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
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2207.00099 | Matthew Jagielski | Matthew Jagielski, Om Thakkar, Florian Tram\`er, Daphne Ippolito,
Katherine Lee, Nicholas Carlini, Eric Wallace, Shuang Song, Abhradeep
Thakurta, Nicolas Papernot, Chiyuan Zhang | Measuring Forgetting of Memorized Training Examples | Appeared at ICLR '23, 22 pages, 12 figures | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Machine learning models exhibit two seemingly contradictory phenomena:
training data memorization, and various forms of forgetting. In memorization,
models overfit specific training examples and become susceptible to privacy
attacks. In forgetting, examples which appeared early in training are forgotten
by the end. In this work, we connect these phenomena. We propose a technique to
measure to what extent models "forget" the specifics of training examples,
becoming less susceptible to privacy attacks on examples they have not seen
recently. We show that, while non-convex models can memorize data forever in
the worst-case, standard image, speech, and language models empirically do
forget examples over time. We identify nondeterminism as a potential
explanation, showing that deterministically trained models do not forget. Our
results suggest that examples seen early when training with extremely large
datasets - for instance those examples used to pre-train a model - may observe
privacy benefits at the expense of examples seen later.
| [
{
"version": "v1",
"created": "Thu, 30 Jun 2022 20:48:26 GMT"
},
{
"version": "v2",
"created": "Tue, 9 May 2023 14:08:17 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Jagielski",
"Matthew",
""
],
[
"Thakkar",
"Om",
""
],
[
"Tramèr",
"Florian",
""
],
[
"Ippolito",
"Daphne",
""
],
[
"Lee",
"Katherine",
""
],
[
"Carlini",
"Nicholas",
""
],
[
"Wallace",
"Eric",
""
],
[
"Song",
"Shuang",
""
],
[
"Thakurta",
"Abhradeep",
""
],
[
"Papernot",
"Nicolas",
""
],
[
"Zhang",
"Chiyuan",
""
]
] | TITLE: Measuring Forgetting of Memorized Training Examples
ABSTRACT: Machine learning models exhibit two seemingly contradictory phenomena:
training data memorization, and various forms of forgetting. In memorization,
models overfit specific training examples and become susceptible to privacy
attacks. In forgetting, examples which appeared early in training are forgotten
by the end. In this work, we connect these phenomena. We propose a technique to
measure to what extent models "forget" the specifics of training examples,
becoming less susceptible to privacy attacks on examples they have not seen
recently. We show that, while non-convex models can memorize data forever in
the worst-case, standard image, speech, and language models empirically do
forget examples over time. We identify nondeterminism as a potential
explanation, showing that deterministically trained models do not forget. Our
results suggest that examples seen early when training with extremely large
datasets - for instance those examples used to pre-train a model - may observe
privacy benefits at the expense of examples seen later.
| no_new_dataset | 0.710879 |
2207.04308 | Taha Belkhouja | Taha Belkhouja, Yan Yan, Janardhan Rao Doppa | Dynamic Time Warping based Adversarial Framework for Time-Series Domain | Accepted for publication at IEEE Transactions on Pattern Analysis and
Machine Intelligence (TPAMI) | null | 10.1109/TPAMI.2022.3224754 | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | Despite the rapid progress on research in adversarial robustness of deep
neural networks (DNNs), there is little principled work for the time-series
domain. Since time-series data arises in diverse applications including mobile
health, finance, and smart grid, it is important to verify and improve the
robustness of DNNs for the time-series domain. In this paper, we propose a
novel framework for the time-series domain referred as {\em Dynamic Time
Warping for Adversarial Robustness (DTW-AR)} using the dynamic time warping
measure. Theoretical and empirical evidence is provided to demonstrate the
effectiveness of DTW over the standard Euclidean distance metric employed in
prior methods for the image domain. We develop a principled algorithm justified
by theoretical analysis to efficiently create diverse adversarial examples
using random alignment paths. Experiments on diverse real-world benchmarks show
the effectiveness of DTW-AR to fool DNNs for time-series data and to improve
their robustness using adversarial training. The source code of DTW-AR
algorithms is available at https://github.com/tahabelkhouja/DTW-AR
| [
{
"version": "v1",
"created": "Sat, 9 Jul 2022 17:23:00 GMT"
},
{
"version": "v2",
"created": "Tue, 22 Nov 2022 17:19:10 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Belkhouja",
"Taha",
""
],
[
"Yan",
"Yan",
""
],
[
"Doppa",
"Janardhan Rao",
""
]
] | TITLE: Dynamic Time Warping based Adversarial Framework for Time-Series Domain
ABSTRACT: Despite the rapid progress on research in adversarial robustness of deep
neural networks (DNNs), there is little principled work for the time-series
domain. Since time-series data arises in diverse applications including mobile
health, finance, and smart grid, it is important to verify and improve the
robustness of DNNs for the time-series domain. In this paper, we propose a
novel framework for the time-series domain referred as {\em Dynamic Time
Warping for Adversarial Robustness (DTW-AR)} using the dynamic time warping
measure. Theoretical and empirical evidence is provided to demonstrate the
effectiveness of DTW over the standard Euclidean distance metric employed in
prior methods for the image domain. We develop a principled algorithm justified
by theoretical analysis to efficiently create diverse adversarial examples
using random alignment paths. Experiments on diverse real-world benchmarks show
the effectiveness of DTW-AR to fool DNNs for time-series data and to improve
their robustness using adversarial training. The source code of DTW-AR
algorithms is available at https://github.com/tahabelkhouja/DTW-AR
| no_new_dataset | 0.710258 |
2207.05921 | Huajun Zhou | Huajun Zhou and Bo Qiao and Lingxiao Yang and Jianhuang Lai and
Xiaohua Xie | Texture-guided Saliency Distilling for Unsupervised Salient Object
Detection | 8 pages, accepted to CVPR 2023 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Deep Learning-based Unsupervised Salient Object Detection (USOD) mainly
relies on the noisy saliency pseudo labels that have been generated from
traditional handcraft methods or pre-trained networks. To cope with the noisy
labels problem, a class of methods focus on only easy samples with reliable
labels but ignore valuable knowledge in hard samples. In this paper, we propose
a novel USOD method to mine rich and accurate saliency knowledge from both easy
and hard samples. First, we propose a Confidence-aware Saliency Distilling
(CSD) strategy that scores samples conditioned on samples' confidences, which
guides the model to distill saliency knowledge from easy samples to hard
samples progressively. Second, we propose a Boundary-aware Texture Matching
(BTM) strategy to refine the boundaries of noisy labels by matching the
textures around the predicted boundary. Extensive experiments on RGB, RGB-D,
RGB-T, and video SOD benchmarks prove that our method achieves state-of-the-art
USOD performance.
| [
{
"version": "v1",
"created": "Wed, 13 Jul 2022 02:01:07 GMT"
},
{
"version": "v2",
"created": "Mon, 6 Mar 2023 10:11:53 GMT"
},
{
"version": "v3",
"created": "Tue, 9 May 2023 04:21:48 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Zhou",
"Huajun",
""
],
[
"Qiao",
"Bo",
""
],
[
"Yang",
"Lingxiao",
""
],
[
"Lai",
"Jianhuang",
""
],
[
"Xie",
"Xiaohua",
""
]
] | TITLE: Texture-guided Saliency Distilling for Unsupervised Salient Object
Detection
ABSTRACT: Deep Learning-based Unsupervised Salient Object Detection (USOD) mainly
relies on the noisy saliency pseudo labels that have been generated from
traditional handcraft methods or pre-trained networks. To cope with the noisy
labels problem, a class of methods focus on only easy samples with reliable
labels but ignore valuable knowledge in hard samples. In this paper, we propose
a novel USOD method to mine rich and accurate saliency knowledge from both easy
and hard samples. First, we propose a Confidence-aware Saliency Distilling
(CSD) strategy that scores samples conditioned on samples' confidences, which
guides the model to distill saliency knowledge from easy samples to hard
samples progressively. Second, we propose a Boundary-aware Texture Matching
(BTM) strategy to refine the boundaries of noisy labels by matching the
textures around the predicted boundary. Extensive experiments on RGB, RGB-D,
RGB-T, and video SOD benchmarks prove that our method achieves state-of-the-art
USOD performance.
| no_new_dataset | 0.709856 |
2208.05320 | Yacine Belal | Yacine Belal and Aur\'elien Bellet and Sonia Ben Mokhtar and Vlad Nitu | PEPPER: Empowering User-Centric Recommender Systems over Gossip Learning | 23 pages, 15 figures | Proceedings of the ACM on Interactive, Mobile, Wearable and
Ubiquitous Technologies 2022 | 10.1145/3550302 | 101 | cs.IR cs.DC cs.LG cs.NI | http://creativecommons.org/licenses/by/4.0/ | Recommender systems are proving to be an invaluable tool for extracting
user-relevant content helping users in their daily activities (e.g., finding
relevant places to visit, content to consume, items to purchase). However, to
be effective, these systems need to collect and analyze large volumes of
personal data (e.g., location check-ins, movie ratings, click rates .. etc.),
which exposes users to numerous privacy threats. In this context, recommender
systems based on Federated Learning (FL) appear to be a promising solution for
enforcing privacy as they compute accurate recommendations while keeping
personal data on the users' devices. However, FL, and therefore FL-based
recommender systems, rely on a central server that can experience scalability
issues besides being vulnerable to attacks. To remedy this, we propose PEPPER,
a decentralized recommender system based on gossip learning principles. In
PEPPER, users gossip model updates and aggregate them asynchronously. At the
heart of PEPPER reside two key components: a personalized peer-sampling
protocol that keeps in the neighborhood of each node, a proportion of nodes
that have similar interests to the former and a simple yet effective model
aggregation function that builds a model that is better suited to each user.
Through experiments on three real datasets implementing two use cases: a
location check-in recommendation and a movie recommendation, we demonstrate
that our solution converges up to 42% faster than with other decentralized
solutions providing up to 9% improvement on average performance metric such as
hit ratio and up to 21% improvement on long tail performance compared to
decentralized competitors.
| [
{
"version": "v1",
"created": "Tue, 9 Aug 2022 14:51:27 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Belal",
"Yacine",
""
],
[
"Bellet",
"Aurélien",
""
],
[
"Mokhtar",
"Sonia Ben",
""
],
[
"Nitu",
"Vlad",
""
]
] | TITLE: PEPPER: Empowering User-Centric Recommender Systems over Gossip Learning
ABSTRACT: Recommender systems are proving to be an invaluable tool for extracting
user-relevant content helping users in their daily activities (e.g., finding
relevant places to visit, content to consume, items to purchase). However, to
be effective, these systems need to collect and analyze large volumes of
personal data (e.g., location check-ins, movie ratings, click rates .. etc.),
which exposes users to numerous privacy threats. In this context, recommender
systems based on Federated Learning (FL) appear to be a promising solution for
enforcing privacy as they compute accurate recommendations while keeping
personal data on the users' devices. However, FL, and therefore FL-based
recommender systems, rely on a central server that can experience scalability
issues besides being vulnerable to attacks. To remedy this, we propose PEPPER,
a decentralized recommender system based on gossip learning principles. In
PEPPER, users gossip model updates and aggregate them asynchronously. At the
heart of PEPPER reside two key components: a personalized peer-sampling
protocol that keeps in the neighborhood of each node, a proportion of nodes
that have similar interests to the former and a simple yet effective model
aggregation function that builds a model that is better suited to each user.
Through experiments on three real datasets implementing two use cases: a
location check-in recommendation and a movie recommendation, we demonstrate
that our solution converges up to 42% faster than with other decentralized
solutions providing up to 9% improvement on average performance metric such as
hit ratio and up to 21% improvement on long tail performance compared to
decentralized competitors.
| no_new_dataset | 0.709384 |
2208.11030 | Mark Goldsmith | Mark Goldsmith, Guillermo Garc\'ia-P\'erez, Joonas Malmi, Matteo A. C.
Rossi, Harto Saarinen, Sabrina Maniscalco | Link prediction with continuous-time classical and quantum walks | 9 pages, 5 figures | null | 10.3390/e25050730 | null | quant-ph cs.LG cs.SI q-bio.MN | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Protein-protein interaction (PPI) networks consist of the physical and/or
functional interactions between the proteins of an organism. Since the
biophysical and high-throughput methods used to form PPI networks are
expensive, time-consuming, and often contain inaccuracies, the resulting
networks are usually incomplete. In order to infer missing interactions in
these networks, we propose a novel class of link prediction methods based on
continuous-time classical and quantum random walks. In the case of quantum
walks, we examine the usage of both the network adjacency and Laplacian
matrices for controlling the walk dynamics. We define a score function based on
the corresponding transition probabilities and perform tests on four real-world
PPI datasets. Our results show that continuous-time classical random walks and
quantum walks using the network adjacency matrix can successfully predict
missing protein-protein interactions, with performance rivalling the state of
the art.
| [
{
"version": "v1",
"created": "Tue, 23 Aug 2022 15:16:13 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Goldsmith",
"Mark",
""
],
[
"García-Pérez",
"Guillermo",
""
],
[
"Malmi",
"Joonas",
""
],
[
"Rossi",
"Matteo A. C.",
""
],
[
"Saarinen",
"Harto",
""
],
[
"Maniscalco",
"Sabrina",
""
]
] | TITLE: Link prediction with continuous-time classical and quantum walks
ABSTRACT: Protein-protein interaction (PPI) networks consist of the physical and/or
functional interactions between the proteins of an organism. Since the
biophysical and high-throughput methods used to form PPI networks are
expensive, time-consuming, and often contain inaccuracies, the resulting
networks are usually incomplete. In order to infer missing interactions in
these networks, we propose a novel class of link prediction methods based on
continuous-time classical and quantum random walks. In the case of quantum
walks, we examine the usage of both the network adjacency and Laplacian
matrices for controlling the walk dynamics. We define a score function based on
the corresponding transition probabilities and perform tests on four real-world
PPI datasets. Our results show that continuous-time classical random walks and
quantum walks using the network adjacency matrix can successfully predict
missing protein-protein interactions, with performance rivalling the state of
the art.
| no_new_dataset | 0.711994 |
2209.01121 | Simone Angarano | Simone Angarano, Mauro Martini, Francesco Salvetti, Vittorio Mazzia,
Marcello Chiaberge | Back-to-Bones: Rediscovering the Role of Backbones in Domain
Generalization | null | null | null | null | cs.CV cs.LG | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Domain Generalization (DG) studies the capability of a deep learning model to
generalize to out-of-training distributions. In the last decade, literature has
been massively filled with training methodologies that claim to obtain more
abstract and robust data representations to tackle domain shifts. Recent
research has provided a reproducible benchmark for DG, pointing out the
effectiveness of naive empirical risk minimization (ERM) over existing
algorithms. Nevertheless, researchers persist in using the same outdated
feature extractors, and no attention has been given to the effects of different
backbones yet. In this paper, we start back to the backbones proposing a
comprehensive analysis of their intrinsic generalization capabilities, which so
far have been ignored by the research community. We evaluate a wide variety of
feature extractors, from standard residual solutions to transformer-based
architectures, finding an evident linear correlation between large-scale
single-domain classification accuracy and DG capability. Our extensive
experimentation shows that by adopting competitive backbones in conjunction
with effective data augmentation, plain ERM outperforms recent DG solutions and
achieves state-of-the-art accuracy. Moreover, our additional qualitative
studies reveal that novel backbones give more similar representations to
same-class samples, separating different domains in the feature space. This
boost in generalization capabilities leaves marginal room for DG algorithms. It
suggests a new paradigm for investigating the problem, placing backbones in the
spotlight and encouraging the development of consistent algorithms on top of
them. The code is available at https://github.com/PIC4SeR/Back-to-Bones.
| [
{
"version": "v1",
"created": "Fri, 2 Sep 2022 15:30:17 GMT"
},
{
"version": "v2",
"created": "Tue, 9 May 2023 14:31:36 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Angarano",
"Simone",
""
],
[
"Martini",
"Mauro",
""
],
[
"Salvetti",
"Francesco",
""
],
[
"Mazzia",
"Vittorio",
""
],
[
"Chiaberge",
"Marcello",
""
]
] | TITLE: Back-to-Bones: Rediscovering the Role of Backbones in Domain
Generalization
ABSTRACT: Domain Generalization (DG) studies the capability of a deep learning model to
generalize to out-of-training distributions. In the last decade, literature has
been massively filled with training methodologies that claim to obtain more
abstract and robust data representations to tackle domain shifts. Recent
research has provided a reproducible benchmark for DG, pointing out the
effectiveness of naive empirical risk minimization (ERM) over existing
algorithms. Nevertheless, researchers persist in using the same outdated
feature extractors, and no attention has been given to the effects of different
backbones yet. In this paper, we start back to the backbones proposing a
comprehensive analysis of their intrinsic generalization capabilities, which so
far have been ignored by the research community. We evaluate a wide variety of
feature extractors, from standard residual solutions to transformer-based
architectures, finding an evident linear correlation between large-scale
single-domain classification accuracy and DG capability. Our extensive
experimentation shows that by adopting competitive backbones in conjunction
with effective data augmentation, plain ERM outperforms recent DG solutions and
achieves state-of-the-art accuracy. Moreover, our additional qualitative
studies reveal that novel backbones give more similar representations to
same-class samples, separating different domains in the feature space. This
boost in generalization capabilities leaves marginal room for DG algorithms. It
suggests a new paradigm for investigating the problem, placing backbones in the
spotlight and encouraging the development of consistent algorithms on top of
them. The code is available at https://github.com/PIC4SeR/Back-to-Bones.
| no_new_dataset | 0.708629 |
2209.04430 | Shashank Bhat | Shashank Sanjay Bhat, Thiagaraj Prabu, Ben Stappers, Atul Ghalame,
Snehanshu Saha, T.S.B Sudarshan, Zafiirah Hosenie | Investigation of a Machine learning methodology for the SKA pulsar
search pipeline | null | Journal of Astronomy and Astrophysics SKA special issue 2022-23
(Under review) | 10.1007/s12036-023-09920-4 | null | astro-ph.IM cs.AI cs.LG | http://creativecommons.org/licenses/by/4.0/ | The SKA pulsar search pipeline will be used for real time detection of
pulsars. Modern radio telescopes such as SKA will be generating petabytes of
data in their full scale of operation. Hence experience-based and data-driven
algorithms become indispensable for applications such as candidate detection.
Here we describe our findings from testing a state of the art object detection
algorithm called Mask R-CNN to detect candidate signatures in the SKA pulsar
search pipeline. We have trained the Mask R-CNN model to detect candidate
images. A custom annotation tool was developed to mark the regions of interest
in large datasets efficiently. We have successfully demonstrated this algorithm
by detecting candidate signatures on a simulation dataset. The paper presents
details of this work with a highlight on the future prospects.
| [
{
"version": "v1",
"created": "Fri, 9 Sep 2022 17:48:36 GMT"
},
{
"version": "v2",
"created": "Sat, 22 Oct 2022 19:51:30 GMT"
},
{
"version": "v3",
"created": "Tue, 17 Jan 2023 05:22:42 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Bhat",
"Shashank Sanjay",
""
],
[
"Prabu",
"Thiagaraj",
""
],
[
"Stappers",
"Ben",
""
],
[
"Ghalame",
"Atul",
""
],
[
"Saha",
"Snehanshu",
""
],
[
"Sudarshan",
"T. S. B",
""
],
[
"Hosenie",
"Zafiirah",
""
]
] | TITLE: Investigation of a Machine learning methodology for the SKA pulsar
search pipeline
ABSTRACT: The SKA pulsar search pipeline will be used for real time detection of
pulsars. Modern radio telescopes such as SKA will be generating petabytes of
data in their full scale of operation. Hence experience-based and data-driven
algorithms become indispensable for applications such as candidate detection.
Here we describe our findings from testing a state of the art object detection
algorithm called Mask R-CNN to detect candidate signatures in the SKA pulsar
search pipeline. We have trained the Mask R-CNN model to detect candidate
images. A custom annotation tool was developed to mark the regions of interest
in large datasets efficiently. We have successfully demonstrated this algorithm
by detecting candidate signatures on a simulation dataset. The paper presents
details of this work with a highlight on the future prospects.
| no_new_dataset | 0.710785 |
2209.06995 | Yue Yu | Yue Yu, Rongzhi Zhang, Ran Xu, Jieyu Zhang, Jiaming Shen, Chao Zhang | Cold-Start Data Selection for Few-shot Language Model Fine-tuning: A
Prompt-Based Uncertainty Propagation Approach | ACL 2023 Main Conference. Code: https://github.com/yueyu1030/Patron | ACL 2023 | null | null | cs.CL cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Large Language Models have demonstrated remarkable few-shot performance, but
the performance can be sensitive to the selection of few-shot instances. We
propose PATRON, a new method that uses prompt-based uncertainty estimation for
data selection for pre-trained language model fine-tuning under cold-start
scenarios, i.e., no initial labeled data are available. In PATRON, we design
(1) a prompt-based uncertainty propagation approach to estimate the importance
of data points and (2) a partition-then-rewrite (PTR) strategy to promote
sample diversity when querying for annotations. Experiments on six text
classification datasets show that PATRON outperforms the strongest cold-start
data selection baselines by up to 6.9%. Besides, with 128 labels only, PATRON
achieves 91.0% and 92.1% of the fully supervised performance based on vanilla
fine-tuning and prompt-based learning respectively. Our implementation of
PATRON is available at \url{https://github.com/yueyu1030/Patron}.
| [
{
"version": "v1",
"created": "Thu, 15 Sep 2022 01:51:22 GMT"
},
{
"version": "v2",
"created": "Mon, 8 May 2023 20:42:26 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Yu",
"Yue",
""
],
[
"Zhang",
"Rongzhi",
""
],
[
"Xu",
"Ran",
""
],
[
"Zhang",
"Jieyu",
""
],
[
"Shen",
"Jiaming",
""
],
[
"Zhang",
"Chao",
""
]
] | TITLE: Cold-Start Data Selection for Few-shot Language Model Fine-tuning: A
Prompt-Based Uncertainty Propagation Approach
ABSTRACT: Large Language Models have demonstrated remarkable few-shot performance, but
the performance can be sensitive to the selection of few-shot instances. We
propose PATRON, a new method that uses prompt-based uncertainty estimation for
data selection for pre-trained language model fine-tuning under cold-start
scenarios, i.e., no initial labeled data are available. In PATRON, we design
(1) a prompt-based uncertainty propagation approach to estimate the importance
of data points and (2) a partition-then-rewrite (PTR) strategy to promote
sample diversity when querying for annotations. Experiments on six text
classification datasets show that PATRON outperforms the strongest cold-start
data selection baselines by up to 6.9%. Besides, with 128 labels only, PATRON
achieves 91.0% and 92.1% of the fully supervised performance based on vanilla
fine-tuning and prompt-based learning respectively. Our implementation of
PATRON is available at \url{https://github.com/yueyu1030/Patron}.
| no_new_dataset | 0.710986 |
2209.15259 | L\^e-Nguy\^en Hoang | El-Mahdi El-Mhamdi, Sadegh Farhadkhani, Rachid Guerraoui, Nirupam
Gupta, L\^e-Nguy\^en Hoang, Rafael Pinot, S\'ebastien Rouault, John Stephan | On the Impossible Safety of Large AI Models | 40 pages | null | null | null | cs.LG cs.AI cs.CR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Large AI Models (LAIMs), of which large language models are the most
prominent recent example, showcase some impressive performance. However they
have been empirically found to pose serious security issues. This paper
systematizes our knowledge about the fundamental impossibility of building
arbitrarily accurate and secure machine learning models. More precisely, we
identify key challenging features of many of today's machine learning settings.
Namely, high accuracy seems to require memorizing large training datasets,
which are often user-generated and highly heterogeneous, with both sensitive
information and fake users. We then survey statistical lower bounds that, we
argue, constitute a compelling case against the possibility of designing
high-accuracy LAIMs with strong security guarantees.
| [
{
"version": "v1",
"created": "Fri, 30 Sep 2022 06:36:49 GMT"
},
{
"version": "v2",
"created": "Tue, 9 May 2023 10:24:28 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"El-Mhamdi",
"El-Mahdi",
""
],
[
"Farhadkhani",
"Sadegh",
""
],
[
"Guerraoui",
"Rachid",
""
],
[
"Gupta",
"Nirupam",
""
],
[
"Hoang",
"Lê-Nguyên",
""
],
[
"Pinot",
"Rafael",
""
],
[
"Rouault",
"Sébastien",
""
],
[
"Stephan",
"John",
""
]
] | TITLE: On the Impossible Safety of Large AI Models
ABSTRACT: Large AI Models (LAIMs), of which large language models are the most
prominent recent example, showcase some impressive performance. However they
have been empirically found to pose serious security issues. This paper
systematizes our knowledge about the fundamental impossibility of building
arbitrarily accurate and secure machine learning models. More precisely, we
identify key challenging features of many of today's machine learning settings.
Namely, high accuracy seems to require memorizing large training datasets,
which are often user-generated and highly heterogeneous, with both sensitive
information and fake users. We then survey statistical lower bounds that, we
argue, constitute a compelling case against the possibility of designing
high-accuracy LAIMs with strong security guarantees.
| no_new_dataset | 0.711205 |
2210.04182 | Enwei Zhu | Enwei Zhu, Yiyang Liu, Jinpeng Li | Deep Span Representations for Named Entity Recognition | Paper accepted to Findings of ACL 2023 | null | null | null | cs.CL cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Span-based models are one of the most straightforward methods for named
entity recognition (NER). Existing span-based NER systems shallowly aggregate
the token representations to span representations. However, this typically
results in significant ineffectiveness for long-span entities, a coupling
between the representations of overlapping spans, and ultimately a performance
degradation. In this study, we propose DSpERT (Deep Span Encoder
Representations from Transformers), which comprises a standard Transformer and
a span Transformer. The latter uses low-layered span representations as
queries, and aggregates the token representations as keys and values, layer by
layer from bottom to top. Thus, DSpERT produces span representations of deep
semantics.
With weight initialization from pretrained language models, DSpERT achieves
performance higher than or competitive with recent state-of-the-art systems on
eight NER benchmarks. Experimental results verify the importance of the depth
for span representations, and show that DSpERT performs particularly well on
long-span entities and nested structures. Further, the deep span
representations are well structured and easily separable in the feature space.
| [
{
"version": "v1",
"created": "Sun, 9 Oct 2022 06:29:04 GMT"
},
{
"version": "v2",
"created": "Tue, 9 May 2023 08:09:50 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Zhu",
"Enwei",
""
],
[
"Liu",
"Yiyang",
""
],
[
"Li",
"Jinpeng",
""
]
] | TITLE: Deep Span Representations for Named Entity Recognition
ABSTRACT: Span-based models are one of the most straightforward methods for named
entity recognition (NER). Existing span-based NER systems shallowly aggregate
the token representations to span representations. However, this typically
results in significant ineffectiveness for long-span entities, a coupling
between the representations of overlapping spans, and ultimately a performance
degradation. In this study, we propose DSpERT (Deep Span Encoder
Representations from Transformers), which comprises a standard Transformer and
a span Transformer. The latter uses low-layered span representations as
queries, and aggregates the token representations as keys and values, layer by
layer from bottom to top. Thus, DSpERT produces span representations of deep
semantics.
With weight initialization from pretrained language models, DSpERT achieves
performance higher than or competitive with recent state-of-the-art systems on
eight NER benchmarks. Experimental results verify the importance of the depth
for span representations, and show that DSpERT performs particularly well on
long-span entities and nested structures. Further, the deep span
representations are well structured and easily separable in the feature space.
| no_new_dataset | 0.709573 |
2210.04260 | Ruomin Huang | Ruomin Huang, Jiawei Huang, Wenjie Liu and Hu Ding | Coresets for Wasserstein Distributionally Robust Optimization Problems | null | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Wasserstein distributionally robust optimization (\textsf{WDRO}) is a popular
model to enhance the robustness of machine learning with ambiguous data.
However, the complexity of \textsf{WDRO} can be prohibitive in practice since
solving its ``minimax'' formulation requires a great amount of computation.
Recently, several fast \textsf{WDRO} training algorithms for some specific
machine learning tasks (e.g., logistic regression) have been developed.
However, the research on designing efficient algorithms for general large-scale
\textsf{WDRO}s is still quite limited, to the best of our knowledge.
\textit{Coreset} is an important tool for compressing large dataset, and thus
it has been widely applied to reduce the computational complexities for many
optimization problems. In this paper, we introduce a unified framework to
construct the $\epsilon$-coreset for the general \textsf{WDRO} problems. Though
it is challenging to obtain a conventional coreset for \textsf{WDRO} due to the
uncertainty issue of ambiguous data, we show that we can compute a ``dual
coreset'' by using the strong duality property of \textsf{WDRO}. Also, the
error introduced by the dual coreset can be theoretically guaranteed for the
original \textsf{WDRO} objective. To construct the dual coreset, we propose a
novel grid sampling approach that is particularly suitable for the dual
formulation of \textsf{WDRO}. Finally, we implement our coreset approach and
illustrate its effectiveness for several \textsf{WDRO} problems in the
experiments.
| [
{
"version": "v1",
"created": "Sun, 9 Oct 2022 13:24:54 GMT"
},
{
"version": "v2",
"created": "Tue, 4 Apr 2023 08:41:47 GMT"
},
{
"version": "v3",
"created": "Tue, 9 May 2023 08:33:44 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Huang",
"Ruomin",
""
],
[
"Huang",
"Jiawei",
""
],
[
"Liu",
"Wenjie",
""
],
[
"Ding",
"Hu",
""
]
] | TITLE: Coresets for Wasserstein Distributionally Robust Optimization Problems
ABSTRACT: Wasserstein distributionally robust optimization (\textsf{WDRO}) is a popular
model to enhance the robustness of machine learning with ambiguous data.
However, the complexity of \textsf{WDRO} can be prohibitive in practice since
solving its ``minimax'' formulation requires a great amount of computation.
Recently, several fast \textsf{WDRO} training algorithms for some specific
machine learning tasks (e.g., logistic regression) have been developed.
However, the research on designing efficient algorithms for general large-scale
\textsf{WDRO}s is still quite limited, to the best of our knowledge.
\textit{Coreset} is an important tool for compressing large dataset, and thus
it has been widely applied to reduce the computational complexities for many
optimization problems. In this paper, we introduce a unified framework to
construct the $\epsilon$-coreset for the general \textsf{WDRO} problems. Though
it is challenging to obtain a conventional coreset for \textsf{WDRO} due to the
uncertainty issue of ambiguous data, we show that we can compute a ``dual
coreset'' by using the strong duality property of \textsf{WDRO}. Also, the
error introduced by the dual coreset can be theoretically guaranteed for the
original \textsf{WDRO} objective. To construct the dual coreset, we propose a
novel grid sampling approach that is particularly suitable for the dual
formulation of \textsf{WDRO}. Finally, we implement our coreset approach and
illustrate its effectiveness for several \textsf{WDRO} problems in the
experiments.
| no_new_dataset | 0.707379 |
2210.13978 | Yinan Huang | Yinan Huang, Xingang Peng, Jianzhu Ma, Muhan Zhang | Boosting the Cycle Counting Power of Graph Neural Networks with
I$^2$-GNNs | null | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Message Passing Neural Networks (MPNNs) are a widely used class of Graph
Neural Networks (GNNs). The limited representational power of MPNNs inspires
the study of provably powerful GNN architectures. However, knowing one model is
more powerful than another gives little insight about what functions they can
or cannot express. It is still unclear whether these models are able to
approximate specific functions such as counting certain graph substructures,
which is essential for applications in biology, chemistry and social network
analysis. Motivated by this, we propose to study the counting power of Subgraph
MPNNs, a recent and popular class of powerful GNN models that extract rooted
subgraphs for each node, assign the root node a unique identifier and encode
the root node's representation within its rooted subgraph. Specifically, we
prove that Subgraph MPNNs fail to count more-than-4-cycles at node level,
implying that node representations cannot correctly encode the surrounding
substructures like ring systems with more than four atoms. To overcome this
limitation, we propose I$^2$-GNNs to extend Subgraph MPNNs by assigning
different identifiers for the root node and its neighbors in each subgraph.
I$^2$-GNNs' discriminative power is shown to be strictly stronger than Subgraph
MPNNs and partially stronger than the 3-WL test. More importantly, I$^2$-GNNs
are proven capable of counting all 3, 4, 5 and 6-cycles, covering common
substructures like benzene rings in organic chemistry, while still keeping
linear complexity. To the best of our knowledge, it is the first linear-time
GNN model that can count 6-cycles with theoretical guarantees. We validate its
counting power in cycle counting tasks and demonstrate its competitive
performance in molecular prediction benchmarks.
| [
{
"version": "v1",
"created": "Sat, 22 Oct 2022 09:40:00 GMT"
},
{
"version": "v2",
"created": "Mon, 3 Apr 2023 17:09:29 GMT"
},
{
"version": "v3",
"created": "Mon, 8 May 2023 20:45:45 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Huang",
"Yinan",
""
],
[
"Peng",
"Xingang",
""
],
[
"Ma",
"Jianzhu",
""
],
[
"Zhang",
"Muhan",
""
]
] | TITLE: Boosting the Cycle Counting Power of Graph Neural Networks with
I$^2$-GNNs
ABSTRACT: Message Passing Neural Networks (MPNNs) are a widely used class of Graph
Neural Networks (GNNs). The limited representational power of MPNNs inspires
the study of provably powerful GNN architectures. However, knowing one model is
more powerful than another gives little insight about what functions they can
or cannot express. It is still unclear whether these models are able to
approximate specific functions such as counting certain graph substructures,
which is essential for applications in biology, chemistry and social network
analysis. Motivated by this, we propose to study the counting power of Subgraph
MPNNs, a recent and popular class of powerful GNN models that extract rooted
subgraphs for each node, assign the root node a unique identifier and encode
the root node's representation within its rooted subgraph. Specifically, we
prove that Subgraph MPNNs fail to count more-than-4-cycles at node level,
implying that node representations cannot correctly encode the surrounding
substructures like ring systems with more than four atoms. To overcome this
limitation, we propose I$^2$-GNNs to extend Subgraph MPNNs by assigning
different identifiers for the root node and its neighbors in each subgraph.
