |
2306.16701 | Subrata Das | Subrata Das, Swaroop Ghosh | TrojanNet: Detecting Trojans in Quantum Circuits using Machine Learning | 9 Pages, 6 Figures, 2 Tables, conference | null | null | null | quant-ph cs.CR | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Quantum computing holds tremendous potential for various applications, but
its security remains a crucial concern. Quantum circuits need high-quality
compilers to optimize the depth and gate count to boost the success probability
on current noisy quantum computers. There is a rise of efficient but
unreliable/untrusted compilers; however, they present a risk of tampering such
as Trojan insertion. We propose TrojanNet, a novel approach to enhance the
security of quantum circuits by detecting and classifying Trojan-inserted
circuits. In particular, we focus on the Quantum Approximate Optimization
Algorithm (QAOA) circuit that is popular in solving a wide range of
optimization problems. We investigate the impact of Trojan insertion on QAOA
circuits and develop a Convolutional Neural Network (CNN) model, referred to as
TrojanNet, to identify their presence accurately. Using the Qiskit framework,
we generate 12 diverse datasets by introducing variations in Trojan gate types,
the number of gates, insertion locations, and compiler backends. These datasets
consist of both original Trojan-free QAOA circuits and their corresponding
Trojan-inserted counterparts. The generated datasets are then utilized for
training and evaluating the TrojanNet model. Experimental results showcase an
average accuracy of 98.80% and an average F1-score of 98.53% in effectively
detecting and classifying Trojan-inserted QAOA circuits. Finally, we conduct a
performance comparison between TrojanNet and existing machine learning-based
Trojan detection methods specifically designed for conventional netlists.
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 05:56:05 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Das",
"Subrata",
""
],
[
"Ghosh",
"Swaroop",
""
]
] | TITLE: TrojanNet: Detecting Trojans in Quantum Circuits using Machine Learning
ABSTRACT: Quantum computing holds tremendous potential for various applications, but
its security remains a crucial concern. Quantum circuits need high-quality
compilers to optimize the depth and gate count to boost the success probability
on current noisy quantum computers. There is a rise of efficient but
unreliable/untrusted compilers; however, they present a risk of tampering such
as Trojan insertion. We propose TrojanNet, a novel approach to enhance the
security of quantum circuits by detecting and classifying Trojan-inserted
circuits. In particular, we focus on the Quantum Approximate Optimization
Algorithm (QAOA) circuit that is popular in solving a wide range of
optimization problems. We investigate the impact of Trojan insertion on QAOA
circuits and develop a Convolutional Neural Network (CNN) model, referred to as
TrojanNet, to identify their presence accurately. Using the Qiskit framework,
we generate 12 diverse datasets by introducing variations in Trojan gate types,
the number of gates, insertion locations, and compiler backends. These datasets
consist of both original Trojan-free QAOA circuits and their corresponding
Trojan-inserted counterparts. The generated datasets are then utilized for
training and evaluating the TrojanNet model. Experimental results showcase an
average accuracy of 98.80% and an average F1-score of 98.53% in effectively
detecting and classifying Trojan-inserted QAOA circuits. Finally, we conduct a
performance comparison between TrojanNet and existing machine learning-based
Trojan detection methods specifically designed for conventional netlists.
| no_new_dataset | 0.704808 |
2306.16707 | Masato Fujitake | Masato Fujitake | DiffusionSTR: Diffusion Model for Scene Text Recognition | Accepted to ICIP 2023 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper presents Diffusion Model for Scene Text Recognition
(DiffusionSTR), an end-to-end text recognition framework using diffusion models
for recognizing text in the wild. While existing studies have viewed the scene
text recognition task as an image-to-text transformation, we rethought it as a
text-text one under images in a diffusion model. We show for the first time
that the diffusion model can be applied to text recognition. Furthermore,
experimental results on publicly available datasets show that the proposed
method achieves competitive accuracy compared to state-of-the-art methods.
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 06:09:32 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Fujitake",
"Masato",
""
]
] | TITLE: DiffusionSTR: Diffusion Model for Scene Text Recognition
ABSTRACT: This paper presents Diffusion Model for Scene Text Recognition
(DiffusionSTR), an end-to-end text recognition framework using diffusion models
for recognizing text in the wild. While existing studies have viewed the scene
text recognition task as an image-to-text transformation, we rethought it as a
text-text one under images in a diffusion model. We show for the first time
that the diffusion model can be applied to text recognition. Furthermore,
experimental results on publicly available datasets show that the proposed
method achieves competitive accuracy compared to state-of-the-art methods.
| no_new_dataset | 0.712476 |
2306.16713 | Abhirama Subramanyam Penamakuri | Abhirama Subramanyam Penamakuri, Manish Gupta, Mithun Das Gupta, Anand
Mishra | Answer Mining from a Pool of Images: Towards Retrieval-Based Visual
Question Answering | Accepted to IJCAI 2023 | null | null | null | cs.CV cs.AI cs.LG | http://creativecommons.org/licenses/by-nc-sa/4.0/ | We study visual question answering in a setting where the answer has to be
mined from a pool of relevant and irrelevant images given as a context. For
such a setting, a model must first retrieve relevant images from the pool and
answer the question from these retrieved images. We refer to this problem as
retrieval-based visual question answering (or RETVQA in short). The RETVQA is
distinctively different and more challenging than the traditionally-studied
Visual Question Answering (VQA), where a given question has to be answered with
a single relevant image in context. Towards solving the RETVQA task, we propose
a unified Multi Image BART (MI-BART) that takes a question and retrieved images
using our relevance encoder for free-form fluent answer generation. Further, we
introduce the largest dataset in this space, namely RETVQA, which has the
following salient features: multi-image and retrieval requirement for VQA,
metadata-independent questions over a pool of heterogeneous images, expecting a
mix of classification-oriented and open-ended generative answers. Our proposed
framework achieves an accuracy of 76.5% and a fluency of 79.3% on the proposed
dataset, namely RETVQA and also outperforms state-of-the-art methods by 4.9%
and 11.8% on the image segment of the publicly available WebQA dataset on the
accuracy and fluency metrics, respectively.
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 06:22:43 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Penamakuri",
"Abhirama Subramanyam",
""
],
[
"Gupta",
"Manish",
""
],
[
"Gupta",
"Mithun Das",
""
],
[
"Mishra",
"Anand",
""
]
] | TITLE: Answer Mining from a Pool of Images: Towards Retrieval-Based Visual
Question Answering
ABSTRACT: We study visual question answering in a setting where the answer has to be
mined from a pool of relevant and irrelevant images given as a context. For
such a setting, a model must first retrieve relevant images from the pool and
answer the question from these retrieved images. We refer to this problem as
retrieval-based visual question answering (or RETVQA in short). The RETVQA is
distinctively different and more challenging than the traditionally-studied
Visual Question Answering (VQA), where a given question has to be answered with
a single relevant image in context. Towards solving the RETVQA task, we propose
a unified Multi Image BART (MI-BART) that takes a question and retrieved images
using our relevance encoder for free-form fluent answer generation. Further, we
introduce the largest dataset in this space, namely RETVQA, which has the
following salient features: multi-image and retrieval requirement for VQA,
metadata-independent questions over a pool of heterogeneous images, expecting a
mix of classification-oriented and open-ended generative answers. Our proposed
framework achieves an accuracy of 76.5% and a fluency of 79.3% on the proposed
dataset, namely RETVQA and also outperforms state-of-the-art methods by 4.9%
and 11.8% on the image segment of the publicly available WebQA dataset on the
accuracy and fluency metrics, respectively.
| new_dataset | 0.71222 |
2306.16714 | Yuming Zhong | Yuming Zhong and Yi Wang | SimPLe: Similarity-Aware Propagation Learning for Weakly-Supervised
Breast Cancer Segmentation in DCE-MRI | null | null | null | null | eess.IV cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Breast dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) plays
an important role in the screening and prognosis assessment of high-risk breast
cancer. The segmentation of cancerous regions is essential useful for the
subsequent analysis of breast MRI. To alleviate the annotation effort to train
the segmentation networks, we propose a weakly-supervised strategy using
extreme points as annotations for breast cancer segmentation. Without using any
bells and whistles, our strategy focuses on fully exploiting the learning
capability of the routine training procedure, i.e., the train - fine-tune -
retrain process. The network first utilizes the pseudo-masks generated using
the extreme points to train itself, by minimizing a contrastive loss, which
encourages the network to learn more representative features for cancerous
voxels. Then the trained network fine-tunes itself by using a similarity-aware
propagation learning (SimPLe) strategy, which leverages feature similarity
between unlabeled and positive voxels to propagate labels. Finally the network
retrains itself by employing the pseudo-masks generated using previous
fine-tuned network. The proposed method is evaluated on our collected DCE-MRI
dataset containing 206 patients with biopsy-proven breast cancers. Experimental
results demonstrate our method effectively fine-tunes the network by using the
SimPLe strategy, and achieves a mean Dice value of 81%.
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 06:22:50 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Zhong",
"Yuming",
""
],
[
"Wang",
"Yi",
""
]
] | TITLE: SimPLe: Similarity-Aware Propagation Learning for Weakly-Supervised
Breast Cancer Segmentation in DCE-MRI
ABSTRACT: Breast dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) plays
an important role in the screening and prognosis assessment of high-risk breast
cancer. The segmentation of cancerous regions is essential useful for the
subsequent analysis of breast MRI. To alleviate the annotation effort to train
the segmentation networks, we propose a weakly-supervised strategy using
extreme points as annotations for breast cancer segmentation. Without using any
bells and whistles, our strategy focuses on fully exploiting the learning
capability of the routine training procedure, i.e., the train - fine-tune -
retrain process. The network first utilizes the pseudo-masks generated using
the extreme points to train itself, by minimizing a contrastive loss, which
encourages the network to learn more representative features for cancerous
voxels. Then the trained network fine-tunes itself by using a similarity-aware
propagation learning (SimPLe) strategy, which leverages feature similarity
between unlabeled and positive voxels to propagate labels. Finally the network
retrains itself by employing the pseudo-masks generated using previous
fine-tuned network. The proposed method is evaluated on our collected DCE-MRI
dataset containing 206 patients with biopsy-proven breast cancers. Experimental
results demonstrate our method effectively fine-tunes the network by using the
SimPLe strategy, and achieves a mean Dice value of 81%.
| no_new_dataset | 0.711875 |
2306.16738 | Jiahao Xie | Jiahao Xie, Chao Zhang, Weijie Liu, Wensong Bai, Hui Qian | Towards Optimal Randomized Strategies in Adversarial Example Game | Extended version of paper https://doi.org/10.1609/aaai.v37i9.26247
which appeared in AAAI 2023 | null | 10.1609/aaai.v37i9.26247 | null | cs.LG cs.CR cs.GT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The vulnerability of deep neural network models to adversarial example
attacks is a practical challenge in many artificial intelligence applications.
A recent line of work shows that the use of randomization in adversarial
training is the key to find optimal strategies against adversarial example
attacks. However, in a fully randomized setting where both the defender and the
attacker can use randomized strategies, there are no efficient algorithm for
finding such an optimal strategy. To fill the gap, we propose the first
algorithm of its kind, called FRAT, which models the problem with a new
infinite-dimensional continuous-time flow on probability distribution spaces.
FRAT maintains a lightweight mixture of models for the defender, with
flexibility to efficiently update mixing weights and model parameters at each
iteration. Furthermore, FRAT utilizes lightweight sampling subroutines to
construct a random strategy for the attacker. We prove that the continuous-time
limit of FRAT converges to a mixed Nash equilibria in a zero-sum game formed by
a defender and an attacker. Experimental results also demonstrate the
efficiency of FRAT on CIFAR-10 and CIFAR-100 datasets.
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 07:29:23 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Xie",
"Jiahao",
""
],
[
"Zhang",
"Chao",
""
],
[
"Liu",
"Weijie",
""
],
[
"Bai",
"Wensong",
""
],
[
"Qian",
"Hui",
""
]
] | TITLE: Towards Optimal Randomized Strategies in Adversarial Example Game
ABSTRACT: The vulnerability of deep neural network models to adversarial example
attacks is a practical challenge in many artificial intelligence applications.
