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
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2302.09789 | Ran Li | Zhong Liu, Ran Li, Shuwei Shao, Xingming Wu and Weihai Chen | Self-Supervised Monocular Depth Estimation with Self-Reference
Distillation and Disparity Offset Refinement | null | null | 10.1109/TCSVT.2023.3275584 | null | cs.CV | http://creativecommons.org/publicdomain/zero/1.0/ | Monocular depth estimation plays a fundamental role in computer vision. Due
to the costly acquisition of depth ground truth, self-supervised methods that
leverage adjacent frames to establish a supervisory signal have emerged as the
most promising paradigms. In this work, we propose two novel ideas to improve
self-supervised monocular depth estimation: 1) self-reference distillation and
2) disparity offset refinement. Specifically, we use a parameter-optimized
model as the teacher updated as the training epochs to provide additional
supervision during the training process. The teacher model has the same
structure as the student model, with weights inherited from the historical
student model. In addition, a multiview check is introduced to filter out the
outliers produced by the teacher model. Furthermore, we leverage the contextual
consistency between high-scale and low-scale features to obtain multiscale
disparity offsets, which are used to refine the disparity output incrementally
by aligning disparity information at different scales. The experimental results
on the KITTI and Make3D datasets show that our method outperforms previous
state-of-the-art competitors.
| [
{
"version": "v1",
"created": "Mon, 20 Feb 2023 06:28:52 GMT"
},
{
"version": "v2",
"created": "Thu, 15 Jun 2023 08:35:44 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Liu",
"Zhong",
""
],
[
"Li",
"Ran",
""
],
[
"Shao",
"Shuwei",
""
],
[
"Wu",
"Xingming",
""
],
[
"Chen",
"Weihai",
""
]
] | TITLE: Self-Supervised Monocular Depth Estimation with Self-Reference
Distillation and Disparity Offset Refinement
ABSTRACT: Monocular depth estimation plays a fundamental role in computer vision. Due
to the costly acquisition of depth ground truth, self-supervised methods that
leverage adjacent frames to establish a supervisory signal have emerged as the
most promising paradigms. In this work, we propose two novel ideas to improve
self-supervised monocular depth estimation: 1) self-reference distillation and
2) disparity offset refinement. Specifically, we use a parameter-optimized
model as the teacher updated as the training epochs to provide additional
supervision during the training process. The teacher model has the same
structure as the student model, with weights inherited from the historical
student model. In addition, a multiview check is introduced to filter out the
outliers produced by the teacher model. Furthermore, we leverage the contextual
consistency between high-scale and low-scale features to obtain multiscale
disparity offsets, which are used to refine the disparity output incrementally
by aligning disparity information at different scales. The experimental results
on the KITTI and Make3D datasets show that our method outperforms previous
state-of-the-art competitors.
|
2302.10183 | Jean-Pierre Fouque | Alessandro Doldi, Yichen Feng, Jean-Pierre Fouque, Marco Frittelli | Multivariate Systemic Risk Measures and Computation by Deep Learning
Algorithms | 4 figures, 5 tables | null | null | null | cs.LG math.PR | http://creativecommons.org/licenses/by/4.0/ | In this work we propose deep learning-based algorithms for the computation of
systemic shortfall risk measures defined via multivariate utility functions. We
discuss the key related theoretical aspects, with a particular focus on the
fairness properties of primal optima and associated risk allocations. The
algorithms we provide allow for learning primal optimizers, optima for the dual
representation and corresponding fair risk allocations. We test our algorithms
by comparison to a benchmark model, based on a paired exponential utility
function, for which we can provide explicit formulas. We also show evidence of
convergence in a case for which explicit formulas are not available.