I$^2$-GNNs' discriminative power is shown to be strictly stronger than Subgraph
MPNNs and partially stronger than the 3-WL test. More importantly, I$^2$-GNNs
are proven capable of counting all 3, 4, 5 and 6-cycles, covering common
substructures like benzene rings in organic chemistry, while still keeping
linear complexity. To the best of our knowledge, it is the first linear-time
GNN model that can count 6-cycles with theoretical guarantees. We validate its
counting power in cycle counting tasks and demonstrate its competitive
performance in molecular prediction benchmarks.
| no_new_dataset | 0.709221 |
2210.14755 | Marie Kune\v{s}ov\'a | Marie Kune\v{s}ov\'a and Zbyn\v{e}k Zaj\'ic | Multitask Detection of Speaker Changes, Overlapping Speech and Voice
Activity Using wav2vec 2.0 | To appear at ICASSP 2023 | null | 10.1109/ICASSP49357.2023.10094972 | null | eess.AS cs.SD | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Self-supervised learning approaches have lately achieved great success on a
broad spectrum of machine learning problems. In the field of speech processing,
one of the most successful recent self-supervised models is wav2vec 2.0. In
this paper, we explore the effectiveness of this model on three basic speech
classification tasks: speaker change detection, overlapped speech detection,
and voice activity detection. First, we concentrate on only one task -- speaker
change detection -- where our proposed system surpasses the previously reported
results on four different corpora, and achieves comparable performance even
when trained on out-of-domain data from an artificially designed dataset. Then
we expand our approach to tackle all three tasks in a single multitask system
with state-of-the-art performance on the AMI corpus. The implementation of the
algorithms in this paper is publicly available at
https://github.com/mkunes/w2v2_audioFrameClassification.
| [
{
"version": "v1",
"created": "Wed, 26 Oct 2022 14:37:05 GMT"
},
{
"version": "v2",
"created": "Fri, 10 Mar 2023 12:33:25 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Kunešová",
"Marie",
""
],
[
"Zajíc",
"Zbyněk",
""
]
] | TITLE: Multitask Detection of Speaker Changes, Overlapping Speech and Voice
Activity Using wav2vec 2.0
ABSTRACT: Self-supervised learning approaches have lately achieved great success on a
broad spectrum of machine learning problems. In the field of speech processing,
one of the most successful recent self-supervised models is wav2vec 2.0. In
this paper, we explore the effectiveness of this model on three basic speech
classification tasks: speaker change detection, overlapped speech detection,
and voice activity detection. First, we concentrate on only one task -- speaker
change detection -- where our proposed system surpasses the previously reported
results on four different corpora, and achieves comparable performance even
when trained on out-of-domain data from an artificially designed dataset. Then
we expand our approach to tackle all three tasks in a single multitask system
with state-of-the-art performance on the AMI corpus. The implementation of the
algorithms in this paper is publicly available at
https://github.com/mkunes/w2v2_audioFrameClassification.
| new_dataset | 0.717198 |
2210.15976 | Chao Fang | Jiayi Tian, Chao Fang, Haonan Wang, Zhongfeng Wang | BEBERT: Efficient and Robust Binary Ensemble BERT | To appear in 2023 IEEE International Conference on Acoustics, Speech,
and Signal Processing (ICASSP 2023) | null | 10.1109/ICASSP49357.2023.10096223 | null | cs.CL cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Pre-trained BERT models have achieved impressive accuracy on natural language
processing (NLP) tasks. However, their excessive amount of parameters hinders
them from efficient deployment on edge devices. Binarization of the BERT models
can significantly alleviate this issue but comes with a severe accuracy drop
compared with their full-precision counterparts. In this paper, we propose an
efficient and robust binary ensemble BERT (BEBERT) to bridge the accuracy gap.
To the best of our knowledge, this is the first work employing ensemble
techniques on binary BERTs, yielding BEBERT, which achieves superior accuracy
while retaining computational efficiency. Furthermore, we remove the knowledge
distillation procedures during ensemble to speed up the training process
without compromising accuracy. Experimental results on the GLUE benchmark show
that the proposed BEBERT significantly outperforms the existing binary BERT
models in accuracy and robustness with a 2x speedup on training time. Moreover,
our BEBERT has only a negligible accuracy loss of 0.3% compared to the
full-precision baseline while saving 15x and 13x in FLOPs and model size,
respectively. In addition, BEBERT also outperforms other compressed BERTs in
accuracy by up to 6.7%.
| [
{
"version": "v1",
"created": "Fri, 28 Oct 2022 08:15:26 GMT"
},
{
"version": "v2",
"created": "Mon, 20 Feb 2023 12:15:46 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Tian",
"Jiayi",
""
],
[
"Fang",
"Chao",
""
],
[
"Wang",
"Haonan",
""
],
[
"Wang",
"Zhongfeng",
""
]
] | TITLE: BEBERT: Efficient and Robust Binary Ensemble BERT
ABSTRACT: Pre-trained BERT models have achieved impressive accuracy on natural language
processing (NLP) tasks. However, their excessive amount of parameters hinders
them from efficient deployment on edge devices. Binarization of the BERT models
can significantly alleviate this issue but comes with a severe accuracy drop
compared with their full-precision counterparts. In this paper, we propose an
efficient and robust binary ensemble BERT (BEBERT) to bridge the accuracy gap.
To the best of our knowledge, this is the first work employing ensemble
techniques on binary BERTs, yielding BEBERT, which achieves superior accuracy
while retaining computational efficiency. Furthermore, we remove the knowledge
distillation procedures during ensemble to speed up the training process
without compromising accuracy. Experimental results on the GLUE benchmark show
that the proposed BEBERT significantly outperforms the existing binary BERT
models in accuracy and robustness with a 2x speedup on training time. Moreover,
our BEBERT has only a negligible accuracy loss of 0.3% compared to the
full-precision baseline while saving 15x and 13x in FLOPs and model size,
respectively. In addition, BEBERT also outperforms other compressed BERTs in
accuracy by up to 6.7%.
| no_new_dataset | 0.709617 |
2210.16559 | Yunqing Zhao | Yunqing Zhao, Keshigeyan Chandrasegaran, Milad Abdollahzadeh, Ngai-Man
Cheung | Few-shot Image Generation via Adaptation-Aware Kernel Modulation | The Thirty-Sixth Annual Conference on Neural Information Processing
Systems (NeurIPS 2022), 14 pages | null | null | null | cs.CV cs.LG eess.IV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Few-shot image generation (FSIG) aims to learn to generate new and diverse
samples given an extremely limited number of samples from a domain, e.g., 10
training samples. Recent work has addressed the problem using transfer learning
approach, leveraging a GAN pretrained on a large-scale source domain dataset
and adapting that model to the target domain based on very limited target
domain samples. Central to recent FSIG methods are knowledge preserving
criteria, which aim to select a subset of source model's knowledge to be
preserved into the adapted model. However, a major limitation of existing
methods is that their knowledge preserving criteria consider only source
domain/source task, and they fail to consider target domain/adaptation task in
selecting source model's knowledge, casting doubt on their suitability for
setups of different proximity between source and target domain. Our work makes
two contributions. As our first contribution, we re-visit recent FSIG works and
their experiments. Our important finding is that, under setups which assumption
of close proximity between source and target domains is relaxed, existing
state-of-the-art (SOTA) methods which consider only source domain/source task
in knowledge preserving perform no better than a baseline fine-tuning method.
To address the limitation of existing methods, as our second contribution, we
propose Adaptation-Aware kernel Modulation (AdAM) to address general FSIG of
different source-target domain proximity. Extensive experimental results show
that the proposed method consistently achieves SOTA performance across
source/target domains of different proximity, including challenging setups when
source and target domains are more apart. Project Page:
https://yunqing-me.github.io/AdAM/
| [
{
"version": "v1",
"created": "Sat, 29 Oct 2022 10:26:40 GMT"
},
{
"version": "v2",
"created": "Sun, 13 Nov 2022 08:00:29 GMT"
},
{
"version": "v3",
"created": "Tue, 9 May 2023 16:42:00 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Zhao",
"Yunqing",
""
],
[
"Chandrasegaran",
"Keshigeyan",
""
],
[
"Abdollahzadeh",
"Milad",
""
],
[
"Cheung",
"Ngai-Man",
""
]
] | TITLE: Few-shot Image Generation via Adaptation-Aware Kernel Modulation
ABSTRACT: Few-shot image generation (FSIG) aims to learn to generate new and diverse
samples given an extremely limited number of samples from a domain, e.g., 10
training samples. Recent work has addressed the problem using transfer learning
approach, leveraging a GAN pretrained on a large-scale source domain dataset
and adapting that model to the target domain based on very limited target
domain samples. Central to recent FSIG methods are knowledge preserving
criteria, which aim to select a subset of source model's knowledge to be
preserved into the adapted model. However, a major limitation of existing
methods is that their knowledge preserving criteria consider only source
domain/source task, and they fail to consider target domain/adaptation task in
selecting source model's knowledge, casting doubt on their suitability for
setups of different proximity between source and target domain. Our work makes
two contributions. As our first contribution, we re-visit recent FSIG works and
their experiments. Our important finding is that, under setups which assumption
of close proximity between source and target domains is relaxed, existing
state-of-the-art (SOTA) methods which consider only source domain/source task
in knowledge preserving perform no better than a baseline fine-tuning method.
To address the limitation of existing methods, as our second contribution, we
propose Adaptation-Aware kernel Modulation (AdAM) to address general FSIG of
different source-target domain proximity. Extensive experimental results show
that the proposed method consistently achieves SOTA performance across
source/target domains of different proximity, including challenging setups when
source and target domains are more apart. Project Page:
https://yunqing-me.github.io/AdAM/
| no_new_dataset | 0.709988 |
2210.17258 | Chao Zhang | Jianjian Qin, Chunzhi Gu, Jun Yu, Chao Zhang | Teacher-Student Network for 3D Point Cloud Anomaly Detection with Few
Normal Samples | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Anomaly detection, which is a critical and popular topic in computer vision,
aims to detect anomalous samples that are different from the normal (i.e.,
non-anomalous) ones. The current mainstream methods focus on anomaly detection
for images, whereas little attention has been paid to 3D point cloud. In this
paper, drawing inspiration from the knowledge transfer ability of
teacher-student architecture and the impressive feature extraction capability
of recent neural networks, we design a teacher-student structured model for 3D
anomaly detection. Specifically, we use feature space alignment, dimension
zoom, and max pooling to extract the features of the point cloud and then
minimize a multi-scale loss between the feature vectors produced by the teacher
and the student networks. Moreover, our method only requires very few normal
samples to train the student network due to the teacher-student distillation
mechanism. Once trained, the teacher-student network pair can be leveraged
jointly to fulfill 3D point cloud anomaly detection based on the calculated
anomaly score. For evaluation, we compare our method against the
reconstruction-based method on the ShapeNet-Part dataset. The experimental
results and ablation studies quantitatively and qualitatively confirm that our
model can achieve higher performance compared with the state of the arts in 3D
anomaly detection with very few training samples.
| [
{
"version": "v1",
"created": "Mon, 31 Oct 2022 12:29:55 GMT"
},
{
"version": "v2",
"created": "Tue, 9 May 2023 13:49:20 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Qin",
"Jianjian",
""
],
[
"Gu",
"Chunzhi",
""
],
[
"Yu",
"Jun",
""
],
[
"Zhang",
"Chao",
""
]
] | TITLE: Teacher-Student Network for 3D Point Cloud Anomaly Detection with Few
Normal Samples
ABSTRACT: Anomaly detection, which is a critical and popular topic in computer vision,
aims to detect anomalous samples that are different from the normal (i.e.,
non-anomalous) ones. The current mainstream methods focus on anomaly detection
for images, whereas little attention has been paid to 3D point cloud. In this
paper, drawing inspiration from the knowledge transfer ability of
teacher-student architecture and the impressive feature extraction capability
of recent neural networks, we design a teacher-student structured model for 3D
anomaly detection. Specifically, we use feature space alignment, dimension
zoom, and max pooling to extract the features of the point cloud and then
minimize a multi-scale loss between the feature vectors produced by the teacher
and the student networks. Moreover, our method only requires very few normal
samples to train the student network due to the teacher-student distillation
mechanism. Once trained, the teacher-student network pair can be leveraged
jointly to fulfill 3D point cloud anomaly detection based on the calculated
anomaly score. For evaluation, we compare our method against the
reconstruction-based method on the ShapeNet-Part dataset. The experimental
results and ablation studies quantitatively and qualitatively confirm that our
model can achieve higher performance compared with the state of the arts in 3D
anomaly detection with very few training samples.
| no_new_dataset | 0.711067 |
2212.03505 | Lejun Zhou | Lejun Zhou, Anke Ye and Simon Hu | A Four-stage Heuristic Algorithm for Solving On-demand Meal Delivery
Routing Problem | Need to be updated | null | null | null | eess.SY cs.SY | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Meal delivery services provided by platforms with integrated delivery systems
are becoming increasingly popular. This paper adopts a rolling horizon approach
to solve the meal delivery routing problem (MDRP). To improve delivery
efficiency in scenarios with high delivery demand, multiple orders are allowed
to be combined into one bundle with orders from different restaurants.
Following this strategy, an optimization-based four-stage heuristic algorithm
is developed to generate an optimal routing plan at each decision point. The
algorithm first generates bundles according to orders' spatial and temporal
distribution. Secondly, we find feasible bundle pairs. Then, routes for
delivering any single bundle or multiple bundles are optimized, respectively.
Finally, the routes are assigned to available couriers. In computational
experiments using instances from open datasets, the system's performance is
evaluated in respect of average click-to-door time and ready-to-pickup time. We
demonstrate that this algorithm can effectively process real-time information
and assign optimal routes to the couriers. By comparing the proposed method
with existing the-state-of-the-art algorithms, the results indicate that our
method can generate solutions with higher service quality and shorter distance.
| [
{
"version": "v1",
"created": "Wed, 7 Dec 2022 08:10:56 GMT"
},
{
"version": "v2",
"created": "Thu, 8 Dec 2022 10:12:04 GMT"
},
{
"version": "v3",
"created": "Tue, 9 May 2023 05:52:33 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Zhou",
"Lejun",
""
],
[
"Ye",
"Anke",
""
],
[
"Hu",
"Simon",
""
]
] | TITLE: A Four-stage Heuristic Algorithm for Solving On-demand Meal Delivery
Routing Problem
ABSTRACT: Meal delivery services provided by platforms with integrated delivery systems
are becoming increasingly popular. This paper adopts a rolling horizon approach
to solve the meal delivery routing problem (MDRP). To improve delivery
efficiency in scenarios with high delivery demand, multiple orders are allowed
to be combined into one bundle with orders from different restaurants.
Following this strategy, an optimization-based four-stage heuristic algorithm
is developed to generate an optimal routing plan at each decision point. The
algorithm first generates bundles according to orders' spatial and temporal
distribution. Secondly, we find feasible bundle pairs. Then, routes for
delivering any single bundle or multiple bundles are optimized, respectively.
Finally, the routes are assigned to available couriers. In computational
experiments using instances from open datasets, the system's performance is
evaluated in respect of average click-to-door time and ready-to-pickup time. We
demonstrate that this algorithm can effectively process real-time information
and assign optimal routes to the couriers. By comparing the proposed method
with existing the-state-of-the-art algorithms, the results indicate that our
method can generate solutions with higher service quality and shorter distance.
| no_new_dataset | 0.709189 |
2212.10218 | Jian Yang | Jian Yang, Shuming Ma, Li Dong, Shaohan Huang, Haoyang Huang, Yuwei
Yin, Dongdong Zhang, Liqun Yang, Furu Wei and Zhoujun Li | GanLM: Encoder-Decoder Pre-training with an Auxiliary Discriminator | ACL 2023 | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Pre-trained models have achieved remarkable success in natural language
processing (NLP). However, existing pre-training methods underutilize the
benefits of language understanding for generation. Inspired by the idea of
Generative Adversarial Networks (GANs), we propose a GAN-style model for
encoder-decoder pre-training by introducing an auxiliary discriminator,
unifying the ability of language understanding and generation in a single
model. Our model, named as GanLM, is trained with two pre-training objectives:
replaced token detection and replaced token denoising. Specifically, given
masked source sentences, the generator outputs the target distribution and the
discriminator predicts whether the target sampled tokens from distribution are
incorrect. The target sentence is replaced with misclassified tokens to
construct noisy previous context, which is used to generate the gold sentence.
In general, both tasks improve the ability of language understanding and
generation by selectively using the denoising data. Extensive experiments in
language generation benchmarks show that GanLM with the powerful language
understanding capability outperforms various strong pre-trained language models
(PLMs) and achieves state-of-the-art performance.
| [
{
"version": "v1",
"created": "Tue, 20 Dec 2022 12:51:11 GMT"
},
{
"version": "v2",
"created": "Tue, 9 May 2023 09:08:33 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Yang",
"Jian",
""
],
[
"Ma",
"Shuming",
""
],
[
"Dong",
"Li",
""
],
[
"Huang",
"Shaohan",
""
],
[
"Huang",
"Haoyang",
""
],
[
"Yin",
"Yuwei",
""
],
[
"Zhang",
"Dongdong",
""
],
[
"Yang",
"Liqun",
""
],
[
"Wei",
"Furu",
""
],
[
"Li",
"Zhoujun",
""
]
] | TITLE: GanLM: Encoder-Decoder Pre-training with an Auxiliary Discriminator
ABSTRACT: Pre-trained models have achieved remarkable success in natural language
processing (NLP). However, existing pre-training methods underutilize the
benefits of language understanding for generation. Inspired by the idea of
Generative Adversarial Networks (GANs), we propose a GAN-style model for
encoder-decoder pre-training by introducing an auxiliary discriminator,
unifying the ability of language understanding and generation in a single
model. Our model, named as GanLM, is trained with two pre-training objectives:
replaced token detection and replaced token denoising. Specifically, given
masked source sentences, the generator outputs the target distribution and the
discriminator predicts whether the target sampled tokens from distribution are
incorrect. The target sentence is replaced with misclassified tokens to
construct noisy previous context, which is used to generate the gold sentence.
In general, both tasks improve the ability of language understanding and
generation by selectively using the denoising data. Extensive experiments in
language generation benchmarks show that GanLM with the powerful language
understanding capability outperforms various strong pre-trained language models
(PLMs) and achieves state-of-the-art performance.
| no_new_dataset | 0.71039 |
2212.10544 | Alexander M. Rush | Junxiong Wang, Jing Nathan Yan, Albert Gu, Alexander M. Rush | Pretraining Without Attention | null | null | null | null | cs.CL cs.LG | http://creativecommons.org/licenses/by/4.0/ | Transformers have been essential to pretraining success in NLP. While other
architectures have been used, downstream accuracy is either significantly
worse, or requires attention layers to match standard benchmarks such as GLUE.
This work explores pretraining without attention by using recent advances in
sequence routing based on state-space models (SSMs). Our proposed model,
Bidirectional Gated SSM (BiGS), combines SSM layers with a multiplicative
gating architecture that has been effective in simplified sequence modeling
architectures. The model learns static layers that do not consider pair-wise
interactions. Even so, BiGS is able to match BERT pretraining accuracy on GLUE
and can be extended to long-form pretraining of 4096 tokens without
approximation. Analysis shows that while the models have similar average
accuracy, the approach has different inductive biases than BERT in terms of
interactions and syntactic representations. All models from this work are
available at https://github.com/jxiw/BiGS.
| [
{
"version": "v1",
"created": "Tue, 20 Dec 2022 18:50:08 GMT"
},
{
"version": "v2",
"created": "Tue, 9 May 2023 01:08:20 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Wang",
"Junxiong",
""
],
[
"Yan",
"Jing Nathan",
""
],
[
"Gu",
"Albert",
""
],
[
"Rush",
"Alexander M.",
""
]
] | TITLE: Pretraining Without Attention
ABSTRACT: Transformers have been essential to pretraining success in NLP. While other
architectures have been used, downstream accuracy is either significantly
worse, or requires attention layers to match standard benchmarks such as GLUE.
This work explores pretraining without attention by using recent advances in
sequence routing based on state-space models (SSMs). Our proposed model,
Bidirectional Gated SSM (BiGS), combines SSM layers with a multiplicative
gating architecture that has been effective in simplified sequence modeling
architectures. The model learns static layers that do not consider pair-wise
interactions. Even so, BiGS is able to match BERT pretraining accuracy on GLUE
and can be extended to long-form pretraining of 4096 tokens without
approximation. Analysis shows that while the models have similar average
accuracy, the approach has different inductive biases than BERT in terms of
interactions and syntactic representations. All models from this work are
available at https://github.com/jxiw/BiGS.
| no_new_dataset | 0.708988 |
2301.02596 | William Bennett | William Bennett, Ryan G. McClarren | Benchmark solutions for radiative transfer with a moving mesh and exact
uncollided source treatments | null | null | null | null | cs.CE | http://creativecommons.org/licenses/by/4.0/ | The set of benchmark solutions used in the thermal radiative transfer
community suffer some coverage gaps, in particular nonlinear, non-equilibrium
problems. Also, there are no non-equilibrium, optically thick benchmarks. These
shortcomings motivated the development of a numerical method free from the
requirement of linearity and easily able to converge on smooth optically thick
problems, a moving mesh Discontinuous Galerkin (DG) framework that utilizes an
uncollided source treatment. Having already proven this method on time
dependent scattering transport problems, we present here solutions to
non-equilibrium thermal radiative transfer problems for familiar linearized
systems together with more physical nonlinear systems in both optically thin
and thick regimes, including both the full transport and the $S_2$/$P_1$
solution. Geometric convergence is observed for smooth sources at all times and
some nonsmooth sources at late times when there is local equilibrium. Also,
accurate solutions are achieved for step sources when the solution is not
smooth.
| [
{
"version": "v1",
"created": "Fri, 6 Jan 2023 16:46:12 GMT"
},
{
"version": "v2",
"created": "Tue, 9 May 2023 17:56:59 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Bennett",
"William",
""
],
[
"McClarren",
"Ryan G.",
""
]
] | TITLE: Benchmark solutions for radiative transfer with a moving mesh and exact
uncollided source treatments
ABSTRACT: The set of benchmark solutions used in the thermal radiative transfer
community suffer some coverage gaps, in particular nonlinear, non-equilibrium
problems. Also, there are no non-equilibrium, optically thick benchmarks. These
shortcomings motivated the development of a numerical method free from the
requirement of linearity and easily able to converge on smooth optically thick
problems, a moving mesh Discontinuous Galerkin (DG) framework that utilizes an
uncollided source treatment. Having already proven this method on time
dependent scattering transport problems, we present here solutions to
non-equilibrium thermal radiative transfer problems for familiar linearized
systems together with more physical nonlinear systems in both optically thin
and thick regimes, including both the full transport and the $S_2$/$P_1$
solution. Geometric convergence is observed for smooth sources at all times and
some nonsmooth sources at late times when there is local equilibrium. Also,
accurate solutions are achieved for step sources when the solution is not
smooth.
| no_new_dataset | 0.707443 |
2301.06998 | Jos\'e Alonso Sol\'is Lemus | Jose Alonso Solis-Lemus, Tiffany Baptiste, Rosie Barrows, Charles
Sillett, Ali Gharaviri, Giulia Raffaele, Orod Razeghi, Marina Strocchi, Iain
Sim, Irum Kotadia, Neil Bodagh, Daniel O'Hare, Mark O'Neill, Steven E
Williams, Caroline Roney, Steven Niederer | Evaluation of an Open-Source Pipeline to Create Patient-Specific Left
Atrial Models: A Reproducibility Study | 17 pages, 7 figures, submitted for review at Journal of Computers in
Biology and Medicine (in press) | null | null | null | physics.med-ph cs.CE | http://creativecommons.org/licenses/by-nc-nd/4.0/ | We present an open-source software pipeline to create patient-specific left
atrial (LA) models with fibre orientations and a fibrosis map, suitable for
electrophysiology simulations. The semi-automatic pipeline takes as input a
contrast enhanced magnetic resonance angiogram, and a late gadolinium enhanced
(LGE) contrast magnetic resonance (CMR). Five operators were allocated 20 cases
each from a set of 50 CMR datasets to create a total of 100 models to evaluate
inter/intra-operator variability. Each output model consisted of (1) a labelled
surface mesh open at the pulmonary veins (PV) and mitral valve (MV), (2) fibre
orientations mapped from a diffusion tensor MRI human atlas, (3) fibrosis map
from the LGE-CMR scan, and (4) simulation of local activation time (LAT) and
phase singularity (PS) mapping. We evaluated reproducibility in our pipeline by
comparing agreement in shape of the output meshes, fibrosis distribution in the
LA body, and fibre orientations; simulations outputs were evaluated comparing
total activation times of LAT maps, mean conduction velocity (CV), and
structural similarity index measure (SSIM) of PS maps. Our workflow allows a
single model to be created in 16.72 +/- 12.25 minutes. Results in this abstract
are reported as inter/intra. Shape only differed noticeably with users'
selection of the MV and the length of the PV from the ostia to the distal end;
fibrosis agreement (0.91/0.99 ICC) and fibre orientation agreement (60.63/71.77
%) were high. LAT maps showed good agreement, the median of the absolute
difference of the total activation times was 2.02ms/1.37ms. The average of the
mean CV difference was -4.04mm/s / 2.1mm/s. PS maps showed a moderately good
agreement with SSIM of 0.648/0.608. Although we found notable differences in
the models due to user input, our tests show that operator variability was
comparable to that of image resolution or fibre estimation.
| [
{
"version": "v1",
"created": "Tue, 17 Jan 2023 16:40:34 GMT"
},
{
"version": "v2",
"created": "Tue, 9 May 2023 15:51:04 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Solis-Lemus",
"Jose Alonso",
""
],
[
"Baptiste",
"Tiffany",
""
],
[
"Barrows",
"Rosie",
""
],
[
"Sillett",
"Charles",
""
],
[
"Gharaviri",
"Ali",
""
],
[
"Raffaele",
"Giulia",
""
],
[
"Razeghi",
"Orod",
""
],
[
"Strocchi",
"Marina",
""
],
[
"Sim",
"Iain",
""
],
[
"Kotadia",
"Irum",
""
],
[
"Bodagh",
"Neil",
""
],
[
"O'Hare",
"Daniel",
""
],
[
"O'Neill",
"Mark",
""
],
[
"Williams",
"Steven E",
""
],
[
"Roney",
"Caroline",
""
],
[
"Niederer",
"Steven",
""
]
] | TITLE: Evaluation of an Open-Source Pipeline to Create Patient-Specific Left
Atrial Models: A Reproducibility Study
ABSTRACT: We present an open-source software pipeline to create patient-specific left
atrial (LA) models with fibre orientations and a fibrosis map, suitable for
electrophysiology simulations. The semi-automatic pipeline takes as input a
contrast enhanced magnetic resonance angiogram, and a late gadolinium enhanced
(LGE) contrast magnetic resonance (CMR). Five operators were allocated 20 cases
each from a set of 50 CMR datasets to create a total of 100 models to evaluate
inter/intra-operator variability. Each output model consisted of (1) a labelled
surface mesh open at the pulmonary veins (PV) and mitral valve (MV), (2) fibre
orientations mapped from a diffusion tensor MRI human atlas, (3) fibrosis map
from the LGE-CMR scan, and (4) simulation of local activation time (LAT) and
phase singularity (PS) mapping. We evaluated reproducibility in our pipeline by
comparing agreement in shape of the output meshes, fibrosis distribution in the
LA body, and fibre orientations; simulations outputs were evaluated comparing
total activation times of LAT maps, mean conduction velocity (CV), and
structural similarity index measure (SSIM) of PS maps. Our workflow allows a
single model to be created in 16.72 +/- 12.25 minutes. Results in this abstract
are reported as inter/intra. Shape only differed noticeably with users'
selection of the MV and the length of the PV from the ostia to the distal end;
fibrosis agreement (0.91/0.99 ICC) and fibre orientation agreement (60.63/71.77
%) were high. LAT maps showed good agreement, the median of the absolute
difference of the total activation times was 2.02ms/1.37ms. The average of the
mean CV difference was -4.04mm/s / 2.1mm/s. PS maps showed a moderately good
agreement with SSIM of 0.648/0.608. Although we found notable differences in
the models due to user input, our tests show that operator variability was
comparable to that of image resolution or fibre estimation.
| no_new_dataset | 0.711875 |
2301.11970 | Mark Keane | Saugat Aryal and Mark T Keane | Even if Explanations: Prior Work, Desiderata & Benchmarks for
Semi-Factual XAI | 14 pages, 4 Figures | 32nd International Joint Conference on Artificial Intelligence
(IJCAI-23), China, Macao, 2023 | null | null | cs.AI | http://creativecommons.org/licenses/by/4.0/ | Recently, eXplainable AI (XAI) research has focused on counterfactual
explanations as post-hoc justifications for AI-system decisions (e.g. a
customer refused a loan might be told: If you asked for a loan with a shorter
term, it would have been approved). Counterfactuals explain what changes to the
input-features of an AI system change the output-decision. However, there is a
sub-type of counterfactual, semi-factuals, that have received less attention in
AI (though the Cognitive Sciences have studied them extensively). This paper
surveys these literatures to summarise historical and recent breakthroughs in
this area. It defines key desiderata for semi-factual XAI and reports benchmark
tests of historical algorithms (along with a novel, naieve method) to provide a
solid basis for future algorithmic developments.
| [
{
"version": "v1",
"created": "Fri, 27 Jan 2023 19:58:12 GMT"
},
{
"version": "v2",
"created": "Mon, 8 May 2023 18:06:32 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Aryal",
"Saugat",
""
],
[
"Keane",
"Mark T",
""
]
] | TITLE: Even if Explanations: Prior Work, Desiderata & Benchmarks for
Semi-Factual XAI
ABSTRACT: Recently, eXplainable AI (XAI) research has focused on counterfactual
explanations as post-hoc justifications for AI-system decisions (e.g. a
customer refused a loan might be told: If you asked for a loan with a shorter
term, it would have been approved). Counterfactuals explain what changes to the
input-features of an AI system change the output-decision. However, there is a
sub-type of counterfactual, semi-factuals, that have received less attention in
AI (though the Cognitive Sciences have studied them extensively). This paper
surveys these literatures to summarise historical and recent breakthroughs in
this area. It defines key desiderata for semi-factual XAI and reports benchmark
tests of historical algorithms (along with a novel, naieve method) to provide a
solid basis for future algorithmic developments.
| no_new_dataset | 0.708824 |
2302.06587 | Minghan Li | Minghan Li, Sheng-Chieh Lin, Xueguang Ma, Jimmy Lin | SLIM: Sparsified Late Interaction for Multi-Vector Retrieval with
Inverted Indexes | null | null | 10.1145/3539618.3591977 | null | cs.IR | http://creativecommons.org/licenses/by/4.0/ | This paper introduces Sparsified Late Interaction for Multi-vector (SLIM)
retrieval with inverted indexes. Multi-vector retrieval methods have
demonstrated their effectiveness on various retrieval datasets, and among them,
ColBERT is the most established method based on the late interaction of
contextualized token embeddings of pre-trained language models. However,
efficient ColBERT implementations require complex engineering and cannot take
advantage of off-the-shelf search libraries, impeding their practical use. To
address this issue, SLIM first maps each contextualized token vector to a
sparse, high-dimensional lexical space before performing late interaction
between these sparse token embeddings. We then introduce an efficient two-stage
retrieval architecture that includes inverted index retrieval followed by a
score refinement module to approximate the sparsified late interaction, which
is fully compatible with off-the-shelf lexical search libraries such as Lucene.
SLIM achieves competitive accuracy on MS MARCO Passages and BEIR compared to
ColBERT while being much smaller and faster on CPUs. To our knowledge, we are
the first to explore using sparse token representations for multi-vector
retrieval. Source code and data are integrated into the Pyserini IR toolkit.
| [
{
"version": "v1",
"created": "Mon, 13 Feb 2023 18:38:14 GMT"
},
{
"version": "v2",
"created": "Mon, 8 May 2023 21:03:03 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Li",
"Minghan",
""
],
[
"Lin",
"Sheng-Chieh",
""
],
[
"Ma",
"Xueguang",
""
],
[
"Lin",
"Jimmy",
""
]
] | TITLE: SLIM: Sparsified Late Interaction for Multi-Vector Retrieval with
Inverted Indexes
ABSTRACT: This paper introduces Sparsified Late Interaction for Multi-vector (SLIM)
retrieval with inverted indexes. Multi-vector retrieval methods have
demonstrated their effectiveness on various retrieval datasets, and among them,
ColBERT is the most established method based on the late interaction of
contextualized token embeddings of pre-trained language models. However,
efficient ColBERT implementations require complex engineering and cannot take
advantage of off-the-shelf search libraries, impeding their practical use. To
address this issue, SLIM first maps each contextualized token vector to a
sparse, high-dimensional lexical space before performing late interaction
between these sparse token embeddings. We then introduce an efficient two-stage
retrieval architecture that includes inverted index retrieval followed by a
score refinement module to approximate the sparsified late interaction, which
is fully compatible with off-the-shelf lexical search libraries such as Lucene.