A recent line of work shows that the use of randomization in adversarial
training is the key to find optimal strategies against adversarial example
attacks. However, in a fully randomized setting where both the defender and the
attacker can use randomized strategies, there are no efficient algorithm for
finding such an optimal strategy. To fill the gap, we propose the first
algorithm of its kind, called FRAT, which models the problem with a new
infinite-dimensional continuous-time flow on probability distribution spaces.
FRAT maintains a lightweight mixture of models for the defender, with
flexibility to efficiently update mixing weights and model parameters at each
iteration. Furthermore, FRAT utilizes lightweight sampling subroutines to
construct a random strategy for the attacker. We prove that the continuous-time
limit of FRAT converges to a mixed Nash equilibria in a zero-sum game formed by
a defender and an attacker. Experimental results also demonstrate the
efficiency of FRAT on CIFAR-10 and CIFAR-100 datasets.
| no_new_dataset | 0.707626 |
2306.16762 | Bowen Yu | Bowen Yu, Cheng Fu, Haiyang Yu, Fei Huang, Yongbin Li | Unified Language Representation for Question Answering over Text,
Tables, and Images | Findings of ACL 2023 | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | When trying to answer complex questions, people often rely on multiple
sources of information, such as visual, textual, and tabular data. Previous
approaches to this problem have focused on designing input features or model
structure in the multi-modal space, which is inflexible for cross-modal
reasoning or data-efficient training. In this paper, we call for an alternative
paradigm, which transforms the images and tables into unified language
representations, so that we can simplify the task into a simpler textual QA
problem that can be solved using three steps: retrieval, ranking, and
generation, all within a language space. This idea takes advantage of the power
of pre-trained language models and is implemented in a framework called Solar.
Our experimental results show that Solar outperforms all existing methods by
10.6-32.3 pts on two datasets, MultimodalQA and MMCoQA, across ten different
metrics. Additionally, Solar achieves the best performance on the WebQA
leaderboard
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 08:02:23 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Yu",
"Bowen",
""
],
[
"Fu",
"Cheng",
""
],
[
"Yu",
"Haiyang",
""
],
[
"Huang",
"Fei",
""
],
[
"Li",
"Yongbin",
""
]
] | TITLE: Unified Language Representation for Question Answering over Text,
Tables, and Images
ABSTRACT: When trying to answer complex questions, people often rely on multiple
sources of information, such as visual, textual, and tabular data. Previous
approaches to this problem have focused on designing input features or model
structure in the multi-modal space, which is inflexible for cross-modal
reasoning or data-efficient training. In this paper, we call for an alternative
paradigm, which transforms the images and tables into unified language
representations, so that we can simplify the task into a simpler textual QA
problem that can be solved using three steps: retrieval, ranking, and
generation, all within a language space. This idea takes advantage of the power
of pre-trained language models and is implemented in a framework called Solar.
Our experimental results show that Solar outperforms all existing methods by
10.6-32.3 pts on two datasets, MultimodalQA and MMCoQA, across ten different
metrics. Additionally, Solar achieves the best performance on the WebQA
leaderboard
| no_new_dataset | 0.708364 |
2306.16770 | Ang Lv | Ang Lv, Jinpeng Li, Yuhan Chen, Xing Gao, Ji Zhang, Rui Yan | DialoGPS: Dialogue Path Sampling in Continuous Semantic Space for Data
Augmentation in Multi-Turn Conversations | ACL 2023 main | null | null | null | cs.CL cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In open-domain dialogue generation tasks, contexts and responses in most
datasets are one-to-one mapped, violating an important many-to-many
characteristic: a context leads to various responses, and a response answers
multiple contexts. Without such patterns, models poorly generalize and prefer
responding safely. Many attempts have been made in either multi-turn settings
from a one-to-many perspective or in a many-to-many perspective but limited to
single-turn settings. The major challenge to many-to-many augment multi-turn
dialogues is that discretely replacing each turn with semantic similarity
breaks fragile context coherence. In this paper, we propose DialoGue Path
Sampling (DialoGPS) method in continuous semantic space, the first many-to-many
augmentation method for multi-turn dialogues. Specifically, we map a dialogue
to our extended Brownian Bridge, a special Gaussian process. We sample latent
variables to form coherent dialogue paths in the continuous space. A dialogue
path corresponds to a new multi-turn dialogue and is used as augmented training
data. We show the effect of DialoGPS with both automatic and human evaluation.
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 08:12:47 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Lv",
"Ang",
""
],
[
"Li",
"Jinpeng",
""
],
[
"Chen",
"Yuhan",
""
],
[
"Gao",
"Xing",
""
],
[
"Zhang",
"Ji",
""
],
[
"Yan",
"Rui",
""
]
] | TITLE: DialoGPS: Dialogue Path Sampling in Continuous Semantic Space for Data
Augmentation in Multi-Turn Conversations
ABSTRACT: In open-domain dialogue generation tasks, contexts and responses in most
datasets are one-to-one mapped, violating an important many-to-many
characteristic: a context leads to various responses, and a response answers
multiple contexts. Without such patterns, models poorly generalize and prefer
responding safely. Many attempts have been made in either multi-turn settings
from a one-to-many perspective or in a many-to-many perspective but limited to
single-turn settings. The major challenge to many-to-many augment multi-turn
dialogues is that discretely replacing each turn with semantic similarity
breaks fragile context coherence. In this paper, we propose DialoGue Path
Sampling (DialoGPS) method in continuous semantic space, the first many-to-many
augmentation method for multi-turn dialogues. Specifically, we map a dialogue
to our extended Brownian Bridge, a special Gaussian process. We sample latent
variables to form coherent dialogue paths in the continuous space. A dialogue
path corresponds to a new multi-turn dialogue and is used as augmented training
data. We show the effect of DialoGPS with both automatic and human evaluation.
| no_new_dataset | 0.71145 |
2306.16780 | Xiang Zhuang | Xiang Zhuang, Qiang Zhang, Bin Wu, Keyan Ding, Yin Fang, Huajun Chen | Graph Sampling-based Meta-Learning for Molecular Property Prediction | Accepted by IJCAI 2023 | null | null | null | cs.LG q-bio.BM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Molecular property is usually observed with a limited number of samples, and
researchers have considered property prediction as a few-shot problem. One
important fact that has been ignored by prior works is that each molecule can
be recorded with several different properties simultaneously. To effectively
utilize many-to-many correlations of molecules and properties, we propose a
Graph Sampling-based Meta-learning (GS-Meta) framework for few-shot molecular
property prediction. First, we construct a Molecule-Property relation Graph
(MPG): molecule and properties are nodes, while property labels decide edges.
Then, to utilize the topological information of MPG, we reformulate an episode
in meta-learning as a subgraph of the MPG, containing a target property node,
molecule nodes, and auxiliary property nodes. Third, as episodes in the form of
subgraphs are no longer independent of each other, we propose to schedule the
subgraph sampling process with a contrastive loss function, which considers the
consistency and discrimination of subgraphs. Extensive experiments on 5
commonly-used benchmarks show GS-Meta consistently outperforms state-of-the-art
methods by 5.71%-6.93% in ROC-AUC and verify the effectiveness of each proposed
module. Our code is available at https://github.com/HICAI-ZJU/GS-Meta.
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 08:34:01 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Zhuang",
"Xiang",
""
],
[
"Zhang",
"Qiang",
""
],
[
"Wu",
"Bin",
""
],
[
"Ding",
"Keyan",
""
],
[
"Fang",
"Yin",
""
],
[
"Chen",
"Huajun",
""
]
] | TITLE: Graph Sampling-based Meta-Learning for Molecular Property Prediction
ABSTRACT: Molecular property is usually observed with a limited number of samples, and
researchers have considered property prediction as a few-shot problem. One
important fact that has been ignored by prior works is that each molecule can
be recorded with several different properties simultaneously. To effectively
utilize many-to-many correlations of molecules and properties, we propose a
Graph Sampling-based Meta-learning (GS-Meta) framework for few-shot molecular
property prediction. First, we construct a Molecule-Property relation Graph
(MPG): molecule and properties are nodes, while property labels decide edges.
Then, to utilize the topological information of MPG, we reformulate an episode
in meta-learning as a subgraph of the MPG, containing a target property node,
molecule nodes, and auxiliary property nodes. Third, as episodes in the form of
subgraphs are no longer independent of each other, we propose to schedule the
subgraph sampling process with a contrastive loss function, which considers the
consistency and discrimination of subgraphs. Extensive experiments on 5
commonly-used benchmarks show GS-Meta consistently outperforms state-of-the-art
methods by 5.71%-6.93% in ROC-AUC and verify the effectiveness of each proposed
module. Our code is available at https://github.com/HICAI-ZJU/GS-Meta.
| no_new_dataset | 0.710641 |
2306.16793 | Joshua Maynez | Joshua Maynez, Priyanka Agrawal, Sebastian Gehrmann | Benchmarking Large Language Model Capabilities for Conditional
Generation | null | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Pre-trained large language models (PLMs) underlie most new developments in
natural language processing. They have shifted the field from
application-specific model pipelines to a single model that is adapted to a
wide range of tasks. Autoregressive PLMs like GPT-3 or PaLM, alongside
techniques like few-shot learning, have additionally shifted the output
modality to generation instead of classification or regression. Despite their
ubiquitous use, the generation quality of language models is rarely evaluated
when these models are introduced. Additionally, it is unclear how existing
generation tasks--while they can be used to compare systems at a high
level--relate to the real world use cases for which people have been adopting
them. In this work, we discuss how to adapt existing application-specific
generation benchmarks to PLMs and provide an in-depth, empirical study of the
limitations and capabilities of PLMs in natural language generation tasks along
dimensions such as scale, architecture, input and output language. Our results
show that PLMs differ in their applicability to different data regimes and
their generalization to multiple languages and inform which PLMs to use for a
given generation task setup. We share best practices to be taken into
consideration when benchmarking generation capabilities during the development
of upcoming PLMs.
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 08:59:40 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Maynez",
"Joshua",
""
],
[
"Agrawal",
"Priyanka",
""
],
[
"Gehrmann",
"Sebastian",
""
]
] | TITLE: Benchmarking Large Language Model Capabilities for Conditional
Generation
ABSTRACT: Pre-trained large language models (PLMs) underlie most new developments in
natural language processing. They have shifted the field from
application-specific model pipelines to a single model that is adapted to a
wide range of tasks. Autoregressive PLMs like GPT-3 or PaLM, alongside
techniques like few-shot learning, have additionally shifted the output
modality to generation instead of classification or regression. Despite their
ubiquitous use, the generation quality of language models is rarely evaluated
when these models are introduced. Additionally, it is unclear how existing
generation tasks--while they can be used to compare systems at a high
level--relate to the real world use cases for which people have been adopting
them. In this work, we discuss how to adapt existing application-specific
generation benchmarks to PLMs and provide an in-depth, empirical study of the
limitations and capabilities of PLMs in natural language generation tasks along
dimensions such as scale, architecture, input and output language. Our results
show that PLMs differ in their applicability to different data regimes and
their generalization to multiple languages and inform which PLMs to use for a
given generation task setup. We share best practices to be taken into
consideration when benchmarking generation capabilities during the development
of upcoming PLMs.
| no_new_dataset | 0.707859 |
2306.16823 | Lyse Naomi Wamba Momo | Lyse Naomi Wamba Momo, Nyalleng Moorosi, Elaine O. Nsoesie, Frank
Rademakers, Bart De Moor | Length of Stay prediction for Hospital Management using Domain
Adaptation | null | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | Inpatient length of stay (LoS) is an important managerial metric which if
known in advance can be used to efficiently plan admissions, allocate resources
and improve care. Using historical patient data and machine learning
techniques, LoS prediction models can be developed. Ethically, these models can
not be used for patient discharge in lieu of unit heads but are of utmost
necessity for hospital management systems in charge of effective hospital
planning. Therefore, the design of the prediction system should be adapted to
work in a true hospital setting. In this study, we predict early hospital LoS
at the granular level of admission units by applying domain adaptation to
leverage information learned from a potential source domain. Time-varying data
from 110,079 and 60,492 patient stays to 8 and 9 intensive care units were
respectively extracted from eICU-CRD and MIMIC-IV. These were fed into a
Long-Short Term Memory and a Fully connected network to train a source domain
model, the weights of which were transferred either partially or fully to
initiate training in target domains. Shapley Additive exPlanations (SHAP)
algorithms were used to study the effect of weight transfer on model
explanability. Compared to the benchmark, the proposed weight transfer model
showed statistically significant gains in prediction accuracy (between 1% and
5%) as well as computation time (up to 2hrs) for some target domains. The
proposed method thus provides an adapted clinical decision support system for
hospital management that can ease processes of data access via ethical
committee, computation infrastructures and time.
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 09:58:21 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Momo",
"Lyse Naomi Wamba",
""
],
[
"Moorosi",
"Nyalleng",
""
],
[
"Nsoesie",
"Elaine O.",
""
],
[
"Rademakers",
"Frank",
""
],
[
"De Moor",
"Bart",
""
]
] | TITLE: Length of Stay prediction for Hospital Management using Domain
Adaptation
ABSTRACT: Inpatient length of stay (LoS) is an important managerial metric which if
known in advance can be used to efficiently plan admissions, allocate resources
and improve care. Using historical patient data and machine learning
techniques, LoS prediction models can be developed. Ethically, these models can
not be used for patient discharge in lieu of unit heads but are of utmost
necessity for hospital management systems in charge of effective hospital
planning. Therefore, the design of the prediction system should be adapted to
work in a true hospital setting. In this study, we predict early hospital LoS
at the granular level of admission units by applying domain adaptation to
leverage information learned from a potential source domain. Time-varying data
from 110,079 and 60,492 patient stays to 8 and 9 intensive care units were
respectively extracted from eICU-CRD and MIMIC-IV. These were fed into a
Long-Short Term Memory and a Fully connected network to train a source domain
model, the weights of which were transferred either partially or fully to
initiate training in target domains. Shapley Additive exPlanations (SHAP)
algorithms were used to study the effect of weight transfer on model
explanability. Compared to the benchmark, the proposed weight transfer model
showed statistically significant gains in prediction accuracy (between 1% and
5%) as well as computation time (up to 2hrs) for some target domains. The
proposed method thus provides an adapted clinical decision support system for
hospital management that can ease processes of data access via ethical
committee, computation infrastructures and time.
| no_new_dataset | 0.709868 |
2306.16916 | Sigrid Passano Hellan | Sigrid Passano Hellan, Huibin Shen, Fran\c{c}ois-Xavier Aubet, David
Salinas and Aaron Klein | Obeying the Order: Introducing Ordered Transfer Hyperparameter
Optimisation | To be presented at the AutoML 2023 Workshop Track | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | We introduce ordered transfer hyperparameter optimisation (OTHPO), a version
of transfer learning for hyperparameter optimisation (HPO) where the tasks
follow a sequential order. Unlike for state-of-the-art transfer HPO, the
assumption is that each task is most correlated to those immediately before it.
This matches many deployed settings, where hyperparameters are retuned as more
data is collected; for instance tuning a sequence of movie recommendation
systems as more movies and ratings are added. We propose a formal definition,
outline the differences to related problems and propose a basic OTHPO method
that outperforms state-of-the-art transfer HPO. We empirically show the
importance of taking order into account using ten benchmarks. The benchmarks
are in the setting of gradually accumulating data, and span XGBoost, random
forest, approximate k-nearest neighbor, elastic net, support vector machines
and a separate real-world motivated optimisation problem. We open source the
benchmarks to foster future research on ordered transfer HPO.
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 13:08:36 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Hellan",
"Sigrid Passano",
""
],
[
"Shen",
"Huibin",
""
],
[
"Aubet",
"François-Xavier",
""
],
[
"Salinas",
"David",
""
],
[
"Klein",
"Aaron",
""
]
] | TITLE: Obeying the Order: Introducing Ordered Transfer Hyperparameter
Optimisation
ABSTRACT: We introduce ordered transfer hyperparameter optimisation (OTHPO), a version
of transfer learning for hyperparameter optimisation (HPO) where the tasks
follow a sequential order. Unlike for state-of-the-art transfer HPO, the
assumption is that each task is most correlated to those immediately before it.
This matches many deployed settings, where hyperparameters are retuned as more
data is collected; for instance tuning a sequence of movie recommendation
systems as more movies and ratings are added. We propose a formal definition,
outline the differences to related problems and propose a basic OTHPO method
that outperforms state-of-the-art transfer HPO. We empirically show the
importance of taking order into account using ten benchmarks. The benchmarks
are in the setting of gradually accumulating data, and span XGBoost, random
forest, approximate k-nearest neighbor, elastic net, support vector machines
and a separate real-world motivated optimisation problem. We open source the
benchmarks to foster future research on ordered transfer HPO.
| no_new_dataset | 0.709667 |
2306.16917 | David Recasens | David Recasens, Martin R. Oswald, Marc Pollefeys, Javier Civera | The Drunkard's Odometry: Estimating Camera Motion in Deforming Scenes | null | null | null | null | cs.CV cs.LG cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Estimating camera motion in deformable scenes poses a complex and open
research challenge. Most existing non-rigid structure from motion techniques
assume to observe also static scene parts besides deforming scene parts in
order to establish an anchoring reference. However, this assumption does not
hold true in certain relevant application cases such as endoscopies. Deformable
odometry and SLAM pipelines, which tackle the most challenging scenario of
exploratory trajectories, suffer from a lack of robustness and proper
quantitative evaluation methodologies. To tackle this issue with a common
benchmark, we introduce the Drunkard's Dataset, a challenging collection of
synthetic data targeting visual navigation and reconstruction in deformable
environments. This dataset is the first large set of exploratory camera
trajectories with ground truth inside 3D scenes where every surface exhibits
non-rigid deformations over time. Simulations in realistic 3D buildings lets us
obtain a vast amount of data and ground truth labels, including camera poses,
RGB images and depth, optical flow and normal maps at high resolution and
quality. We further present a novel deformable odometry method, dubbed the
Drunkard's Odometry, which decomposes optical flow estimates into rigid-body
camera motion and non-rigid scene deformations. In order to validate our data,
our work contains an evaluation of several baselines as well as a novel
tracking error metric which does not require ground truth data. Dataset and
code: https://davidrecasens.github.io/TheDrunkard'sOdometry/
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 13:09:31 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Recasens",
"David",
""
],
[
"Oswald",
"Martin R.",
""
],
[
"Pollefeys",
"Marc",
""
],
[
"Civera",
"Javier",
""
]
] | TITLE: The Drunkard's Odometry: Estimating Camera Motion in Deforming Scenes
ABSTRACT: Estimating camera motion in deformable scenes poses a complex and open
research challenge. Most existing non-rigid structure from motion techniques
assume to observe also static scene parts besides deforming scene parts in
order to establish an anchoring reference. However, this assumption does not
hold true in certain relevant application cases such as endoscopies. Deformable
odometry and SLAM pipelines, which tackle the most challenging scenario of
exploratory trajectories, suffer from a lack of robustness and proper
quantitative evaluation methodologies. To tackle this issue with a common
benchmark, we introduce the Drunkard's Dataset, a challenging collection of
synthetic data targeting visual navigation and reconstruction in deformable
environments. This dataset is the first large set of exploratory camera
trajectories with ground truth inside 3D scenes where every surface exhibits
non-rigid deformations over time. Simulations in realistic 3D buildings lets us
obtain a vast amount of data and ground truth labels, including camera poses,
RGB images and depth, optical flow and normal maps at high resolution and
quality. We further present a novel deformable odometry method, dubbed the
Drunkard's Odometry, which decomposes optical flow estimates into rigid-body
camera motion and non-rigid scene deformations. In order to validate our data,
our work contains an evaluation of several baselines as well as a novel
tracking error metric which does not require ground truth data. Dataset and
code: https://davidrecasens.github.io/TheDrunkard'sOdometry/
| new_dataset | 0.714846 |
2306.16918 | Tao Wang | Tao Wang, Xinlin Zhang, Yuanbo Zhou, Junlin Lan, Tao Tan, Min Du,
Qinquan Gao and Tong Tong | PCDAL: A Perturbation Consistency-Driven Active Learning Approach for
Medical Image Segmentation and Classification | null | null | null | null | eess.IV cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In recent years, deep learning has become a breakthrough technique in
assisting medical image diagnosis. Supervised learning using convolutional
neural networks (CNN) provides state-of-the-art performance and has served as a
benchmark for various medical image segmentation and classification. However,
supervised learning deeply relies on large-scale annotated data, which is
expensive, time-consuming, and even impractical to acquire in medical imaging
applications. Active Learning (AL) methods have been widely applied in natural
image classification tasks to reduce annotation costs by selecting more
valuable examples from the unlabeled data pool. However, their application in
medical image segmentation tasks is limited, and there is currently no
effective and universal AL-based method specifically designed for 3D medical
image segmentation. To address this limitation, we propose an AL-based method
that can be simultaneously applied to 2D medical image classification,
segmentation, and 3D medical image segmentation tasks. We extensively validated
our proposed active learning method on three publicly available and challenging
medical image datasets, Kvasir Dataset, COVID-19 Infection Segmentation
Dataset, and BraTS2019 Dataset. The experimental results demonstrate that our
PCDAL can achieve significantly improved performance with fewer annotations in
2D classification and segmentation and 3D segmentation tasks. The codes of this
study are available at https://github.com/ortonwang/PCDAL.
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 13:11:46 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Wang",
"Tao",
""
],
[
"Zhang",
"Xinlin",
""
],
[
"Zhou",
"Yuanbo",
""
],
[
"Lan",
"Junlin",
""
],
[
"Tan",
"Tao",
""
],
[
"Du",
"Min",
""
],
[
"Gao",
"Qinquan",
""
],
[
"Tong",
"Tong",
""
]
] | TITLE: PCDAL: A Perturbation Consistency-Driven Active Learning Approach for
Medical Image Segmentation and Classification
ABSTRACT: In recent years, deep learning has become a breakthrough technique in
assisting medical image diagnosis. Supervised learning using convolutional
neural networks (CNN) provides state-of-the-art performance and has served as a
benchmark for various medical image segmentation and classification. However,
supervised learning deeply relies on large-scale annotated data, which is
expensive, time-consuming, and even impractical to acquire in medical imaging
applications. Active Learning (AL) methods have been widely applied in natural
image classification tasks to reduce annotation costs by selecting more
valuable examples from the unlabeled data pool. However, their application in
medical image segmentation tasks is limited, and there is currently no
effective and universal AL-based method specifically designed for 3D medical
image segmentation. To address this limitation, we propose an AL-based method
that can be simultaneously applied to 2D medical image classification,
segmentation, and 3D medical image segmentation tasks. We extensively validated
our proposed active learning method on three publicly available and challenging
medical image datasets, Kvasir Dataset, COVID-19 Infection Segmentation
Dataset, and BraTS2019 Dataset. The experimental results demonstrate that our
PCDAL can achieve significantly improved performance with fewer annotations in
2D classification and segmentation and 3D segmentation tasks. The codes of this
study are available at https://github.com/ortonwang/PCDAL.
| no_new_dataset | 0.710245 |
2306.16925 | Guotai Wang | Guotai Wang, Jianghao Wu, Xiangde Luo, Xinglong Liu, Kang Li, Shaoting
Zhang | MIS-FM: 3D Medical Image Segmentation using Foundation Models Pretrained
on a Large-Scale Unannotated Dataset | 13 pages, 8 figures | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Pretraining with large-scale 3D volumes has a potential for improving the
segmentation performance on a target medical image dataset where the training
images and annotations are limited. Due to the high cost of acquiring
pixel-level segmentation annotations on the large-scale pretraining dataset,
pretraining with unannotated images is highly desirable. In this work, we
propose a novel self-supervised learning strategy named Volume Fusion (VF) for
pretraining 3D segmentation models. It fuses several random patches from a
foreground sub-volume to a background sub-volume based on a predefined set of
discrete fusion coefficients, and forces the model to predict the fusion
coefficient of each voxel, which is formulated as a self-supervised
segmentation task without manual annotations. Additionally, we propose a novel
network architecture based on parallel convolution and transformer blocks that
is suitable to be transferred to different downstream segmentation tasks with
various scales of organs and lesions. The proposed model was pretrained with
110k unannotated 3D CT volumes, and experiments with different downstream
segmentation targets including head and neck organs, thoracic/abdominal organs
showed that our pretrained model largely outperformed training from scratch and
several state-of-the-art self-supervised training methods and segmentation
models. The code and pretrained model are available at
https://github.com/openmedlab/MIS-FM.