SLIM achieves competitive accuracy on MS MARCO Passages and BEIR compared to
ColBERT while being much smaller and faster on CPUs. To our knowledge, we are
the first to explore using sparse token representations for multi-vector
retrieval. Source code and data are integrated into the Pyserini IR toolkit.
| no_new_dataset | 0.70866 |
2303.01017 | Mohammad Amin Zarrabian | Mohammad Amin Zarrabian, Ni Ding, Parastoo Sadeghi | On the Lift, Related Privacy Measures, and Applications to
Privacy-Utility Tradeoffs | null | null | 10.3390/e25040679 | null | cs.IT math.IT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper investigates lift, the likelihood ratio between the posterior and
prior belief about sensitive features in a dataset. Maximum and minimum lifts
over sensitive features quantify the adversary's knowledge gain and should be
bounded to protect privacy. We demonstrate that max and min lifts have a
distinct range of values and probability of appearance in the dataset, referred
to as \emph{lift asymmetry}. We propose asymmetric local information privacy
(ALIP) as a compatible privacy notion with lift asymmetry, where different
bounds can be applied to min and max lifts. We use ALIP in the watchdog and
optimal random response (ORR) mechanisms, the main methods to achieve
lift-based privacy. It is shown that ALIP enhances utility in these methods
compared to existing local information privacy, which ensures the same
(symmetric) bounds on both max and min lifts. We propose subset merging for the
watchdog mechanism to improve data utility and subset random response for the
ORR to reduce complexity. We then investigate the related lift-based measures,
including $\ell_1$-norm, $\chi^2$-privacy criterion, and $\alpha$-lift. We
reveal that they can only restrict max-lift, resulting in significant min-lift
leakage. To overcome this problem, we propose corresponding lift-inverse
measures to restrict the min-lift. We apply these lift-based and lift-inverse
measures in the watchdog mechanism. We show that they can be considered as
relaxations of ALIP, where a higher utility can be achieved by bounding only
average max and min lifts.
| [
{
"version": "v1",
"created": "Thu, 2 Mar 2023 07:10:22 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Zarrabian",
"Mohammad Amin",
""
],
[
"Ding",
"Ni",
""
],
[
"Sadeghi",
"Parastoo",
""
]
] | TITLE: On the Lift, Related Privacy Measures, and Applications to
Privacy-Utility Tradeoffs
ABSTRACT: This paper investigates lift, the likelihood ratio between the posterior and
prior belief about sensitive features in a dataset. Maximum and minimum lifts
over sensitive features quantify the adversary's knowledge gain and should be
bounded to protect privacy. We demonstrate that max and min lifts have a
distinct range of values and probability of appearance in the dataset, referred
to as \emph{lift asymmetry}. We propose asymmetric local information privacy
(ALIP) as a compatible privacy notion with lift asymmetry, where different
bounds can be applied to min and max lifts. We use ALIP in the watchdog and
optimal random response (ORR) mechanisms, the main methods to achieve
lift-based privacy. It is shown that ALIP enhances utility in these methods
compared to existing local information privacy, which ensures the same
(symmetric) bounds on both max and min lifts. We propose subset merging for the
watchdog mechanism to improve data utility and subset random response for the
ORR to reduce complexity. We then investigate the related lift-based measures,
including $\ell_1$-norm, $\chi^2$-privacy criterion, and $\alpha$-lift. We
reveal that they can only restrict max-lift, resulting in significant min-lift
leakage. To overcome this problem, we propose corresponding lift-inverse
measures to restrict the min-lift. We apply these lift-based and lift-inverse
measures in the watchdog mechanism. We show that they can be considered as
relaxations of ALIP, where a higher utility can be achieved by bounding only
average max and min lifts.
| no_new_dataset | 0.708572 |
2303.01911 | Rachel Bawden | Rachel Bawden and Fran\c{c}ois Yvon | Investigating the Translation Performance of a Large Multilingual
Language Model: the Case of BLOOM | Accepted at EAMT 2023 | null | null | null | cs.CL | http://creativecommons.org/licenses/by-sa/4.0/ | The NLP community recently saw the release of a new large open-access
multilingual language model, BLOOM (BigScience et al., 2022) covering 46
languages. We focus on BLOOM's multilingual ability by evaluating its machine
translation performance across several datasets (WMT, Flores-101 and DiaBLa)
and language pairs (high- and low-resourced). Our results show that 0-shot
performance suffers from overgeneration and generating in the wrong language,
but this is greatly improved in the few-shot setting, with very good results
for a number of language pairs. We study several aspects including prompt
design, model sizes, cross-lingual transfer and the use of discursive context.
| [
{
"version": "v1",
"created": "Fri, 3 Mar 2023 13:23:42 GMT"
},
{
"version": "v2",
"created": "Tue, 9 May 2023 12:21:58 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Bawden",
"Rachel",
""
],
[
"Yvon",
"François",
""
]
] | TITLE: Investigating the Translation Performance of a Large Multilingual
Language Model: the Case of BLOOM
ABSTRACT: The NLP community recently saw the release of a new large open-access
multilingual language model, BLOOM (BigScience et al., 2022) covering 46
languages. We focus on BLOOM's multilingual ability by evaluating its machine
translation performance across several datasets (WMT, Flores-101 and DiaBLa)
and language pairs (high- and low-resourced). Our results show that 0-shot
performance suffers from overgeneration and generating in the wrong language,
but this is greatly improved in the few-shot setting, with very good results
for a number of language pairs. We study several aspects including prompt
design, model sizes, cross-lingual transfer and the use of discursive context.
| no_new_dataset | 0.707398 |
2303.02632 | Guy Farjon | Guy Farjon, Liu Huijun, Yael Edan | Deep-Learning-based Counting Methods, Datasets, and Applications in
Agriculture -- A Review | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | The number of objects is considered an important factor in a variety of tasks
in the agricultural domain. Automated counting can improve farmers decisions
regarding yield estimation, stress detection, disease prevention, and more. In
recent years, deep learning has been increasingly applied to many
agriculture-related applications, complementing conventional computer-vision
algorithms for counting agricultural objects. This article reviews progress in
the past decade and the state of the art for counting methods in agriculture,
focusing on deep-learning methods. It presents an overview of counting
algorithms, metrics, platforms, and sensors, a list of all publicly available
datasets, and an in-depth discussion of various deep-learning methods used for
counting. Finally, it discusses open challenges in object counting using deep
learning and gives a glimpse into new directions and future perspectives for
counting research. The review reveals a major leap forward in object counting
in agriculture in the past decade, led by the penetration of deep learning
methods into counting platforms.
| [
{
"version": "v1",
"created": "Sun, 5 Mar 2023 10:17:10 GMT"
},
{
"version": "v2",
"created": "Tue, 9 May 2023 08:29:16 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Farjon",
"Guy",
""
],
[
"Huijun",
"Liu",
""
],
[
"Edan",
"Yael",
""
]
] | TITLE: Deep-Learning-based Counting Methods, Datasets, and Applications in
Agriculture -- A Review
ABSTRACT: The number of objects is considered an important factor in a variety of tasks
in the agricultural domain. Automated counting can improve farmers decisions
regarding yield estimation, stress detection, disease prevention, and more. In
recent years, deep learning has been increasingly applied to many
agriculture-related applications, complementing conventional computer-vision
algorithms for counting agricultural objects. This article reviews progress in
the past decade and the state of the art for counting methods in agriculture,
focusing on deep-learning methods. It presents an overview of counting
algorithms, metrics, platforms, and sensors, a list of all publicly available
datasets, and an in-depth discussion of various deep-learning methods used for
counting. Finally, it discusses open challenges in object counting using deep
learning and gives a glimpse into new directions and future perspectives for
counting research. The review reveals a major leap forward in object counting
in agriculture in the past decade, led by the penetration of deep learning
methods into counting platforms.
| no_new_dataset | 0.711863 |
2303.03509 | Gagandeep Singh | Gagandeep Singh, Alireza Khodamoradi, Kristof Denolf, Jack Lo, Juan
G\'omez-Luna, Joseph Melber, Andra Bisca, Henk Corporaal, Onur Mutlu | SPARTA: Spatial Acceleration for Efficient and Scalable Horizontal
Diffusion Weather Stencil Computation | null | null | null | null | cs.AR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Fast and accurate climate simulations and weather predictions are critical
for understanding and preparing for the impact of climate change. Real-world
weather and climate modeling consist of complex compound stencil kernels that
do not perform well on conventional architectures. Horizontal diffusion is one
such important compound stencil found in many climate and weather prediction
models. Recent works propose using FPGAs as an alternative to traditional CPU
and GPU-based systems to accelerate compound stencil kernels. However, we
observe that compound stencil computations cannot leverage the bit-level
flexibility available on an FPGA because of its complex memory access patterns,
leading to high hardware resource utilization and low peak performance. We
introduce SPARTA, a novel spatial accelerator for horizontal diffusion weather
stencil computation. We exploit the two-dimensional spatial architecture to
efficiently accelerate horizontal diffusion stencil by designing the first
scaled-out spatial accelerator using MLIR (Multi-Level Intermediate
Representation) compiler framework. We evaluate its performance on a real
cutting-edge AMD-Xilinx Versal AI Engine spatial architecture. Our real-system
evaluation results demonstrate that SPARTA outperforms the state-of-the-art
CPU, GPU, and FPGA implementations by 17.1x, 1.2x, and 2.1x, respectively. Our
results reveal that balancing workload across the available processing
resources is crucial in achieving high performance on spatial architectures. We
also implement and evaluate five elementary stencils that are commonly used as
benchmarks for stencil computation research. We freely open-source all our
implementations to aid future research in stencil computation and spatial
computing systems at https://github.com/CMU-SAFARI/SPARTA.
| [
{
"version": "v1",
"created": "Mon, 6 Mar 2023 21:33:36 GMT"
},
{
"version": "v2",
"created": "Tue, 9 May 2023 14:45:38 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Singh",
"Gagandeep",
""
],
[
"Khodamoradi",
"Alireza",
""
],
[
"Denolf",
"Kristof",
""
],
[
"Lo",
"Jack",
""
],
[
"Gómez-Luna",
"Juan",
""
],
[
"Melber",
"Joseph",
""
],
[
"Bisca",
"Andra",
""
],
[
"Corporaal",
"Henk",
""
],
[
"Mutlu",
"Onur",
""
]
] | TITLE: SPARTA: Spatial Acceleration for Efficient and Scalable Horizontal
Diffusion Weather Stencil Computation
ABSTRACT: Fast and accurate climate simulations and weather predictions are critical
for understanding and preparing for the impact of climate change. Real-world
weather and climate modeling consist of complex compound stencil kernels that
do not perform well on conventional architectures. Horizontal diffusion is one
such important compound stencil found in many climate and weather prediction
models. Recent works propose using FPGAs as an alternative to traditional CPU
and GPU-based systems to accelerate compound stencil kernels. However, we
observe that compound stencil computations cannot leverage the bit-level
flexibility available on an FPGA because of its complex memory access patterns,
leading to high hardware resource utilization and low peak performance. We
introduce SPARTA, a novel spatial accelerator for horizontal diffusion weather
stencil computation. We exploit the two-dimensional spatial architecture to
efficiently accelerate horizontal diffusion stencil by designing the first
scaled-out spatial accelerator using MLIR (Multi-Level Intermediate
Representation) compiler framework. We evaluate its performance on a real
cutting-edge AMD-Xilinx Versal AI Engine spatial architecture. Our real-system
evaluation results demonstrate that SPARTA outperforms the state-of-the-art
CPU, GPU, and FPGA implementations by 17.1x, 1.2x, and 2.1x, respectively. Our
results reveal that balancing workload across the available processing
resources is crucial in achieving high performance on spatial architectures. We
also implement and evaluate five elementary stencils that are commonly used as
benchmarks for stencil computation research. We freely open-source all our
implementations to aid future research in stencil computation and spatial
computing systems at https://github.com/CMU-SAFARI/SPARTA.
| no_new_dataset | 0.709239 |
2304.02431 | Darren Tsai | Darren Tsai, Julie Stephany Berrio, Mao Shan, Eduardo Nebot and
Stewart Worrall | MS3D: Leveraging Multiple Detectors for Unsupervised Domain Adaptation
in 3D Object Detection | Our code is available at https://github.com/darrenjkt/MS3D | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We introduce Multi-Source 3D (MS3D), a new self-training pipeline for
unsupervised domain adaptation in 3D object detection. Despite the remarkable
accuracy of 3D detectors, they often overfit to specific domain biases, leading
to suboptimal performance in various sensor setups and environments. Existing
methods typically focus on adapting a single detector to the target domain,
overlooking the fact that different detectors possess distinct expertise on
different unseen domains. MS3D leverages this by combining different
pre-trained detectors from multiple source domains and incorporating temporal
information to produce high-quality pseudo-labels for fine-tuning. Our proposed
Kernel-Density Estimation (KDE) Box Fusion method fuses box proposals from
multiple domains to obtain pseudo-labels that surpass the performance of the
best source domain detectors. MS3D exhibits greater robustness to domain shift
and produces accurate pseudo-labels over greater distances, making it
well-suited for high-to-low beam domain adaptation and vice versa. Our method
achieved state-of-the-art performance on all evaluated datasets, and we
demonstrate that the pre-trained detector's source dataset has minimal impact
on the fine-tuned result, making MS3D suitable for real-world applications.
| [
{
"version": "v1",
"created": "Wed, 5 Apr 2023 13:29:21 GMT"
},
{
"version": "v2",
"created": "Mon, 10 Apr 2023 07:17:49 GMT"
},
{
"version": "v3",
"created": "Tue, 9 May 2023 01:23:24 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Tsai",
"Darren",
""
],
[
"Berrio",
"Julie Stephany",
""
],
[
"Shan",
"Mao",
""
],
[
"Nebot",
"Eduardo",
""
],
[
"Worrall",
"Stewart",
""
]
] | TITLE: MS3D: Leveraging Multiple Detectors for Unsupervised Domain Adaptation
in 3D Object Detection
ABSTRACT: We introduce Multi-Source 3D (MS3D), a new self-training pipeline for
unsupervised domain adaptation in 3D object detection. Despite the remarkable
accuracy of 3D detectors, they often overfit to specific domain biases, leading
to suboptimal performance in various sensor setups and environments. Existing
methods typically focus on adapting a single detector to the target domain,
overlooking the fact that different detectors possess distinct expertise on
different unseen domains. MS3D leverages this by combining different
pre-trained detectors from multiple source domains and incorporating temporal
information to produce high-quality pseudo-labels for fine-tuning. Our proposed
Kernel-Density Estimation (KDE) Box Fusion method fuses box proposals from
multiple domains to obtain pseudo-labels that surpass the performance of the
best source domain detectors. MS3D exhibits greater robustness to domain shift
and produces accurate pseudo-labels over greater distances, making it
well-suited for high-to-low beam domain adaptation and vice versa. Our method
achieved state-of-the-art performance on all evaluated datasets, and we
demonstrate that the pre-trained detector's source dataset has minimal impact
on the fine-tuned result, making MS3D suitable for real-world applications.
| no_new_dataset | 0.709856 |
2304.03910 | Yazhou Yao | Gensheng Pei, Yazhou Yao, Fumin Shen, Dan Huang, Xingguo Huang, and
Heng-Tao Shen | Co-attention Propagation Network for Zero-Shot Video Object Segmentation | accepted by IEEE Transactions on Image Processing | null | 10.1109/TIP.2023.3267244 | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Zero-shot video object segmentation (ZS-VOS) aims to segment foreground
objects in a video sequence without prior knowledge of these objects. However,
existing ZS-VOS methods often struggle to distinguish between foreground and
background or to keep track of the foreground in complex scenarios. The common
practice of introducing motion information, such as optical flow, can lead to
overreliance on optical flow estimation. To address these challenges, we
propose an encoder-decoder-based hierarchical co-attention propagation network
(HCPN) capable of tracking and segmenting objects. Specifically, our model is
built upon multiple collaborative evolutions of the parallel co-attention
module (PCM) and the cross co-attention module (CCM). PCM captures common
foreground regions among adjacent appearance and motion features, while CCM
further exploits and fuses cross-modal motion features returned by PCM. Our
method is progressively trained to achieve hierarchical spatio-temporal feature
propagation across the entire video. Experimental results demonstrate that our
HCPN outperforms all previous methods on public benchmarks, showcasing its
effectiveness for ZS-VOS.
| [
{
"version": "v1",
"created": "Sat, 8 Apr 2023 04:45:48 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Pei",
"Gensheng",
""
],
[
"Yao",
"Yazhou",
""
],
[
"Shen",
"Fumin",
""
],
[
"Huang",
"Dan",
""
],
[
"Huang",
"Xingguo",
""
],
[
"Shen",
"Heng-Tao",
""
]
] | TITLE: Co-attention Propagation Network for Zero-Shot Video Object Segmentation
ABSTRACT: Zero-shot video object segmentation (ZS-VOS) aims to segment foreground
objects in a video sequence without prior knowledge of these objects. However,
existing ZS-VOS methods often struggle to distinguish between foreground and
background or to keep track of the foreground in complex scenarios. The common
practice of introducing motion information, such as optical flow, can lead to
overreliance on optical flow estimation. To address these challenges, we
propose an encoder-decoder-based hierarchical co-attention propagation network
(HCPN) capable of tracking and segmenting objects. Specifically, our model is
built upon multiple collaborative evolutions of the parallel co-attention
module (PCM) and the cross co-attention module (CCM). PCM captures common
foreground regions among adjacent appearance and motion features, while CCM
further exploits and fuses cross-modal motion features returned by PCM. Our
method is progressively trained to achieve hierarchical spatio-temporal feature
propagation across the entire video. Experimental results demonstrate that our
HCPN outperforms all previous methods on public benchmarks, showcasing its
effectiveness for ZS-VOS.
| no_new_dataset | 0.709177 |
2304.07459 | Anh-Khoa Nguyen Vu | Anh-Khoa Nguyen Vu, Thanh-Toan Do, Nhat-Duy Nguyen, Vinh-Tiep Nguyen,
Thanh Duc Ngo, Tam V. Nguyen | Instance-level Few-shot Learning with Class Hierarchy Mining | accepted by IEEE Transactions on Image Processing | null | 10.1109/TIP.2023.3267621 | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Few-shot learning is proposed to tackle the problem of scarce training data
in novel classes. However, prior works in instance-level few-shot learning have
paid less attention to effectively utilizing the relationship between
categories. In this paper, we exploit the hierarchical information to leverage
discriminative and relevant features of base classes to effectively classify
novel objects. These features are extracted from abundant data of base classes,
which could be utilized to reasonably describe classes with scarce data.
Specifically, we propose a novel superclass approach that automatically creates
a hierarchy considering base and novel classes as fine-grained classes for
few-shot instance segmentation (FSIS). Based on the hierarchical information,
we design a novel framework called Soft Multiple Superclass (SMS) to extract
relevant features or characteristics of classes in the same superclass. A new
class assigned to the superclass is easier to classify by leveraging these
relevant features. Besides, in order to effectively train the
hierarchy-based-detector in FSIS, we apply the label refinement to further
describe the associations between fine-grained classes. The extensive
experiments demonstrate the effectiveness of our method on FSIS benchmarks.
Code is available online.
| [
{
"version": "v1",
"created": "Sat, 15 Apr 2023 02:55:08 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Vu",
"Anh-Khoa Nguyen",
""
],
[
"Do",
"Thanh-Toan",
""
],
[
"Nguyen",
"Nhat-Duy",
""
],
[
"Nguyen",
"Vinh-Tiep",
""
],
[
"Ngo",
"Thanh Duc",
""
],
[
"Nguyen",
"Tam V.",
""
]
] | TITLE: Instance-level Few-shot Learning with Class Hierarchy Mining
ABSTRACT: Few-shot learning is proposed to tackle the problem of scarce training data
in novel classes. However, prior works in instance-level few-shot learning have
paid less attention to effectively utilizing the relationship between
categories. In this paper, we exploit the hierarchical information to leverage
discriminative and relevant features of base classes to effectively classify
novel objects. These features are extracted from abundant data of base classes,
which could be utilized to reasonably describe classes with scarce data.
Specifically, we propose a novel superclass approach that automatically creates
a hierarchy considering base and novel classes as fine-grained classes for
few-shot instance segmentation (FSIS). Based on the hierarchical information,
we design a novel framework called Soft Multiple Superclass (SMS) to extract
relevant features or characteristics of classes in the same superclass. A new
class assigned to the superclass is easier to classify by leveraging these
relevant features. Besides, in order to effectively train the
hierarchy-based-detector in FSIS, we apply the label refinement to further
describe the associations between fine-grained classes. The extensive
experiments demonstrate the effectiveness of our method on FSIS benchmarks.
Code is available online.
| no_new_dataset | 0.710672 |
2304.08990 | Zhaoming Kong | Zhaoming Kong, Fangxi Deng, Haomin Zhuang, Jun Yu, Lifang He and
Xiaowei Yang | A Comparison of Image Denoising Methods | In this paper, we intend to collect and compare various denoising
methods to investigate their effectiveness, efficiency, applicability and
generalization ability with both synthetic and real-world experiments. arXiv
admin note: substantial text overlap with arXiv:2011.03462 | null | null | null | eess.IV cs.CV | http://creativecommons.org/licenses/by/4.0/ | The advancement of imaging devices and countless images generated everyday
pose an increasingly high demand on image denoising, which still remains a
challenging task in terms of both effectiveness and efficiency. To improve
denoising quality, numerous denoising techniques and approaches have been
proposed in the past decades, including different transforms, regularization
terms, algebraic representations and especially advanced deep neural network
(DNN) architectures. Despite their sophistication, many methods may fail to
achieve desirable results for simultaneous noise removal and fine detail
preservation. In this paper, to investigate the applicability of existing
denoising techniques, we compare a variety of denoising methods on both
synthetic and real-world datasets for different applications. We also introduce
a new dataset for benchmarking, and the evaluations are performed from four
different perspectives including quantitative metrics, visual effects, human
ratings and computational cost. Our experiments demonstrate: (i) the
effectiveness and efficiency of representative traditional denoisers for
various denoising tasks, (ii) a simple matrix-based algorithm may be able to
produce similar results compared with its tensor counterparts, and (iii) the
notable achievements of DNN models, which exhibit impressive generalization
ability and show state-of-the-art performance on various datasets. In spite of
the progress in recent years, we discuss shortcomings and possible extensions
of existing techniques. Datasets, code and results are made publicly available
and will be continuously updated at
https://github.com/ZhaomingKong/Denoising-Comparison.
| [
{
"version": "v1",
"created": "Tue, 18 Apr 2023 13:41:42 GMT"
},
{
"version": "v2",
"created": "Tue, 9 May 2023 05:57:21 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Kong",
"Zhaoming",
""
],
[
"Deng",
"Fangxi",
""
],
[
"Zhuang",
"Haomin",
""
],
[
"Yu",
"Jun",
""
],
[
"He",
"Lifang",
""
],
[
"Yang",
"Xiaowei",
""
]
] | TITLE: A Comparison of Image Denoising Methods
ABSTRACT: The advancement of imaging devices and countless images generated everyday
pose an increasingly high demand on image denoising, which still remains a
challenging task in terms of both effectiveness and efficiency. To improve
denoising quality, numerous denoising techniques and approaches have been
proposed in the past decades, including different transforms, regularization
terms, algebraic representations and especially advanced deep neural network
(DNN) architectures. Despite their sophistication, many methods may fail to
achieve desirable results for simultaneous noise removal and fine detail
preservation. In this paper, to investigate the applicability of existing
denoising techniques, we compare a variety of denoising methods on both
synthetic and real-world datasets for different applications. We also introduce
a new dataset for benchmarking, and the evaluations are performed from four
different perspectives including quantitative metrics, visual effects, human
ratings and computational cost. Our experiments demonstrate: (i) the
effectiveness and efficiency of representative traditional denoisers for
various denoising tasks, (ii) a simple matrix-based algorithm may be able to
produce similar results compared with its tensor counterparts, and (iii) the
notable achievements of DNN models, which exhibit impressive generalization
ability and show state-of-the-art performance on various datasets. In spite of
the progress in recent years, we discuss shortcomings and possible extensions
of existing techniques. Datasets, code and results are made publicly available
and will be continuously updated at
https://github.com/ZhaomingKong/Denoising-Comparison.
| new_dataset | 0.714634 |
2304.11584 | Jiahao Nie | Jiahao Nie, Zhiwei He, Yuxiang Yang, Zhengyi Bao, Mingyu Gao, Jing
Zhang | OSP2B: One-Stage Point-to-Box Network for 3D Siamese Tracking | Accepted to IJCAI'23. Code will be available at
https://github.com/haooozi/OSP2B | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Two-stage point-to-box network acts as a critical role in the recent popular
3D Siamese tracking paradigm, which first generates proposals and then predicts
corresponding proposal-wise scores. However, such a network suffers from
tedious hyper-parameter tuning and task misalignment, limiting the tracking
performance. Towards these concerns, we propose a simple yet effective
one-stage point-to-box network for point cloud-based 3D single object tracking.
It synchronizes 3D proposal generation and center-ness score prediction by a
parallel predictor without tedious hyper-parameters. To guide a task-aligned
score ranking of proposals, a center-aware focal loss is proposed to supervise
the training of the center-ness branch, which enhances the network's
discriminative ability to distinguish proposals of different quality. Besides,
we design a binary target classifier to identify target-relevant points. By
integrating the derived classification scores with the center-ness scores, the
resulting network can effectively suppress interference proposals and further
mitigate task misalignment. Finally, we present a novel one-stage Siamese
tracker OSP2B equipped with the designed network. Extensive experiments on
challenging benchmarks including KITTI and Waymo SOT Dataset show that our
OSP2B achieves leading performance with a considerable real-time speed.Code
will be available at https://github.com/haooozi/OSP2B.
| [
{
"version": "v1",
"created": "Sun, 23 Apr 2023 08:52:36 GMT"
},
{
"version": "v2",
"created": "Tue, 9 May 2023 02:27:49 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Nie",
"Jiahao",
""
],
[
"He",
"Zhiwei",
""
],
[
"Yang",
"Yuxiang",
""
],
[
"Bao",
"Zhengyi",
""
],
[
"Gao",
"Mingyu",
""
],
[
"Zhang",
"Jing",
""
]
] | TITLE: OSP2B: One-Stage Point-to-Box Network for 3D Siamese Tracking
ABSTRACT: Two-stage point-to-box network acts as a critical role in the recent popular
3D Siamese tracking paradigm, which first generates proposals and then predicts
corresponding proposal-wise scores. However, such a network suffers from
tedious hyper-parameter tuning and task misalignment, limiting the tracking
performance. Towards these concerns, we propose a simple yet effective
one-stage point-to-box network for point cloud-based 3D single object tracking.
It synchronizes 3D proposal generation and center-ness score prediction by a
parallel predictor without tedious hyper-parameters. To guide a task-aligned
score ranking of proposals, a center-aware focal loss is proposed to supervise
the training of the center-ness branch, which enhances the network's
discriminative ability to distinguish proposals of different quality. Besides,
we design a binary target classifier to identify target-relevant points. By
integrating the derived classification scores with the center-ness scores, the
resulting network can effectively suppress interference proposals and further
mitigate task misalignment. Finally, we present a novel one-stage Siamese
tracker OSP2B equipped with the designed network. Extensive experiments on
challenging benchmarks including KITTI and Waymo SOT Dataset show that our
OSP2B achieves leading performance with a considerable real-time speed.Code
will be available at https://github.com/haooozi/OSP2B.
| no_new_dataset | 0.709648 |
2304.14560 | Yasaman Haghighi | Yasaman Haghighi, Suryansh Kumar, Jean-Philippe Thiran, Luc Van Gool | Neural Implicit Dense Semantic SLAM | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Visual Simultaneous Localization and Mapping (vSLAM) is a widely used
technique in robotics and computer vision that enables a robot to create a map
of an unfamiliar environment using a camera sensor while simultaneously
tracking its position over time. In this paper, we propose a novel RGBD vSLAM
algorithm that can learn a memory-efficient, dense 3D geometry, and semantic
segmentation of an indoor scene in an online manner. Our pipeline combines
classical 3D vision-based tracking and loop closing with neural fields-based
mapping. The mapping network learns the SDF of the scene as well as RGB, depth,
and semantic maps of any novel view using only a set of keyframes.
Additionally, we extend our pipeline to large scenes by using multiple local
mapping networks. Extensive experiments on well-known benchmark datasets
confirm that our approach provides robust tracking, mapping, and semantic
labeling even with noisy, sparse, or no input depth. Overall, our proposed
algorithm can greatly enhance scene perception and assist with a range of robot
control problems.
| [
{
"version": "v1",
"created": "Thu, 27 Apr 2023 23:03:52 GMT"
},
{
"version": "v2",
"created": "Tue, 9 May 2023 13:58:15 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Haghighi",
"Yasaman",
""
],
[
"Kumar",
"Suryansh",
""
],
[
"Thiran",
"Jean-Philippe",
""
],
[
"Van Gool",
"Luc",
""
]
] | TITLE: Neural Implicit Dense Semantic SLAM
ABSTRACT: Visual Simultaneous Localization and Mapping (vSLAM) is a widely used
technique in robotics and computer vision that enables a robot to create a map
of an unfamiliar environment using a camera sensor while simultaneously
tracking its position over time. In this paper, we propose a novel RGBD vSLAM
algorithm that can learn a memory-efficient, dense 3D geometry, and semantic
segmentation of an indoor scene in an online manner. Our pipeline combines
classical 3D vision-based tracking and loop closing with neural fields-based
mapping. The mapping network learns the SDF of the scene as well as RGB, depth,
and semantic maps of any novel view using only a set of keyframes.
Additionally, we extend our pipeline to large scenes by using multiple local
mapping networks. Extensive experiments on well-known benchmark datasets
confirm that our approach provides robust tracking, mapping, and semantic
labeling even with noisy, sparse, or no input depth. Overall, our proposed
algorithm can greatly enhance scene perception and assist with a range of robot
control problems.
| no_new_dataset | 0.711481 |
2305.00663 | Li-Yue Sun | John Chiang | Activation Functions Not To Active: A Plausible Theory on Interpreting
Neural Networks | 11 pages, 3 figures | null | null | null | cs.LG cs.NE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Researchers commonly believe that neural networks model a high-dimensional
space but cannot give a clear definition of this space. What is this space?
What is its dimension? And does it has finite dimensions? In this paper, we
develop a plausible theory on interpreting neural networks in terms of the role
of activation functions in neural networks and define a high-dimensional (more
precisely, an infinite-dimensional) space that neural networks including
deep-learning networks could create. We show that the activation function acts
as a magnifying function that maps the low-dimensional linear space into an
infinite-dimensional space, which can distinctly identify the polynomial
approximation of any multivariate continuous function of the variable values
being the same features of the given dataset. Given a dataset with each example
of $d$ features $f_1$, $f_2$, $\cdots$, $f_d$, we believe that neural networks
model a special space with infinite dimensions, each of which is a monomial
$$\prod_{i_1, i_2, \cdots, i_d} f_1^{i_1} f_2^{i_2} \cdots f_d^{i_d}$$ for some
non-negative integers ${i_1, i_2, \cdots, i_d} \in
\mathbb{Z}_{0}^{+}=\{0,1,2,3,\ldots\} $. We term such an infinite-dimensional
space a $\textit{ Super Space (SS)}$. We see such a dimension as the minimum
information unit. Every neuron node previously through an activation layer in
neural networks is a $\textit{ Super Plane (SP) }$, which is actually a
polynomial of infinite degree. This $\textit{ Super Space }$ is something like
a coordinate system, in which every multivalue function can be represented by a
$\textit{ Super Plane }$. We also show that training NNs could at least be
reduced to solving a system of nonlinear equations. %solve sets of nonlinear
equations
| [
{
"version": "v1",
"created": "Mon, 1 May 2023 05:23:58 GMT"
},
{
"version": "v2",
"created": "Tue, 9 May 2023 07:10:11 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Chiang",
"John",
""
]
] | TITLE: Activation Functions Not To Active: A Plausible Theory on Interpreting
Neural Networks
ABSTRACT: Researchers commonly believe that neural networks model a high-dimensional
space but cannot give a clear definition of this space. What is this space?
What is its dimension? And does it has finite dimensions? In this paper, we
develop a plausible theory on interpreting neural networks in terms of the role
of activation functions in neural networks and define a high-dimensional (more
precisely, an infinite-dimensional) space that neural networks including
deep-learning networks could create. We show that the activation function acts
as a magnifying function that maps the low-dimensional linear space into an
infinite-dimensional space, which can distinctly identify the polynomial
approximation of any multivariate continuous function of the variable values
being the same features of the given dataset. Given a dataset with each example
of $d$ features $f_1$, $f_2$, $\cdots$, $f_d$, we believe that neural networks
model a special space with infinite dimensions, each of which is a monomial
$$\prod_{i_1, i_2, \cdots, i_d} f_1^{i_1} f_2^{i_2} \cdots f_d^{i_d}$$ for some
non-negative integers ${i_1, i_2, \cdots, i_d} \in
\mathbb{Z}_{0}^{+}=\{0,1,2,3,\ldots\} $. We term such an infinite-dimensional
space a $\textit{ Super Space (SS)}$. We see such a dimension as the minimum
information unit. Every neuron node previously through an activation layer in
neural networks is a $\textit{ Super Plane (SP) }$, which is actually a
polynomial of infinite degree. This $\textit{ Super Space }$ is something like
a coordinate system, in which every multivalue function can be represented by a
$\textit{ Super Plane }$. We also show that training NNs could at least be
reduced to solving a system of nonlinear equations. %solve sets of nonlinear
equations
| no_new_dataset | 0.709372 |
2305.01118 | Gengchen Mai | Gengchen Mai, Ni Lao, Yutong He, Jiaming Song, Stefano Ermon | CSP: Self-Supervised Contrastive Spatial Pre-Training for
Geospatial-Visual Representations | In: ICML 2023, Jul 23 - 29, 2023, Honolulu, Hawaii, USA | null | null | null | cs.CV cs.AI cs.LG | http://creativecommons.org/publicdomain/zero/1.0/ | Geo-tagged images are publicly available in large quantities, whereas labels
such as object classes are rather scarce and expensive to collect. Meanwhile,
contrastive learning has achieved tremendous success in various natural image
and language tasks with limited labeled data. However, existing methods fail to
fully leverage geospatial information, which can be paramount to distinguishing
objects that are visually similar. To directly leverage the abundant geospatial
information associated with images in pre-training, fine-tuning, and inference
stages, we present Contrastive Spatial Pre-Training (CSP), a self-supervised
learning framework for geo-tagged images. We use a dual-encoder to separately
encode the images and their corresponding geo-locations, and use contrastive
objectives to learn effective location representations from images, which can
be transferred to downstream supervised tasks such as image classification.
Experiments show that CSP can improve model performance on both iNat2018 and
fMoW datasets. Especially, on iNat2018, CSP significantly boosts the model
performance with 10-34% relative improvement with various labeled training data
sampling ratios.
| [
{
"version": "v1",
"created": "Mon, 1 May 2023 23:11:18 GMT"
},
{
"version": "v2",
"created": "Tue, 9 May 2023 01:29:35 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Mai",
"Gengchen",
""
],
[
"Lao",
"Ni",
""
],
[
"He",
"Yutong",
""
],
[
"Song",
"Jiaming",
""
],
[
"Ermon",
"Stefano",
""
]
] | TITLE: CSP: Self-Supervised Contrastive Spatial Pre-Training for
Geospatial-Visual Representations
ABSTRACT: Geo-tagged images are publicly available in large quantities, whereas labels
such as object classes are rather scarce and expensive to collect. Meanwhile,
contrastive learning has achieved tremendous success in various natural image
and language tasks with limited labeled data. However, existing methods fail to
fully leverage geospatial information, which can be paramount to distinguishing
objects that are visually similar. To directly leverage the abundant geospatial
information associated with images in pre-training, fine-tuning, and inference
stages, we present Contrastive Spatial Pre-Training (CSP), a self-supervised
learning framework for geo-tagged images. We use a dual-encoder to separately
encode the images and their corresponding geo-locations, and use contrastive
objectives to learn effective location representations from images, which can
be transferred to downstream supervised tasks such as image classification.
Experiments show that CSP can improve model performance on both iNat2018 and
fMoW datasets. Especially, on iNat2018, CSP significantly boosts the model
performance with 10-34% relative improvement with various labeled training data
sampling ratios.
| no_new_dataset | 0.710848 |
2305.01279 | Fangjian Lin | Fangjian Lin, Yizhe Ma, Shengwei Tian | Exploring vision transformer layer choosing for semantic segmentation | Accepted by IEEE ICASSP | null | 10.1109/ICASSP49357.2023.10096645 | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Extensive work has demonstrated the effectiveness of Vision Transformers. The
plain Vision Transformer tends to obtain multi-scale features by selecting
fixed layers, or the last layer of features aiming to achieve higher
performance in dense prediction tasks. However, this selection is often based
on manual operation. And different samples often exhibit different features at
different layers (e.g., edge, structure, texture, detail, etc.). This requires
us to seek a dynamic adaptive fusion method to filter different layer features.
In this paper, unlike previous encoder and decoder work, we design a neck
network for adaptive fusion and feature selection, called ViTController. We
validate the effectiveness of our method on different datasets and models and
surpass previous state-of-the-art methods. Finally, our method can also be used
as a plug-in module and inserted into different networks.
| [
{
"version": "v1",
"created": "Tue, 2 May 2023 09:29:12 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Lin",
"Fangjian",
""
],
[
"Ma",
"Yizhe",
""
],
[
"Tian",
"Shengwei",
""
]
] | TITLE: Exploring vision transformer layer choosing for semantic segmentation
ABSTRACT: Extensive work has demonstrated the effectiveness of Vision Transformers. The
plain Vision Transformer tends to obtain multi-scale features by selecting
fixed layers, or the last layer of features aiming to achieve higher
performance in dense prediction tasks. However, this selection is often based
on manual operation. And different samples often exhibit different features at
different layers (e.g., edge, structure, texture, detail, etc.). This requires
us to seek a dynamic adaptive fusion method to filter different layer features.