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 13:22:13 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Wang",
"Guotai",
""
],
[
"Wu",
"Jianghao",
""
],
[
"Luo",
"Xiangde",
""
],
[
"Liu",
"Xinglong",
""
],
[
"Li",
"Kang",
""
],
[
"Zhang",
"Shaoting",
""
]
] | TITLE: MIS-FM: 3D Medical Image Segmentation using Foundation Models Pretrained
on a Large-Scale Unannotated Dataset
ABSTRACT: Pretraining with large-scale 3D volumes has a potential for improving the
segmentation performance on a target medical image dataset where the training
images and annotations are limited. Due to the high cost of acquiring
pixel-level segmentation annotations on the large-scale pretraining dataset,
pretraining with unannotated images is highly desirable. In this work, we
propose a novel self-supervised learning strategy named Volume Fusion (VF) for
pretraining 3D segmentation models. It fuses several random patches from a
foreground sub-volume to a background sub-volume based on a predefined set of
discrete fusion coefficients, and forces the model to predict the fusion
coefficient of each voxel, which is formulated as a self-supervised
segmentation task without manual annotations. Additionally, we propose a novel
network architecture based on parallel convolution and transformer blocks that
is suitable to be transferred to different downstream segmentation tasks with
various scales of organs and lesions. The proposed model was pretrained with
110k unannotated 3D CT volumes, and experiments with different downstream
segmentation targets including head and neck organs, thoracic/abdominal organs
showed that our pretrained model largely outperformed training from scratch and
several state-of-the-art self-supervised training methods and segmentation
models. The code and pretrained model are available at
https://github.com/openmedlab/MIS-FM.
| no_new_dataset | 0.71145 |
2306.16931 | Junda Wang | Junda Wang, Zonghai Yao, Avijit Mitra, Samuel Osebe, Zhichao Yang,
Hong Yu | UMASS_BioNLP at MEDIQA-Chat 2023: Can LLMs generate high-quality
synthetic note-oriented doctor-patient conversations? | null | null | null | null | cs.CL cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper presents UMASS_BioNLP team participation in the MEDIQA-Chat 2023
shared task for Task-A and Task-C. We focus especially on Task-C and propose a
novel LLMs cooperation system named a doctor-patient loop to generate
high-quality conversation data sets. The experiment results demonstrate that
our approaches yield reasonable performance as evaluated by automatic metrics
such as ROUGE, medical concept recall, BLEU, and Self-BLEU. Furthermore, we
conducted a comparative analysis between our proposed method and ChatGPT and
GPT-4. This analysis also investigates the potential of utilizing cooperation
LLMs to generate high-quality datasets.
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 13:30:41 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Wang",
"Junda",
""
],
[
"Yao",
"Zonghai",
""
],
[
"Mitra",
"Avijit",
""
],
[
"Osebe",
"Samuel",
""
],
[
"Yang",
"Zhichao",
""
],
[
"Yu",
"Hong",
""
]
] | TITLE: UMASS_BioNLP at MEDIQA-Chat 2023: Can LLMs generate high-quality
synthetic note-oriented doctor-patient conversations?
ABSTRACT: This paper presents UMASS_BioNLP team participation in the MEDIQA-Chat 2023
shared task for Task-A and Task-C. We focus especially on Task-C and propose a
novel LLMs cooperation system named a doctor-patient loop to generate
high-quality conversation data sets. The experiment results demonstrate that
our approaches yield reasonable performance as evaluated by automatic metrics
such as ROUGE, medical concept recall, BLEU, and Self-BLEU. Furthermore, we
conducted a comparative analysis between our proposed method and ChatGPT and
GPT-4. This analysis also investigates the potential of utilizing cooperation
LLMs to generate high-quality datasets.
| no_new_dataset | 0.710019 |
2306.16938 | Xiao-Shan Gao | Yihan Wang and Lijia Yu and Xiao-Shan Gao | Restore Translation Using Equivariant Neural Networks | null | null | null | null | cs.LG cs.CV | http://creativecommons.org/licenses/by/4.0/ | Invariance to spatial transformations such as translations and rotations is a
desirable property and a basic design principle for classification neural
networks. However, the commonly used convolutional neural networks (CNNs) are
actually very sensitive to even small translations. There exist vast works to
achieve exact or approximate transformation invariance by designing
transformation-invariant models or assessing the transformations. These works
usually make changes to the standard CNNs and harm the performance on standard
datasets. In this paper, rather than modifying the classifier, we propose a
pre-classifier restorer to recover translated (or even rotated) inputs to the
original ones which will be fed into any classifier for the same dataset. The
restorer is based on a theoretical result which gives a sufficient and
necessary condition for an affine operator to be translational equivariant on a
tensor space.
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 13:34:35 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Wang",
"Yihan",
""
],
[
"Yu",
"Lijia",
""
],
[
"Gao",
"Xiao-Shan",
""
]
] | TITLE: Restore Translation Using Equivariant Neural Networks
ABSTRACT: Invariance to spatial transformations such as translations and rotations is a
desirable property and a basic design principle for classification neural
networks. However, the commonly used convolutional neural networks (CNNs) are
actually very sensitive to even small translations. There exist vast works to
achieve exact or approximate transformation invariance by designing
transformation-invariant models or assessing the transformations. These works
usually make changes to the standard CNNs and harm the performance on standard
datasets. In this paper, rather than modifying the classifier, we propose a
pre-classifier restorer to recover translated (or even rotated) inputs to the
original ones which will be fed into any classifier for the same dataset. The
restorer is based on a theoretical result which gives a sufficient and
necessary condition for an affine operator to be translational equivariant on a
tensor space.
| no_new_dataset | 0.711825 |
2306.16940 | Priyanka Patel | Michael J. Black, Priyanka Patel, Joachim Tesch, Jinlong Yang | BEDLAM: A Synthetic Dataset of Bodies Exhibiting Detailed Lifelike
Animated Motion | null | CVPR 2023 | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We show, for the first time, that neural networks trained only on synthetic
data achieve state-of-the-art accuracy on the problem of 3D human pose and
shape (HPS) estimation from real images. Previous synthetic datasets have been
small, unrealistic, or lacked realistic clothing. Achieving sufficient realism
is non-trivial and we show how to do this for full bodies in motion.
Specifically, our BEDLAM dataset contains monocular RGB videos with
ground-truth 3D bodies in SMPL-X format. It includes a diversity of body
shapes, motions, skin tones, hair, and clothing. The clothing is realistically
simulated on the moving bodies using commercial clothing physics simulation. We
render varying numbers of people in realistic scenes with varied lighting and
camera motions. We then train various HPS regressors using BEDLAM and achieve
state-of-the-art accuracy on real-image benchmarks despite training with
synthetic data. We use BEDLAM to gain insights into what model design choices
are important for accuracy. With good synthetic training data, we find that a
basic method like HMR approaches the accuracy of the current SOTA method
(CLIFF). BEDLAM is useful for a variety of tasks and all images, ground truth
bodies, 3D clothing, support code, and more are available for research
purposes. Additionally, we provide detailed information about our synthetic
data generation pipeline, enabling others to generate their own datasets. See
the project page: https://bedlam.is.tue.mpg.de/.
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 13:35:16 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Black",
"Michael J.",
""
],
[
"Patel",
"Priyanka",
""
],
[
"Tesch",
"Joachim",
""
],
[
"Yang",
"Jinlong",
""
]
] | TITLE: BEDLAM: A Synthetic Dataset of Bodies Exhibiting Detailed Lifelike
Animated Motion
ABSTRACT: We show, for the first time, that neural networks trained only on synthetic
data achieve state-of-the-art accuracy on the problem of 3D human pose and
shape (HPS) estimation from real images. Previous synthetic datasets have been
small, unrealistic, or lacked realistic clothing. Achieving sufficient realism
is non-trivial and we show how to do this for full bodies in motion.
Specifically, our BEDLAM dataset contains monocular RGB videos with
ground-truth 3D bodies in SMPL-X format. It includes a diversity of body
shapes, motions, skin tones, hair, and clothing. The clothing is realistically
simulated on the moving bodies using commercial clothing physics simulation. We
render varying numbers of people in realistic scenes with varied lighting and
camera motions. We then train various HPS regressors using BEDLAM and achieve
state-of-the-art accuracy on real-image benchmarks despite training with
synthetic data. We use BEDLAM to gain insights into what model design choices
are important for accuracy. With good synthetic training data, we find that a
basic method like HMR approaches the accuracy of the current SOTA method
(CLIFF). BEDLAM is useful for a variety of tasks and all images, ground truth
bodies, 3D clothing, support code, and more are available for research
purposes. Additionally, we provide detailed information about our synthetic
data generation pipeline, enabling others to generate their own datasets. See
the project page: https://bedlam.is.tue.mpg.de/.
| new_dataset | 0.719153 |
2306.16951 | Fedor Pavutnitskiy | Kirill Brilliantov, Fedor Pavutnitskiy, Dmitry Pasechnyuk, German
Magai | Applying language models to algebraic topology: generating simplicial
cycles using multi-labeling in Wu's formula | 20 pages | null | null | null | math.AT cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Computing homotopy groups of spheres has long been a fundamental objective in
algebraic topology. Various theoretical and algorithmic approaches have been
developed to tackle this problem. In this paper we take a step towards the goal
of comprehending the group-theoretic structure of the generators of these
homotopy groups by leveraging the power of machine learning. Specifically, in
the simplicial group setting of Wu's formula, we reformulate the problem of
generating simplicial cycles as a problem of sampling from the intersection of
algorithmic datasets related to Dyck languages. We present and evaluate
language modelling approaches that employ multi-label information for input
sequences, along with the necessary group-theoretic toolkit and non-neural
baselines.
| [
{
"version": "v1",
"created": "Thu, 1 Jun 2023 12:23:14 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Brilliantov",
"Kirill",
""
],
[
"Pavutnitskiy",
"Fedor",
""
],
[
"Pasechnyuk",
"Dmitry",
""
],
[
"Magai",
"German",
""
]
] | TITLE: Applying language models to algebraic topology: generating simplicial
cycles using multi-labeling in Wu's formula
ABSTRACT: Computing homotopy groups of spheres has long been a fundamental objective in
algebraic topology. Various theoretical and algorithmic approaches have been
developed to tackle this problem. In this paper we take a step towards the goal
of comprehending the group-theoretic structure of the generators of these
homotopy groups by leveraging the power of machine learning. Specifically, in
the simplicial group setting of Wu's formula, we reformulate the problem of
generating simplicial cycles as a problem of sampling from the intersection of
algorithmic datasets related to Dyck languages. We present and evaluate
language modelling approaches that employ multi-label information for input
sequences, along with the necessary group-theoretic toolkit and non-neural
baselines.
| no_new_dataset | 0.710427 |
2306.16955 | Francesco Foscarin | Francesco Foscarin, Daniel Harasim, Gerhard Widmer | Predicting Music Hierarchies with a Graph-Based Neural Decoder | To be published in the Proceedings of the International Society for
Music Information Retrieval Conference (ISMIR) | null | null | null | cs.SD cs.CL eess.AS | http://creativecommons.org/licenses/by/4.0/ | This paper describes a data-driven framework to parse musical sequences into
dependency trees, which are hierarchical structures used in music cognition
research and music analysis. The parsing involves two steps. First, the input
sequence is passed through a transformer encoder to enrich it with contextual
information. Then, a classifier filters the graph of all possible dependency
arcs to produce the dependency tree. One major benefit of this system is that
it can be easily integrated into modern deep-learning pipelines. Moreover,
since it does not rely on any particular symbolic grammar, it can consider
multiple musical features simultaneously, make use of sequential context
information, and produce partial results for noisy inputs. We test our approach
on two datasets of musical trees -- time-span trees of monophonic note
sequences and harmonic trees of jazz chord sequences -- and show that our
approach outperforms previous methods.
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 13:59:18 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Foscarin",
"Francesco",
""
],
[
"Harasim",
"Daniel",
""
],
[
"Widmer",
"Gerhard",
""
]
] | TITLE: Predicting Music Hierarchies with a Graph-Based Neural Decoder
ABSTRACT: This paper describes a data-driven framework to parse musical sequences into
dependency trees, which are hierarchical structures used in music cognition
research and music analysis. The parsing involves two steps. First, the input
sequence is passed through a transformer encoder to enrich it with contextual
information. Then, a classifier filters the graph of all possible dependency
arcs to produce the dependency tree. One major benefit of this system is that
it can be easily integrated into modern deep-learning pipelines. Moreover,
since it does not rely on any particular symbolic grammar, it can consider
multiple musical features simultaneously, make use of sequential context
information, and produce partial results for noisy inputs. We test our approach
on two datasets of musical trees -- time-span trees of monophonic note
sequences and harmonic trees of jazz chord sequences -- and show that our
approach outperforms previous methods.
| no_new_dataset | 0.70883 |
2306.16956 | Hongjie Cai | Hongjie Cai, Nan Song, Zengzhi Wang, Qiming Xie, Qiankun Zhao, Ke Li,
Siwei Wu, Shijie Liu, Jianfei Yu, Rui Xia | MEMD-ABSA: A Multi-Element Multi-Domain Dataset for Aspect-Based
Sentiment Analysis | null | null | null | null | cs.CL cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Aspect-based sentiment analysis is a long-standing research interest in the
field of opinion mining, and in recent years, researchers have gradually
shifted their focus from simple ABSA subtasks to end-to-end multi-element ABSA
tasks. However, the datasets currently used in the research are limited to
individual elements of specific tasks, usually focusing on in-domain settings,
ignoring implicit aspects and opinions, and with a small data scale. To address
these issues, we propose a large-scale Multi-Element Multi-Domain dataset
(MEMD) that covers the four elements across five domains, including nearly
20,000 review sentences and 30,000 quadruples annotated with explicit and
implicit aspects and opinions for ABSA research. Meanwhile, we evaluate
generative and non-generative baselines on multiple ABSA subtasks under the
open domain setting, and the results show that open domain ABSA as well as
mining implicit aspects and opinions remain ongoing challenges to be addressed.
The datasets are publicly released at \url{https://github.com/NUSTM/MEMD-ABSA}.
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 14:03:49 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Cai",
"Hongjie",
""
],
[
"Song",
"Nan",
""
],
[
"Wang",
"Zengzhi",
""
],
[
"Xie",
"Qiming",
""
],
[
"Zhao",
"Qiankun",
""
],
[
"Li",
"Ke",
""
],
[
"Wu",
"Siwei",
""
],
[
"Liu",
"Shijie",
""
],
[
"Yu",
"Jianfei",
""
],
[
"Xia",
"Rui",
""
]
] | TITLE: MEMD-ABSA: A Multi-Element Multi-Domain Dataset for Aspect-Based
Sentiment Analysis
ABSTRACT: Aspect-based sentiment analysis is a long-standing research interest in the
field of opinion mining, and in recent years, researchers have gradually
shifted their focus from simple ABSA subtasks to end-to-end multi-element ABSA
tasks. However, the datasets currently used in the research are limited to
individual elements of specific tasks, usually focusing on in-domain settings,
ignoring implicit aspects and opinions, and with a small data scale. To address
these issues, we propose a large-scale Multi-Element Multi-Domain dataset
(MEMD) that covers the four elements across five domains, including nearly
20,000 review sentences and 30,000 quadruples annotated with explicit and
implicit aspects and opinions for ABSA research. Meanwhile, we evaluate
generative and non-generative baselines on multiple ABSA subtasks under the
open domain setting, and the results show that open domain ABSA as well as
mining implicit aspects and opinions remain ongoing challenges to be addressed.
The datasets are publicly released at \url{https://github.com/NUSTM/MEMD-ABSA}.
| new_dataset | 0.714217 |
2306.16957 | Yushun Tang | Yushun Tang, Qinghai Guo, and Zhihai He | Cross-Inferential Networks for Source-free Unsupervised Domain
Adaptation | ICIP2023 accepted | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | One central challenge in source-free unsupervised domain adaptation (UDA) is
the lack of an effective approach to evaluate the prediction results of the
adapted network model in the target domain. To address this challenge, we
propose to explore a new method called cross-inferential networks (CIN). Our
main idea is that, when we adapt the network model to predict the sample labels
from encoded features, we use these prediction results to construct new
training samples with derived labels to learn a new examiner network that
performs a different but compatible task in the target domain. Specifically, in
this work, the base network model is performing image classification while the
examiner network is tasked to perform relative ordering of triplets of samples
whose training labels are carefully constructed from the prediction results of
the base network model. Two similarity measures, cross-network correlation
matrix similarity and attention consistency, are then developed to provide
important guidance for the UDA process. Our experimental results on benchmark
datasets demonstrate that our proposed CIN approach can significantly improve
the performance of source-free UDA.
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 14:04:24 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Tang",
"Yushun",
""
],
[
"Guo",
"Qinghai",
""
],
[
"He",
"Zhihai",
""
]
] | TITLE: Cross-Inferential Networks for Source-free Unsupervised Domain
Adaptation
ABSTRACT: One central challenge in source-free unsupervised domain adaptation (UDA) is
the lack of an effective approach to evaluate the prediction results of the
adapted network model in the target domain. To address this challenge, we
propose to explore a new method called cross-inferential networks (CIN). Our
main idea is that, when we adapt the network model to predict the sample labels
from encoded features, we use these prediction results to construct new
training samples with derived labels to learn a new examiner network that
performs a different but compatible task in the target domain. Specifically, in
this work, the base network model is performing image classification while the
examiner network is tasked to perform relative ordering of triplets of samples
whose training labels are carefully constructed from the prediction results of
the base network model. Two similarity measures, cross-network correlation
matrix similarity and attention consistency, are then developed to provide
important guidance for the UDA process. Our experimental results on benchmark
datasets demonstrate that our proposed CIN approach can significantly improve
the performance of source-free UDA.
| no_new_dataset | 0.710069 |
2306.16962 | Felix Burkhardt | Felix Burkhardt, Johannes Wagner, Hagen Wierstorf, Florian Eyben,
Bj\"orn Schuller | Speech-based Age and Gender Prediction with Transformers | 5 pages, submitted to 15th ITG Conference on Speech Communication | null | null | null | cs.SD eess.AS | http://creativecommons.org/licenses/by/4.0/ | We report on the curation of several publicly available datasets for age and
gender prediction. Furthermore, we present experiments to predict age and
gender with models based on a pre-trained wav2vec 2.0. Depending on the
dataset, we achieve an MAE between 7.1 years and 10.8 years for age, and at
least 91.1% ACC for gender (female, male, child). Compared to a modelling
approach built on handcrafted features, our proposed system shows an
improvement of 9% UAR for age and 4% UAR for gender. To make our findings
reproducible, we release the best performing model to the community as well as
the sample lists of the data splits.
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 14:13:15 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Burkhardt",
"Felix",
""
],
[
"Wagner",
"Johannes",
""
],
[
"Wierstorf",
"Hagen",
""
],
[
"Eyben",
"Florian",
""
],
[
"Schuller",
"Björn",
""
]
] | TITLE: Speech-based Age and Gender Prediction with Transformers
ABSTRACT: We report on the curation of several publicly available datasets for age and
gender prediction. Furthermore, we present experiments to predict age and
gender with models based on a pre-trained wav2vec 2.0. Depending on the
dataset, we achieve an MAE between 7.1 years and 10.8 years for age, and at
least 91.1% ACC for gender (female, male, child). Compared to a modelling
approach built on handcrafted features, our proposed system shows an
improvement of 9% UAR for age and 4% UAR for gender. To make our findings
reproducible, we release the best performing model to the community as well as
the sample lists of the data splits.
| no_new_dataset | 0.711406 |
2306.16991 | Weide Liu | Weide Liu, Xiaoyang Zhong, Jingwen Hou, Shaohua Li, Haozhe Huang and
Yuming Fang | Integrating Large Pre-trained Models into Multimodal Named Entity
Recognition with Evidential Fusion | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Multimodal Named Entity Recognition (MNER) is a crucial task for information
extraction from social media platforms such as Twitter. Most current methods
rely on attention weights to extract information from both text and images but
are often unreliable and lack interpretability. To address this problem, we
propose incorporating uncertainty estimation into the MNER task, producing
trustworthy predictions. Our proposed algorithm models the distribution of each
modality as a Normal-inverse Gamma distribution, and fuses them into a unified
distribution with an evidential fusion mechanism, enabling hierarchical
characterization of uncertainties and promotion of prediction accuracy and
trustworthiness. Additionally, we explore the potential of pre-trained large
foundation models in MNER and propose an efficient fusion approach that
leverages their robust feature representations. Experiments on two datasets
demonstrate that our proposed method outperforms the baselines and achieves new
state-of-the-art performance.
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 14:50:23 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Liu",
"Weide",
""
],
[
"Zhong",
"Xiaoyang",
""
],
[
"Hou",
"Jingwen",
""
],
[
"Li",
"Shaohua",
""
],
[
"Huang",
"Haozhe",
""
],
[
"Fang",
"Yuming",
""
]
] | TITLE: Integrating Large Pre-trained Models into Multimodal Named Entity
Recognition with Evidential Fusion
ABSTRACT: Multimodal Named Entity Recognition (MNER) is a crucial task for information
extraction from social media platforms such as Twitter. Most current methods
rely on attention weights to extract information from both text and images but
are often unreliable and lack interpretability. To address this problem, we
propose incorporating uncertainty estimation into the MNER task, producing
trustworthy predictions. Our proposed algorithm models the distribution of each
modality as a Normal-inverse Gamma distribution, and fuses them into a unified
distribution with an evidential fusion mechanism, enabling hierarchical
characterization of uncertainties and promotion of prediction accuracy and
trustworthiness. Additionally, we explore the potential of pre-trained large
foundation models in MNER and propose an efficient fusion approach that
leverages their robust feature representations. Experiments on two datasets
demonstrate that our proposed method outperforms the baselines and achieves new
state-of-the-art performance.
| no_new_dataset | 0.709202 |
2306.17000 | Ce Zhang Dr. | Ce Zhang, Chengjie Zhang, Yiluan Guo, Lingji Chen, Michael Happold | MotionTrack: End-to-End Transformer-based Multi-Object Tracing with
LiDAR-Camera Fusion | This paper is accepted by CVPR WAD 2023 | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Multiple Object Tracking (MOT) is crucial to autonomous vehicle perception.
End-to-end transformer-based algorithms, which detect and track objects
simultaneously, show great potential for the MOT task. However, most existing
methods focus on image-based tracking with a single object category. In this
paper, we propose an end-to-end transformer-based MOT algorithm (MotionTrack)
with multi-modality sensor inputs to track objects with multiple classes. Our
objective is to establish a transformer baseline for the MOT in an autonomous
driving environment. The proposed algorithm consists of a transformer-based
data association (DA) module and a transformer-based query enhancement module
to achieve MOT and Multiple Object Detection (MOD) simultaneously. The
MotionTrack and its variations achieve better results (AMOTA score at 0.55) on
the nuScenes dataset compared with other classical baseline models, such as the
AB3DMOT, the CenterTrack, and the probabilistic 3D Kalman filter. In addition,
we prove that a modified attention mechanism can be utilized for DA to
accomplish the MOT, and aggregate history features to enhance the MOD
performance.
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 15:00:12 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Zhang",
"Ce",
""
],
[
"Zhang",
"Chengjie",
""
],
[
"Guo",
"Yiluan",
""
],
[
"Chen",
"Lingji",
""
],
[
"Happold",
"Michael",
""
]
] | TITLE: MotionTrack: End-to-End Transformer-based Multi-Object Tracing with
LiDAR-Camera Fusion
ABSTRACT: Multiple Object Tracking (MOT) is crucial to autonomous vehicle perception.