In this paper, unlike previous encoder and decoder work, we design a neck
network for adaptive fusion and feature selection, called ViTController. We
validate the effectiveness of our method on different datasets and models and
surpass previous state-of-the-art methods. Finally, our method can also be used
as a plug-in module and inserted into different networks.
| no_new_dataset | 0.709466 |
2305.02644 | Steffen Czolbe | Steffen Czolbe and Adrian V. Dalca | Neuralizer: General Neuroimage Analysis without Re-Training | Presented at CVPR 2023 Available on github:
https://github.com/SteffenCzolbe/neuralizer | null | null | null | eess.IV cs.CV | http://creativecommons.org/licenses/by/4.0/ | Neuroimage processing tasks like segmentation, reconstruction, and
registration are central to the study of neuroscience. Robust deep learning
strategies and architectures used to solve these tasks are often similar. Yet,
when presented with a new task or a dataset with different visual
characteristics, practitioners most often need to train a new model, or
fine-tune an existing one. This is a time-consuming process that poses a
substantial barrier for the thousands of neuroscientists and clinical
researchers who often lack the resources or machine-learning expertise to train
deep learning models. In practice, this leads to a lack of adoption of deep
learning, and neuroscience tools being dominated by classical frameworks.
We introduce Neuralizer, a single model that generalizes to previously unseen
neuroimaging tasks and modalities without the need for re-training or
fine-tuning. Tasks do not have to be known a priori, and generalization happens
in a single forward pass during inference. The model can solve processing tasks
across multiple image modalities, acquisition methods, and datasets, and
generalize to tasks and modalities it has not been trained on. Our experiments
on coronal slices show that when few annotated subjects are available, our
multi-task network outperforms task-specific baselines without training on the
task.
| [
{
"version": "v1",
"created": "Thu, 4 May 2023 08:26:53 GMT"
},
{
"version": "v2",
"created": "Tue, 9 May 2023 13:39:04 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Czolbe",
"Steffen",
""
],
[
"Dalca",
"Adrian V.",
""
]
] | TITLE: Neuralizer: General Neuroimage Analysis without Re-Training
ABSTRACT: Neuroimage processing tasks like segmentation, reconstruction, and
registration are central to the study of neuroscience. Robust deep learning
strategies and architectures used to solve these tasks are often similar. Yet,
when presented with a new task or a dataset with different visual
characteristics, practitioners most often need to train a new model, or
fine-tune an existing one. This is a time-consuming process that poses a
substantial barrier for the thousands of neuroscientists and clinical
researchers who often lack the resources or machine-learning expertise to train
deep learning models. In practice, this leads to a lack of adoption of deep
learning, and neuroscience tools being dominated by classical frameworks.
We introduce Neuralizer, a single model that generalizes to previously unseen
neuroimaging tasks and modalities without the need for re-training or
fine-tuning. Tasks do not have to be known a priori, and generalization happens
in a single forward pass during inference. The model can solve processing tasks
across multiple image modalities, acquisition methods, and datasets, and
generalize to tasks and modalities it has not been trained on. Our experiments
on coronal slices show that when few annotated subjects are available, our
multi-task network outperforms task-specific baselines without training on the
task.
| no_new_dataset | 0.708023 |
2305.02718 | Sai Zhang | Sai Zhang, Yuwei Hu, Xiaojie Wang and Caixia Yuan | An Asynchronous Updating Reinforcement Learning Framework for
Task-oriented Dialog System | Accepted by ICASSP 2023 | null | 10.1109/ICASSP49357.2023.10096940 | null | cs.CL cs.AI cs.HC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Reinforcement learning has been applied to train the dialog systems in many
works. Previous approaches divide the dialog system into multiple modules
including DST (dialog state tracking) and DP (dialog policy), and train these
modules simultaneously. However, different modules influence each other during
training. The errors from DST might misguide the dialog policy, and the system
action brings extra difficulties for the DST module. To alleviate this problem,
we propose Asynchronous Updating Reinforcement Learning framework (AURL) that
updates the DST module and the DP module asynchronously under a cooperative
setting. Furthermore, curriculum learning is implemented to address the problem
of unbalanced data distribution during reinforcement learning sampling, and
multiple user models are introduced to increase the dialog diversity. Results
on the public SSD-PHONE dataset show that our method achieves a compelling
result with a 31.37% improvement on the dialog success rate. The code is
publicly available via https://github.com/shunjiu/AURL.
| [
{
"version": "v1",
"created": "Thu, 4 May 2023 10:39:17 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Zhang",
"Sai",
""
],
[
"Hu",
"Yuwei",
""
],
[
"Wang",
"Xiaojie",
""
],
[
"Yuan",
"Caixia",
""
]
] | TITLE: An Asynchronous Updating Reinforcement Learning Framework for
Task-oriented Dialog System
ABSTRACT: Reinforcement learning has been applied to train the dialog systems in many
works. Previous approaches divide the dialog system into multiple modules
including DST (dialog state tracking) and DP (dialog policy), and train these
modules simultaneously. However, different modules influence each other during
training. The errors from DST might misguide the dialog policy, and the system
action brings extra difficulties for the DST module. To alleviate this problem,
we propose Asynchronous Updating Reinforcement Learning framework (AURL) that
updates the DST module and the DP module asynchronously under a cooperative
setting. Furthermore, curriculum learning is implemented to address the problem
of unbalanced data distribution during reinforcement learning sampling, and
multiple user models are introduced to increase the dialog diversity. Results
on the public SSD-PHONE dataset show that our method achieves a compelling
result with a 31.37% improvement on the dialog success rate. The code is
publicly available via https://github.com/shunjiu/AURL.
| no_new_dataset | 0.710007 |
2305.03953 | Xu Chen | Xu Chen and Zida Cheng and Shuai Xiao and Xiaoyi Zeng and Weilin Huang | Cross-domain Augmentation Networks for Click-Through Rate Prediction | null | null | null | null | cs.IR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Data sparsity is an important issue for click-through rate (CTR) prediction,
particularly when user-item interactions is too sparse to learn a reliable
model. Recently, many works on cross-domain CTR (CDCTR) prediction have been
developed in an effort to leverage meaningful data from a related domain.
However, most existing CDCTR works have an impractical limitation that requires
homogeneous inputs (\textit{i.e.} shared feature fields) across domains, and
CDCTR with heterogeneous inputs (\textit{i.e.} varying feature fields) across
domains has not been widely explored but is an urgent and important research
problem. In this work, we propose a cross-domain augmentation network (CDAnet)
being able to perform knowledge transfer between two domains with
\textit{heterogeneous inputs}. Specifically, CDAnet contains a designed
translation network and an augmentation network which are trained sequentially.
The translation network is able to compute features from two domains with
heterogeneous inputs separately by designing two independent branches, and then
learn meaningful cross-domain knowledge using a designed cross-supervised
feature translator. Later the augmentation network encodes the learned
cross-domain knowledge via feature translation performed in the latent space
and fine-tune the model for final CTR prediction. Through extensive experiments
on two public benchmarks and one industrial production dataset, we show CDAnet
can learn meaningful translated features and largely improve the performance of
CTR prediction. CDAnet has been conducted online A/B test in image2product
retrieval at Taobao app over 20days, bringing an absolute \textbf{0.11 point}
CTR improvement and a relative \textbf{1.26\%} GMV increase.
| [
{
"version": "v1",
"created": "Sat, 6 May 2023 06:37:52 GMT"
},
{
"version": "v2",
"created": "Tue, 9 May 2023 08:43:29 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Chen",
"Xu",
""
],
[
"Cheng",
"Zida",
""
],
[
"Xiao",
"Shuai",
""
],
[
"Zeng",
"Xiaoyi",
""
],
[
"Huang",
"Weilin",
""
]
] | TITLE: Cross-domain Augmentation Networks for Click-Through Rate Prediction
ABSTRACT: Data sparsity is an important issue for click-through rate (CTR) prediction,
particularly when user-item interactions is too sparse to learn a reliable
model. Recently, many works on cross-domain CTR (CDCTR) prediction have been
developed in an effort to leverage meaningful data from a related domain.
However, most existing CDCTR works have an impractical limitation that requires
homogeneous inputs (\textit{i.e.} shared feature fields) across domains, and
CDCTR with heterogeneous inputs (\textit{i.e.} varying feature fields) across
domains has not been widely explored but is an urgent and important research
problem. In this work, we propose a cross-domain augmentation network (CDAnet)
being able to perform knowledge transfer between two domains with
\textit{heterogeneous inputs}. Specifically, CDAnet contains a designed
translation network and an augmentation network which are trained sequentially.
The translation network is able to compute features from two domains with
heterogeneous inputs separately by designing two independent branches, and then
learn meaningful cross-domain knowledge using a designed cross-supervised
feature translator. Later the augmentation network encodes the learned
cross-domain knowledge via feature translation performed in the latent space
and fine-tune the model for final CTR prediction. Through extensive experiments
on two public benchmarks and one industrial production dataset, we show CDAnet
can learn meaningful translated features and largely improve the performance of
CTR prediction. CDAnet has been conducted online A/B test in image2product
retrieval at Taobao app over 20days, bringing an absolute \textbf{0.11 point}
CTR improvement and a relative \textbf{1.26\%} GMV increase.
| no_new_dataset | 0.709629 |
2305.04166 | Nghia Hieu Nguyen | Doanh C. Bui, Nghia Hieu Nguyen, Khang Nguyen | UIT-OpenViIC: A Novel Benchmark for Evaluating Image Captioning in
Vietnamese | 10 pages, 7 figures, submitted to Elsevier | null | null | null | cs.CV cs.CL | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Image Captioning is one of the vision-language tasks that still interest the
research community worldwide in the 2020s. MS-COCO Caption benchmark is
commonly used to evaluate the performance of advanced captioning models,
although it was published in 2015. Recent captioning models trained on the
MS-COCO Caption dataset only have good performance in language patterns of
English; they do not have such good performance in contexts captured in Vietnam
or fluently caption images using Vietnamese. To contribute to the low-resources
research community as in Vietnam, we introduce a novel image captioning dataset
in Vietnamese, the Open-domain Vietnamese Image Captioning dataset
(UIT-OpenViIC). The introduced dataset includes complex scenes captured in
Vietnam and manually annotated by Vietnamese under strict rules and
supervision. In this paper, we present in more detail the dataset creation
process. From preliminary analysis, we show that our dataset is challenging to
recent state-of-the-art (SOTA) Transformer-based baselines, which performed
well on the MS COCO dataset. Then, the modest results prove that UIT-OpenViIC
has room to grow, which can be one of the standard benchmarks in Vietnamese for
the research community to evaluate their captioning models. Furthermore, we
present a CAMO approach that effectively enhances the image representation
ability by a multi-level encoder output fusion mechanism, which helps improve
the quality of generated captions compared to previous captioning models.
| [
{
"version": "v1",
"created": "Sun, 7 May 2023 02:48:47 GMT"
},
{
"version": "v2",
"created": "Tue, 9 May 2023 12:46:06 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Bui",
"Doanh C.",
""
],
[
"Nguyen",
"Nghia Hieu",
""
],
[
"Nguyen",
"Khang",
""
]
] | TITLE: UIT-OpenViIC: A Novel Benchmark for Evaluating Image Captioning in
Vietnamese
ABSTRACT: Image Captioning is one of the vision-language tasks that still interest the
research community worldwide in the 2020s. MS-COCO Caption benchmark is
commonly used to evaluate the performance of advanced captioning models,
although it was published in 2015. Recent captioning models trained on the
MS-COCO Caption dataset only have good performance in language patterns of
English; they do not have such good performance in contexts captured in Vietnam
or fluently caption images using Vietnamese. To contribute to the low-resources
research community as in Vietnam, we introduce a novel image captioning dataset
in Vietnamese, the Open-domain Vietnamese Image Captioning dataset
(UIT-OpenViIC). The introduced dataset includes complex scenes captured in
Vietnam and manually annotated by Vietnamese under strict rules and
supervision. In this paper, we present in more detail the dataset creation
process. From preliminary analysis, we show that our dataset is challenging to
recent state-of-the-art (SOTA) Transformer-based baselines, which performed
well on the MS COCO dataset. Then, the modest results prove that UIT-OpenViIC
has room to grow, which can be one of the standard benchmarks in Vietnamese for
the research community to evaluate their captioning models. Furthermore, we
present a CAMO approach that effectively enhances the image representation
ability by a multi-level encoder output fusion mechanism, which helps improve
the quality of generated captions compared to previous captioning models.
| new_dataset | 0.71922 |
2305.04206 | Yu-Min Zhang | Yu-Ming Zhang, Jun-Wei Hsieh, Chun-Chieh Lee, Kuo-Chin Fan | RATs-NAS: Redirection of Adjacent Trails on GCN for Neural Architecture
Search | null | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by/4.0/ | Various hand-designed CNN architectures have been developed, such as VGG,
ResNet, DenseNet, etc., and achieve State-of-the-Art (SoTA) levels on different
tasks. Neural Architecture Search (NAS) now focuses on automatically finding
the best CNN architecture to handle the above tasks. However, the verification
of a searched architecture is very time-consuming and makes predictor-based
methods become an essential and important branch of NAS. Two commonly used
techniques to build predictors are graph-convolution networks (GCN) and
multilayer perceptron (MLP). In this paper, we consider the difference between
GCN and MLP on adjacent operation trails and then propose the Redirected
Adjacent Trails NAS (RATs-NAS) to quickly search for the desired neural network
architecture. The RATs-NAS consists of two components: the Redirected Adjacent
Trails GCN (RATs-GCN) and the Predictor-based Search Space Sampling (P3S)
module. RATs-GCN can change trails and their strengths to search for a better
neural network architecture. P3S can rapidly focus on tighter intervals of
FLOPs in the search space. Based on our observations on cell-based NAS, we
believe that architectures with similar FLOPs will perform similarly. Finally,
the RATs-NAS consisting of RATs-GCN and P3S beats WeakNAS, Arch-Graph, and
others by a significant margin on three sub-datasets of NASBench-201.
| [
{
"version": "v1",
"created": "Sun, 7 May 2023 07:13:33 GMT"
},
{
"version": "v2",
"created": "Tue, 9 May 2023 01:12:25 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Zhang",
"Yu-Ming",
""
],
[
"Hsieh",
"Jun-Wei",
""
],
[
"Lee",
"Chun-Chieh",
""
],
[
"Fan",
"Kuo-Chin",
""
]
] | TITLE: RATs-NAS: Redirection of Adjacent Trails on GCN for Neural Architecture
Search
ABSTRACT: Various hand-designed CNN architectures have been developed, such as VGG,
ResNet, DenseNet, etc., and achieve State-of-the-Art (SoTA) levels on different
tasks. Neural Architecture Search (NAS) now focuses on automatically finding
the best CNN architecture to handle the above tasks. However, the verification
of a searched architecture is very time-consuming and makes predictor-based
methods become an essential and important branch of NAS. Two commonly used
techniques to build predictors are graph-convolution networks (GCN) and
multilayer perceptron (MLP). In this paper, we consider the difference between
GCN and MLP on adjacent operation trails and then propose the Redirected
Adjacent Trails NAS (RATs-NAS) to quickly search for the desired neural network
architecture. The RATs-NAS consists of two components: the Redirected Adjacent
Trails GCN (RATs-GCN) and the Predictor-based Search Space Sampling (P3S)
module. RATs-GCN can change trails and their strengths to search for a better
neural network architecture. P3S can rapidly focus on tighter intervals of
FLOPs in the search space. Based on our observations on cell-based NAS, we
believe that architectures with similar FLOPs will perform similarly. Finally,
the RATs-NAS consisting of RATs-GCN and P3S beats WeakNAS, Arch-Graph, and
others by a significant margin on three sub-datasets of NASBench-201.
| no_new_dataset | 0.712795 |
2305.04553 | Denis Antipov | Alexandra Ivanova, Denis Antipov, Benjamin Doerr | Larger Offspring Populations Help the $(1 + (\lambda, \lambda))$ Genetic
Algorithm to Overcome the Noise | Author-generated version of the same paper published at GECCO 2023 | null | null | null | cs.NE cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Evolutionary algorithms are known to be robust to noise in the evaluation of
the fitness. In particular, larger offspring population sizes often lead to
strong robustness. We analyze to what extent the $(1+(\lambda,\lambda))$
genetic algorithm is robust to noise. This algorithm also works with larger
offspring population sizes, but an intermediate selection step and a
non-standard use of crossover as repair mechanism could render this algorithm
less robust than, e.g., the simple $(1+\lambda)$ evolutionary algorithm. Our
experimental analysis on several classic benchmark problems shows that this
difficulty does not arise. Surprisingly, in many situations this algorithm is
even more robust to noise than the $(1+\lambda)$~EA.
| [
{
"version": "v1",
"created": "Mon, 8 May 2023 08:49:01 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Ivanova",
"Alexandra",
""
],
[
"Antipov",
"Denis",
""
],
[
"Doerr",
"Benjamin",
""
]
] | TITLE: Larger Offspring Populations Help the $(1 + (\lambda, \lambda))$ Genetic
Algorithm to Overcome the Noise
ABSTRACT: Evolutionary algorithms are known to be robust to noise in the evaluation of
the fitness. In particular, larger offspring population sizes often lead to
strong robustness. We analyze to what extent the $(1+(\lambda,\lambda))$
genetic algorithm is robust to noise. This algorithm also works with larger
offspring population sizes, but an intermediate selection step and a
non-standard use of crossover as repair mechanism could render this algorithm
less robust than, e.g., the simple $(1+\lambda)$ evolutionary algorithm. Our
experimental analysis on several classic benchmark problems shows that this
difficulty does not arise. Surprisingly, in many situations this algorithm is
even more robust to noise than the $(1+\lambda)$~EA.
| no_new_dataset | 0.710189 |
2305.04883 | Mahmood Khalsan | Mahmood Khalsan, Mu Mu, Eman Salih Al-Shamery, Lee Machado, Suraj
Ajit, Michael Opoku Agyeman | Fuzzy Gene Selection and Cancer Classification Based on Deep Learning
Model | Journal of Intelligent Information Systems (25,17) | null | null | null | q-bio.GN cs.LG | http://creativecommons.org/licenses/by-sa/4.0/ | Machine learning (ML) approaches have been used to develop highly accurate
and efficient applications in many fields including bio-medical science.
However, even with advanced ML techniques, cancer classification using gene
expression data is still complicated because of the high dimensionality of the
datasets employed. We developed a new fuzzy gene selection technique (FGS) to
identify informative genes to facilitate cancer classification and reduce the
dimensionality of the available gene expression data. Three feature selection
methods (Mutual Information, F-ClassIf, and Chi-squared) were evaluated and
employed to obtain the score and rank for each gene. Then, using Fuzzification
and Defuzzification methods to obtain the best single score for each gene,
which aids in the identification of significant genes. Our study applied the
fuzzy measures to six gene expression datasets including four Microarray and
two RNA-seq datasets for evaluating the proposed algorithm. With our
FGS-enhanced method, the cancer classification model achieved 96.5%,96.2%,96%,
and 95.9% for accuracy, precision, recall, and f1-score respectively, which is
significantly higher than 69.2% accuracy, 57.8% precision, 66% recall, and
58.2% f1-score when the standard MLP method was used. In examining the six
datasets that were used, the proposed model demonstrates it's capacity to
classify cancer effectively.
| [
{
"version": "v1",
"created": "Thu, 4 May 2023 21:52:57 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Khalsan",
"Mahmood",
""
],
[
"Mu",
"Mu",
""
],
[
"Al-Shamery",
"Eman Salih",
""
],
[
"Machado",
"Lee",
""
],
[
"Ajit",
"Suraj",
""
],
[
"Agyeman",
"Michael Opoku",
""
]
] | TITLE: Fuzzy Gene Selection and Cancer Classification Based on Deep Learning
Model
ABSTRACT: Machine learning (ML) approaches have been used to develop highly accurate
and efficient applications in many fields including bio-medical science.
However, even with advanced ML techniques, cancer classification using gene
expression data is still complicated because of the high dimensionality of the
datasets employed. We developed a new fuzzy gene selection technique (FGS) to
identify informative genes to facilitate cancer classification and reduce the
dimensionality of the available gene expression data. Three feature selection
methods (Mutual Information, F-ClassIf, and Chi-squared) were evaluated and
employed to obtain the score and rank for each gene. Then, using Fuzzification
and Defuzzification methods to obtain the best single score for each gene,
which aids in the identification of significant genes. Our study applied the
fuzzy measures to six gene expression datasets including four Microarray and
two RNA-seq datasets for evaluating the proposed algorithm. With our
FGS-enhanced method, the cancer classification model achieved 96.5%,96.2%,96%,
and 95.9% for accuracy, precision, recall, and f1-score respectively, which is
significantly higher than 69.2% accuracy, 57.8% precision, 66% recall, and
58.2% f1-score when the standard MLP method was used. In examining the six
datasets that were used, the proposed model demonstrates it's capacity to
classify cancer effectively.
| no_new_dataset | 0.712389 |
2305.04961 | Richard Luo | Richard Luo, Austin Peng, Heidi Yap and Koby Beard | Joint Moment Retrieval and Highlight Detection Via Natural Language
Queries | null | null | null | null | cs.CV cs.CL cs.LG | http://creativecommons.org/licenses/by/4.0/ | Video summarization has become an increasingly important task in the field of
computer vision due to the vast amount of video content available on the
internet. In this project, we propose a new method for natural language query
based joint video summarization and highlight detection using multi-modal
transformers. This approach will use both visual and audio cues to match a
user's natural language query to retrieve the most relevant and interesting
moments from a video. Our approach employs multiple recent techniques used in
Vision Transformers (ViTs) to create a transformer-like encoder-decoder model.
We evaluated our approach on multiple datasets such as YouTube Highlights and
TVSum to demonstrate the flexibility of our proposed method.
| [
{
"version": "v1",
"created": "Mon, 8 May 2023 18:00:33 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Luo",
"Richard",
""
],
[
"Peng",
"Austin",
""
],
[
"Yap",
"Heidi",
""
],
[
"Beard",
"Koby",
""
]
] | TITLE: Joint Moment Retrieval and Highlight Detection Via Natural Language
Queries
ABSTRACT: Video summarization has become an increasingly important task in the field of
computer vision due to the vast amount of video content available on the
internet. In this project, we propose a new method for natural language query
based joint video summarization and highlight detection using multi-modal
transformers. This approach will use both visual and audio cues to match a
user's natural language query to retrieve the most relevant and interesting
moments from a video. Our approach employs multiple recent techniques used in
Vision Transformers (ViTs) to create a transformer-like encoder-decoder model.
We evaluated our approach on multiple datasets such as YouTube Highlights and
TVSum to demonstrate the flexibility of our proposed method.
| no_new_dataset | 0.712013 |
2305.04963 | Cai Zhou | Cai Zhou, Xiyuan Wang, Muhan Zhang | From Relational Pooling to Subgraph GNNs: A Universal Framework for More
Expressive Graph Neural Networks | To be published in ICML 2023. 27 pages, 5 figures | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by/4.0/ | Relational pooling is a framework for building more expressive and
permutation-invariant graph neural networks. However, there is limited
understanding of the exact enhancement in the expressivity of RP and its
connection with the Weisfeiler Lehman hierarchy. Starting from RP, we propose
to explicitly assign labels to nodes as additional features to improve
expressive power of message passing neural networks. The method is then
extended to higher dimensional WL, leading to a novel $k,l$-WL algorithm, a
more general framework than $k$-WL. Theoretically, we analyze the expressivity
of $k,l$-WL with respect to $k$ and $l$ and unifies it with a great number of
subgraph GNNs. Complexity reduction methods are also systematically discussed
to build powerful and practical $k,l$-GNN instances. We theoretically and
experimentally prove that our method is universally compatible and capable of
improving the expressivity of any base GNN model. Our $k,l$-GNNs achieve
superior performance on many synthetic and real-world datasets, which verifies
the effectiveness of our framework.
| [
{
"version": "v1",
"created": "Mon, 8 May 2023 18:00:50 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Zhou",
"Cai",
""
],
[
"Wang",
"Xiyuan",
""
],
[
"Zhang",
"Muhan",
""
]
] | TITLE: From Relational Pooling to Subgraph GNNs: A Universal Framework for More
Expressive Graph Neural Networks
ABSTRACT: Relational pooling is a framework for building more expressive and
permutation-invariant graph neural networks. However, there is limited
understanding of the exact enhancement in the expressivity of RP and its
connection with the Weisfeiler Lehman hierarchy. Starting from RP, we propose
to explicitly assign labels to nodes as additional features to improve
expressive power of message passing neural networks. The method is then
extended to higher dimensional WL, leading to a novel $k,l$-WL algorithm, a
more general framework than $k$-WL. Theoretically, we analyze the expressivity
of $k,l$-WL with respect to $k$ and $l$ and unifies it with a great number of
subgraph GNNs. Complexity reduction methods are also systematically discussed
to build powerful and practical $k,l$-GNN instances. We theoretically and
experimentally prove that our method is universally compatible and capable of
improving the expressivity of any base GNN model. Our $k,l$-GNNs achieve
superior performance on many synthetic and real-world datasets, which verifies
the effectiveness of our framework.
| no_new_dataset | 0.706981 |
2305.04989 | Kaushik Roy | Kaushik Roy, Tarun Garg, Vedant Palit, Yuxin Zi, Vignesh Narayanan,
Amit Sheth | Knowledge Graph Guided Semantic Evaluation of Language Models For User
Trust | null | null | null | null | cs.CL cs.AI | http://creativecommons.org/licenses/by/4.0/ | A fundamental question in natural language processing is - what kind of
language structure and semantics is the language model capturing? Graph formats
such as knowledge graphs are easy to evaluate as they explicitly express
language semantics and structure. This study evaluates the semantics encoded in
the self-attention transformers by leveraging explicit knowledge graph
structures. We propose novel metrics to measure the reconstruction error when
providing graph path sequences from a knowledge graph and trying to
reproduce/reconstruct the same from the outputs of the self-attention
transformer models. The opacity of language models has an immense bearing on
societal issues of trust and explainable decision outcomes. Our findings
suggest that language models are models of stochastic control processes for
plausible language pattern generation. However, they do not ascribe object and
concept-level meaning and semantics to the learned stochastic patterns such as
those described in knowledge graphs. Furthermore, to enable robust evaluation
of concept understanding by language models, we construct and make public an
augmented language understanding benchmark built on the General Language
Understanding Evaluation (GLUE) benchmark. This has significant
application-level user trust implications as stochastic patterns without a
strong sense of meaning cannot be trusted in high-stakes applications.
| [
{
"version": "v1",
"created": "Mon, 8 May 2023 18:53:14 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Roy",
"Kaushik",
""
],
[
"Garg",
"Tarun",
""
],
[
"Palit",
"Vedant",
""
],
[
"Zi",
"Yuxin",
""
],
[
"Narayanan",
"Vignesh",
""
],
[
"Sheth",
"Amit",
""
]
] | TITLE: Knowledge Graph Guided Semantic Evaluation of Language Models For User
Trust
ABSTRACT: A fundamental question in natural language processing is - what kind of
language structure and semantics is the language model capturing? Graph formats
such as knowledge graphs are easy to evaluate as they explicitly express
language semantics and structure. This study evaluates the semantics encoded in
the self-attention transformers by leveraging explicit knowledge graph
structures. We propose novel metrics to measure the reconstruction error when
providing graph path sequences from a knowledge graph and trying to
reproduce/reconstruct the same from the outputs of the self-attention
transformer models. The opacity of language models has an immense bearing on
societal issues of trust and explainable decision outcomes. Our findings
suggest that language models are models of stochastic control processes for
plausible language pattern generation. However, they do not ascribe object and
concept-level meaning and semantics to the learned stochastic patterns such as
those described in knowledge graphs. Furthermore, to enable robust evaluation
of concept understanding by language models, we construct and make public an
augmented language understanding benchmark built on the General Language
Understanding Evaluation (GLUE) benchmark. This has significant
application-level user trust implications as stochastic patterns without a
strong sense of meaning cannot be trusted in high-stakes applications.
| no_new_dataset | 0.580411 |
2305.04992 | Tsvetan Yordanov | Tsvetan R. Yordanov, Ameen Abu-Hanna, Anita CJ Ravelli, Iacopo
Vagliano | Autoencoder-based prediction of ICU clinical codes | Extended version of 5-page short paper submitted to AIME23 conference | null | null | null | cs.LG cs.IR | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Availability of diagnostic codes in Electronic Health Records (EHRs) is
crucial for patient care as well as reimbursement purposes. However, entering
them in the EHR is tedious, and some clinical codes may be overlooked. Given an
in-complete list of clinical codes, we investigate the performance of ML
methods on predicting the complete ones, and assess the added predictive value
of including other clinical patient data in this task. We used the MIMIC-III
dataset and frame the task of completing the clinical codes as a recommendation
problem. We con-sider various autoencoder approaches plus two strong baselines;
item co-occurrence and Singular Value Decomposition (SVD). Inputs are 1) a
record's known clinical codes, 2) the codes plus variables. The
co-occurrence-based ap-proach performed slightly better (F1 score=0.26, Mean
Average Precision [MAP]=0.19) than the SVD (F1=0.24, MAP=0.18). However, the
adversarial autoencoder achieved the best performance when using the codes plus
variables (F1=0.32, MAP=0.25). Adversarial autoencoders performed best in terms
of F1 and were equal to vanilla and denoising autoencoders in term of MAP.
Using clinical variables in addition to the incomplete codes list, improves the
predictive performance of the models.
| [
{
"version": "v1",
"created": "Mon, 8 May 2023 18:56:37 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Yordanov",
"Tsvetan R.",
""
],
[
"Abu-Hanna",
"Ameen",
""
],
[
"Ravelli",
"Anita CJ",
""
],
[
"Vagliano",
"Iacopo",
""
]
] | TITLE: Autoencoder-based prediction of ICU clinical codes
ABSTRACT: Availability of diagnostic codes in Electronic Health Records (EHRs) is
crucial for patient care as well as reimbursement purposes. However, entering
them in the EHR is tedious, and some clinical codes may be overlooked. Given an
in-complete list of clinical codes, we investigate the performance of ML
methods on predicting the complete ones, and assess the added predictive value
of including other clinical patient data in this task. We used the MIMIC-III
dataset and frame the task of completing the clinical codes as a recommendation
problem. We con-sider various autoencoder approaches plus two strong baselines;
item co-occurrence and Singular Value Decomposition (SVD). Inputs are 1) a
record's known clinical codes, 2) the codes plus variables. The
co-occurrence-based ap-proach performed slightly better (F1 score=0.26, Mean
Average Precision [MAP]=0.19) than the SVD (F1=0.24, MAP=0.18). However, the
adversarial autoencoder achieved the best performance when using the codes plus
variables (F1=0.32, MAP=0.25). Adversarial autoencoders performed best in terms
of F1 and were equal to vanilla and denoising autoencoders in term of MAP.
Using clinical variables in addition to the incomplete codes list, improves the
predictive performance of the models.
| no_new_dataset | 0.709629 |
2305.04994 | Rapha\"el d'Andrimont | Momchil Yordanov, Raphael d'Andrimont, Laura Martinez-Sanchez, Guido
Lemoine, Dominique Fasbender, Marijn van der Velde | Crop identification using deep learning on LUCAS crop cover photos | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Crop classification via deep learning on ground imagery can deliver timely
and accurate crop-specific information to various stakeholders. Dedicated
ground-based image acquisition exercises can help to collect data in data
scarce regions, improve control on timing of collection, or when study areas
are to small to monitor via satellite. Automatic labelling is essential when
collecting large volumes of data. One such data collection is the EU's Land Use
Cover Area frame Survey (LUCAS), and in particular, the recently published
LUCAS Cover photos database. The aim of this paper is to select and publish a
subset of LUCAS Cover photos for 12 mature major crops across the EU, to
deploy, benchmark, and identify the best configuration of Mobile-net for the
classification task, to showcase the possibility of using entropy-based metrics
for post-processing of results, and finally to show the applications and
limitations of the model in a practical and policy relevant context. In
particular, the usefulness of automatically identifying crops on geo-tagged
photos is illustrated in the context of the EU's Common Agricultural Policy.
The work has produced a dataset of 169,460 images of mature crops for the 12
classes, out of which 15,876 were manually selected as representing a clean
sample without any foreign objects or unfavorable conditions. The best
performing model achieved a Macro F1 (M-F1) of 0.75 on an imbalanced test
dataset of 8,642 photos. Using metrics from information theory, namely - the
Equivalence Reference Probability, resulted in achieving an increase of 6%. The
most unfavorable conditions for taking such images, across all crop classes,
were found to be too early or late in the season. The proposed methodology
shows the possibility for using minimal auxiliary data, outside the images
themselves, in order to achieve a M-F1 of 0.817 for labelling between 12 major
European crops.
| [
{
"version": "v1",
"created": "Mon, 8 May 2023 19:03:21 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Yordanov",
"Momchil",
""
],
[
"d'Andrimont",
"Raphael",
""
],
[
"Martinez-Sanchez",
"Laura",
""
],
[
"Lemoine",
"Guido",
""
],
[
"Fasbender",
"Dominique",
""
],
[
"van der Velde",
"Marijn",
""
]
] | TITLE: Crop identification using deep learning on LUCAS crop cover photos
ABSTRACT: Crop classification via deep learning on ground imagery can deliver timely
and accurate crop-specific information to various stakeholders. Dedicated
ground-based image acquisition exercises can help to collect data in data
scarce regions, improve control on timing of collection, or when study areas
are to small to monitor via satellite. Automatic labelling is essential when
collecting large volumes of data. One such data collection is the EU's Land Use
Cover Area frame Survey (LUCAS), and in particular, the recently published
LUCAS Cover photos database. The aim of this paper is to select and publish a
subset of LUCAS Cover photos for 12 mature major crops across the EU, to
deploy, benchmark, and identify the best configuration of Mobile-net for the
classification task, to showcase the possibility of using entropy-based metrics
for post-processing of results, and finally to show the applications and
limitations of the model in a practical and policy relevant context. In
particular, the usefulness of automatically identifying crops on geo-tagged
photos is illustrated in the context of the EU's Common Agricultural Policy.
The work has produced a dataset of 169,460 images of mature crops for the 12
classes, out of which 15,876 were manually selected as representing a clean
sample without any foreign objects or unfavorable conditions. The best
performing model achieved a Macro F1 (M-F1) of 0.75 on an imbalanced test
dataset of 8,642 photos. Using metrics from information theory, namely - the
Equivalence Reference Probability, resulted in achieving an increase of 6%. The
most unfavorable conditions for taking such images, across all crop classes,
were found to be too early or late in the season. The proposed methodology
shows the possibility for using minimal auxiliary data, outside the images
themselves, in order to achieve a M-F1 of 0.817 for labelling between 12 major
European crops.
| no_new_dataset | 0.698394 |
2305.05010 | Jiaming Shen | Rongzhi Zhang, Jiaming Shen, Tianqi Liu, Jialu Liu, Michael Bendersky,
Marc Najork, Chao Zhang | Do Not Blindly Imitate the Teacher: Using Perturbed Loss for Knowledge
Distillation | 16 pages | null | null | null | cs.LG cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Knowledge distillation is a popular technique to transfer knowledge from
large teacher models to a small student model. Typically, the student learns to
imitate the teacher by minimizing the KL divergence of its output distribution
with the teacher's output distribution. In this work, we argue that such a
learning objective is sub-optimal because there exists a discrepancy between
the teacher's output distribution and the ground truth label distribution.