End-to-end transformer-based algorithms, which detect and track objects
simultaneously, show great potential for the MOT task. However, most existing
methods focus on image-based tracking with a single object category. In this
paper, we propose an end-to-end transformer-based MOT algorithm (MotionTrack)
with multi-modality sensor inputs to track objects with multiple classes. Our
objective is to establish a transformer baseline for the MOT in an autonomous
driving environment. The proposed algorithm consists of a transformer-based
data association (DA) module and a transformer-based query enhancement module
to achieve MOT and Multiple Object Detection (MOD) simultaneously. The
MotionTrack and its variations achieve better results (AMOTA score at 0.55) on
the nuScenes dataset compared with other classical baseline models, such as the
AB3DMOT, the CenterTrack, and the probabilistic 3D Kalman filter. In addition,
we prove that a modified attention mechanism can be utilized for DA to
accomplish the MOT, and aggregate history features to enhance the MOD
performance.
| no_new_dataset | 0.709982 |
2306.17034 | Tao He | Tao He, Ming Liu, Yixin Cao, Zekun Wang, Zihao Zheng, Zheng Chu, and
Bing Qin | Exploring & Exploiting High-Order Graph Structure for Sparse Knowledge
Graph Completion | 12 pages, 5 figures | null | null | null | cs.AI cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Sparse knowledge graph (KG) scenarios pose a challenge for previous Knowledge
Graph Completion (KGC) methods, that is, the completion performance decreases
rapidly with the increase of graph sparsity. This problem is also exacerbated
because of the widespread existence of sparse KGs in practical applications. To
alleviate this challenge, we present a novel framework, LR-GCN, that is able to
automatically capture valuable long-range dependency among entities to
supplement insufficient structure features and distill logical reasoning
knowledge for sparse KGC. The proposed approach comprises two main components:
a GNN-based predictor and a reasoning path distiller. The reasoning path
distiller explores high-order graph structures such as reasoning paths and
encodes them as rich-semantic edges, explicitly compositing long-range
dependencies into the predictor. This step also plays an essential role in
densifying KGs, effectively alleviating the sparse issue. Furthermore, the path
distiller further distills logical reasoning knowledge from these mined
reasoning paths into the predictor. These two components are jointly optimized
using a well-designed variational EM algorithm. Extensive experiments and
analyses on four sparse benchmarks demonstrate the effectiveness of our
proposed method.
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 15:35:34 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"He",
"Tao",
""
],
[
"Liu",
"Ming",
""
],
[
"Cao",
"Yixin",
""
],
[
"Wang",
"Zekun",
""
],
[
"Zheng",
"Zihao",
""
],
[
"Chu",
"Zheng",
""
],
[
"Qin",
"Bing",
""
]
] | TITLE: Exploring & Exploiting High-Order Graph Structure for Sparse Knowledge
Graph Completion
ABSTRACT: Sparse knowledge graph (KG) scenarios pose a challenge for previous Knowledge
Graph Completion (KGC) methods, that is, the completion performance decreases
rapidly with the increase of graph sparsity. This problem is also exacerbated
because of the widespread existence of sparse KGs in practical applications. To
alleviate this challenge, we present a novel framework, LR-GCN, that is able to
automatically capture valuable long-range dependency among entities to
supplement insufficient structure features and distill logical reasoning
knowledge for sparse KGC. The proposed approach comprises two main components:
a GNN-based predictor and a reasoning path distiller. The reasoning path
distiller explores high-order graph structures such as reasoning paths and
encodes them as rich-semantic edges, explicitly compositing long-range
dependencies into the predictor. This step also plays an essential role in
densifying KGs, effectively alleviating the sparse issue. Furthermore, the path
distiller further distills logical reasoning knowledge from these mined
reasoning paths into the predictor. These two components are jointly optimized
using a well-designed variational EM algorithm. Extensive experiments and
analyses on four sparse benchmarks demonstrate the effectiveness of our
proposed method.
| no_new_dataset | 0.70883 |
2306.17062 | Nabeel Bhat | Nabeel Nisar Bhat, Rafael Berkvens, Jeroen Famaey | Gesture Recognition with mmWave Wi-Fi Access Points: Lessons Learned | null | null | null | null | cs.LG eess.SP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In recent years, channel state information (CSI) at sub-6 GHz has been widely
exploited for Wi-Fi sensing, particularly for activity and gesture recognition.
In this work, we instead explore mmWave (60 GHz) Wi-Fi signals for gesture
recognition/pose estimation. Our focus is on the mmWave Wi-Fi signals so that
they can be used not only for high data rate communication but also for
improved sensing e.g., for extended reality (XR) applications. For this reason,
we extract spatial beam signal-to-noise ratios (SNRs) from the periodic beam
training employed by IEEE 802.11ad devices. We consider a set of 10
gestures/poses motivated by XR applications. We conduct experiments in two
environments and with three people.As a comparison, we also collect CSI from
IEEE 802.11ac devices. To extract features from the CSI and the beam SNR, we
leverage a deep neural network (DNN). The DNN classifier achieves promising
results on the beam SNR task with state-of-the-art 96.7% accuracy in a single
environment, even with a limited dataset. We also investigate the robustness of
the beam SNR against CSI across different environments. Our experiments reveal
that features from the CSI generalize without additional re-training, while
those from beam SNRs do not. Therefore, re-training is required in the latter
case.
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 16:10:07 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Bhat",
"Nabeel Nisar",
""
],
[
"Berkvens",
"Rafael",
""
],
[
"Famaey",
"Jeroen",
""
]
] | TITLE: Gesture Recognition with mmWave Wi-Fi Access Points: Lessons Learned
ABSTRACT: In recent years, channel state information (CSI) at sub-6 GHz has been widely
exploited for Wi-Fi sensing, particularly for activity and gesture recognition.
In this work, we instead explore mmWave (60 GHz) Wi-Fi signals for gesture
recognition/pose estimation. Our focus is on the mmWave Wi-Fi signals so that
they can be used not only for high data rate communication but also for
improved sensing e.g., for extended reality (XR) applications. For this reason,
we extract spatial beam signal-to-noise ratios (SNRs) from the periodic beam
training employed by IEEE 802.11ad devices. We consider a set of 10
gestures/poses motivated by XR applications. We conduct experiments in two
environments and with three people.As a comparison, we also collect CSI from
IEEE 802.11ac devices. To extract features from the CSI and the beam SNR, we
leverage a deep neural network (DNN). The DNN classifier achieves promising
results on the beam SNR task with state-of-the-art 96.7% accuracy in a single
environment, even with a limited dataset. We also investigate the robustness of
the beam SNR against CSI across different environments. Our experiments reveal
that features from the CSI generalize without additional re-training, while
those from beam SNRs do not. Therefore, re-training is required in the latter
case.
| no_new_dataset | 0.712063 |
2306.17074 | Zhongwei Qiu | Zhongwei Qiu, Qiansheng Yang, Jian Wang, Xiyu Wang, Chang Xu, Dongmei
Fu, Kun Yao, Junyu Han, Errui Ding, Jingdong Wang | Learning Structure-Guided Diffusion Model for 2D Human Pose Estimation | null | null | null | null | cs.CV cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | One of the mainstream schemes for 2D human pose estimation (HPE) is learning
keypoints heatmaps by a neural network. Existing methods typically improve the
quality of heatmaps by customized architectures, such as high-resolution
representation and vision Transformers. In this paper, we propose
\textbf{DiffusionPose}, a new scheme that formulates 2D HPE as a keypoints
heatmaps generation problem from noised heatmaps. During training, the
keypoints are diffused to random distribution by adding noises and the
diffusion model learns to recover ground-truth heatmaps from noised heatmaps
with respect to conditions constructed by image feature. During inference, the
diffusion model generates heatmaps from initialized heatmaps in a progressive
denoising way. Moreover, we further explore improving the performance of
DiffusionPose with conditions from human structural information. Extensive
experiments show the prowess of our DiffusionPose, with improvements of 1.6,
1.2, and 1.2 mAP on widely-used COCO, CrowdPose, and AI Challenge datasets,
respectively.
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 16:24:32 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Qiu",
"Zhongwei",
""
],
[
"Yang",
"Qiansheng",
""
],
[
"Wang",
"Jian",
""
],
[
"Wang",
"Xiyu",
""
],
[
"Xu",
"Chang",
""
],
[
"Fu",
"Dongmei",
""
],
[
"Yao",
"Kun",
""
],
[
"Han",
"Junyu",
""
],
[
"Ding",
"Errui",
""
],
[
"Wang",
"Jingdong",
""
]
] | TITLE: Learning Structure-Guided Diffusion Model for 2D Human Pose Estimation
ABSTRACT: One of the mainstream schemes for 2D human pose estimation (HPE) is learning
keypoints heatmaps by a neural network. Existing methods typically improve the
quality of heatmaps by customized architectures, such as high-resolution
representation and vision Transformers. In this paper, we propose
\textbf{DiffusionPose}, a new scheme that formulates 2D HPE as a keypoints
heatmaps generation problem from noised heatmaps. During training, the
keypoints are diffused to random distribution by adding noises and the
diffusion model learns to recover ground-truth heatmaps from noised heatmaps
with respect to conditions constructed by image feature. During inference, the
diffusion model generates heatmaps from initialized heatmaps in a progressive
denoising way. Moreover, we further explore improving the performance of
DiffusionPose with conditions from human structural information. Extensive
experiments show the prowess of our DiffusionPose, with improvements of 1.6,
1.2, and 1.2 mAP on widely-used COCO, CrowdPose, and AI Challenge datasets,
respectively.
| no_new_dataset | 0.708408 |
2306.17075 | Zitong Yu | Yingxin Lai, Zhiming Luo, Zitong Yu | Detect Any Deepfakes: Segment Anything Meets Face Forgery Detection and
Localization | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The rapid advancements in computer vision have stimulated remarkable progress
in face forgery techniques, capturing the dedicated attention of researchers
committed to detecting forgeries and precisely localizing manipulated areas.
Nonetheless, with limited fine-grained pixel-wise supervision labels, deepfake
detection models perform unsatisfactorily on precise forgery detection and
localization. To address this challenge, we introduce the well-trained vision
segmentation foundation model, i.e., Segment Anything Model (SAM) in face
forgery detection and localization. Based on SAM, we propose the Detect Any
Deepfakes (DADF) framework with the Multiscale Adapter, which can capture
short- and long-range forgery contexts for efficient fine-tuning. Moreover, to
better identify forged traces and augment the model's sensitivity towards
forgery regions, Reconstruction Guided Attention (RGA) module is proposed. The
proposed framework seamlessly integrates end-to-end forgery localization and
detection optimization. Extensive experiments on three benchmark datasets
demonstrate the superiority of our approach for both forgery detection and
localization. The codes will be released soon at
https://github.com/laiyingxin2/DADF.
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 16:25:04 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Lai",
"Yingxin",
""
],
[
"Luo",
"Zhiming",
""
],
[
"Yu",
"Zitong",
""
]
] | TITLE: Detect Any Deepfakes: Segment Anything Meets Face Forgery Detection and
Localization
ABSTRACT: The rapid advancements in computer vision have stimulated remarkable progress
in face forgery techniques, capturing the dedicated attention of researchers
committed to detecting forgeries and precisely localizing manipulated areas.