Therefore, forcing the student to blindly imitate the unreliable teacher output
distribution leads to inferior performance. To this end, we propose a novel
knowledge distillation objective PTLoss by first representing the vanilla
KL-based distillation loss function via a Maclaurin series and then perturbing
the leading-order terms in this series. This perturbed loss implicitly
transforms the original teacher into a proxy teacher with a distribution closer
to the ground truth distribution. We establish the theoretical connection
between this "distribution closeness" and the student model generalizability,
which enables us to select the PTLoss's perturbation coefficients in a
principled way. Extensive experiments on five datasets demonstrate PTLoss can
significantly improve the distillation effectiveness for teachers of various
scales.
| [
{
"version": "v1",
"created": "Mon, 8 May 2023 19:31:09 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Zhang",
"Rongzhi",
""
],
[
"Shen",
"Jiaming",
""
],
[
"Liu",
"Tianqi",
""
],
[
"Liu",
"Jialu",
""
],
[
"Bendersky",
"Michael",
""
],
[
"Najork",
"Marc",
""
],
[
"Zhang",
"Chao",
""
]
] | TITLE: Do Not Blindly Imitate the Teacher: Using Perturbed Loss for Knowledge
Distillation
ABSTRACT: Knowledge distillation is a popular technique to transfer knowledge from
large teacher models to a small student model. Typically, the student learns to
imitate the teacher by minimizing the KL divergence of its output distribution
with the teacher's output distribution. In this work, we argue that such a
learning objective is sub-optimal because there exists a discrepancy between
the teacher's output distribution and the ground truth label distribution.
Therefore, forcing the student to blindly imitate the unreliable teacher output
distribution leads to inferior performance. To this end, we propose a novel
knowledge distillation objective PTLoss by first representing the vanilla
KL-based distillation loss function via a Maclaurin series and then perturbing
the leading-order terms in this series. This perturbed loss implicitly
transforms the original teacher into a proxy teacher with a distribution closer
to the ground truth distribution. We establish the theoretical connection
between this "distribution closeness" and the student model generalizability,
which enables us to select the PTLoss's perturbation coefficients in a
principled way. Extensive experiments on five datasets demonstrate PTLoss can
significantly improve the distillation effectiveness for teachers of various
scales.
| no_new_dataset | 0.708975 |
2305.05026 | Li Jiang | Li Jiang, Zetong Yang, Shaoshuai Shi, Vladislav Golyanik, Dengxin Dai,
Bernt Schiele | Self-supervised Pre-training with Masked Shape Prediction for 3D Scene
Understanding | CVPR 2023 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Masked signal modeling has greatly advanced self-supervised pre-training for
language and 2D images. However, it is still not fully explored in 3D scene
understanding. Thus, this paper introduces Masked Shape Prediction (MSP), a new
framework to conduct masked signal modeling in 3D scenes. MSP uses the
essential 3D semantic cue, i.e., geometric shape, as the prediction target for
masked points. The context-enhanced shape target consisting of explicit shape
context and implicit deep shape feature is proposed to facilitate exploiting
contextual cues in shape prediction. Meanwhile, the pre-training architecture
in MSP is carefully designed to alleviate the masked shape leakage from point
coordinates. Experiments on multiple 3D understanding tasks on both indoor and
outdoor datasets demonstrate the effectiveness of MSP in learning good feature
representations to consistently boost downstream performance.
| [
{
"version": "v1",
"created": "Mon, 8 May 2023 20:09:19 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Jiang",
"Li",
""
],
[
"Yang",
"Zetong",
""
],
[
"Shi",
"Shaoshuai",
""
],
[
"Golyanik",
"Vladislav",
""
],
[
"Dai",
"Dengxin",
""
],
[
"Schiele",
"Bernt",
""
]
] | TITLE: Self-supervised Pre-training with Masked Shape Prediction for 3D Scene
Understanding
ABSTRACT: Masked signal modeling has greatly advanced self-supervised pre-training for
language and 2D images. However, it is still not fully explored in 3D scene
understanding. Thus, this paper introduces Masked Shape Prediction (MSP), a new
framework to conduct masked signal modeling in 3D scenes. MSP uses the
essential 3D semantic cue, i.e., geometric shape, as the prediction target for
masked points. The context-enhanced shape target consisting of explicit shape
context and implicit deep shape feature is proposed to facilitate exploiting
contextual cues in shape prediction. Meanwhile, the pre-training architecture
in MSP is carefully designed to alleviate the masked shape leakage from point
coordinates. Experiments on multiple 3D understanding tasks on both indoor and
outdoor datasets demonstrate the effectiveness of MSP in learning good feature
representations to consistently boost downstream performance.
| no_new_dataset | 0.709831 |
2305.05033 | Alexandros Daglis | Albert Cho and Anish Saxena and Moinuddin Qureshi and Alexandros
Daglis | A Case for CXL-Centric Server Processors | null | null | null | null | cs.AR | http://creativecommons.org/licenses/by/4.0/ | The memory system is a major performance determinant for server processors.
Ever-growing core counts and datasets demand higher bandwidth and capacity as
well as lower latency from the memory system. To keep up with growing demands,
DDR--the dominant processor interface to memory over the past two decades--has
offered higher bandwidth with every generation. However, because each parallel
DDR interface requires a large number of on-chip pins, the processor's memory
bandwidth is ultimately restrained by its pin-count, which is a scarce
resource. With limited bandwidth, multiple memory requests typically contend
for each memory channel, resulting in significant queuing delays that often
overshadow DRAM's service time and degrade performance.
We present CoaXiaL, a server design that overcomes memory bandwidth
limitations by replacing \textit{all} DDR interfaces to the processor with the
more pin-efficient CXL interface. The widespread adoption and industrial
momentum of CXL makes such a transition possible, offering $4\times$ higher
bandwidth per pin compared to DDR at a modest latency overhead. We demonstrate
that, for a broad range of workloads, CXL's latency premium is more than offset
by its higher bandwidth. As CoaXiaL distributes memory requests across more
channels, it drastically reduces queuing delays and thereby both the average
value and variance of memory access latency. Our evaluation with a variety of
workloads shows that CoaXiaL improves the performance of manycore
throughput-oriented servers by $1.52\times$ on average and by up to $3\times$.
| [
{
"version": "v1",
"created": "Mon, 8 May 2023 20:21:39 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Cho",
"Albert",
""
],
[
"Saxena",
"Anish",
""
],
[
"Qureshi",
"Moinuddin",
""
],
[
"Daglis",
"Alexandros",
""
]
] | TITLE: A Case for CXL-Centric Server Processors
ABSTRACT: The memory system is a major performance determinant for server processors.
Ever-growing core counts and datasets demand higher bandwidth and capacity as
well as lower latency from the memory system. To keep up with growing demands,
DDR--the dominant processor interface to memory over the past two decades--has
offered higher bandwidth with every generation. However, because each parallel
DDR interface requires a large number of on-chip pins, the processor's memory
bandwidth is ultimately restrained by its pin-count, which is a scarce
resource. With limited bandwidth, multiple memory requests typically contend
for each memory channel, resulting in significant queuing delays that often
overshadow DRAM's service time and degrade performance.
We present CoaXiaL, a server design that overcomes memory bandwidth
limitations by replacing \textit{all} DDR interfaces to the processor with the
more pin-efficient CXL interface. The widespread adoption and industrial
momentum of CXL makes such a transition possible, offering $4\times$ higher
bandwidth per pin compared to DDR at a modest latency overhead. We demonstrate
that, for a broad range of workloads, CXL's latency premium is more than offset
by its higher bandwidth. As CoaXiaL distributes memory requests across more
channels, it drastically reduces queuing delays and thereby both the average
value and variance of memory access latency. Our evaluation with a variety of
workloads shows that CoaXiaL improves the performance of manycore
throughput-oriented servers by $1.52\times$ on average and by up to $3\times$.
| no_new_dataset | 0.70978 |
2305.05079 | Neeraj Varshney | Neeraj Varshney, Himanshu Gupta, Eric Robertson, Bing Liu, Chitta
Baral | A Unified Evaluation Framework for Novelty Detection and Accommodation
in NLP with an Instantiation in Authorship Attribution | Findings of ACL 2023 | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | State-of-the-art natural language processing models have been shown to
achieve remarkable performance in 'closed-world' settings where all the labels
in the evaluation set are known at training time. However, in real-world
settings, 'novel' instances that do not belong to any known class are often
observed. This renders the ability to deal with novelties crucial. To initiate
a systematic research in this important area of 'dealing with novelties', we
introduce 'NoveltyTask', a multi-stage task to evaluate a system's performance
on pipelined novelty 'detection' and 'accommodation' tasks. We provide
mathematical formulation of NoveltyTask and instantiate it with the authorship
attribution task that pertains to identifying the correct author of a given
text. We use Amazon reviews corpus and compile a large dataset (consisting of
250k instances across 200 authors/labels) for NoveltyTask. We conduct
comprehensive experiments and explore several baseline methods for the task.
Our results show that the methods achieve considerably low performance making
the task challenging and leaving sufficient room for improvement. Finally, we
believe our work will encourage research in this underexplored area of dealing
with novelties, an important step en route to developing robust systems.
| [
{
"version": "v1",
"created": "Mon, 8 May 2023 22:37:30 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Varshney",
"Neeraj",
""
],
[
"Gupta",
"Himanshu",
""
],
[
"Robertson",
"Eric",
""
],
[
"Liu",
"Bing",
""
],
[
"Baral",
"Chitta",
""
]
] | TITLE: A Unified Evaluation Framework for Novelty Detection and Accommodation
in NLP with an Instantiation in Authorship Attribution
ABSTRACT: State-of-the-art natural language processing models have been shown to
achieve remarkable performance in 'closed-world' settings where all the labels
in the evaluation set are known at training time. However, in real-world
settings, 'novel' instances that do not belong to any known class are often
observed. This renders the ability to deal with novelties crucial. To initiate
a systematic research in this important area of 'dealing with novelties', we
introduce 'NoveltyTask', a multi-stage task to evaluate a system's performance
on pipelined novelty 'detection' and 'accommodation' tasks. We provide
mathematical formulation of NoveltyTask and instantiate it with the authorship
attribution task that pertains to identifying the correct author of a given
text. We use Amazon reviews corpus and compile a large dataset (consisting of
250k instances across 200 authors/labels) for NoveltyTask. We conduct
comprehensive experiments and explore several baseline methods for the task.
Our results show that the methods achieve considerably low performance making
the task challenging and leaving sufficient room for improvement. Finally, we
believe our work will encourage research in this underexplored area of dealing
with novelties, an important step en route to developing robust systems.
| new_dataset | 0.716026 |
2305.05082 | Jing Xiong | Jing Xiong and Yu Zhang | A Unifying Framework of Attention-based Neural Load Forecasting | null | null | 10.1109/ACCESS.2023.3275095 | null | cs.LG eess.SP | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Accurate load forecasting is critical for reliable and efficient planning and
operation of electric power grids. In this paper, we propose a unifying deep
learning framework for load forecasting, which includes time-varying feature
weighting, hierarchical temporal attention, and feature-reinforced error
correction. Our framework adopts a modular design with good generalization
capability. First, the feature-weighting mechanism assigns input features with
temporal weights. Second, a recurrent encoder-decoder structure with
hierarchical attention is developed as a load predictor. The hierarchical
attention enables a similar day selection, which re-evaluates the importance of
historical information at each time step. Third, we develop an error correction
module that explores the errors and learned feature hidden information to
further improve the model's forecasting performance. Experimental results
demonstrate that our proposed framework outperforms existing methods on two
public datasets and performance metrics, with the feature weighting mechanism
and error correction module being critical to achieving superior performance.
Our framework provides an effective solution to the electric load forecasting
problem, which can be further adapted to many other forecasting tasks.
| [
{
"version": "v1",
"created": "Mon, 8 May 2023 22:46:54 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Xiong",
"Jing",
""
],
[
"Zhang",
"Yu",
""
]
] | TITLE: A Unifying Framework of Attention-based Neural Load Forecasting
ABSTRACT: Accurate load forecasting is critical for reliable and efficient planning and
operation of electric power grids. In this paper, we propose a unifying deep
learning framework for load forecasting, which includes time-varying feature
weighting, hierarchical temporal attention, and feature-reinforced error
correction. Our framework adopts a modular design with good generalization
capability. First, the feature-weighting mechanism assigns input features with
temporal weights. Second, a recurrent encoder-decoder structure with
hierarchical attention is developed as a load predictor. The hierarchical
attention enables a similar day selection, which re-evaluates the importance of
historical information at each time step. Third, we develop an error correction
module that explores the errors and learned feature hidden information to
further improve the model's forecasting performance. Experimental results
demonstrate that our proposed framework outperforms existing methods on two
public datasets and performance metrics, with the feature weighting mechanism
and error correction module being critical to achieving superior performance.
Our framework provides an effective solution to the electric load forecasting
problem, which can be further adapted to many other forecasting tasks.
| no_new_dataset | 0.709025 |
2305.05095 | Liangliang Cao | Liangliang Cao, Bowen Zhang, Chen Chen, Yinfei Yang, Xianzhi Du,
Wencong Zhang, Zhiyun Lu, Yantao Zheng | Less is More: Removing Text-regions Improves CLIP Training Efficiency
and Robustness | 10 pages, 8 figures | null | null | null | cs.CV cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The CLIP (Contrastive Language-Image Pre-training) model and its variants are
becoming the de facto backbone in many applications. However, training a CLIP
model from hundreds of millions of image-text pairs can be prohibitively
expensive. Furthermore, the conventional CLIP model doesn't differentiate
between the visual semantics and meaning of text regions embedded in images.
This can lead to non-robustness when the text in the embedded region doesn't
match the image's visual appearance. In this paper, we discuss two effective
approaches to improve the efficiency and robustness of CLIP training: (1)
augmenting the training dataset while maintaining the same number of
optimization steps, and (2) filtering out samples that contain text regions in
the image. By doing so, we significantly improve the classification and
retrieval accuracy on public benchmarks like ImageNet and CoCo. Filtering out
images with text regions also protects the model from typographic attacks. To
verify this, we build a new dataset named ImageNet with Adversarial Text
Regions (ImageNet-Attr). Our filter-based CLIP model demonstrates a top-1
accuracy of 68.78\%, outperforming previous models whose accuracy was all below
50\%.
| [
{
"version": "v1",
"created": "Mon, 8 May 2023 23:47:07 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Cao",
"Liangliang",
""
],
[
"Zhang",
"Bowen",
""
],
[
"Chen",
"Chen",
""
],
[
"Yang",
"Yinfei",
""
],
[
"Du",
"Xianzhi",
""
],
[
"Zhang",
"Wencong",
""
],
[
"Lu",
"Zhiyun",
""
],
[
"Zheng",
"Yantao",
""
]
] | TITLE: Less is More: Removing Text-regions Improves CLIP Training Efficiency
and Robustness
ABSTRACT: The CLIP (Contrastive Language-Image Pre-training) model and its variants are
becoming the de facto backbone in many applications. However, training a CLIP
model from hundreds of millions of image-text pairs can be prohibitively
expensive. Furthermore, the conventional CLIP model doesn't differentiate
between the visual semantics and meaning of text regions embedded in images.
This can lead to non-robustness when the text in the embedded region doesn't
match the image's visual appearance. In this paper, we discuss two effective
approaches to improve the efficiency and robustness of CLIP training: (1)
augmenting the training dataset while maintaining the same number of
optimization steps, and (2) filtering out samples that contain text regions in
the image. By doing so, we significantly improve the classification and
retrieval accuracy on public benchmarks like ImageNet and CoCo. Filtering out
images with text regions also protects the model from typographic attacks. To
verify this, we build a new dataset named ImageNet with Adversarial Text
Regions (ImageNet-Attr). Our filter-based CLIP model demonstrates a top-1
accuracy of 68.78\%, outperforming previous models whose accuracy was all below
50\%.
| new_dataset | 0.71202 |
2305.05111 | Mir Riyanul Islam | Mobyen Uddin Ahmed and Shaibal Barua and Shahina Begum and Mir Riyanul
Islam and Rosina O Weber | When a CBR in Hand is Better than Twins in the Bush | The version of this paper published in ICCBR XCBR '22 contained an
erroneous sum in Equation 3 that we have corrected in this version | ICCBR XCBR '22: 4th Workshop on XCBR: Case-based Reasoning for the
Explanation of Intelligent Systems at ICCBR-2022, September, 2022, Nancy,
France | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by/4.0/ | AI methods referred to as interpretable are often discredited as inaccurate
by supporters of the existence of a trade-off between interpretability and
accuracy. In many problem contexts however this trade-off does not hold. This
paper discusses a regression problem context to predict flight take-off delays
where the most accurate data regression model was trained via the XGBoost
implementation of gradient boosted decision trees. While building an XGB-CBR
Twin and converting the XGBoost feature importance into global weights in the
CBR model, the resultant CBR model alone provides the most accurate local
prediction, maintains the global importance to provide a global explanation of
the model, and offers the most interpretable representation for local
explanations. This resultant CBR model becomes a benchmark of accuracy and
interpretability for this problem context, and hence it is used to evaluate the
two additive feature attribute methods SHAP and LIME to explain the XGBoost
regression model. The results with respect to local accuracy and feature
attribution lead to potentially valuable future work.
| [
{
"version": "v1",
"created": "Tue, 9 May 2023 00:55:09 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Ahmed",
"Mobyen Uddin",
""
],
[
"Barua",
"Shaibal",
""
],
[
"Begum",
"Shahina",
""
],
[
"Islam",
"Mir Riyanul",
""
],
[
"Weber",
"Rosina O",
""
]
] | TITLE: When a CBR in Hand is Better than Twins in the Bush
ABSTRACT: AI methods referred to as interpretable are often discredited as inaccurate
by supporters of the existence of a trade-off between interpretability and
accuracy. In many problem contexts however this trade-off does not hold. This
paper discusses a regression problem context to predict flight take-off delays
where the most accurate data regression model was trained via the XGBoost
implementation of gradient boosted decision trees. While building an XGB-CBR
Twin and converting the XGBoost feature importance into global weights in the
CBR model, the resultant CBR model alone provides the most accurate local
prediction, maintains the global importance to provide a global explanation of
the model, and offers the most interpretable representation for local
explanations. This resultant CBR model becomes a benchmark of accuracy and
interpretability for this problem context, and hence it is used to evaluate the
two additive feature attribute methods SHAP and LIME to explain the XGBoost
regression model. The results with respect to local accuracy and feature
attribution lead to potentially valuable future work.
| no_new_dataset | 0.710396 |
2305.05116 | Yang Yu | Lei Yuan, Feng Chen, Zhongzhang Zhang, Yang Yu | Communication-Robust Multi-Agent Learning by Adaptable Auxiliary
Multi-Agent Adversary Generation | null | null | null | null | cs.LG | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Communication can promote coordination in cooperative Multi-Agent
Reinforcement Learning (MARL). Nowadays, existing works mainly focus on
improving the communication efficiency of agents, neglecting that real-world
communication is much more challenging as there may exist noise or potential
attackers. Thus the robustness of the communication-based policies becomes an
emergent and severe issue that needs more exploration. In this paper, we posit
that the ego system trained with auxiliary adversaries may handle this
limitation and propose an adaptable method of Multi-Agent Auxiliary Adversaries
Generation for robust Communication, dubbed MA3C, to obtain a robust
communication-based policy. In specific, we introduce a novel message-attacking
approach that models the learning of the auxiliary attacker as a cooperative
problem under a shared goal to minimize the coordination ability of the ego
system, with which every information channel may suffer from distinct message
attacks. Furthermore, as naive adversarial training may impede the
generalization ability of the ego system, we design an attacker population
generation approach based on evolutionary learning. Finally, the ego system is
paired with an attacker population and then alternatively trained against the
continuously evolving attackers to improve its robustness, meaning that both
the ego system and the attackers are adaptable. Extensive experiments on
multiple benchmarks indicate that our proposed MA3C provides comparable or
better robustness and generalization ability than other baselines.
| [
{
"version": "v1",
"created": "Tue, 9 May 2023 01:29:46 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Yuan",
"Lei",
""
],
[
"Chen",
"Feng",
""
],
[
"Zhang",
"Zhongzhang",
""
],
[
"Yu",
"Yang",
""
]
] | TITLE: Communication-Robust Multi-Agent Learning by Adaptable Auxiliary
Multi-Agent Adversary Generation
ABSTRACT: Communication can promote coordination in cooperative Multi-Agent
Reinforcement Learning (MARL). Nowadays, existing works mainly focus on
improving the communication efficiency of agents, neglecting that real-world
communication is much more challenging as there may exist noise or potential
attackers. Thus the robustness of the communication-based policies becomes an
emergent and severe issue that needs more exploration. In this paper, we posit
that the ego system trained with auxiliary adversaries may handle this
limitation and propose an adaptable method of Multi-Agent Auxiliary Adversaries
Generation for robust Communication, dubbed MA3C, to obtain a robust
communication-based policy. In specific, we introduce a novel message-attacking
approach that models the learning of the auxiliary attacker as a cooperative
problem under a shared goal to minimize the coordination ability of the ego
system, with which every information channel may suffer from distinct message
attacks. Furthermore, as naive adversarial training may impede the
generalization ability of the ego system, we design an attacker population
generation approach based on evolutionary learning. Finally, the ego system is
paired with an attacker population and then alternatively trained against the
continuously evolving attackers to improve its robustness, meaning that both
the ego system and the attackers are adaptable. Extensive experiments on
multiple benchmarks indicate that our proposed MA3C provides comparable or
better robustness and generalization ability than other baselines.
| no_new_dataset | 0.708395 |
2305.05128 | Ziheng Geng | Ziheng Geng, Chao Zhang, Yuhao Ren, Minxiang Zhu, Renpeng Chen, and
Hongzhan Cheng | A Kriging-Random Forest Hybrid Model for Real-time Ground Property
Prediction during Earth Pressure Balance Shield Tunneling | null | null | null | null | cs.LG cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A kriging-random forest hybrid model is developed for real-time ground
property prediction ahead of the earth pressure balanced shield by integrating
Kriging extrapolation and random forest, which can guide shield operating
parameter selection thereby mitigate construction risks. The proposed KRF
algorithm synergizes two types of information: prior information and real-time
information. The previously predicted ground properties with EPB operating
parameters are extrapolated via the Kriging algorithm to provide prior
information for the prediction of currently being excavated ground properties.
The real-time information refers to the real-time operating parameters of the
EPB shield, which are input into random forest to provide a real-time
prediction of ground properties. The integration of these two predictions is
achieved by assigning weights to each prediction according to their
uncertainties, ensuring the prediction of KRF with minimum uncertainty. The
performance of the KRF algorithm is assessed via a case study of the Changsha
Metro Line 4 project. It reveals that the proposed KRF algorithm can predict
ground properties with an accuracy of 93%, overperforming the existing
algorithms of LightGBM, AdaBoost-CART, and DNN by 29%, 8%, and 12%,
respectively. Another dataset from Shenzhen Metro Line 13 project is utilized
to further evaluate the model generalization performance, revealing that the
model can transfer its learned knowledge from one region to another with an
accuracy of 89%.
| [
{
"version": "v1",
"created": "Tue, 9 May 2023 02:16:48 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Geng",
"Ziheng",
""
],
[
"Zhang",
"Chao",
""
],
[
"Ren",
"Yuhao",
""
],
[
"Zhu",
"Minxiang",
""
],
[
"Chen",
"Renpeng",
""
],
[
"Cheng",
"Hongzhan",
""
]
] | TITLE: A Kriging-Random Forest Hybrid Model for Real-time Ground Property
Prediction during Earth Pressure Balance Shield Tunneling
ABSTRACT: A kriging-random forest hybrid model is developed for real-time ground
property prediction ahead of the earth pressure balanced shield by integrating
Kriging extrapolation and random forest, which can guide shield operating
parameter selection thereby mitigate construction risks. The proposed KRF
algorithm synergizes two types of information: prior information and real-time
information. The previously predicted ground properties with EPB operating
parameters are extrapolated via the Kriging algorithm to provide prior
information for the prediction of currently being excavated ground properties.
The real-time information refers to the real-time operating parameters of the
EPB shield, which are input into random forest to provide a real-time
prediction of ground properties. The integration of these two predictions is
achieved by assigning weights to each prediction according to their
uncertainties, ensuring the prediction of KRF with minimum uncertainty. The
performance of the KRF algorithm is assessed via a case study of the Changsha
Metro Line 4 project. It reveals that the proposed KRF algorithm can predict
ground properties with an accuracy of 93%, overperforming the existing
algorithms of LightGBM, AdaBoost-CART, and DNN by 29%, 8%, and 12%,
respectively. Another dataset from Shenzhen Metro Line 13 project is utilized
to further evaluate the model generalization performance, revealing that the
model can transfer its learned knowledge from one region to another with an
accuracy of 89%.
| no_new_dataset | 0.71123 |
2305.05136 | Shamna Pootheri | Shamna Pootheri and Govindan V K | Localisation of Mammographic masses by Greedy Backtracking of
Activations in the Stacked Auto-Encoders | null | null | null | null | cs.CV cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Mammographic image analysis requires accurate localisation of salient
mammographic masses. In mammographic computer-aided diagnosis, mass or Region
of Interest (ROI) is often marked by physicians and features are extracted from
the marked ROI. In this paper, we present a novel mammographic mass
localisation framework, based on the maximal class activations of the stacked
auto-encoders. We hypothesize that the image regions activating abnormal
classes in mammographic images will be the breast masses which causes the
anomaly. The experiment is conducted using randomly selected 200 mammographic
images (100 normal and 100 abnormal) from IRMA mammographic dataset. Abnormal
mass regions marked by an expert radiologist are used as the ground truth. The
proposed method outperforms existing Deep Convolutional Neural Network (DCNN)
based techniques in terms of salient region detection accuracy. The proposed
greedy backtracking method is more efficient and does not require a vast number
of labelled training images as in DCNN based method. Such automatic
localisation method will assist physicians to make accurate decisions on biopsy
recommendations and treatment evaluations.
| [
{
"version": "v1",
"created": "Tue, 9 May 2023 02:46:13 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Pootheri",
"Shamna",
""
],
[
"K",
"Govindan V",
""
]
] | TITLE: Localisation of Mammographic masses by Greedy Backtracking of
Activations in the Stacked Auto-Encoders
ABSTRACT: Mammographic image analysis requires accurate localisation of salient
mammographic masses. In mammographic computer-aided diagnosis, mass or Region
of Interest (ROI) is often marked by physicians and features are extracted from
the marked ROI. In this paper, we present a novel mammographic mass
localisation framework, based on the maximal class activations of the stacked
auto-encoders. We hypothesize that the image regions activating abnormal
classes in mammographic images will be the breast masses which causes the
anomaly. The experiment is conducted using randomly selected 200 mammographic
images (100 normal and 100 abnormal) from IRMA mammographic dataset. Abnormal
mass regions marked by an expert radiologist are used as the ground truth. The
proposed method outperforms existing Deep Convolutional Neural Network (DCNN)
based techniques in terms of salient region detection accuracy. The proposed
greedy backtracking method is more efficient and does not require a vast number
of labelled training images as in DCNN based method. Such automatic
localisation method will assist physicians to make accurate decisions on biopsy
recommendations and treatment evaluations.
| no_new_dataset | 0.71227 |
2305.05162 | Guodong Liu | Guodong Liu | Effective Medical Code Prediction via Label Internal Alignment | null | null | null | null | cs.LG cs.CL | http://creativecommons.org/licenses/by/4.0/ | The clinical notes are usually typed into the system by physicians. They are
typically required to be marked by standard medical codes, and each code
represents a diagnosis or medical treatment procedure. Annotating these notes
is time consuming and prone to error. In this paper, we proposed a multi-view
attention based Neural network to predict medical codes from clinical texts.
Our method incorporates three aspects of information, the semantic context of
the clinical text, the relationship among the label (medical codes) space, and
the alignment between each pair of a clinical text and medical code. Our method
is verified to be effective on the open source dataset. The experimental result
shows that our method achieves better performance against the prior
state-of-art on multiple metrics.
| [
{
"version": "v1",
"created": "Tue, 9 May 2023 04:14:20 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Liu",
"Guodong",
""
]
] | TITLE: Effective Medical Code Prediction via Label Internal Alignment
ABSTRACT: The clinical notes are usually typed into the system by physicians. They are
typically required to be marked by standard medical codes, and each code
represents a diagnosis or medical treatment procedure. Annotating these notes
is time consuming and prone to error. In this paper, we proposed a multi-view
attention based Neural network to predict medical codes from clinical texts.
Our method incorporates three aspects of information, the semantic context of
the clinical text, the relationship among the label (medical codes) space, and
the alignment between each pair of a clinical text and medical code. Our method
is verified to be effective on the open source dataset. The experimental result
shows that our method achieves better performance against the prior
state-of-art on multiple metrics.
| no_new_dataset | 0.711067 |
2305.05171 | Lesly Miculicich | Lesly Miculicich, Yujia Xie, Song Wang, Pengcheng He | Summarization with Precise Length Control | null | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Many applications of text generation such as summarization benefit from
accurately controlling the text length. Existing approaches on
length-controlled summarization either result in degraded performance or can
only control the length approximately. In this work, we present a framework to
generate summaries with precisely the specified number of tokens or sentences,
while maintaining or even improving the text quality. In addition, we jointly
train the models to predict the lengths, so our model can generate summaries
with optimal length. We evaluate the proposed framework on the CNNDM dataset
and show improved performance compared to existing methods.
| [
{
"version": "v1",
"created": "Tue, 9 May 2023 04:45:24 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Miculicich",
"Lesly",
""
],
[
"Xie",
"Yujia",
""
],
[
"Wang",
"Song",
""
],
[
"He",
"Pengcheng",
""
]
] | TITLE: Summarization with Precise Length Control
ABSTRACT: Many applications of text generation such as summarization benefit from
accurately controlling the text length. Existing approaches on
length-controlled summarization either result in degraded performance or can
only control the length approximately. In this work, we present a framework to
generate summaries with precisely the specified number of tokens or sentences,
while maintaining or even improving the text quality. In addition, we jointly
train the models to predict the lengths, so our model can generate summaries
with optimal length. We evaluate the proposed framework on the CNNDM dataset
and show improved performance compared to existing methods.
| no_new_dataset | 0.710396 |
2305.05183 | Bo Sun | Bo Sun, Baoxin Wang, Yixuan Wang, Wanxiang Che, Dayong Wu, Shijin Wang
and Ting Liu | CSED: A Chinese Semantic Error Diagnosis Corpus | 12 pages. arXiv admin note: text overlap with arXiv:2204.07464 | null | null | null | cs.CL cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recently, much Chinese text error correction work has focused on Chinese
Spelling Check (CSC) and Chinese Grammatical Error Diagnosis (CGED). In
contrast, little attention has been paid to the complicated problem of Chinese
Semantic Error Diagnosis (CSED), which lacks relevant datasets. The study of
semantic errors is important because they are very common and may lead to
syntactic irregularities or even problems of comprehension. To investigate
this, we build the CSED corpus, which includes two datasets. The one is for the
CSED-Recognition (CSED-R) task. The other is for the CSED-Correction (CSED-C)
task. Our annotation guarantees high-quality data through quality assurance
mechanisms. Our experiments show that powerful pre-trained models perform
poorly on this corpus. We also find that the CSED task is challenging, as
evidenced by the fact that even humans receive a low score. This paper proposes
syntax-aware models to specifically adapt to the CSED task. The experimental
results show that the introduction of the syntax-aware approach is meaningful.
| [
{
"version": "v1",
"created": "Tue, 9 May 2023 05:33:31 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Sun",
"Bo",
""
],
[
"Wang",
"Baoxin",
""
],
[
"Wang",
"Yixuan",
""
],
[
"Che",
"Wanxiang",
""
],
[
"Wu",
"Dayong",
""
],
[
"Wang",
"Shijin",
""
],
[
"Liu",
"Ting",
""
]
] | TITLE: CSED: A Chinese Semantic Error Diagnosis Corpus
ABSTRACT: Recently, much Chinese text error correction work has focused on Chinese
Spelling Check (CSC) and Chinese Grammatical Error Diagnosis (CGED). In
contrast, little attention has been paid to the complicated problem of Chinese
Semantic Error Diagnosis (CSED), which lacks relevant datasets. The study of
semantic errors is important because they are very common and may lead to
syntactic irregularities or even problems of comprehension. To investigate
this, we build the CSED corpus, which includes two datasets. The one is for the
CSED-Recognition (CSED-R) task. The other is for the CSED-Correction (CSED-C)
task. Our annotation guarantees high-quality data through quality assurance
mechanisms. Our experiments show that powerful pre-trained models perform
poorly on this corpus. We also find that the CSED task is challenging, as
evidenced by the fact that even humans receive a low score. This paper proposes
syntax-aware models to specifically adapt to the CSED task. The experimental
results show that the introduction of the syntax-aware approach is meaningful.
| new_dataset | 0.700114 |
2305.05198 | Han Wang | Minh Duc Vu, Han Wang, Zhuang Li, Gholamreza Haffari, Zhenchang Xing,
Chunyang Chen | Voicify Your UI: Towards Android App Control with Voice Commands | null | Proceedings of the ACM on Interactive, Mobile, Wearable and
Ubiquitous Technologies 7, no. 1 (2023): 1-22 | 10.1145/3581998 | null | cs.HC | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Nowadays, voice assistants help users complete tasks on the smartphone with
voice commands, replacing traditional touchscreen interactions when such
interactions are inhibited. However, the usability of those tools remains
moderate due to the problems in understanding rich language variations in human
commands, along with efficiency and comprehensibility issues. Therefore, we
introduce Voicify, an Android virtual assistant that allows users to interact
with on-screen elements in mobile apps through voice commands. Using a novel
deep learning command parser, Voicify interprets human verbal input and
performs matching with UI elements. In addition, the tool can directly open a
specific feature from installed applications by fetching application code
information to explore the set of in-app components. Our command parser
achieved 90\% accuracy on the human command dataset. Furthermore, the direct
feature invocation module achieves better feature coverage in comparison to
Google Assistant. The user study demonstrates the usefulness of Voicify in
real-world scenarios.