Nonetheless, with limited fine-grained pixel-wise supervision labels, deepfake
detection models perform unsatisfactorily on precise forgery detection and
localization. To address this challenge, we introduce the well-trained vision
segmentation foundation model, i.e., Segment Anything Model (SAM) in face
forgery detection and localization. Based on SAM, we propose the Detect Any
Deepfakes (DADF) framework with the Multiscale Adapter, which can capture
short- and long-range forgery contexts for efficient fine-tuning. Moreover, to
better identify forged traces and augment the model's sensitivity towards
forgery regions, Reconstruction Guided Attention (RGA) module is proposed. The
proposed framework seamlessly integrates end-to-end forgery localization and
detection optimization. Extensive experiments on three benchmark datasets
demonstrate the superiority of our approach for both forgery detection and
localization. The codes will be released soon at
https://github.com/laiyingxin2/DADF.
| no_new_dataset | 0.708811 |
2306.17090 | Truong Son Hy | Ngoc-Dung Do, Truong Son Hy, Duy Khuong Nguyen | Sparsity exploitation via discovering graphical models in multi-variate
time-series forecasting | null | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | Graph neural networks (GNNs) have been widely applied in multi-variate
time-series forecasting (MTSF) tasks because of their capability in capturing
the correlations among different time-series. These graph-based learning
approaches improve the forecasting performance by discovering and understanding
the underlying graph structures, which represent the data correlation. When the
explicit prior graph structures are not available, most existing works cannot
guarantee the sparsity of the generated graphs that make the overall model
computational expensive and less interpretable. In this work, we propose a
decoupled training method, which includes a graph generating module and a GNNs
forecasting module. First, we use Graphical Lasso (or GraphLASSO) to directly
exploit the sparsity pattern from data to build graph structures in both static
and time-varying cases. Second, we fit these graph structures and the input
data into a Graph Convolutional Recurrent Network (GCRN) to train a forecasting
model. The experimental results on three real-world datasets show that our
novel approach has competitive performance against existing state-of-the-art
forecasting algorithms while providing sparse, meaningful and explainable graph
structures and reducing training time by approximately 40%. Our PyTorch
implementation is publicly available at https://github.com/HySonLab/GraphLASSO
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 16:48:00 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Do",
"Ngoc-Dung",
""
],
[
"Hy",
"Truong Son",
""
],
[
"Nguyen",
"Duy Khuong",
""
]
] | TITLE: Sparsity exploitation via discovering graphical models in multi-variate
time-series forecasting
ABSTRACT: Graph neural networks (GNNs) have been widely applied in multi-variate
time-series forecasting (MTSF) tasks because of their capability in capturing
the correlations among different time-series. These graph-based learning
approaches improve the forecasting performance by discovering and understanding
the underlying graph structures, which represent the data correlation. When the
explicit prior graph structures are not available, most existing works cannot
guarantee the sparsity of the generated graphs that make the overall model
computational expensive and less interpretable. In this work, we propose a
decoupled training method, which includes a graph generating module and a GNNs
forecasting module. First, we use Graphical Lasso (or GraphLASSO) to directly
exploit the sparsity pattern from data to build graph structures in both static
and time-varying cases. Second, we fit these graph structures and the input
data into a Graph Convolutional Recurrent Network (GCRN) to train a forecasting
model. The experimental results on three real-world datasets show that our
novel approach has competitive performance against existing state-of-the-art
forecasting algorithms while providing sparse, meaningful and explainable graph
structures and reducing training time by approximately 40%. Our PyTorch
implementation is publicly available at https://github.com/HySonLab/GraphLASSO
| no_new_dataset | 0.709831 |
2306.17101 | Wanming Yu | Wanming Yu, Chuanyu Yang, Christopher McGreavy, Eleftherios
Triantafyllidis, Guillaume Bellegarda, Milad Shafiee, Auke Jan Ijspeert, and
Zhibin Li | Identifying Important Sensory Feedback for Learning Locomotion Skills | null | null | null | null | cs.RO cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Robot motor skills can be learned through deep reinforcement learning (DRL)
by neural networks as state-action mappings. While the selection of state
observations is crucial, there has been a lack of quantitative analysis to
date. Here, we present a systematic saliency analysis that quantitatively
evaluates the relative importance of different feedback states for motor skills
learned through DRL. Our approach can identify the most essential feedback
states for locomotion skills, including balance recovery, trotting, bounding,
pacing and galloping. By using only key states including joint positions,
gravity vector, base linear and angular velocities, we demonstrate that a
simulated quadruped robot can achieve robust performance in various test
scenarios across these distinct skills. The benchmarks using task performance
metrics show that locomotion skills learned with key states can achieve
comparable performance to those with all states, and the task performance or
learning success rate will drop significantly if key states are missing. This
work provides quantitative insights into the relationship between state
observations and specific types of motor skills, serving as a guideline for
robot motor learning. The proposed method is applicable to differentiable
state-action mapping, such as neural network based control policies, enabling
the learning of a wide range of motor skills with minimal sensing dependencies.
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 16:58:08 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Yu",
"Wanming",
""
],
[
"Yang",
"Chuanyu",
""
],
[
"McGreavy",
"Christopher",
""
],
[
"Triantafyllidis",
"Eleftherios",
""
],
[
"Bellegarda",
"Guillaume",
""
],
[
"Shafiee",
"Milad",
""
],
[
"Ijspeert",
"Auke Jan",
""
],
[
"Li",
"Zhibin",
""
]
] | TITLE: Identifying Important Sensory Feedback for Learning Locomotion Skills
ABSTRACT: Robot motor skills can be learned through deep reinforcement learning (DRL)
by neural networks as state-action mappings. While the selection of state
observations is crucial, there has been a lack of quantitative analysis to
date. Here, we present a systematic saliency analysis that quantitatively
evaluates the relative importance of different feedback states for motor skills
learned through DRL. Our approach can identify the most essential feedback
states for locomotion skills, including balance recovery, trotting, bounding,
pacing and galloping. By using only key states including joint positions,
gravity vector, base linear and angular velocities, we demonstrate that a
simulated quadruped robot can achieve robust performance in various test
scenarios across these distinct skills. The benchmarks using task performance
metrics show that locomotion skills learned with key states can achieve
comparable performance to those with all states, and the task performance or
learning success rate will drop significantly if key states are missing. This
work provides quantitative insights into the relationship between state
observations and specific types of motor skills, serving as a guideline for
robot motor learning. The proposed method is applicable to differentiable
state-action mapping, such as neural network based control policies, enabling
the learning of a wide range of motor skills with minimal sensing dependencies.
| no_new_dataset | 0.71039 |
2306.17116 | Piotr W\'ojcik | Piotr W\'ojcik, Hussein Naji, Adrian Simon, Reinhard B\"uttner,
Katarzyna Bo\.zek | Learning Nuclei Representations with Masked Image Modelling | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Masked image modelling (MIM) is a powerful self-supervised representation
learning paradigm, whose potential has not been widely demonstrated in medical
image analysis. In this work, we show the capacity of MIM to capture rich
semantic representations of Haemotoxylin & Eosin (H&E)-stained images at the
nuclear level. Inspired by Bidirectional Encoder representation from Image
Transformers (BEiT), we split the images into smaller patches and generate
corresponding discrete visual tokens. In addition to the regular grid-based
patches, typically used in visual Transformers, we introduce patches of
individual cell nuclei. We propose positional encoding of the irregular
distribution of these structures within an image. We pre-train the model in a
self-supervised manner on H&E-stained whole-slide images of diffuse large
B-cell lymphoma, where cell nuclei have been segmented. The pre-training
objective is to recover the original discrete visual tokens of the masked image
on the one hand, and to reconstruct the visual tokens of the masked object
instances on the other. Coupling these two pre-training tasks allows us to
build powerful, context-aware representations of nuclei. Our model generalizes
well and can be fine-tuned on downstream classification tasks, achieving
improved cell classification accuracy on PanNuke dataset by more than 5%
compared to current instance segmentation methods.
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 17:20:05 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Wójcik",
"Piotr",
""
],
[
"Naji",
"Hussein",
""
],
[
"Simon",
"Adrian",
""
],
[
"Büttner",
"Reinhard",
""
],
[
"Bożek",
"Katarzyna",
""
]
] | TITLE: Learning Nuclei Representations with Masked Image Modelling
ABSTRACT: Masked image modelling (MIM) is a powerful self-supervised representation
learning paradigm, whose potential has not been widely demonstrated in medical
image analysis. In this work, we show the capacity of MIM to capture rich
semantic representations of Haemotoxylin & Eosin (H&E)-stained images at the
nuclear level. Inspired by Bidirectional Encoder representation from Image
Transformers (BEiT), we split the images into smaller patches and generate
corresponding discrete visual tokens. In addition to the regular grid-based
patches, typically used in visual Transformers, we introduce patches of
individual cell nuclei. We propose positional encoding of the irregular
distribution of these structures within an image. We pre-train the model in a
self-supervised manner on H&E-stained whole-slide images of diffuse large
B-cell lymphoma, where cell nuclei have been segmented. The pre-training
objective is to recover the original discrete visual tokens of the masked image
on the one hand, and to reconstruct the visual tokens of the masked object
instances on the other. Coupling these two pre-training tasks allows us to
build powerful, context-aware representations of nuclei. Our model generalizes
well and can be fine-tuned on downstream classification tasks, achieving
improved cell classification accuracy on PanNuke dataset by more than 5%
compared to current instance segmentation methods.
| no_new_dataset | 0.711193 |
2306.17123 | Kai-En Lin | Kai-En Lin and Alex Trevithick and Keli Cheng and Michel Sarkis and
Mohsen Ghafoorian and Ning Bi and Gerhard Reitmayr and Ravi Ramamoorthi | PVP: Personalized Video Prior for Editable Dynamic Portraits using
StyleGAN | Project website:
https://cseweb.ucsd.edu//~viscomp/projects/EGSR23PVP/ | null | null | null | cs.CV cs.GR | http://creativecommons.org/licenses/by/4.0/ | Portrait synthesis creates realistic digital avatars which enable users to
interact with others in a compelling way. Recent advances in StyleGAN and its
extensions have shown promising results in synthesizing photorealistic and
accurate reconstruction of human faces. However, previous methods often focus
on frontal face synthesis and most methods are not able to handle large head
rotations due to the training data distribution of StyleGAN. In this work, our
goal is to take as input a monocular video of a face, and create an editable
dynamic portrait able to handle extreme head poses. The user can create novel
viewpoints, edit the appearance, and animate the face. Our method utilizes
pivotal tuning inversion (PTI) to learn a personalized video prior from a
monocular video sequence. Then we can input pose and expression coefficients to
MLPs and manipulate the latent vectors to synthesize different viewpoints and
expressions of the subject. We also propose novel loss functions to further
disentangle pose and expression in the latent space. Our algorithm shows much
better performance over previous approaches on monocular video datasets, and it
is also capable of running in real-time at 54 FPS on an RTX 3080.
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 17:26:51 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Lin",
"Kai-En",
""
],
[
"Trevithick",
"Alex",
""
],
[
"Cheng",
"Keli",
""
],
[
"Sarkis",
"Michel",
""
],
[
"Ghafoorian",
"Mohsen",
""
],
[
"Bi",
"Ning",
""
],
[
"Reitmayr",
"Gerhard",
""
],
[
"Ramamoorthi",
"Ravi",
""
]
] | TITLE: PVP: Personalized Video Prior for Editable Dynamic Portraits using
StyleGAN
ABSTRACT: Portrait synthesis creates realistic digital avatars which enable users to
interact with others in a compelling way. Recent advances in StyleGAN and its
extensions have shown promising results in synthesizing photorealistic and
accurate reconstruction of human faces. However, previous methods often focus
on frontal face synthesis and most methods are not able to handle large head
rotations due to the training data distribution of StyleGAN. In this work, our
goal is to take as input a monocular video of a face, and create an editable
dynamic portrait able to handle extreme head poses. The user can create novel
viewpoints, edit the appearance, and animate the face. Our method utilizes
pivotal tuning inversion (PTI) to learn a personalized video prior from a
monocular video sequence. Then we can input pose and expression coefficients to
MLPs and manipulate the latent vectors to synthesize different viewpoints and
expressions of the subject. We also propose novel loss functions to further
disentangle pose and expression in the latent space. Our algorithm shows much
better performance over previous approaches on monocular video datasets, and it
is also capable of running in real-time at 54 FPS on an RTX 3080.
| no_new_dataset | 0.709585 |
2306.17138 | Nishchal Dwivedi | Yashvir Tiberwal and Nishchal Dwivedi | Orbit Classification of asteroids using implementation of radial Basis
Function on Support Vector Machines | 15 pages, 8 figures, 3 tables | null | null | null | astro-ph.EP astro-ph.IM cs.LG physics.data-an | http://creativecommons.org/licenses/by/4.0/ | This research paper focuses on the implementation of radial Basis Function
(RBF) Support Vector Machines (SVM) for classifying asteroid orbits. Asteroids
are important astronomical objects, and their orbits play a crucial role in
understanding the dynamics of the solar system. The International Astronomical
Union maintains data archives that provide a playground to experiment with
various machine-learning techniques. In this study, we explore the application
of RBF SVM algorithm to classify asteroids. The results show that the RBF SVM
algorithm provides a good efficiency and accuracy to the dataset. We also
analyze the impact of various parameters on the performance of the RBF SVM
algorithm and present the optimal parameter settings. Our study highlights the
importance of using machine learning techniques for classifying asteroid orbits
and the effectiveness of the RBF SVM algorithm in this regard.