| [
{
"version": "v1",
"created": "Tue, 9 May 2023 06:23:47 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Vu",
"Minh Duc",
""
],
[
"Wang",
"Han",
""
],
[
"Li",
"Zhuang",
""
],
[
"Haffari",
"Gholamreza",
""
],
[
"Xing",
"Zhenchang",
""
],
[
"Chen",
"Chunyang",
""
]
] | TITLE: Voicify Your UI: Towards Android App Control with Voice Commands
ABSTRACT: Nowadays, voice assistants help users complete tasks on the smartphone with
voice commands, replacing traditional touchscreen interactions when such
interactions are inhibited. However, the usability of those tools remains
moderate due to the problems in understanding rich language variations in human
commands, along with efficiency and comprehensibility issues. Therefore, we
introduce Voicify, an Android virtual assistant that allows users to interact
with on-screen elements in mobile apps through voice commands. Using a novel
deep learning command parser, Voicify interprets human verbal input and
performs matching with UI elements. In addition, the tool can directly open a
specific feature from installed applications by fetching application code
information to explore the set of in-app components. Our command parser
achieved 90\% accuracy on the human command dataset. Furthermore, the direct
feature invocation module achieves better feature coverage in comparison to
Google Assistant. The user study demonstrates the usefulness of Voicify in
real-world scenarios.
| no_new_dataset | 0.694523 |
2305.05200 | Shanshan Zhong | Shanshan Zhong, Wushao Wen, Jinghui Qin, Qiangpu Chen, Zhongzhan Huang | LSAS: Lightweight Sub-attention Strategy for Alleviating Attention Bias
Problem | null | null | null | null | cs.CV cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In computer vision, the performance of deep neural networks (DNNs) is highly
related to the feature extraction ability, i.e., the ability to recognize and
focus on key pixel regions in an image. However, in this paper, we
quantitatively and statistically illustrate that DNNs have a serious attention
bias problem on many samples from some popular datasets: (1) Position bias:
DNNs fully focus on label-independent regions; (2) Range bias: The focused
regions from DNN are not completely contained in the ideal region. Moreover, we
find that the existing self-attention modules can alleviate these biases to a
certain extent, but the biases are still non-negligible. To further mitigate
them, we propose a lightweight sub-attention strategy (LSAS), which utilizes
high-order sub-attention modules to improve the original self-attention
modules. The effectiveness of LSAS is demonstrated by extensive experiments on
widely-used benchmark datasets and popular attention networks. We release our
code to help other researchers to reproduce the results of
LSAS~\footnote{https://github.com/Qrange-group/LSAS}.
| [
{
"version": "v1",
"created": "Tue, 9 May 2023 06:25:59 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Zhong",
"Shanshan",
""
],
[
"Wen",
"Wushao",
""
],
[
"Qin",
"Jinghui",
""
],
[
"Chen",
"Qiangpu",
""
],
[
"Huang",
"Zhongzhan",
""
]
] | TITLE: LSAS: Lightweight Sub-attention Strategy for Alleviating Attention Bias
Problem
ABSTRACT: In computer vision, the performance of deep neural networks (DNNs) is highly
related to the feature extraction ability, i.e., the ability to recognize and
focus on key pixel regions in an image. However, in this paper, we
quantitatively and statistically illustrate that DNNs have a serious attention
bias problem on many samples from some popular datasets: (1) Position bias:
DNNs fully focus on label-independent regions; (2) Range bias: The focused
regions from DNN are not completely contained in the ideal region. Moreover, we
find that the existing self-attention modules can alleviate these biases to a
certain extent, but the biases are still non-negligible. To further mitigate
them, we propose a lightweight sub-attention strategy (LSAS), which utilizes
high-order sub-attention modules to improve the original self-attention
modules. The effectiveness of LSAS is demonstrated by extensive experiments on
widely-used benchmark datasets and popular attention networks. We release our
code to help other researchers to reproduce the results of
LSAS~\footnote{https://github.com/Qrange-group/LSAS}.
| no_new_dataset | 0.71081 |
2305.05204 | Yuanhao Liu | Yuanhao Liu, Qi Cao, Huawei Shen, Yunfan Wu, Shuchang Tao and Xueqi
Cheng | Popularity Debiasing from Exposure to Interaction in Collaborative
Filtering | Published as a SIGIR'23 short paper | null | 10.1145/3539618.3591947 | null | cs.IR cs.SI | http://creativecommons.org/licenses/by/4.0/ | Recommender systems often suffer from popularity bias, where popular items
are overly recommended while sacrificing unpopular items. Existing researches
generally focus on ensuring the number of recommendations exposure of each item
is equal or proportional, using inverse propensity weighting, causal
intervention, or adversarial training. However, increasing the exposure of
unpopular items may not bring more clicks or interactions, resulting in skewed
benefits and failing in achieving real reasonable popularity debiasing. In this
paper, we propose a new criterion for popularity debiasing, i.e., in an
unbiased recommender system, both popular and unpopular items should receive
Interactions Proportional to the number of users who Like it, namely IPL
criterion. Under the guidance of the criterion, we then propose a debiasing
framework with IPL regularization term which is theoretically shown to achieve
a win-win situation of both popularity debiasing and recommendation
performance. Experiments conducted on four public datasets demonstrate that
when equipping two representative collaborative filtering models with our
framework, the popularity bias is effectively alleviated while maintaining the
recommendation performance.
| [
{
"version": "v1",
"created": "Tue, 9 May 2023 06:44:56 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Liu",
"Yuanhao",
""
],
[
"Cao",
"Qi",
""
],
[
"Shen",
"Huawei",
""
],
[
"Wu",
"Yunfan",
""
],
[
"Tao",
"Shuchang",
""
],
[
"Cheng",
"Xueqi",
""
]
] | TITLE: Popularity Debiasing from Exposure to Interaction in Collaborative
Filtering
ABSTRACT: Recommender systems often suffer from popularity bias, where popular items
are overly recommended while sacrificing unpopular items. Existing researches
generally focus on ensuring the number of recommendations exposure of each item
is equal or proportional, using inverse propensity weighting, causal
intervention, or adversarial training. However, increasing the exposure of
unpopular items may not bring more clicks or interactions, resulting in skewed
benefits and failing in achieving real reasonable popularity debiasing. In this
paper, we propose a new criterion for popularity debiasing, i.e., in an
unbiased recommender system, both popular and unpopular items should receive
Interactions Proportional to the number of users who Like it, namely IPL
criterion. Under the guidance of the criterion, we then propose a debiasing
framework with IPL regularization term which is theoretically shown to achieve
a win-win situation of both popularity debiasing and recommendation
performance. Experiments conducted on four public datasets demonstrate that
when equipping two representative collaborative filtering models with our
framework, the popularity bias is effectively alleviated while maintaining the
recommendation performance.
| no_new_dataset | 0.709416 |
2305.05260 | Bo Yuan | Bo Yuan, Yao Jiang, Keren Fu, Qijun Zhao | Guided Focal Stack Refinement Network for Light Field Salient Object
Detection | Accepted by ICME 2023 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Light field salient object detection (SOD) is an emerging research direction
attributed to the richness of light field data. However, most existing methods
lack effective handling of focal stacks, therefore making the latter involved
in a lot of interfering information and degrade the performance of SOD. To
address this limitation, we propose to utilize multi-modal features to refine
focal stacks in a guided manner, resulting in a novel guided focal stack
refinement network called GFRNet. To this end, we propose a guided refinement
and fusion module (GRFM) to refine focal stacks and aggregate multi-modal
features. In GRFM, all-in-focus (AiF) and depth modalities are utilized to
refine focal stacks separately, leading to two novel sub-modules for different
modalities, namely AiF-based refinement module (ARM) and depth-based refinement
module (DRM). Such refinement modules enhance structural and positional
information of salient objects in focal stacks, and are able to improve SOD
accuracy. Experimental results on four benchmark datasets demonstrate the
superiority of our GFRNet model against 12 state-of-the-art models.
| [
{
"version": "v1",
"created": "Tue, 9 May 2023 08:32:06 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Yuan",
"Bo",
""
],
[
"Jiang",
"Yao",
""
],
[
"Fu",
"Keren",
""
],
[
"Zhao",
"Qijun",
""
]
] | TITLE: Guided Focal Stack Refinement Network for Light Field Salient Object
Detection
ABSTRACT: Light field salient object detection (SOD) is an emerging research direction
attributed to the richness of light field data. However, most existing methods
lack effective handling of focal stacks, therefore making the latter involved
in a lot of interfering information and degrade the performance of SOD. To
address this limitation, we propose to utilize multi-modal features to refine
focal stacks in a guided manner, resulting in a novel guided focal stack
refinement network called GFRNet. To this end, we propose a guided refinement
and fusion module (GRFM) to refine focal stacks and aggregate multi-modal
features. In GRFM, all-in-focus (AiF) and depth modalities are utilized to
refine focal stacks separately, leading to two novel sub-modules for different
modalities, namely AiF-based refinement module (ARM) and depth-based refinement
module (DRM). Such refinement modules enhance structural and positional
information of salient objects in focal stacks, and are able to improve SOD
accuracy. Experimental results on four benchmark datasets demonstrate the
superiority of our GFRNet model against 12 state-of-the-art models.
| no_new_dataset | 0.709831 |
2305.05271 | Xuandi Fu | Xuandi Fu, Kanthashree Mysore Sathyendra, Ankur Gandhe, Jing Liu,
Grant P. Strimel, Ross McGowan, Athanasios Mouchtaris | Robust Acoustic and Semantic Contextual Biasing in Neural Transducers
for Speech Recognition | Accepted at ICASSP 2023 | null | null | null | cs.CL cs.LG cs.SD eess.AS | http://creativecommons.org/licenses/by/4.0/ | Attention-based contextual biasing approaches have shown significant
improvements in the recognition of generic and/or personal rare-words in
End-to-End Automatic Speech Recognition (E2E ASR) systems like neural
transducers. These approaches employ cross-attention to bias the model towards
specific contextual entities injected as bias-phrases to the model. Prior
approaches typically relied on subword encoders for encoding the bias phrases.
However, subword tokenizations are coarse and fail to capture granular
pronunciation information which is crucial for biasing based on acoustic
similarity. In this work, we propose to use lightweight character
representations to encode fine-grained pronunciation features to improve
contextual biasing guided by acoustic similarity between the audio and the
contextual entities (termed acoustic biasing). We further integrate pretrained
neural language model (NLM) based encoders to encode the utterance's semantic
context along with contextual entities to perform biasing informed by the
utterance's semantic context (termed semantic biasing). Experiments using a
Conformer Transducer model on the Librispeech dataset show a 4.62% - 9.26%
relative WER improvement on different biasing list sizes over the baseline
contextual model when incorporating our proposed acoustic and semantic biasing
approach. On a large-scale in-house dataset, we observe 7.91% relative WER
improvement compared to our baseline model. On tail utterances, the
improvements are even more pronounced with 36.80% and 23.40% relative WER
improvements on Librispeech rare words and an in-house testset respectively.
| [
{
"version": "v1",
"created": "Tue, 9 May 2023 08:51:44 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Fu",
"Xuandi",
""
],
[
"Sathyendra",
"Kanthashree Mysore",
""
],
[
"Gandhe",
"Ankur",
""
],
[
"Liu",
"Jing",
""
],
[
"Strimel",
"Grant P.",
""
],
[
"McGowan",
"Ross",
""
],
[
"Mouchtaris",
"Athanasios",
""
]
] | TITLE: Robust Acoustic and Semantic Contextual Biasing in Neural Transducers
for Speech Recognition
ABSTRACT: Attention-based contextual biasing approaches have shown significant
improvements in the recognition of generic and/or personal rare-words in
End-to-End Automatic Speech Recognition (E2E ASR) systems like neural
transducers. These approaches employ cross-attention to bias the model towards
specific contextual entities injected as bias-phrases to the model. Prior
approaches typically relied on subword encoders for encoding the bias phrases.
However, subword tokenizations are coarse and fail to capture granular
pronunciation information which is crucial for biasing based on acoustic
similarity. In this work, we propose to use lightweight character
representations to encode fine-grained pronunciation features to improve
contextual biasing guided by acoustic similarity between the audio and the
contextual entities (termed acoustic biasing). We further integrate pretrained
neural language model (NLM) based encoders to encode the utterance's semantic
context along with contextual entities to perform biasing informed by the
utterance's semantic context (termed semantic biasing). Experiments using a
Conformer Transducer model on the Librispeech dataset show a 4.62% - 9.26%
relative WER improvement on different biasing list sizes over the baseline
contextual model when incorporating our proposed acoustic and semantic biasing
approach. On a large-scale in-house dataset, we observe 7.91% relative WER
improvement compared to our baseline model. On tail utterances, the
improvements are even more pronounced with 36.80% and 23.40% relative WER
improvements on Librispeech rare words and an in-house testset respectively.
| no_new_dataset | 0.711857 |
2305.05302 | Eliya Habba | Eliya Habba, Renana Keydar, Dan Bareket, Gabriel Stanovsky | The Perfect Victim: Computational Analysis of Judicial Attitudes towards
Victims of Sexual Violence | null | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | We develop computational models to analyze court statements in order to
assess judicial attitudes toward victims of sexual violence in the Israeli
court system. The study examines the resonance of "rape myths" in the criminal
justice system's response to sex crimes, in particular in judicial assessment
of victim's credibility. We begin by formulating an ontology for evaluating
judicial attitudes toward victim's credibility, with eight ordinal labels and
binary categorizations. Second, we curate a manually annotated dataset for
judicial assessments of victim's credibility in the Hebrew language, as well as
a model that can extract credibility labels from court cases. The dataset
consists of 855 verdict decision documents in sexual assault cases from
1990-2021, annotated with the help of legal experts and trained law students.
The model uses a combined approach of syntactic and latent structures to find
sentences that convey the judge's attitude towards the victim and classify them
according to the credibility label set. Our ontology, data, and models will be
made available upon request, in the hope they spur future progress in this
judicial important task.
| [
{
"version": "v1",
"created": "Tue, 9 May 2023 09:45:44 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Habba",
"Eliya",
""
],
[
"Keydar",
"Renana",
""
],
[
"Bareket",
"Dan",
""
],
[
"Stanovsky",
"Gabriel",
""
]
] | TITLE: The Perfect Victim: Computational Analysis of Judicial Attitudes towards
Victims of Sexual Violence
ABSTRACT: We develop computational models to analyze court statements in order to
assess judicial attitudes toward victims of sexual violence in the Israeli
court system. The study examines the resonance of "rape myths" in the criminal
justice system's response to sex crimes, in particular in judicial assessment
of victim's credibility. We begin by formulating an ontology for evaluating
judicial attitudes toward victim's credibility, with eight ordinal labels and
binary categorizations. Second, we curate a manually annotated dataset for
judicial assessments of victim's credibility in the Hebrew language, as well as
a model that can extract credibility labels from court cases. The dataset
consists of 855 verdict decision documents in sexual assault cases from
1990-2021, annotated with the help of legal experts and trained law students.
The model uses a combined approach of syntactic and latent structures to find
sentences that convey the judge's attitude towards the victim and classify them
according to the credibility label set. Our ontology, data, and models will be
made available upon request, in the hope they spur future progress in this
judicial important task.
| new_dataset | 0.716188 |
2305.05311 | Daniel Fern\'andez-Gonz\'alez | Daniel Fern\'andez-Gonz\'alez | Structured Sentiment Analysis as Transition-based Dependency Parsing | null | null | null | null | cs.CL cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Structured sentiment analysis (SSA) aims to automatically extract people's
opinions from a text in natural language and adequately represent that
information in a graph structure. One of the most accurate methods for
performing SSA was recently proposed and consists of approaching it as a
dependency parsing task. Although we can find in the literature how
transition-based algorithms excel in dependency parsing in terms of accuracy
and efficiency, all proposed attempts to tackle SSA following that approach
were based on graph-based models. In this article, we present the first
transition-based method to address SSA as dependency parsing. Specifically, we
design a transition system that processes the input text in a left-to-right
pass, incrementally generating the graph structure containing all identified
opinions. To effectively implement our final transition-based model, we resort
to a Pointer Network architecture as a backbone. From an extensive evaluation,
we demonstrate that our model offers the best performance to date in
practically all cases among prior dependency-based methods, and surpass recent
task-specific techniques on the most challenging datasets. We additionally
include an in-depth analysis and empirically prove that the overall
time-complexity cost of our approach is quadratic in the sentence length, being
more efficient than top-performing graph-based parsers.
| [
{
"version": "v1",
"created": "Tue, 9 May 2023 10:03:34 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Fernández-González",
"Daniel",
""
]
] | TITLE: Structured Sentiment Analysis as Transition-based Dependency Parsing
ABSTRACT: Structured sentiment analysis (SSA) aims to automatically extract people's
opinions from a text in natural language and adequately represent that
information in a graph structure. One of the most accurate methods for
performing SSA was recently proposed and consists of approaching it as a
dependency parsing task. Although we can find in the literature how
transition-based algorithms excel in dependency parsing in terms of accuracy
and efficiency, all proposed attempts to tackle SSA following that approach
were based on graph-based models. In this article, we present the first
transition-based method to address SSA as dependency parsing. Specifically, we
design a transition system that processes the input text in a left-to-right
pass, incrementally generating the graph structure containing all identified
opinions. To effectively implement our final transition-based model, we resort
to a Pointer Network architecture as a backbone. From an extensive evaluation,
we demonstrate that our model offers the best performance to date in
practically all cases among prior dependency-based methods, and surpass recent
task-specific techniques on the most challenging datasets. We additionally
include an in-depth analysis and empirically prove that the overall
time-complexity cost of our approach is quadratic in the sentence length, being
more efficient than top-performing graph-based parsers.
| no_new_dataset | 0.709189 |
2305.05321 | Joy Odimegwu | Marvellous Ajala, Blessing Oko, David Oba-Fidelis, Joycelyn Iyasele,
Joy I. Odimegwu | Application of Artificial Intelligence in the Classification of
Microscopical Starch Images for Drug Formulation | 22 pages, 10 figures | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by-sa/4.0/ | Starches are important energy sources found in plants with many uses in the
pharmaceutical industry such as binders, disintegrants, bulking agents in drugs
and thus require very careful physicochemical analysis for proper
identification and verification which includes microscopy. In this work, we
applied artificial intelligence techniques (using transfer learning and deep
convolution neural network CNNs to microscopical images obtained from 9 starch
samples of different botanical sources. Our approach obtained an accuracy of
61% when the machine learning model was pretrained on microscopic images from
MicroNet dataset. However the accuracy jumped to 81% for model pretrained on
random day to day images obtained from Imagenet dataset. The model pretrained
on the imagenet dataset also showed a better precision, recall and f1 score
than that pretrained on the imagenet dataset.
| [
{
"version": "v1",
"created": "Tue, 9 May 2023 10:16:02 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Ajala",
"Marvellous",
""
],
[
"Oko",
"Blessing",
""
],
[
"Oba-Fidelis",
"David",
""
],
[
"Iyasele",
"Joycelyn",
""
],
[
"Odimegwu",
"Joy I.",
""
]
] | TITLE: Application of Artificial Intelligence in the Classification of
Microscopical Starch Images for Drug Formulation
ABSTRACT: Starches are important energy sources found in plants with many uses in the
pharmaceutical industry such as binders, disintegrants, bulking agents in drugs
and thus require very careful physicochemical analysis for proper
identification and verification which includes microscopy. In this work, we
applied artificial intelligence techniques (using transfer learning and deep
convolution neural network CNNs to microscopical images obtained from 9 starch
samples of different botanical sources. Our approach obtained an accuracy of
61% when the machine learning model was pretrained on microscopic images from
MicroNet dataset. However the accuracy jumped to 81% for model pretrained on
random day to day images obtained from Imagenet dataset. The model pretrained
on the imagenet dataset also showed a better precision, recall and f1 score
than that pretrained on the imagenet dataset.
| no_new_dataset | 0.714068 |
2305.05322 | Tianlun Zheng | Tianlun Zheng, Zhineng Chen, Jinfeng Bai, Hongtao Xie, Yu-Gang Jiang | TPS++: Attention-Enhanced Thin-Plate Spline for Scene Text Recognition | Accepted by IJCAI 2023 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Text irregularities pose significant challenges to scene text recognizers.
Thin-Plate Spline (TPS)-based rectification is widely regarded as an effective
means to deal with them. Currently, the calculation of TPS transformation
parameters purely depends on the quality of regressed text borders. It ignores
the text content and often leads to unsatisfactory rectified results for
severely distorted text. In this work, we introduce TPS++, an
attention-enhanced TPS transformation that incorporates the attention mechanism
to text rectification for the first time. TPS++ formulates the parameter
calculation as a joint process of foreground control point regression and
content-based attention score estimation, which is computed by a dedicated
designed gated-attention block. TPS++ builds a more flexible content-aware
rectifier, generating a natural text correction that is easier to read by the
subsequent recognizer. Moreover, TPS++ shares the feature backbone with the
recognizer in part and implements the rectification at feature-level rather
than image-level, incurring only a small overhead in terms of parameters and
inference time. Experiments on public benchmarks show that TPS++ consistently
improves the recognition and achieves state-of-the-art accuracy. Meanwhile, it
generalizes well on different backbones and recognizers. Code is at
https://github.com/simplify23/TPS_PP.
| [
{
"version": "v1",
"created": "Tue, 9 May 2023 10:16:43 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Zheng",
"Tianlun",
""
],
[
"Chen",
"Zhineng",
""
],
[
"Bai",
"Jinfeng",
""
],
[
"Xie",
"Hongtao",
""
],
[
"Jiang",
"Yu-Gang",
""
]
] | TITLE: TPS++: Attention-Enhanced Thin-Plate Spline for Scene Text Recognition
ABSTRACT: Text irregularities pose significant challenges to scene text recognizers.
Thin-Plate Spline (TPS)-based rectification is widely regarded as an effective
means to deal with them. Currently, the calculation of TPS transformation
parameters purely depends on the quality of regressed text borders. It ignores
the text content and often leads to unsatisfactory rectified results for
severely distorted text. In this work, we introduce TPS++, an
attention-enhanced TPS transformation that incorporates the attention mechanism
to text rectification for the first time. TPS++ formulates the parameter
calculation as a joint process of foreground control point regression and
content-based attention score estimation, which is computed by a dedicated
designed gated-attention block. TPS++ builds a more flexible content-aware
rectifier, generating a natural text correction that is easier to read by the
subsequent recognizer. Moreover, TPS++ shares the feature backbone with the
recognizer in part and implements the rectification at feature-level rather
than image-level, incurring only a small overhead in terms of parameters and
inference time. Experiments on public benchmarks show that TPS++ consistently
improves the recognition and achieves state-of-the-art accuracy. Meanwhile, it
generalizes well on different backbones and recognizers. Code is at
https://github.com/simplify23/TPS_PP.
| no_new_dataset | 0.71202 |
2305.05334 | Sougata Saha | Sougata Saha and Rohini Srihari | ArgU: A Controllable Factual Argument Generator | null | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Effective argumentation is essential towards a purposeful conversation with a
satisfactory outcome. For example, persuading someone to reconsider smoking
might involve empathetic, well founded arguments based on facts and expert
opinions about its ill-effects and the consequences on one's family. However,
the automatic generation of high-quality factual arguments can be challenging.
Addressing existing controllability issues can make the recent advances in
computational models for argument generation a potential solution. In this
paper, we introduce ArgU: a neural argument generator capable of producing
factual arguments from input facts and real-world concepts that can be
explicitly controlled for stance and argument structure using Walton's argument
scheme-based control codes. Unfortunately, computational argument generation is
a relatively new field and lacks datasets conducive to training. Hence, we have
compiled and released an annotated corpora of 69,428 arguments spanning six
topics and six argument schemes, making it the largest publicly available
corpus for identifying argument schemes; the paper details our annotation and
dataset creation framework. We further experiment with an argument generation
strategy that establishes an inference strategy by generating an ``argument
template'' before actual argument generation. Our results demonstrate that it
is possible to automatically generate diverse arguments exhibiting different
inference patterns for the same set of facts by using control codes based on
argument schemes and stance.
| [
{
"version": "v1",
"created": "Tue, 9 May 2023 10:49:45 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Saha",
"Sougata",
""
],
[
"Srihari",
"Rohini",
""
]
] | TITLE: ArgU: A Controllable Factual Argument Generator
ABSTRACT: Effective argumentation is essential towards a purposeful conversation with a
satisfactory outcome. For example, persuading someone to reconsider smoking
might involve empathetic, well founded arguments based on facts and expert
opinions about its ill-effects and the consequences on one's family. However,
the automatic generation of high-quality factual arguments can be challenging.
Addressing existing controllability issues can make the recent advances in
computational models for argument generation a potential solution. In this
paper, we introduce ArgU: a neural argument generator capable of producing
factual arguments from input facts and real-world concepts that can be
explicitly controlled for stance and argument structure using Walton's argument
scheme-based control codes. Unfortunately, computational argument generation is
a relatively new field and lacks datasets conducive to training. Hence, we have
compiled and released an annotated corpora of 69,428 arguments spanning six
topics and six argument schemes, making it the largest publicly available
corpus for identifying argument schemes; the paper details our annotation and
dataset creation framework. We further experiment with an argument generation
strategy that establishes an inference strategy by generating an ``argument
template'' before actual argument generation. Our results demonstrate that it
is possible to automatically generate diverse arguments exhibiting different
inference patterns for the same set of facts by using control codes based on
argument schemes and stance.
| new_dataset | 0.718156 |
2305.05377 | David Noever | David Noever and Matt Ciolino | Professional Certification Benchmark Dataset: The First 500 Jobs For
Large Language Models | null | null | null | null | cs.AI cs.LG | http://creativecommons.org/licenses/by-sa/4.0/ | The research creates a professional certification survey to test large
language models and evaluate their employable skills. It compares the
performance of two AI models, GPT-3 and Turbo-GPT3.5, on a benchmark dataset of
1149 professional certifications, emphasizing vocational readiness rather than
academic performance. GPT-3 achieved a passing score (>70% correct) in 39% of
the professional certifications without fine-tuning or exam preparation. The
models demonstrated qualifications in various computer-related fields, such as
cloud and virtualization, business analytics, cybersecurity, network setup and
repair, and data analytics. Turbo-GPT3.5 scored 100% on the valuable Offensive
Security Certified Professional (OSCP) exam. The models also displayed
competence in other professional domains, including nursing, licensed
counseling, pharmacy, and teaching. Turbo-GPT3.5 passed the Financial Industry
Regulatory Authority (FINRA) Series 6 exam with a 70% grade without
preparation. Interestingly, Turbo-GPT3.5 performed well on customer service
tasks, suggesting potential applications in human augmentation for chatbots in
call centers and routine advice services. The models also score well on sensory
and experience-based tests such as wine sommelier, beer taster, emotional
quotient, and body language reader. The OpenAI model improvement from Babbage
to Turbo resulted in a median 60% better-graded performance in less than a few
years. This progress suggests that focusing on the latest model's shortcomings
could lead to a highly performant AI capable of mastering the most demanding
professional certifications. We open-source the benchmark to expand the range
of testable professional skills as the models improve or gain emergent
capabilities.
| [
{
"version": "v1",
"created": "Sun, 7 May 2023 00:56:58 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Noever",
"David",
""
],
[
"Ciolino",
"Matt",
""
]
] | TITLE: Professional Certification Benchmark Dataset: The First 500 Jobs For
Large Language Models
ABSTRACT: The research creates a professional certification survey to test large
language models and evaluate their employable skills. It compares the
performance of two AI models, GPT-3 and Turbo-GPT3.5, on a benchmark dataset of
1149 professional certifications, emphasizing vocational readiness rather than
academic performance. GPT-3 achieved a passing score (>70% correct) in 39% of
the professional certifications without fine-tuning or exam preparation. The
models demonstrated qualifications in various computer-related fields, such as
cloud and virtualization, business analytics, cybersecurity, network setup and
repair, and data analytics. Turbo-GPT3.5 scored 100% on the valuable Offensive
Security Certified Professional (OSCP) exam. The models also displayed
competence in other professional domains, including nursing, licensed
counseling, pharmacy, and teaching. Turbo-GPT3.5 passed the Financial Industry
Regulatory Authority (FINRA) Series 6 exam with a 70% grade without
preparation. Interestingly, Turbo-GPT3.5 performed well on customer service
tasks, suggesting potential applications in human augmentation for chatbots in
call centers and routine advice services. The models also score well on sensory
and experience-based tests such as wine sommelier, beer taster, emotional
quotient, and body language reader. The OpenAI model improvement from Babbage
to Turbo resulted in a median 60% better-graded performance in less than a few
years. This progress suggests that focusing on the latest model's shortcomings
could lead to a highly performant AI capable of mastering the most demanding
professional certifications. We open-source the benchmark to expand the range
of testable professional skills as the models improve or gain emergent
capabilities.
| no_new_dataset | 0.698856 |
2305.05378 | Jingbo Zhou | Qiwei Lang, Jingbo Zhou, Haoyi Wang, Shiqi Lyu, Rui Zhang | PLM-GNN: A Webpage Classification Method based on Joint Pre-trained
Language Model and Graph Neural Network | null | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The number of web pages is growing at an exponential rate, accumulating
massive amounts of data on the web. It is one of the key processes to classify
webpages in web information mining. Some classical methods are based on
manually building features of web pages and training classifiers based on
machine learning or deep learning. However, building features manually requires
specific domain knowledge and usually takes a long time to validate the
validity of features. Considering webpages generated by the combination of text
and HTML Document Object Model(DOM) trees, we propose a representation and
classification method based on a pre-trained language model and graph neural
network, named PLM-GNN. It is based on the joint encoding of text and HTML DOM
trees in the web pages. It performs well on the KI-04 and SWDE datasets and on
practical dataset AHS for the project of scholar's homepage crawling.
| [
{
"version": "v1",
"created": "Tue, 9 May 2023 12:19:10 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Lang",
"Qiwei",
""
],
[
"Zhou",
"Jingbo",
""
],
[
"Wang",
"Haoyi",
""
],
[
"Lyu",
"Shiqi",
""
],
[
"Zhang",
"Rui",
""
]
] | TITLE: PLM-GNN: A Webpage Classification Method based on Joint Pre-trained
Language Model and Graph Neural Network
ABSTRACT: The number of web pages is growing at an exponential rate, accumulating
massive amounts of data on the web. It is one of the key processes to classify
webpages in web information mining. Some classical methods are based on
manually building features of web pages and training classifiers based on
machine learning or deep learning. However, building features manually requires
specific domain knowledge and usually takes a long time to validate the
validity of features. Considering webpages generated by the combination of text
and HTML Document Object Model(DOM) trees, we propose a representation and
classification method based on a pre-trained language model and graph neural
network, named PLM-GNN. It is based on the joint encoding of text and HTML DOM
trees in the web pages. It performs well on the KI-04 and SWDE datasets and on
practical dataset AHS for the project of scholar's homepage crawling.
| no_new_dataset | 0.710842 |
2305.05383 | Shuai Lu | Chenxiao Liu, Shuai Lu, Weizhu Chen, Daxin Jiang, Alexey Svyatkovskiy,
Shengyu Fu, Neel Sundaresan and Nan Duan | Code Execution with Pre-trained Language Models | Accepted to the Findings of ACL 2023 | null | null | null | cs.PL cs.AI cs.CL cs.SE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Code execution is a fundamental aspect of programming language semantics that
reflects the exact behavior of the code. However, most pre-trained models for
code intelligence ignore the execution trace and only rely on source code and
syntactic structures. In this paper, we investigate how well pre-trained models
can understand and perform code execution. We develop a mutation-based data
augmentation technique to create a large-scale and realistic Python dataset and
task for code execution, which challenges existing models such as Codex. We
then present CodeExecutor, a Transformer model that leverages code execution
pre-training and curriculum learning to enhance its semantic comprehension. We
evaluate CodeExecutor on code execution and show its promising performance and
limitations. We also demonstrate its potential benefits for code intelligence
tasks such as zero-shot code-to-code search and text-to-code generation. Our
analysis provides insights into the learning and generalization abilities of
pre-trained models for code execution.
| [
{
"version": "v1",
"created": "Mon, 8 May 2023 10:00:05 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Liu",
"Chenxiao",
""
],
[
"Lu",
"Shuai",
""
],
[
"Chen",
"Weizhu",
""
],
[
"Jiang",
"Daxin",
""
],
[
"Svyatkovskiy",
"Alexey",
""
],
[
"Fu",
"Shengyu",
""
],
[
"Sundaresan",
"Neel",
""
],
[
"Duan",
"Nan",
""
]
] | TITLE: Code Execution with Pre-trained Language Models
ABSTRACT: Code execution is a fundamental aspect of programming language semantics that
reflects the exact behavior of the code. However, most pre-trained models for
code intelligence ignore the execution trace and only rely on source code and
syntactic structures. In this paper, we investigate how well pre-trained models
can understand and perform code execution. We develop a mutation-based data
augmentation technique to create a large-scale and realistic Python dataset and
task for code execution, which challenges existing models such as Codex. We
then present CodeExecutor, a Transformer model that leverages code execution
pre-training and curriculum learning to enhance its semantic comprehension. We
evaluate CodeExecutor on code execution and show its promising performance and
limitations. We also demonstrate its potential benefits for code intelligence
tasks such as zero-shot code-to-code search and text-to-code generation. Our
analysis provides insights into the learning and generalization abilities of
pre-trained models for code execution.
| no_new_dataset | 0.673433 |
2305.05393 | Yixiao Ma | Yixiao Ma, Yueyue Wu, Weihang Su, Qingyao Ai, Yiqun Liu | CaseEncoder: A Knowledge-enhanced Pre-trained Model for Legal Case
Encoding | 5 pages | null | null | null | cs.IR cs.CL | http://creativecommons.org/licenses/by/4.0/ | Legal case retrieval is a critical process for modern legal information
systems. While recent studies have utilized pre-trained language models (PLMs)
based on the general domain self-supervised pre-training paradigm to build
models for legal case retrieval, there are limitations in using general domain
PLMs as backbones. Specifically, these models may not fully capture the
underlying legal features in legal case documents. To address this issue, we
propose CaseEncoder, a legal document encoder that leverages fine-grained legal
knowledge in both the data sampling and pre-training phases. In the data
sampling phase, we enhance the quality of the training data by utilizing
fine-grained law article information to guide the selection of positive and
negative examples. In the pre-training phase, we design legal-specific
pre-training tasks that align with the judging criteria of relevant legal
cases. Based on these tasks, we introduce an innovative loss function called
Biased Circle Loss to enhance the model's ability to recognize case relevance
in fine grains. Experimental results on multiple benchmarks demonstrate that
CaseEncoder significantly outperforms both existing general pre-training models
and legal-specific pre-training models in zero-shot legal case retrieval.
| [
{
"version": "v1",
"created": "Tue, 9 May 2023 12:40:19 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Ma",
"Yixiao",
""
],
[
"Wu",
"Yueyue",
""
],
[
"Su",
"Weihang",
""
],
[
"Ai",
"Qingyao",
""
],
[
"Liu",
"Yiqun",
""
]
] | TITLE: CaseEncoder: A Knowledge-enhanced Pre-trained Model for Legal Case
Encoding
ABSTRACT: Legal case retrieval is a critical process for modern legal information
systems. While recent studies have utilized pre-trained language models (PLMs)
based on the general domain self-supervised pre-training paradigm to build
models for legal case retrieval, there are limitations in using general domain
PLMs as backbones. Specifically, these models may not fully capture the
underlying legal features in legal case documents. To address this issue, we
propose CaseEncoder, a legal document encoder that leverages fine-grained legal
knowledge in both the data sampling and pre-training phases. In the data
sampling phase, we enhance the quality of the training data by utilizing
fine-grained law article information to guide the selection of positive and
negative examples. In the pre-training phase, we design legal-specific
pre-training tasks that align with the judging criteria of relevant legal
cases. Based on these tasks, we introduce an innovative loss function called
Biased Circle Loss to enhance the model's ability to recognize case relevance
in fine grains. Experimental results on multiple benchmarks demonstrate that
CaseEncoder significantly outperforms both existing general pre-training models
and legal-specific pre-training models in zero-shot legal case retrieval.
| no_new_dataset | 0.709585 |
2305.05423 | Alireza Ebrahimi Dorcheh | Amanda Issac (1), Alireza Ebrahimi (2), Javad Mohammadpour Velni (2),
Glen Rains (3) ((1) School of Electrical and Computer Engineering, University
of Georgia, (2) Department of Mechanical Engineering, Clemson University, (3)
Department of Entomology, University of Georgia) | High-throughput Cotton Phenotyping Big Data Pipeline Lambda Architecture
Computer Vision Deep Neural Networks | null | null | null | null | cs.CV cs.LG | http://creativecommons.org/licenses/by/4.0/ | In this study, we propose a big data pipeline for cotton bloom detection
using a Lambda architecture, which enables real-time and batch processing of
data. Our proposed approach leverages Azure resources such as Data Factory,
Event Grids, Rest APIs, and Databricks. This work is the first to develop and
demonstrate the implementation of such a pipeline for plant phenotyping through
Azure's cloud computing service. The proposed pipeline consists of data
preprocessing, object detection using a YOLOv5 neural network model trained
through Azure AutoML, and visualization of object detection bounding boxes on
output images. The trained model achieves a mean Average Precision (mAP) score
of 0.96, demonstrating its high performance for cotton bloom classification. We
evaluate our Lambda architecture pipeline using 9000 images yielding an
optimized runtime of 34 minutes. The results illustrate the scalability of the
proposed pipeline as a solution for deep learning object detection, with the
potential for further expansion through additional Azure processing cores. This
work advances the scientific research field by providing a new method for
cotton bloom detection on a large dataset and demonstrates the potential of
utilizing cloud computing resources, specifically Azure, for efficient and
accurate big data processing in precision agriculture.
| [
{
"version": "v1",
"created": "Tue, 9 May 2023 13:15:19 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Issac",
"Amanda",
""
],
[
"Ebrahimi",
"Alireza",
""
],
[
"Velni",
"Javad Mohammadpour",
""
],
[
"Rains",
"Glen",
""
]
] | TITLE: High-throughput Cotton Phenotyping Big Data Pipeline Lambda Architecture
Computer Vision Deep Neural Networks
ABSTRACT: In this study, we propose a big data pipeline for cotton bloom detection
using a Lambda architecture, which enables real-time and batch processing of
data. Our proposed approach leverages Azure resources such as Data Factory,
Event Grids, Rest APIs, and Databricks. This work is the first to develop and
demonstrate the implementation of such a pipeline for plant phenotyping through
Azure's cloud computing service. The proposed pipeline consists of data
preprocessing, object detection using a YOLOv5 neural network model trained
through Azure AutoML, and visualization of object detection bounding boxes on
output images. The trained model achieves a mean Average Precision (mAP) score
of 0.96, demonstrating its high performance for cotton bloom classification. We
evaluate our Lambda architecture pipeline using 9000 images yielding an
optimized runtime of 34 minutes. The results illustrate the scalability of the
proposed pipeline as a solution for deep learning object detection, with the
potential for further expansion through additional Azure processing cores. This
work advances the scientific research field by providing a new method for
cotton bloom detection on a large dataset and demonstrates the potential of
utilizing cloud computing resources, specifically Azure, for efficient and
accurate big data processing in precision agriculture.
| no_new_dataset | 0.710829 |
2305.05461 | Kris Cao | Kris Cao | What is the best recipe for character-level encoder-only modelling? | accepted at ACL 2023 | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | This paper aims to benchmark recent progress in language understanding models
that output contextualised representations at the character level. Many such
modelling architectures and methods to train those architectures have been
proposed, but it is currently unclear what the relative contributions of the
architecture vs. the pretraining objective are to final model performance. We
explore the design space of such models, comparing architectural innovations
and a variety of different pretraining objectives on a suite of evaluation
tasks with a fixed training procedure in order to find the currently optimal
way to build and train character-level BERT-like models. We find that our best
performing character-level model exceeds the performance of a token-based model
trained with the same settings on the same data, suggesting that
character-level models are ready for more widespread adoption. Unfortunately,
the best method to train character-level models still relies on a subword-level
tokeniser during pretraining, and final model performance is highly dependent
on tokeniser quality. We believe our results demonstrate the readiness of
character-level models for multilingual language representation, and encourage
NLP practitioners to try them as drop-in replacements for token-based models.
| [
{
"version": "v1",
"created": "Tue, 9 May 2023 14:00:15 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Cao",
"Kris",
""
]
] | TITLE: What is the best recipe for character-level encoder-only modelling?