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 17:39:01 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Tiberwal",
"Yashvir",
""
],
[
"Dwivedi",
"Nishchal",
""
]
] | TITLE: Orbit Classification of asteroids using implementation of radial Basis
Function on Support Vector Machines
ABSTRACT: This research paper focuses on the implementation of radial Basis Function
(RBF) Support Vector Machines (SVM) for classifying asteroid orbits. Asteroids
are important astronomical objects, and their orbits play a crucial role in
understanding the dynamics of the solar system. The International Astronomical
Union maintains data archives that provide a playground to experiment with
various machine-learning techniques. In this study, we explore the application
of RBF SVM algorithm to classify asteroids. The results show that the RBF SVM
algorithm provides a good efficiency and accuracy to the dataset. We also
analyze the impact of various parameters on the performance of the RBF SVM
algorithm and present the optimal parameter settings. Our study highlights the
importance of using machine learning techniques for classifying asteroid orbits
and the effectiveness of the RBF SVM algorithm in this regard.
| no_new_dataset | 0.71262 |
2306.17162 | Simeon Adebola | Simeon Adebola, Rishi Parikh, Mark Presten, Satvik Sharma, Shrey
Aeron, Ananth Rao, Sandeep Mukherjee, Tomson Qu, Christina Wistrom, Eugen
Solowjow, Ken Goldberg | Can Machines Garden? Systematically Comparing the AlphaGarden vs.
Professional Horticulturalists | International Conference on Robotics and Automation(ICRA) 2023 Oral | null | null | null | cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The AlphaGarden is an automated testbed for indoor polyculture farming which
combines a first-order plant simulator, a gantry robot, a seed planting
algorithm, plant phenotyping and tracking algorithms, irrigation sensors and
algorithms, and custom pruning tools and algorithms. In this paper, we
systematically compare the performance of the AlphaGarden to professional
horticulturalists on the staff of the UC Berkeley Oxford Tract Greenhouse. The
humans and the machine tend side-by-side polyculture gardens with the same seed
arrangement. We compare performance in terms of canopy coverage, plant
diversity, and water consumption. Results from two 60-day cycles suggest that
the automated AlphaGarden performs comparably to professional horticulturalists
in terms of coverage and diversity, and reduces water consumption by as much as
44%. Code, videos, and datasets are available at
https://sites.google.com/berkeley.edu/systematiccomparison.
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 17:59:05 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Adebola",
"Simeon",
""
],
[
"Parikh",
"Rishi",
""
],
[
"Presten",
"Mark",
""
],
[
"Sharma",
"Satvik",
""
],
[
"Aeron",
"Shrey",
""
],
[
"Rao",
"Ananth",
""
],
[
"Mukherjee",
"Sandeep",
""
],
[
"Qu",
"Tomson",
""
],
[
"Wistrom",
"Christina",
""
],
[
"Solowjow",
"Eugen",
""
],
[
"Goldberg",
"Ken",
""
]
] | TITLE: Can Machines Garden? Systematically Comparing the AlphaGarden vs.
Professional Horticulturalists
ABSTRACT: The AlphaGarden is an automated testbed for indoor polyculture farming which
combines a first-order plant simulator, a gantry robot, a seed planting
algorithm, plant phenotyping and tracking algorithms, irrigation sensors and
algorithms, and custom pruning tools and algorithms. In this paper, we
systematically compare the performance of the AlphaGarden to professional
horticulturalists on the staff of the UC Berkeley Oxford Tract Greenhouse. The
humans and the machine tend side-by-side polyculture gardens with the same seed
arrangement. We compare performance in terms of canopy coverage, plant
diversity, and water consumption. Results from two 60-day cycles suggest that
the automated AlphaGarden performs comparably to professional horticulturalists
in terms of coverage and diversity, and reduces water consumption by as much as
44%. Code, videos, and datasets are available at
https://sites.google.com/berkeley.edu/systematiccomparison.
| no_new_dataset | 0.712263 |
2306.17165 | Chuang Gan | Zitian Chen, Mingyu Ding, Yikang Shen, Wei Zhan, Masayoshi Tomizuka,
Erik Learned-Miller, Chuang Gan | An Efficient General-Purpose Modular Vision Model via Multi-Task
Heterogeneous Training | null | null | null | null | cs.CV cs.AI cs.LG | http://creativecommons.org/publicdomain/zero/1.0/ | We present a model that can perform multiple vision tasks and can be adapted
to other downstream tasks efficiently. Despite considerable progress in
multi-task learning, most efforts focus on learning from multi-label data: a
single image set with multiple task labels. Such multi-label data sets are
rare, small, and expensive. We say heterogeneous to refer to image sets with
different task labels, or to combinations of single-task datasets. Few have
explored training on such heterogeneous datasets. General-purpose vision models
are still dominated by single-task pretraining, and it remains unclear how to
scale up multi-task models by leveraging mainstream vision datasets designed
for different purposes. The challenges lie in managing large intrinsic
differences among vision tasks, including data distribution, architectures,
task-specific modules, dataset scales, and sampling strategies. To address
these challenges, we propose to modify and scale up mixture-of-experts (MoE)
vision transformers, so that they can simultaneously learn classification,
detection, and segmentation on diverse mainstream vision datasets including
ImageNet, COCO, and ADE20K. Our approach achieves comparable results to
single-task state-of-the-art models and demonstrates strong generalization on
downstream tasks. Due to its emergent modularity, this general-purpose model
decomposes into high-performing components, efficiently adapting to downstream
tasks. We can fine-tune it with fewer training parameters, fewer model
parameters, and less computation. Additionally, its modularity allows for easy
expansion in continual-learning-without-forgetting scenarios. Finally, these
functions can be controlled and combined to meet various demands of downstream
tasks.
| [
{
"version": "v1",
"created": "Thu, 29 Jun 2023 17:59:57 GMT"
}
] | 2023-06-30T00:00:00 | [
[
"Chen",
"Zitian",
""
],
[
"Ding",
"Mingyu",
""
],
[
"Shen",
"Yikang",
""
],
[
"Zhan",
"Wei",
""
],
[
"Tomizuka",
"Masayoshi",
""
],
[
"Learned-Miller",
"Erik",
""
],
[
"Gan",
"Chuang",
""
]
] | TITLE: An Efficient General-Purpose Modular Vision Model via Multi-Task
Heterogeneous Training
ABSTRACT: We present a model that can perform multiple vision tasks and can be adapted
to other downstream tasks efficiently. Despite considerable progress in
multi-task learning, most efforts focus on learning from multi-label data: a
single image set with multiple task labels. Such multi-label data sets are
rare, small, and expensive. We say heterogeneous to refer to image sets with
different task labels, or to combinations of single-task datasets. Few have
explored training on such heterogeneous datasets. General-purpose vision models
are still dominated by single-task pretraining, and it remains unclear how to
scale up multi-task models by leveraging mainstream vision datasets designed
for different purposes. The challenges lie in managing large intrinsic
differences among vision tasks, including data distribution, architectures,
task-specific modules, dataset scales, and sampling strategies. To address
these challenges, we propose to modify and scale up mixture-of-experts (MoE)
vision transformers, so that they can simultaneously learn classification,
detection, and segmentation on diverse mainstream vision datasets including
ImageNet, COCO, and ADE20K. Our approach achieves comparable results to
single-task state-of-the-art models and demonstrates strong generalization on
downstream tasks. Due to its emergent modularity, this general-purpose model
decomposes into high-performing components, efficiently adapting to downstream
tasks. We can fine-tune it with fewer training parameters, fewer model
parameters, and less computation. Additionally, its modularity allows for easy
expansion in continual-learning-without-forgetting scenarios. Finally, these
functions can be controlled and combined to meet various demands of downstream
tasks.
| no_new_dataset | 0.705823 |
1909.03354 | Soufiane Belharbi | J\'er\^ome Rony, Soufiane Belharbi, Jose Dolz, Ismail Ben Ayed, Luke
McCaffrey, Eric Granger | Deep Weakly-Supervised Learning Methods for Classification and
Localization in Histology Images: A Survey | Accepted for publication at the Journal of Machine Learning for
Biomedical Imaging (MELBA) https://melba-journal.org/2023:004 | Machine.Learning.for.Biomedical.Imaging. 2 (2023) | 10.59275/j.melba.2023-5g54 | null | cs.CV cs.LG eess.IV | http://creativecommons.org/licenses/by/4.0/ | Using deep learning models to diagnose cancer from histology data presents
several challenges. Cancer grading and localization of regions of interest
(ROIs) in these images normally relies on both image- and pixel-level labels,
the latter requiring a costly annotation process. Deep weakly-supervised object
localization (WSOL) methods provide different strategies for low-cost training
of deep learning models. Using only image-class annotations, these methods can
be trained to classify an image, and yield class activation maps (CAMs) for ROI
localization. This paper provides a review of state-of-art DL methods for WSOL.
We propose a taxonomy where these methods are divided into bottom-up and
top-down methods according to the information flow in models. Although the
latter have seen limited progress, recent bottom-up methods are currently
driving much progress with deep WSOL methods. Early works focused on designing
different spatial pooling functions. However, these methods reached limited
localization accuracy, and unveiled a major limitation -- the under-activation
of CAMs which leads to high false negative localization. Subsequent works aimed
to alleviate this issue and recover complete object. Representative methods
from our taxonomy are evaluated and compared in terms of classification and
localization accuracy on two challenging histology datasets. Overall, the
results indicate poor localization performance, particularly for generic
methods that were initially designed to process natural images. Methods
designed to address the challenges of histology data yielded good results.
However, all methods suffer from high false positive/negative localization.
Four key challenges are identified for the application of deep WSOL methods in
histology -- under/over activation of CAMs, sensitivity to thresholding, and
model selection.
| [
{
"version": "v1",
"created": "Sun, 8 Sep 2019 00:01:37 GMT"
},
{
"version": "v2",
"created": "Thu, 26 Sep 2019 14:12:21 GMT"
},
{
"version": "v3",
"created": "Mon, 18 May 2020 04:16:09 GMT"
},
{
"version": "v4",
"created": "Thu, 31 Dec 2020 05:41:51 GMT"
},
{
"version": "v5",
"created": "Tue, 26 Apr 2022 00:12:01 GMT"
},
{
"version": "v6",
"created": "Mon, 18 Jul 2022 04:15:23 GMT"
},
{
"version": "v7",
"created": "Fri, 3 Mar 2023 16:04:31 GMT"
}
] | 2023-06-29T00:00:00 | [
[
"Rony",
"Jérôme",
""
],
[
"Belharbi",
"Soufiane",
""
],
[
"Dolz",
"Jose",
""
],
[
"Ayed",
"Ismail Ben",
""
],
[
"McCaffrey",
"Luke",
""
],
[
"Granger",
"Eric",
""
]
] | TITLE: Deep Weakly-Supervised Learning Methods for Classification and
Localization in Histology Images: A Survey
ABSTRACT: Using deep learning models to diagnose cancer from histology data presents
several challenges. Cancer grading and localization of regions of interest
(ROIs) in these images normally relies on both image- and pixel-level labels,
the latter requiring a costly annotation process. Deep weakly-supervised object
localization (WSOL) methods provide different strategies for low-cost training
of deep learning models. Using only image-class annotations, these methods can
be trained to classify an image, and yield class activation maps (CAMs) for ROI
localization. This paper provides a review of state-of-art DL methods for WSOL.
We propose a taxonomy where these methods are divided into bottom-up and
top-down methods according to the information flow in models. Although the
latter have seen limited progress, recent bottom-up methods are currently
driving much progress with deep WSOL methods. Early works focused on designing
different spatial pooling functions. However, these methods reached limited
localization accuracy, and unveiled a major limitation -- the under-activation
of CAMs which leads to high false negative localization. Subsequent works aimed
to alleviate this issue and recover complete object. Representative methods
from our taxonomy are evaluated and compared in terms of classification and
localization accuracy on two challenging histology datasets. Overall, the
results indicate poor localization performance, particularly for generic
methods that were initially designed to process natural images. Methods
designed to address the challenges of histology data yielded good results.
However, all methods suffer from high false positive/negative localization.
Four key challenges are identified for the application of deep WSOL methods in
histology -- under/over activation of CAMs, sensitivity to thresholding, and
model selection.
| no_new_dataset | 0.710823 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.