ABSTRACT: This paper aims to benchmark recent progress in language understanding models
that output contextualised representations at the character level. Many such
modelling architectures and methods to train those architectures have been
proposed, but it is currently unclear what the relative contributions of the
architecture vs. the pretraining objective are to final model performance. We
explore the design space of such models, comparing architectural innovations
and a variety of different pretraining objectives on a suite of evaluation
tasks with a fixed training procedure in order to find the currently optimal
way to build and train character-level BERT-like models. We find that our best
performing character-level model exceeds the performance of a token-based model
trained with the same settings on the same data, suggesting that
character-level models are ready for more widespread adoption. Unfortunately,
the best method to train character-level models still relies on a subword-level
tokeniser during pretraining, and final model performance is highly dependent
on tokeniser quality. We believe our results demonstrate the readiness of
character-level models for multilingual language representation, and encourage
NLP practitioners to try them as drop-in replacements for token-based models.
| no_new_dataset | 0.707998 |
2305.05464 | Nisha Huang | Nisha Huang, Yuxin Zhang, Weiming Dong | Style-A-Video: Agile Diffusion for Arbitrary Text-based Video Style
Transfer | null | null | null | null | cs.CV cs.MM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Large-scale text-to-video diffusion models have demonstrated an exceptional
ability to synthesize diverse videos. However, due to the lack of extensive
text-to-video datasets and the necessary computational resources for training,
directly applying these models for video stylization remains difficult. Also,
given that the noise addition process on the input content is random and
destructive, fulfilling the style transfer task's content preservation criteria
is challenging. This paper proposes a zero-shot video stylization method named
Style-A-Video, which utilizes a generative pre-trained transformer with an
image latent diffusion model to achieve a concise text-controlled video
stylization. We improve the guidance condition in the denoising process,
establishing a balance between artistic expression and structure preservation.
Furthermore, to decrease inter-frame flicker and avoid the formation of
additional artifacts, we employ a sampling optimization and a temporal
consistency module. Extensive experiments show that we can attain superior
content preservation and stylistic performance while incurring less consumption
than previous solutions. Code will be available at
https://github.com/haha-lisa/Style-A-Video.
| [
{
"version": "v1",
"created": "Tue, 9 May 2023 14:03:27 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Huang",
"Nisha",
""
],
[
"Zhang",
"Yuxin",
""
],
[
"Dong",
"Weiming",
""
]
] | TITLE: Style-A-Video: Agile Diffusion for Arbitrary Text-based Video Style
Transfer
ABSTRACT: Large-scale text-to-video diffusion models have demonstrated an exceptional
ability to synthesize diverse videos. However, due to the lack of extensive
text-to-video datasets and the necessary computational resources for training,
directly applying these models for video stylization remains difficult. Also,
given that the noise addition process on the input content is random and
destructive, fulfilling the style transfer task's content preservation criteria
is challenging. This paper proposes a zero-shot video stylization method named
Style-A-Video, which utilizes a generative pre-trained transformer with an
image latent diffusion model to achieve a concise text-controlled video
stylization. We improve the guidance condition in the denoising process,
establishing a balance between artistic expression and structure preservation.
Furthermore, to decrease inter-frame flicker and avoid the formation of
additional artifacts, we employ a sampling optimization and a temporal
consistency module. Extensive experiments show that we can attain superior
content preservation and stylistic performance while incurring less consumption
than previous solutions. Code will be available at
https://github.com/haha-lisa/Style-A-Video.
| no_new_dataset | 0.709831 |
2305.05474 | Piotr Rybak | Aleksandra Chrabrowa, Tsimur Hadeliya, Dariusz Kajtoch, Robert
Mroczkowski, Piotr Rybak | Going beyond research datasets: Novel intent discovery in the industry
setting | Accepted to Findings of EACL 2023 | null | null | null | cs.CL cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Novel intent discovery automates the process of grouping similar messages
(questions) to identify previously unknown intents. However, current research
focuses on publicly available datasets which have only the question field and
significantly differ from real-life datasets. This paper proposes methods to
improve the intent discovery pipeline deployed in a large e-commerce platform.
We show the benefit of pre-training language models on in-domain data: both
self-supervised and with weak supervision. We also devise the best method to
utilize the conversational structure (i.e., question and answer) of real-life
datasets during fine-tuning for clustering tasks, which we call Conv. All our
methods combined to fully utilize real-life datasets give up to 33pp
performance boost over state-of-the-art Constrained Deep Adaptive Clustering
(CDAC) model for question only. By comparison CDAC model for the question data
only gives only up to 13pp performance boost over the naive baseline.
| [
{
"version": "v1",
"created": "Tue, 9 May 2023 14:21:29 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Chrabrowa",
"Aleksandra",
""
],
[
"Hadeliya",
"Tsimur",
""
],
[
"Kajtoch",
"Dariusz",
""
],
[
"Mroczkowski",
"Robert",
""
],
[
"Rybak",
"Piotr",
""
]
] | TITLE: Going beyond research datasets: Novel intent discovery in the industry
setting
ABSTRACT: Novel intent discovery automates the process of grouping similar messages
(questions) to identify previously unknown intents. However, current research
focuses on publicly available datasets which have only the question field and
significantly differ from real-life datasets. This paper proposes methods to
improve the intent discovery pipeline deployed in a large e-commerce platform.
We show the benefit of pre-training language models on in-domain data: both
self-supervised and with weak supervision. We also devise the best method to
utilize the conversational structure (i.e., question and answer) of real-life
datasets during fine-tuning for clustering tasks, which we call Conv. All our
methods combined to fully utilize real-life datasets give up to 33pp
performance boost over state-of-the-art Constrained Deep Adaptive Clustering
(CDAC) model for question only. By comparison CDAC model for the question data
only gives only up to 13pp performance boost over the naive baseline.
| no_new_dataset | 0.712076 |
2305.05486 | Piotr Rybak | Piotr Rybak | MAUPQA: Massive Automatically-created Polish Question Answering Dataset | null | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recently, open-domain question answering systems have begun to rely heavily
on annotated datasets to train neural passage retrievers. However, manually
annotating such datasets is both difficult and time-consuming, which limits
their availability for less popular languages. In this work, we experiment with
several methods for automatically collecting weakly labeled datasets and show
how they affect the performance of the neural passage retrieval models. As a
result of our work, we publish the MAUPQA dataset, consisting of nearly 400,000
question-passage pairs for Polish, as well as the HerBERT-QA neural retriever.
| [
{
"version": "v1",
"created": "Tue, 9 May 2023 14:36:04 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Rybak",
"Piotr",
""
]
] | TITLE: MAUPQA: Massive Automatically-created Polish Question Answering Dataset
ABSTRACT: Recently, open-domain question answering systems have begun to rely heavily
on annotated datasets to train neural passage retrievers. However, manually
annotating such datasets is both difficult and time-consuming, which limits
their availability for less popular languages. In this work, we experiment with
several methods for automatically collecting weakly labeled datasets and show
how they affect the performance of the neural passage retrieval models. As a
result of our work, we publish the MAUPQA dataset, consisting of nearly 400,000
question-passage pairs for Polish, as well as the HerBERT-QA neural retriever.
| new_dataset | 0.714392 |
2305.05490 | Katia Jodogne-Del Litto | Katia Jodogne-Del Litto, Guillaume-Alexandre Bilodeau | Real-time instance segmentation with polygons using an
Intersection-over-Union loss | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Predicting a binary mask for an object is more accurate but also more
computationally expensive than a bounding box. Polygonal masks as developed in
CenterPoly can be a good compromise. In this paper, we improve over CenterPoly
by enhancing the classical regression L1 loss with a novel region-based loss
and a novel order loss, as well as with a new training process for the vertices
prediction head. Moreover, the previous methods that predict polygonal masks
use different coordinate systems, but it is not clear if one is better than
another, if we abstract the architecture requirement. We therefore investigate
their impact on the prediction. We also use a new evaluation protocol with
oracle predictions for the detection head, to further isolate the segmentation
process and better compare the polygonal masks with binary masks. Our instance
segmentation method is trained and tested with challenging datasets containing
urban scenes, with a high density of road users. Experiments show, in
particular, that using a combination of a regression loss and a region-based
loss allows significant improvements on the Cityscapes and IDD test set
compared to CenterPoly. Moreover the inference stage remains fast enough to
reach real-time performance with an average of 0.045 s per frame for
2048$\times$1024 images on a single RTX 2070 GPU. The code is available
$\href{https://github.com/KatiaJDL/CenterPoly-v2}{\text{here}}$.
| [
{
"version": "v1",
"created": "Tue, 9 May 2023 14:43:38 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Litto",
"Katia Jodogne-Del",
""
],
[
"Bilodeau",
"Guillaume-Alexandre",
""
]
] | TITLE: Real-time instance segmentation with polygons using an
Intersection-over-Union loss
ABSTRACT: Predicting a binary mask for an object is more accurate but also more
computationally expensive than a bounding box. Polygonal masks as developed in
CenterPoly can be a good compromise. In this paper, we improve over CenterPoly
by enhancing the classical regression L1 loss with a novel region-based loss
and a novel order loss, as well as with a new training process for the vertices
prediction head. Moreover, the previous methods that predict polygonal masks
use different coordinate systems, but it is not clear if one is better than
another, if we abstract the architecture requirement. We therefore investigate
their impact on the prediction. We also use a new evaluation protocol with
oracle predictions for the detection head, to further isolate the segmentation
process and better compare the polygonal masks with binary masks. Our instance
segmentation method is trained and tested with challenging datasets containing
urban scenes, with a high density of road users. Experiments show, in
particular, that using a combination of a regression loss and a region-based
loss allows significant improvements on the Cityscapes and IDD test set
compared to CenterPoly. Moreover the inference stage remains fast enough to
reach real-time performance with an average of 0.045 s per frame for
2048$\times$1024 images on a single RTX 2070 GPU. The code is available
$\href{https://github.com/KatiaJDL/CenterPoly-v2}{\text{here}}$.
| no_new_dataset | 0.711281 |
2305.05495 | Boje Deforce | Boje Deforce, Bart Baesens, Jan Diels, Estefan\'ia Serral Asensio | Self-Supervised Anomaly Detection of Rogue Soil Moisture Sensors | null | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | IoT data is a central element in the successful digital transformation of
agriculture. However, IoT data comes with its own set of challenges. E.g., the
risk of data contamination due to rogue sensors. A sensor is considered rogue
when it provides incorrect measurements over time. To ensure correct analytical
results, an essential preprocessing step when working with IoT data is the
detection of such rogue sensors. Existing methods assume that well-behaving
sensors are known or that a large majority of the sensors is well-behaving.
However, real-world data is often completely unlabeled and voluminous, calling
for self-supervised methods that can detect rogue sensors without prior
information. We present a self-supervised anomalous sensor detector based on a
neural network with a contrastive loss, followed by DBSCAN. A core contribution
of our paper is the use of Dynamic Time Warping in the negative sampling for
the triplet loss. This novelty makes the use of triplet networks feasible for
anomalous sensor detection. Our method shows promising results on a challenging
dataset of soil moisture sensors deployed in multiple pear orchards.
| [
{
"version": "v1",
"created": "Tue, 9 May 2023 14:47:16 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Deforce",
"Boje",
""
],
[
"Baesens",
"Bart",
""
],
[
"Diels",
"Jan",
""
],
[
"Asensio",
"Estefanía Serral",
""
]
] | TITLE: Self-Supervised Anomaly Detection of Rogue Soil Moisture Sensors
ABSTRACT: IoT data is a central element in the successful digital transformation of
agriculture. However, IoT data comes with its own set of challenges. E.g., the
risk of data contamination due to rogue sensors. A sensor is considered rogue
when it provides incorrect measurements over time. To ensure correct analytical
results, an essential preprocessing step when working with IoT data is the
detection of such rogue sensors. Existing methods assume that well-behaving
sensors are known or that a large majority of the sensors is well-behaving.
However, real-world data is often completely unlabeled and voluminous, calling
for self-supervised methods that can detect rogue sensors without prior
information. We present a self-supervised anomalous sensor detector based on a
neural network with a contrastive loss, followed by DBSCAN. A core contribution
of our paper is the use of Dynamic Time Warping in the negative sampling for
the triplet loss. This novelty makes the use of triplet networks feasible for
anomalous sensor detection. Our method shows promising results on a challenging
dataset of soil moisture sensors deployed in multiple pear orchards.
| no_new_dataset | 0.708767 |
2305.05503 | Shiyi Qi | Shiyi Qi and Yuanhang Yang and Shuzhzeng Gao and Cuiyun Gao and
Zenglin Xu | BadCS: A Backdoor Attack Framework for Code search | null | null | null | null | cs.SE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | With the development of deep learning (DL), DL-based code search models have
achieved state-of-the-art performance and have been widely used by developers
during software development. However, the security issue, e.g., recommending
vulnerable code, has not received sufficient attention, which will bring
potential harm to software development. Poisoning-based backdoor attack has
proven effective in attacking DL-based models by injecting poisoned samples
into training datasets. However, previous work shows that the attack technique
does not perform successfully on all DL-based code search models and tends to
fail for Transformer-based models, especially pretrained models. Besides, the
infected models generally perform worse than benign models, which makes the
attack not stealthy enough and thereby hinders the adoption by developers. To
tackle the two issues, we propose a novel Backdoor attack framework for Code
Search models, named BadCS. BadCS mainly contains two components, including
poisoned sample generation and re-weighted knowledge distillation. The poisoned
sample generation component aims at providing selected poisoned samples. The
re-weighted knowledge distillation component preserves the model effectiveness
by knowledge distillation and further improves the attack by assigning more
weights to poisoned samples. Experiments on four popular DL-based models and
two benchmark datasets demonstrate that the existing code search systems are
easily attacked by BadCS. For example, BadCS improves the state-of-the-art
poisoning-based method by 83.03%-99.98% and 75.98%-99.90% on Python and Java
datasets, respectively. Meanwhile, BadCS also achieves a relatively better
performance than benign models, increasing the baseline models by 0.49% and
0.46% on average, respectively.
| [
{
"version": "v1",
"created": "Tue, 9 May 2023 14:52:38 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Qi",
"Shiyi",
""
],
[
"Yang",
"Yuanhang",
""
],
[
"Gao",
"Shuzhzeng",
""
],
[
"Gao",
"Cuiyun",
""
],
[
"Xu",
"Zenglin",
""
]
] | TITLE: BadCS: A Backdoor Attack Framework for Code search
ABSTRACT: With the development of deep learning (DL), DL-based code search models have
achieved state-of-the-art performance and have been widely used by developers
during software development. However, the security issue, e.g., recommending
vulnerable code, has not received sufficient attention, which will bring
potential harm to software development. Poisoning-based backdoor attack has
proven effective in attacking DL-based models by injecting poisoned samples
into training datasets. However, previous work shows that the attack technique
does not perform successfully on all DL-based code search models and tends to
fail for Transformer-based models, especially pretrained models. Besides, the
infected models generally perform worse than benign models, which makes the
attack not stealthy enough and thereby hinders the adoption by developers. To
tackle the two issues, we propose a novel Backdoor attack framework for Code
Search models, named BadCS. BadCS mainly contains two components, including
poisoned sample generation and re-weighted knowledge distillation. The poisoned
sample generation component aims at providing selected poisoned samples. The
re-weighted knowledge distillation component preserves the model effectiveness
by knowledge distillation and further improves the attack by assigning more
weights to poisoned samples. Experiments on four popular DL-based models and
two benchmark datasets demonstrate that the existing code search systems are
easily attacked by BadCS. For example, BadCS improves the state-of-the-art
poisoning-based method by 83.03%-99.98% and 75.98%-99.90% on Python and Java
datasets, respectively. Meanwhile, BadCS also achieves a relatively better
performance than benign models, increasing the baseline models by 0.49% and
0.46% on average, respectively.
| no_new_dataset | 0.71086 |
2305.05505 | Rawwad Alhejaili | Rawwad Alhejaili (1 and 2 and 3), Motaz Alfarraj (1 and 2 and 3),
Hamzah Luqman (1 and 4), Ali Al-Shaikhi (1 and 2) ((1) King Fahd University
of Petroleum and Minerals, (2) Electrical Engineering Department, (3)
SDAIA-KFUPM Joint Research Center for Artificial Intelligence, (4)
Information and Computer Science Department) | Recursions Are All You Need: Towards Efficient Deep Unfolding Networks | Accepted to ECV 2023 CVPR workshop | null | null | null | cs.CV cs.LG eess.IV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The use of deep unfolding networks in compressive sensing (CS) has seen wide
success as they provide both simplicity and interpretability. However, since
most deep unfolding networks are iterative, this incurs significant
redundancies in the network. In this work, we propose a novel recursion-based
framework to enhance the efficiency of deep unfolding models. First, recursions
are used to effectively eliminate the redundancies in deep unfolding networks.
Secondly, we randomize the number of recursions during training to decrease the
overall training time. Finally, to effectively utilize the power of recursions,
we introduce a learnable unit to modulate the features of the model based on
both the total number of iterations and the current iteration index. To
evaluate the proposed framework, we apply it to both ISTA-Net+ and COAST.
Extensive testing shows that our proposed framework allows the network to cut
down as much as 75% of its learnable parameters while mostly maintaining its
performance, and at the same time, it cuts around 21% and 42% from the training
time for ISTA-Net+ and COAST respectively. Moreover, when presented with a
limited training dataset, the recursive models match or even outperform their
respective non-recursive baseline. Codes and pretrained models are available at
https://github.com/Rawwad-Alhejaili/Recursions-Are-All-You-Need .
| [
{
"version": "v1",
"created": "Tue, 9 May 2023 14:54:41 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Alhejaili",
"Rawwad",
"",
"1 and 2 and 3"
],
[
"Alfarraj",
"Motaz",
"",
"1 and 2 and 3"
],
[
"Luqman",
"Hamzah",
"",
"1 and 4"
],
[
"Al-Shaikhi",
"Ali",
"",
"1 and 2"
]
] | TITLE: Recursions Are All You Need: Towards Efficient Deep Unfolding Networks
ABSTRACT: The use of deep unfolding networks in compressive sensing (CS) has seen wide
success as they provide both simplicity and interpretability. However, since
most deep unfolding networks are iterative, this incurs significant
redundancies in the network. In this work, we propose a novel recursion-based
framework to enhance the efficiency of deep unfolding models. First, recursions
are used to effectively eliminate the redundancies in deep unfolding networks.
Secondly, we randomize the number of recursions during training to decrease the
overall training time. Finally, to effectively utilize the power of recursions,
we introduce a learnable unit to modulate the features of the model based on
both the total number of iterations and the current iteration index. To
evaluate the proposed framework, we apply it to both ISTA-Net+ and COAST.
Extensive testing shows that our proposed framework allows the network to cut
down as much as 75% of its learnable parameters while mostly maintaining its
performance, and at the same time, it cuts around 21% and 42% from the training
time for ISTA-Net+ and COAST respectively. Moreover, when presented with a
limited training dataset, the recursive models match or even outperform their
respective non-recursive baseline. Codes and pretrained models are available at
https://github.com/Rawwad-Alhejaili/Recursions-Are-All-You-Need .
| no_new_dataset | 0.711186 |
2305.05523 | Yini Fang | Yini Fang, Didan Deng, Liang Wu, Frederic Jumelle, Bertram Shi | RMES: Real-Time Micro-Expression Spotting Using Phase From Riesz Pyramid | This paper will be published in ICME 2023 | null | null | null | cs.CV | http://creativecommons.org/licenses/by-sa/4.0/ | Micro-expressions (MEs) are involuntary and subtle facial expressions that
are thought to reveal feelings people are trying to hide. ME spotting detects
the temporal intervals containing MEs in videos. Detecting such quick and
subtle motions from long videos is difficult. Recent works leverage detailed
facial motion representations, such as the optical flow, and deep learning
models, leading to high computational complexity. To reduce computational
complexity and achieve real-time operation, we propose RMES, a real-time ME
spotting framework. We represent motion using phase computed by Riesz Pyramid,
and feed this motion representation into a three-stream shallow CNN, which
predicts the likelihood of each frame belonging to an ME. In comparison to
optical flow, phase provides more localized motion estimates, which are
essential for ME spotting, resulting in higher performance. Using phase also
reduces the required computation of the ME spotting pipeline by 77.8%. Despite
its relative simplicity and low computational complexity, our framework
achieves state-of-the-art performance on two public datasets: CAS(ME)2 and SAMM
Long Videos.
| [
{
"version": "v1",
"created": "Tue, 9 May 2023 15:22:18 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Fang",
"Yini",
""
],
[
"Deng",
"Didan",
""
],
[
"Wu",
"Liang",
""
],
[
"Jumelle",
"Frederic",
""
],
[
"Shi",
"Bertram",
""
]
] | TITLE: RMES: Real-Time Micro-Expression Spotting Using Phase From Riesz Pyramid
ABSTRACT: Micro-expressions (MEs) are involuntary and subtle facial expressions that
are thought to reveal feelings people are trying to hide. ME spotting detects
the temporal intervals containing MEs in videos. Detecting such quick and
subtle motions from long videos is difficult. Recent works leverage detailed
facial motion representations, such as the optical flow, and deep learning
models, leading to high computational complexity. To reduce computational
complexity and achieve real-time operation, we propose RMES, a real-time ME
spotting framework. We represent motion using phase computed by Riesz Pyramid,
and feed this motion representation into a three-stream shallow CNN, which
predicts the likelihood of each frame belonging to an ME. In comparison to
optical flow, phase provides more localized motion estimates, which are
essential for ME spotting, resulting in higher performance. Using phase also
reduces the required computation of the ME spotting pipeline by 77.8%. Despite
its relative simplicity and low computational complexity, our framework
achieves state-of-the-art performance on two public datasets: CAS(ME)2 and SAMM
Long Videos.
| no_new_dataset | 0.713856 |
2305.05525 | Min Hun Lee | Min Hun Lee, Yi Jing Choy | Exploring a Gradient-based Explainable AI Technique for Time-Series
Data: A Case Study of Assessing Stroke Rehabilitation Exercises | ICLR 2023 Workshop on Time Series Representation Learning for Health | null | null | null | cs.LG cs.AI cs.HC | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Explainable artificial intelligence (AI) techniques are increasingly being
explored to provide insights into why AI and machine learning (ML) models
provide a certain outcome in various applications. However, there has been
limited exploration of explainable AI techniques on time-series data,
especially in the healthcare context. In this paper, we describe a
threshold-based method that utilizes a weakly supervised model and a
gradient-based explainable AI technique (i.e. saliency map) and explore its
feasibility to identify salient frames of time-series data. Using the dataset
from 15 post-stroke survivors performing three upper-limb exercises and labels
on whether a compensatory motion is observed or not, we implemented a
feed-forward neural network model and utilized gradients of each input on model
outcomes to identify salient frames that involve compensatory motions.
According to the evaluation using frame-level annotations, our approach
achieved a recall of 0.96 and an F2-score of 0.91. Our results demonstrated the
potential of a gradient-based explainable AI technique (e.g. saliency map) for
time-series data, such as highlighting the frames of a video that therapists
should focus on reviewing and reducing the efforts on frame-level labeling for
model training.
| [
{
"version": "v1",
"created": "Mon, 8 May 2023 08:30:05 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Lee",
"Min Hun",
""
],
[
"Choy",
"Yi Jing",
""
]
] | TITLE: Exploring a Gradient-based Explainable AI Technique for Time-Series
Data: A Case Study of Assessing Stroke Rehabilitation Exercises
ABSTRACT: Explainable artificial intelligence (AI) techniques are increasingly being
explored to provide insights into why AI and machine learning (ML) models
provide a certain outcome in various applications. However, there has been
limited exploration of explainable AI techniques on time-series data,
especially in the healthcare context. In this paper, we describe a
threshold-based method that utilizes a weakly supervised model and a
gradient-based explainable AI technique (i.e. saliency map) and explore its
feasibility to identify salient frames of time-series data. Using the dataset
from 15 post-stroke survivors performing three upper-limb exercises and labels
on whether a compensatory motion is observed or not, we implemented a
feed-forward neural network model and utilized gradients of each input on model
outcomes to identify salient frames that involve compensatory motions.
According to the evaluation using frame-level annotations, our approach
achieved a recall of 0.96 and an F2-score of 0.91. Our results demonstrated the
potential of a gradient-based explainable AI technique (e.g. saliency map) for
time-series data, such as highlighting the frames of a video that therapists
should focus on reviewing and reducing the efforts on frame-level labeling for
model training.
| no_new_dataset | 0.710057 |
2305.05526 | Haldun Balim | Haldun Balim, Seonwook Park, Xi Wang, Xucong Zhang, Otmar Hilliges | EFE: End-to-end Frame-to-Gaze Estimation | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Despite the recent development of learning-based gaze estimation methods,
most methods require one or more eye or face region crops as inputs and produce
a gaze direction vector as output. Cropping results in a higher resolution in
the eye regions and having fewer confounding factors (such as clothing and
hair) is believed to benefit the final model performance. However, this
eye/face patch cropping process is expensive, erroneous, and
implementation-specific for different methods. In this paper, we propose a
frame-to-gaze network that directly predicts both 3D gaze origin and 3D gaze
direction from the raw frame out of the camera without any face or eye
cropping. Our method demonstrates that direct gaze regression from the raw
downscaled frame, from FHD/HD to VGA/HVGA resolution, is possible despite the
challenges of having very few pixels in the eye region. The proposed method
achieves comparable results to state-of-the-art methods in Point-of-Gaze (PoG)
estimation on three public gaze datasets: GazeCapture, MPIIFaceGaze, and EVE,
and generalizes well to extreme camera view changes.
| [
{
"version": "v1",
"created": "Tue, 9 May 2023 15:25:45 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Balim",
"Haldun",
""
],
[
"Park",
"Seonwook",
""
],
[
"Wang",
"Xi",
""
],
[
"Zhang",
"Xucong",
""
],
[
"Hilliges",
"Otmar",
""
]
] | TITLE: EFE: End-to-end Frame-to-Gaze Estimation
ABSTRACT: Despite the recent development of learning-based gaze estimation methods,
most methods require one or more eye or face region crops as inputs and produce
a gaze direction vector as output. Cropping results in a higher resolution in
the eye regions and having fewer confounding factors (such as clothing and
hair) is believed to benefit the final model performance. However, this
eye/face patch cropping process is expensive, erroneous, and
implementation-specific for different methods. In this paper, we propose a
frame-to-gaze network that directly predicts both 3D gaze origin and 3D gaze
direction from the raw frame out of the camera without any face or eye
cropping. Our method demonstrates that direct gaze regression from the raw
downscaled frame, from FHD/HD to VGA/HVGA resolution, is possible despite the
challenges of having very few pixels in the eye region. The proposed method
achieves comparable results to state-of-the-art methods in Point-of-Gaze (PoG)
estimation on three public gaze datasets: GazeCapture, MPIIFaceGaze, and EVE,
and generalizes well to extreme camera view changes.
| no_new_dataset | 0.712389 |
2305.05531 | Teodor Rares Begu | Teodor Rares Begu | Modelling Concurrency Bugs Using Machine Learning | 52 pages, 41 figures | null | null | null | cs.SE cs.LG cs.PL | http://creativecommons.org/publicdomain/zero/1.0/ | Artificial Intelligence has gained a lot of traction in the recent years,
with machine learning notably starting to see more applications across a varied
range of fields. One specific machine learning application that is of interest
to us is that of software safety and security, especially in the context of
parallel programs. The issue of being able to detect concurrency bugs
automatically has intrigued programmers for a long time, as the added layer of
complexity makes concurrent programs more prone to failure. The development of
such automatic detection tools provides considerable benefits to programmers in
terms of saving time while debugging, as well as reducing the number of
unexpected bugs. We believe machine learning may help achieve this goal by
providing additional advantages over current approaches, in terms of both
overall tool accuracy as well as programming language flexibility. However, due
to the presence of numerous challenges specific to the machine learning
approach (correctly labelling a sufficiently large dataset, finding the best
model types/architectures and so forth), we have to approach each issue of
developing such a tool separately. Therefore, the focus of this project is on
comparing both common and recent machine learning approaches. We abstract away
the complexity of procuring a labelled dataset of concurrent programs under the
form of a synthetic dataset that we define and generate with the scope of
simulating real-life (concurrent) programs. We formulate hypotheses about
fundamental limits of various machine learning model types which we then
validate by running extensive tests on our synthetic dataset. We hope that our
findings provide more insight in the advantages and disadvantages of various
model types when modelling programs using machine learning, as well as any
other related field (e.g. NLP).
| [
{
"version": "v1",
"created": "Mon, 8 May 2023 17:30:24 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Begu",
"Teodor Rares",
""
]
] | TITLE: Modelling Concurrency Bugs Using Machine Learning
ABSTRACT: Artificial Intelligence has gained a lot of traction in the recent years,
with machine learning notably starting to see more applications across a varied
range of fields. One specific machine learning application that is of interest
to us is that of software safety and security, especially in the context of
parallel programs. The issue of being able to detect concurrency bugs
automatically has intrigued programmers for a long time, as the added layer of
complexity makes concurrent programs more prone to failure. The development of
such automatic detection tools provides considerable benefits to programmers in
terms of saving time while debugging, as well as reducing the number of
unexpected bugs. We believe machine learning may help achieve this goal by
providing additional advantages over current approaches, in terms of both
overall tool accuracy as well as programming language flexibility. However, due
to the presence of numerous challenges specific to the machine learning
approach (correctly labelling a sufficiently large dataset, finding the best
model types/architectures and so forth), we have to approach each issue of
developing such a tool separately. Therefore, the focus of this project is on
comparing both common and recent machine learning approaches. We abstract away
the complexity of procuring a labelled dataset of concurrent programs under the
form of a synthetic dataset that we define and generate with the scope of
simulating real-life (concurrent) programs. We formulate hypotheses about
fundamental limits of various machine learning model types which we then
validate by running extensive tests on our synthetic dataset. We hope that our
findings provide more insight in the advantages and disadvantages of various
model types when modelling programs using machine learning, as well as any
other related field (e.g. NLP).
| new_dataset | 0.718162 |
2305.05534 | Yini Fang | Yini Fang, Liang Wu, Frederic Jumelle, Bertram Shi | Integrating Holistic and Local Information to Estimate Emotional
Reaction Intensity | This paper will be published in CVPRW 2023 | null | null | null | cs.CV cs.LG | http://creativecommons.org/licenses/by-sa/4.0/ | Video-based Emotional Reaction Intensity (ERI) estimation measures the
intensity of subjects' reactions to stimuli along several emotional dimensions
from videos of the subject as they view the stimuli. We propose a multi-modal
architecture for video-based ERI combining video and audio information. Video
input is encoded spatially first, frame-by-frame, combining features encoding
holistic aspects of the subjects' facial expressions and features encoding
spatially localized aspects of their expressions. Input is then combined across
time: from frame-to-frame using gated recurrent units (GRUs), then globally by
a transformer. We handle variable video length with a regression token that
accumulates information from all frames into a fixed-dimensional vector
independent of video length. Audio information is handled similarly: spectral
information extracted within each frame is integrated across time by a cascade
of GRUs and a transformer with regression token. The video and audio regression
tokens' outputs are merged by concatenation, then input to a final fully
connected layer producing intensity estimates. Our architecture achieved
excellent performance on the Hume-Reaction dataset in the ERI Esimation
Challenge of the Fifth Competition on Affective Behavior Analysis in-the-Wild
(ABAW5). The Pearson Correlation Coefficients between estimated and subject
self-reported scores, averaged across all emotions, were 0.455 on the
validation dataset and 0.4547 on the test dataset, well above the baselines.
The transformer's self-attention mechanism enables our architecture to focus on
the most critical video frames regardless of length. Ablation experiments
establish the advantages of combining holistic/local features and of
multi-modal integration. Code available at https://github.com/HKUST-NISL/ABAW5.
| [
{
"version": "v1",
"created": "Tue, 9 May 2023 15:28:24 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Fang",
"Yini",
""
],
[
"Wu",
"Liang",
""
],
[
"Jumelle",
"Frederic",
""
],
[
"Shi",
"Bertram",
""
]
] | TITLE: Integrating Holistic and Local Information to Estimate Emotional
Reaction Intensity
ABSTRACT: Video-based Emotional Reaction Intensity (ERI) estimation measures the
intensity of subjects' reactions to stimuli along several emotional dimensions
from videos of the subject as they view the stimuli. We propose a multi-modal
architecture for video-based ERI combining video and audio information. Video
input is encoded spatially first, frame-by-frame, combining features encoding
holistic aspects of the subjects' facial expressions and features encoding
spatially localized aspects of their expressions. Input is then combined across
time: from frame-to-frame using gated recurrent units (GRUs), then globally by
a transformer. We handle variable video length with a regression token that
accumulates information from all frames into a fixed-dimensional vector
independent of video length. Audio information is handled similarly: spectral
information extracted within each frame is integrated across time by a cascade
of GRUs and a transformer with regression token. The video and audio regression
tokens' outputs are merged by concatenation, then input to a final fully
connected layer producing intensity estimates. Our architecture achieved
excellent performance on the Hume-Reaction dataset in the ERI Esimation
Challenge of the Fifth Competition on Affective Behavior Analysis in-the-Wild
(ABAW5). The Pearson Correlation Coefficients between estimated and subject
self-reported scores, averaged across all emotions, were 0.455 on the
validation dataset and 0.4547 on the test dataset, well above the baselines.
The transformer's self-attention mechanism enables our architecture to focus on
the most critical video frames regardless of length. Ablation experiments
establish the advantages of combining holistic/local features and of
multi-modal integration. Code available at https://github.com/HKUST-NISL/ABAW5.
| no_new_dataset | 0.713226 |
2305.05538 | Len Feremans | Len Feremans, Boris Cule, Bart Goethals | Efficient pattern-based anomaly detection in a network of multivariate
devices | null | null | null | null | cs.SI cs.AI cs.LG cs.NI | http://creativecommons.org/licenses/by/4.0/ | Many organisations manage service quality and monitor a large set devices and
servers where each entity is associated with telemetry or physical sensor data
series. Recently, various methods have been proposed to detect behavioural
anomalies, however existing approaches focus on multivariate time series and
ignore communication between entities. Moreover, we aim to support end-users in
not only in locating entities and sensors causing an anomaly at a certain
period, but also explain this decision. We propose a scalable approach to
detect anomalies using a two-step approach. First, we recover relations between
entities in the network, since relations are often dynamic in nature and caused
by an unknown underlying process. Next, we report anomalies based on an
embedding of sequential patterns. Pattern mining is efficient and supports
interpretation, i.e. patterns represent frequent occurring behaviour in time
series. We extend pattern mining to filter sequential patterns based on
frequency, temporal constraints and minimum description length. We collect and
release two public datasets for international broadcasting and X from an
Internet company. \textit{BAD} achieves an overall F1-Score of 0.78 on 9
benchmark datasets, significantly outperforming the best baseline by 3\%.
Additionally, \textit{BAD} is also an order-of-magnitude faster than
state-of-the-art anomaly detection methods.
| [
{
"version": "v1",
"created": "Sun, 7 May 2023 16:05:30 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Feremans",
"Len",
""
],
[
"Cule",
"Boris",
""
],
[
"Goethals",
"Bart",
""
]
] | TITLE: Efficient pattern-based anomaly detection in a network of multivariate
devices
ABSTRACT: Many organisations manage service quality and monitor a large set devices and
servers where each entity is associated with telemetry or physical sensor data
series. Recently, various methods have been proposed to detect behavioural
anomalies, however existing approaches focus on multivariate time series and
ignore communication between entities. Moreover, we aim to support end-users in
not only in locating entities and sensors causing an anomaly at a certain
period, but also explain this decision. We propose a scalable approach to
detect anomalies using a two-step approach. First, we recover relations between
entities in the network, since relations are often dynamic in nature and caused
by an unknown underlying process. Next, we report anomalies based on an
embedding of sequential patterns. Pattern mining is efficient and supports
interpretation, i.e. patterns represent frequent occurring behaviour in time
series. We extend pattern mining to filter sequential patterns based on
frequency, temporal constraints and minimum description length. We collect and
release two public datasets for international broadcasting and X from an
Internet company. \textit{BAD} achieves an overall F1-Score of 0.78 on 9
benchmark datasets, significantly outperforming the best baseline by 3\%.
Additionally, \textit{BAD} is also an order-of-magnitude faster than
state-of-the-art anomaly detection methods.
| no_new_dataset | 0.709007 |
2305.05548 | Wei Lu | Wei Lu, Hua Ma, and Tien-Ping Tan | CIT-EmotionNet: CNN Interactive Transformer Network for EEG Emotion
Recognition | 10 pages,3 tables | null | null | null | eess.SP cs.LG | http://creativecommons.org/licenses/by/4.0/ | Emotion recognition using Electroencephalogram (EEG) signals has emerged as a
significant research challenge in affective computing and intelligent
interaction. However, effectively combining global and local features of EEG
signals to improve performance in emotion recognition is still a difficult
task. In this study, we propose a novel CNN Interactive Transformer Network for
EEG Emotion Recognition, known as CIT-EmotionNet, which efficiently integrates
global and local features of EEG signals. Initially, we convert raw EEG signals
into spatial-frequency representations, which serve as inputs. Then, we
integrate Convolutional Neural Network (CNN) and Transformer within a single
framework in a parallel manner. Finally, we design a CNN interactive
Transformer module, which facilitates the interaction and fusion of local and
global features, thereby enhancing the model's ability to extract both types of
features from EEG spatial-frequency representations. The proposed
CIT-EmotionNet outperforms state-of-the-art methods, achieving an average
recognition accuracy of 98.57\% and 92.09\% on two publicly available datasets,
SEED and SEED-IV, respectively.
| [
{
"version": "v1",
"created": "Sun, 7 May 2023 16:27:09 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Lu",
"Wei",
""
],
[
"Ma",
"Hua",
""
],
[
"Tan",
"Tien-Ping",
""
]
] | TITLE: CIT-EmotionNet: CNN Interactive Transformer Network for EEG Emotion
Recognition
ABSTRACT: Emotion recognition using Electroencephalogram (EEG) signals has emerged as a
significant research challenge in affective computing and intelligent
interaction. However, effectively combining global and local features of EEG
signals to improve performance in emotion recognition is still a difficult
task. In this study, we propose a novel CNN Interactive Transformer Network for
EEG Emotion Recognition, known as CIT-EmotionNet, which efficiently integrates
global and local features of EEG signals. Initially, we convert raw EEG signals
into spatial-frequency representations, which serve as inputs. Then, we
integrate Convolutional Neural Network (CNN) and Transformer within a single
framework in a parallel manner. Finally, we design a CNN interactive
Transformer module, which facilitates the interaction and fusion of local and
global features, thereby enhancing the model's ability to extract both types of
features from EEG spatial-frequency representations. The proposed
CIT-EmotionNet outperforms state-of-the-art methods, achieving an average
recognition accuracy of 98.57\% and 92.09\% on two publicly available datasets,
SEED and SEED-IV, respectively.
| no_new_dataset | 0.711387 |
2305.05566 | Adam Michalski | Adam Michalski, Filippos Christianos, Stefano V. Albrecht | SMAClite: A Lightweight Environment for Multi-Agent Reinforcement
Learning | null | null | null | null | cs.LG cs.AI cs.MA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | There is a lack of standard benchmarks for Multi-Agent Reinforcement Learning
(MARL) algorithms. The Starcraft Multi-Agent Challenge (SMAC) has been widely
used in MARL research, but is built on top of a heavy, closed-source computer
game, StarCraft II. Thus, SMAC is computationally expensive and requires
knowledge and the use of proprietary tools specific to the game for any
meaningful alteration or contribution to the environment. We introduce SMAClite
-- a challenge based on SMAC that is both decoupled from Starcraft II and
open-source, along with a framework which makes it possible to create new
content for SMAClite without any special knowledge. We conduct experiments to
show that SMAClite is equivalent to SMAC, by training MARL algorithms on
SMAClite and reproducing SMAC results. We then show that SMAClite outperforms
SMAC in both runtime speed and memory.
| [
{
"version": "v1",
"created": "Tue, 9 May 2023 15:55:19 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Michalski",
"Adam",
""
],
[
"Christianos",
"Filippos",
""
],
[
"Albrecht",
"Stefano V.",
""
]
] | TITLE: SMAClite: A Lightweight Environment for Multi-Agent Reinforcement
Learning
ABSTRACT: There is a lack of standard benchmarks for Multi-Agent Reinforcement Learning
(MARL) algorithms. The Starcraft Multi-Agent Challenge (SMAC) has been widely
used in MARL research, but is built on top of a heavy, closed-source computer
game, StarCraft II. Thus, SMAC is computationally expensive and requires
knowledge and the use of proprietary tools specific to the game for any
meaningful alteration or contribution to the environment. We introduce SMAClite
-- a challenge based on SMAC that is both decoupled from Starcraft II and
open-source, along with a framework which makes it possible to create new
content for SMAClite without any special knowledge. We conduct experiments to
show that SMAClite is equivalent to SMAC, by training MARL algorithms on
SMAClite and reproducing SMAC results. We then show that SMAClite outperforms
SMAC in both runtime speed and memory.
| no_new_dataset | 0.705392 |
2305.05577 | Alexandre Duval | Alexandre Duval, Victor Schmidt, Alex Hernandez Garcia, Santiago
Miret, Fragkiskos D. Malliaros, Yoshua Bengio, David Rolnick | FAENet: Frame Averaging Equivariant GNN for Materials Modeling | Accepted at ICML 2023 | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | Applications of machine learning techniques for materials modeling typically
involve functions known to be equivariant or invariant to specific symmetries.
While graph neural networks (GNNs) have proven successful in such tasks, they
enforce symmetries via the model architecture, which often reduces their
expressivity, scalability and comprehensibility. In this paper, we introduce
(1) a flexible framework relying on stochastic frame-averaging (SFA) to make
any model E(3)-equivariant or invariant through data transformations. (2)
FAENet: a simple, fast and expressive GNN, optimized for SFA, that processes
geometric information without any symmetrypreserving design constraints. We
prove the validity of our method theoretically and empirically demonstrate its
superior accuracy and computational scalability in materials modeling on the
OC20 dataset (S2EF, IS2RE) as well as common molecular modeling tasks (QM9,
QM7-X). A package implementation is available at https://faenet.readthedocs.io.
| [
{
"version": "v1",
"created": "Fri, 28 Apr 2023 21:48:31 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Duval",
"Alexandre",
""
],
[
"Schmidt",
"Victor",
""
],
[
"Garcia",
"Alex Hernandez",
""
],
[
"Miret",
"Santiago",
""
],
[
"Malliaros",
"Fragkiskos D.",
""
],
[
"Bengio",
"Yoshua",
""
],
[
"Rolnick",
"David",
""
]
] | TITLE: FAENet: Frame Averaging Equivariant GNN for Materials Modeling
ABSTRACT: Applications of machine learning techniques for materials modeling typically
involve functions known to be equivariant or invariant to specific symmetries.
While graph neural networks (GNNs) have proven successful in such tasks, they
enforce symmetries via the model architecture, which often reduces their
expressivity, scalability and comprehensibility. In this paper, we introduce
(1) a flexible framework relying on stochastic frame-averaging (SFA) to make
any model E(3)-equivariant or invariant through data transformations. (2)
FAENet: a simple, fast and expressive GNN, optimized for SFA, that processes
geometric information without any symmetrypreserving design constraints. We
prove the validity of our method theoretically and empirically demonstrate its
superior accuracy and computational scalability in materials modeling on the
OC20 dataset (S2EF, IS2RE) as well as common molecular modeling tasks (QM9,
QM7-X). A package implementation is available at https://faenet.readthedocs.io.
| no_new_dataset | 0.70939 |
2305.05580 | Enric Moreu | Enric Moreu, Alex Martinelli, Martina Naughton, Philip Kelly, Noel E.
O'Connor | Fashion CUT: Unsupervised domain adaptation for visual pattern
classification in clothes using synthetic data and pseudo-labels | null | null | 10.1007/978-3-031-31435-3_21 | null | cs.CV cs.LG | http://creativecommons.org/licenses/by/4.0/ | Accurate product information is critical for e-commerce stores to allow
customers to browse, filter, and search for products. Product data quality is
affected by missing or incorrect information resulting in poor customer
experience. While machine learning can be used to correct inaccurate or missing
information, achieving high performance on fashion image classification tasks
requires large amounts of annotated data, but it is expensive to generate due
to labeling costs. One solution can be to generate synthetic data which
requires no manual labeling. However, training a model with a dataset of solely
synthetic images can lead to poor generalization when performing inference on
real-world data because of the domain shift. We introduce a new unsupervised
domain adaptation technique that converts images from the synthetic domain into
the real-world domain. Our approach combines a generative neural network and a
classifier that are jointly trained to produce realistic images while
preserving the synthetic label information. We found that using real-world
pseudo-labels during training helps the classifier to generalize in the
real-world domain, reducing the synthetic bias. We successfully train a visual
pattern classification model in the fashion domain without real-world
annotations. Experiments show that our method outperforms other unsupervised
domain adaptation algorithms.
| [
{
"version": "v1",
"created": "Tue, 9 May 2023 16:14:57 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Moreu",
"Enric",
""
],
[
"Martinelli",
"Alex",
""
],
[
"Naughton",
"Martina",
""
],
[
"Kelly",
"Philip",
""
],
[
"O'Connor",
"Noel E.",
""
]
] | TITLE: Fashion CUT: Unsupervised domain adaptation for visual pattern
classification in clothes using synthetic data and pseudo-labels
ABSTRACT: Accurate product information is critical for e-commerce stores to allow
customers to browse, filter, and search for products. Product data quality is
affected by missing or incorrect information resulting in poor customer
experience. While machine learning can be used to correct inaccurate or missing
information, achieving high performance on fashion image classification tasks
requires large amounts of annotated data, but it is expensive to generate due
to labeling costs. One solution can be to generate synthetic data which
requires no manual labeling. However, training a model with a dataset of solely
synthetic images can lead to poor generalization when performing inference on
real-world data because of the domain shift. We introduce a new unsupervised
domain adaptation technique that converts images from the synthetic domain into
the real-world domain. Our approach combines a generative neural network and a
classifier that are jointly trained to produce realistic images while
preserving the synthetic label information. We found that using real-world
pseudo-labels during training helps the classifier to generalize in the
real-world domain, reducing the synthetic bias. We successfully train a visual
pattern classification model in the fashion domain without real-world
annotations. Experiments show that our method outperforms other unsupervised
domain adaptation algorithms.
| no_new_dataset | 0.708799 |
2305.05583 | Youliang Zhang | Youliang Zhang, Zhuo Zhou, Wenxuan Liu, Danni Xu, Zheng Wang | Group Activity Recognition via Dynamic Composition and Interaction | 13 pages, 6 figures | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Previous group activity recognition approaches were limited to reasoning
using human relations or finding important subgroups and tended to ignore
indispensable group composition and human-object interactions. This absence
makes a partial interpretation of the scene and increases the interference of
irrelevant actions on the results. Therefore, we propose our DynamicFormer with
Dynamic composition Module (DcM) and Dynamic interaction Module (DiM) to model
relations and locations of persons and discriminate the contribution of
participants, respectively. Our findings on group composition and human-object
interaction inspire our core idea. Group composition tells us the location of
people and their relations inside the group, while interaction reflects the
relation between humans and objects outside the group. We utilize spatial and
temporal encoders in DcM to model our dynamic composition and build DiM to
explore interaction with a novel GCN, which has a transformer inside to
consider the temporal neighbors of human/object. Also, a Multi-level Dynamic
Integration is employed to integrate features from different levels. We conduct
extensive experiments on two public datasets and show that our method achieves
state-of-the-art.
| [
{
"version": "v1",
"created": "Tue, 9 May 2023 16:18:18 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Zhang",
"Youliang",
""
],
[
"Zhou",
"Zhuo",
""
],
[
"Liu",
"Wenxuan",
""
],
[
"Xu",
"Danni",
""
],
[
"Wang",
"Zheng",
""
]
] | TITLE: Group Activity Recognition via Dynamic Composition and Interaction
ABSTRACT: Previous group activity recognition approaches were limited to reasoning
using human relations or finding important subgroups and tended to ignore
indispensable group composition and human-object interactions. This absence
makes a partial interpretation of the scene and increases the interference of
irrelevant actions on the results. Therefore, we propose our DynamicFormer with
Dynamic composition Module (DcM) and Dynamic interaction Module (DiM) to model
relations and locations of persons and discriminate the contribution of
participants, respectively. Our findings on group composition and human-object
interaction inspire our core idea. Group composition tells us the location of
people and their relations inside the group, while interaction reflects the
relation between humans and objects outside the group. We utilize spatial and
temporal encoders in DcM to model our dynamic composition and build DiM to
explore interaction with a novel GCN, which has a transformer inside to
consider the temporal neighbors of human/object. Also, a Multi-level Dynamic
Integration is employed to integrate features from different levels. We conduct
extensive experiments on two public datasets and show that our method achieves
state-of-the-art.
| no_new_dataset | 0.708799 |
2305.05585 | Xiangyuan Liu | Xin Xin, Xiangyuan Liu, Hanbing Wang, Pengjie Ren, Zhumin Chen,
Jiahuan Lei, Xinlei Shi, Hengliang Luo, Joemon Jose, Maarten de Rijke and
Zhaochun Ren | Improving Implicit Feedback-Based Recommendation through Multi-Behavior
Alignment | null | null | 10.1145/3539618.3591697 | null | cs.IR | http://creativecommons.org/licenses/by/4.0/ | Recommender systems that learn from implicit feedback often use large volumes
of a single type of implicit user feedback, such as clicks, to enhance the
prediction of sparse target behavior such as purchases. Using multiple types of
implicit user feedback for such target behavior prediction purposes is still an
open question. Existing studies that attempted to learn from multiple types of
user behavior often fail to: (i) learn universal and accurate user preferences
from different behavioral data distributions, and (ii) overcome the noise and
bias in observed implicit user feedback. To address the above problems, we
propose multi-behavior alignment (MBA), a novel recommendation framework that
learns from implicit feedback by using multiple types of behavioral data. We
conjecture that multiple types of behavior from the same user (e.g., clicks and
purchases) should reflect similar preferences of that user. To this end, we
regard the underlying universal user preferences as a latent variable. The
variable is inferred by maximizing the likelihood of multiple observed
behavioral data distributions and, at the same time, minimizing the
Kullback-Leibler divergence (KL-divergence) between user models learned from
auxiliary behavior (such as clicks or views) and the target behavior
separately. MBA infers universal user preferences from multi-behavior data and
performs data denoising to enable effective knowledge transfer. We conduct
experiments on three datasets, including a dataset collected from an
operational e-commerce platform. Empirical results demonstrate the
effectiveness of our proposed method in utilizing multiple types of behavioral
data to enhance the prediction of the target behavior.
| [
{
"version": "v1",
"created": "Tue, 9 May 2023 16:19:07 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Xin",
"Xin",
""
],
[
"Liu",
"Xiangyuan",
""
],
[
"Wang",
"Hanbing",
""
],
[
"Ren",
"Pengjie",
""
],
[
"Chen",
"Zhumin",
""
],
[
"Lei",
"Jiahuan",
""
],
[
"Shi",
"Xinlei",
""
],
[
"Luo",
"Hengliang",
""
],
[
"Jose",
"Joemon",
""
],
[
"de Rijke",
"Maarten",
""
],
[
"Ren",
"Zhaochun",
""
]
] | TITLE: Improving Implicit Feedback-Based Recommendation through Multi-Behavior
Alignment
ABSTRACT: Recommender systems that learn from implicit feedback often use large volumes
of a single type of implicit user feedback, such as clicks, to enhance the
prediction of sparse target behavior such as purchases. Using multiple types of
implicit user feedback for such target behavior prediction purposes is still an
open question. Existing studies that attempted to learn from multiple types of
user behavior often fail to: (i) learn universal and accurate user preferences
from different behavioral data distributions, and (ii) overcome the noise and
bias in observed implicit user feedback. To address the above problems, we
propose multi-behavior alignment (MBA), a novel recommendation framework that
learns from implicit feedback by using multiple types of behavioral data. We
conjecture that multiple types of behavior from the same user (e.g., clicks and
purchases) should reflect similar preferences of that user. To this end, we
regard the underlying universal user preferences as a latent variable. The
variable is inferred by maximizing the likelihood of multiple observed
behavioral data distributions and, at the same time, minimizing the
Kullback-Leibler divergence (KL-divergence) between user models learned from
auxiliary behavior (such as clicks or views) and the target behavior
separately. MBA infers universal user preferences from multi-behavior data and
performs data denoising to enable effective knowledge transfer. We conduct
experiments on three datasets, including a dataset collected from an
operational e-commerce platform. Empirical results demonstrate the
effectiveness of our proposed method in utilizing multiple types of behavioral
data to enhance the prediction of the target behavior.
| no_new_dataset | 0.712001 |
2305.05589 | Anant Khandelwal | Anant Khandelwal | DomainInv: Domain Invariant Fine Tuning and Adversarial Label Correction
For QA Domain Adaptation | 13 pages, 2 Tables, 2 Figures | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Existing Question Answering (QA) systems limited by the capability of
answering questions from unseen domain or any out-of-domain distributions
making them less reliable for deployment to real scenarios. Most importantly
all the existing QA domain adaptation methods are either based on generating
synthetic data or pseudo labeling the target domain data. The domain adaptation
methods based on synthetic data and pseudo labeling suffers either from the
requirement of computational resources or an extra overhead of carefully
selecting the confidence threshold to separate the noisy examples from being in
the training dataset. In this paper, we propose the unsupervised domain
adaptation for unlabeled target domain by transferring the target
representation near to source domain while still using the supervision from
source domain. Towards that we proposed the idea of domain invariant fine
tuning along with adversarial label correction to identify the target instances
which lie far apart from the source domain, so that the feature encoder can be
learnt to minimize the distance between such target instances and source
instances class wisely, removing the possibility of learning the features of
target domain which are still near to source support but are ambiguous.
Evaluation of our QA domain adaptation method namely, DomainInv on multiple
target QA dataset reveal the performance improvement over the strongest
baseline.
| [
{
"version": "v1",
"created": "Thu, 4 May 2023 18:13:17 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Khandelwal",
"Anant",
""
]
] | TITLE: DomainInv: Domain Invariant Fine Tuning and Adversarial Label Correction
For QA Domain Adaptation
ABSTRACT: Existing Question Answering (QA) systems limited by the capability of
answering questions from unseen domain or any out-of-domain distributions
making them less reliable for deployment to real scenarios. Most importantly
all the existing QA domain adaptation methods are either based on generating
synthetic data or pseudo labeling the target domain data. The domain adaptation
methods based on synthetic data and pseudo labeling suffers either from the
requirement of computational resources or an extra overhead of carefully
selecting the confidence threshold to separate the noisy examples from being in
the training dataset. In this paper, we propose the unsupervised domain
adaptation for unlabeled target domain by transferring the target
representation near to source domain while still using the supervision from
source domain. Towards that we proposed the idea of domain invariant fine
tuning along with adversarial label correction to identify the target instances
which lie far apart from the source domain, so that the feature encoder can be
learnt to minimize the distance between such target instances and source
instances class wisely, removing the possibility of learning the features of
target domain which are still near to source support but are ambiguous.
Evaluation of our QA domain adaptation method namely, DomainInv on multiple
target QA dataset reveal the performance improvement over the strongest
baseline.
| no_new_dataset | 0.709982 |
2305.05594 | Yiqun Wang | Yiqun Wang, Ivan Skorokhodov, Peter Wonka | PET-NeuS: Positional Encoding Tri-Planes for Neural Surfaces | CVPR 2023; 20 Pages; Project page:
\url{https://github.com/yiqun-wang/PET-NeuS} | null | null | null | cs.CV cs.AI cs.GR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A signed distance function (SDF) parametrized by an MLP is a common
ingredient of neural surface reconstruction. We build on the successful recent
method NeuS to extend it by three new components. The first component is to
borrow the tri-plane representation from EG3D and represent signed distance
fields as a mixture of tri-planes and MLPs instead of representing it with MLPs
only. Using tri-planes leads to a more expressive data structure but will also
introduce noise in the reconstructed surface. The second component is to use a
new type of positional encoding with learnable weights to combat noise in the
reconstruction process. We divide the features in the tri-plane into multiple
frequency scales and modulate them with sin and cos functions of different
frequencies. The third component is to use learnable convolution operations on
the tri-plane features using self-attention convolution to produce features
with different frequency bands. The experiments show that PET-NeuS achieves
high-fidelity surface reconstruction on standard datasets. Following previous
work and using the Chamfer metric as the most important way to measure surface
reconstruction quality, we are able to improve upon the NeuS baseline by 57% on
Nerf-synthetic (0.84 compared to 1.97) and by 15.5% on DTU (0.71 compared to
0.84). The qualitative evaluation reveals how our method can better control the
interference of high-frequency noise. Code available at
\url{https://github.com/yiqun-wang/PET-NeuS}.
| [
{
"version": "v1",
"created": "Tue, 9 May 2023 16:35:39 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Wang",
"Yiqun",
""
],
[
"Skorokhodov",
"Ivan",
""
],
[
"Wonka",
"Peter",
""
]
] | TITLE: PET-NeuS: Positional Encoding Tri-Planes for Neural Surfaces
ABSTRACT: A signed distance function (SDF) parametrized by an MLP is a common
ingredient of neural surface reconstruction. We build on the successful recent
method NeuS to extend it by three new components. The first component is to
borrow the tri-plane representation from EG3D and represent signed distance
fields as a mixture of tri-planes and MLPs instead of representing it with MLPs
only. Using tri-planes leads to a more expressive data structure but will also
introduce noise in the reconstructed surface. The second component is to use a
new type of positional encoding with learnable weights to combat noise in the
reconstruction process. We divide the features in the tri-plane into multiple
frequency scales and modulate them with sin and cos functions of different
frequencies. The third component is to use learnable convolution operations on
the tri-plane features using self-attention convolution to produce features
with different frequency bands. The experiments show that PET-NeuS achieves
high-fidelity surface reconstruction on standard datasets. Following previous
work and using the Chamfer metric as the most important way to measure surface
reconstruction quality, we are able to improve upon the NeuS baseline by 57% on
Nerf-synthetic (0.84 compared to 1.97) and by 15.5% on DTU (0.71 compared to
0.84). The qualitative evaluation reveals how our method can better control the
interference of high-frequency noise. Code available at
\url{https://github.com/yiqun-wang/PET-NeuS}.
| no_new_dataset | 0.711017 |
2305.05599 | Jun Chen | Jun Chen, Wei Rao, Zilin Wang, Jiuxin Lin, Zhiyong Wu, Yannan Wang,
Shidong Shang, Helen Meng | Inter-SubNet: Speech Enhancement with Subband Interaction | Accepted by ICASSP 2023 | null | null | null | cs.SD cs.HC eess.AS | http://creativecommons.org/licenses/by-sa/4.0/ | Subband-based approaches process subbands in parallel through the model with
shared parameters to learn the commonality of local spectrums for noise
reduction. In this way, they have achieved remarkable results with fewer
parameters. However, in some complex environments, the lack of global spectral
information has a negative impact on the performance of these subband-based
approaches. To this end, this paper introduces the subband interaction as a new
way to complement the subband model with the global spectral information such
as cross-band dependencies and global spectral patterns, and proposes a new
lightweight single-channel speech enhancement framework called Interactive
Subband Network (Inter-SubNet). Experimental results on DNS Challenge -
Interspeech 2021 dataset show that the proposed Inter-SubNet yields a
significant improvement over the subband model and outperforms other
state-of-the-art speech enhancement approaches, which demonstrate the
effectiveness of subband interaction.
| [
{
"version": "v1",
"created": "Tue, 9 May 2023 16:47:20 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Chen",
"Jun",
""
],
[
"Rao",
"Wei",
""
],
[
"Wang",
"Zilin",
""
],
[
"Lin",
"Jiuxin",
""
],
[
"Wu",
"Zhiyong",
""
],
[
"Wang",
"Yannan",
""
],
[
"Shang",
"Shidong",
""
],
[
"Meng",
"Helen",
""
]
] | TITLE: Inter-SubNet: Speech Enhancement with Subband Interaction
ABSTRACT: Subband-based approaches process subbands in parallel through the model with
shared parameters to learn the commonality of local spectrums for noise
reduction. In this way, they have achieved remarkable results with fewer
parameters. However, in some complex environments, the lack of global spectral
information has a negative impact on the performance of these subband-based
approaches. To this end, this paper introduces the subband interaction as a new
way to complement the subband model with the global spectral information such
as cross-band dependencies and global spectral patterns, and proposes a new
lightweight single-channel speech enhancement framework called Interactive
Subband Network (Inter-SubNet). Experimental results on DNS Challenge -
Interspeech 2021 dataset show that the proposed Inter-SubNet yields a
significant improvement over the subband model and outperforms other
state-of-the-art speech enhancement approaches, which demonstrate the
effectiveness of subband interaction.
| no_new_dataset | 0.71086 |
2305.05609 | Timothy Poterucha | Timothy Poterucha, Pierre Elias, Christopher M. Haggerty | The Case Records of ChatGPT: Language Models and Complex Clinical
Questions | 9 pages, 2 figures | null | null | null | cs.CL stat.AP | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Background: Artificial intelligence language models have shown promise in
various applications, including assisting with clinical decision-making as
demonstrated by strong performance of large language models on medical
licensure exams. However, their ability to solve complex, open-ended cases,
which may be representative of clinical practice, remains unexplored. Methods:
In this study, the accuracy of large language AI models GPT4 and GPT3.5 in
diagnosing complex clinical cases was investigated using published Case Records
of the Massachusetts General Hospital. A total of 50 cases requiring a
diagnosis and diagnostic test published from January 1, 2022 to April 16, 2022
were identified. For each case, models were given a prompt requesting the top
three specific diagnoses and associated diagnostic tests, followed by case
text, labs, and figure legends. Model outputs were assessed in comparison to
the final clinical diagnosis and whether the model-predicted test would result
in a correct diagnosis. Results: GPT4 and GPT3.5 accurately provided the
correct diagnosis in 26% and 22% of cases in one attempt, and 46% and 42%
within three attempts, respectively. GPT4 and GPT3.5 provided a correct
essential diagnostic test in 28% and 24% of cases in one attempt, and 44% and
50% within three attempts, respectively. No significant differences were found
between the two models, and multiple trials with identical prompts using the
GPT3.5 model provided similar results. Conclusions: In summary, these models
demonstrate potential usefulness in generating differential diagnoses but
remain limited in their ability to provide a single unifying diagnosis in
complex, open-ended cases. Future research should focus on evaluating model
performance in larger datasets of open-ended clinical challenges and exploring
potential human-AI collaboration strategies to enhance clinical
decision-making.
| [
{
"version": "v1",
"created": "Tue, 9 May 2023 16:58:32 GMT"
}
] | 2023-05-10T00:00:00 | [
[
"Poterucha",
"Timothy",
""
],
[
"Elias",
"Pierre",
""
],
[
"Haggerty",
"Christopher M.",
""
]
] | TITLE: The Case Records of ChatGPT: Language Models and Complex Clinical
Questions
ABSTRACT: Background: Artificial intelligence language models have shown promise in
various applications, including assisting with clinical decision-making as
demonstrated by strong performance of large language models on medical
licensure exams. However, their ability to solve complex, open-ended cases,
which may be representative of clinical practice, remains unexplored. Methods:
In this study, the accuracy of large language AI models GPT4 and GPT3.5 in
diagnosing complex clinical cases was investigated using published Case Records
of the Massachusetts General Hospital. A total of 50 cases requiring a
diagnosis and diagnostic test published from January 1, 2022 to April 16, 2022
were identified. For each case, models were given a prompt requesting the top
three specific diagnoses and associated diagnostic tests, followed by case
text, labs, and figure legends. Model outputs were assessed in comparison to
the final clinical diagnosis and whether the model-predicted test would result
in a correct diagnosis. Results: GPT4 and GPT3.5 accurately provided the
correct diagnosis in 26% and 22% of cases in one attempt, and 46% and 42%
within three attempts, respectively. GPT4 and GPT3.5 provided a correct
essential diagnostic test in 28% and 24% of cases in one attempt, and 44% and
50% within three attempts, respectively. No significant differences were found
between the two models, and multiple trials with identical prompts using the
GPT3.5 model provided similar results. Conclusions: In summary, these models
demonstrate potential usefulness in generating differential diagnoses but
remain limited in their ability to provide a single unifying diagnosis in
complex, open-ended cases. Future research should focus on evaluating model
performance in larger datasets of open-ended clinical challenges and exploring
potential human-AI collaboration strategies to enhance clinical
decision-making.
| no_new_dataset | 0.70779 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.