| [
{
"version": "v1",
"created": "Thu, 2 Feb 2023 22:16:49 GMT"
},
{
"version": "v2",
"created": "Wed, 14 Jun 2023 19:19:36 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Doldi",
"Alessandro",
""
],
[
"Feng",
"Yichen",
""
],
[
"Fouque",
"Jean-Pierre",
""
],
[
"Frittelli",
"Marco",
""
]
] | TITLE: Multivariate Systemic Risk Measures and Computation by Deep Learning
Algorithms
ABSTRACT: In this work we propose deep learning-based algorithms for the computation of
systemic shortfall risk measures defined via multivariate utility functions. We
discuss the key related theoretical aspects, with a particular focus on the
fairness properties of primal optima and associated risk allocations. The
algorithms we provide allow for learning primal optimizers, optima for the dual
representation and corresponding fair risk allocations. We test our algorithms
by comparison to a benchmark model, based on a paired exponential utility
function, for which we can provide explicit formulas. We also show evidence of
convergence in a case for which explicit formulas are not available.
|
2302.11756 | Yijia Zheng | Yijia Zheng, Tong He, Yixuan Qiu, David Wipf | Learning Manifold Dimensions with Conditional Variational Autoencoders | Published in NeurIPS 2022 | null | null | null | cs.LG cs.AI stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Although the variational autoencoder (VAE) and its conditional extension
(CVAE) are capable of state-of-the-art results across multiple domains, their
precise behavior is still not fully understood, particularly in the context of
data (like images) that lie on or near a low-dimensional manifold. For example,
while prior work has suggested that the globally optimal VAE solution can learn
the correct manifold dimension, a necessary (but not sufficient) condition for
producing samples from the true data distribution, this has never been
rigorously proven. Moreover, it remains unclear how such considerations would
change when various types of conditioning variables are introduced, or when the
data support is extended to a union of manifolds (e.g., as is likely the case
for MNIST digits and related). In this work, we address these points by first
proving that VAE global minima are indeed capable of recovering the correct
manifold dimension. We then extend this result to more general CVAEs,
demonstrating practical scenarios whereby the conditioning variables allow the
model to adaptively learn manifolds of varying dimension across samples. Our
analyses, which have practical implications for various CVAE design choices,
are also supported by numerical results on both synthetic and real-world
datasets.
| [
{
"version": "v1",
"created": "Thu, 23 Feb 2023 03:04:00 GMT"
},
{
"version": "v2",
"created": "Tue, 13 Jun 2023 18:50:18 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Zheng",
"Yijia",
""
],
[
"He",
"Tong",
""
],
[
"Qiu",
"Yixuan",
""
],
[
"Wipf",
"David",
""
]
] | TITLE: Learning Manifold Dimensions with Conditional Variational Autoencoders
ABSTRACT: Although the variational autoencoder (VAE) and its conditional extension
(CVAE) are capable of state-of-the-art results across multiple domains, their
precise behavior is still not fully understood, particularly in the context of
data (like images) that lie on or near a low-dimensional manifold. For example,
while prior work has suggested that the globally optimal VAE solution can learn
the correct manifold dimension, a necessary (but not sufficient) condition for
producing samples from the true data distribution, this has never been
rigorously proven. Moreover, it remains unclear how such considerations would
change when various types of conditioning variables are introduced, or when the
data support is extended to a union of manifolds (e.g., as is likely the case
for MNIST digits and related). In this work, we address these points by first
proving that VAE global minima are indeed capable of recovering the correct
manifold dimension. We then extend this result to more general CVAEs,
demonstrating practical scenarios whereby the conditioning variables allow the
model to adaptively learn manifolds of varying dimension across samples. Our
analyses, which have practical implications for various CVAE design choices,
are also supported by numerical results on both synthetic and real-world
datasets.
|
2302.11793 | Callum Rhys Tilbury | Callum Rhys Tilbury, Filippos Christianos, Stefano V. Albrecht | Revisiting the Gumbel-Softmax in MADDPG | Presented at AAMAS Workshop on Adaptive and Learning Agents, 2023 | null | null | null | cs.LG cs.AI cs.MA stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | MADDPG is an algorithm in multi-agent reinforcement learning (MARL) that
extends the popular single-agent method, DDPG, to multi-agent scenarios.
Importantly, DDPG is an algorithm designed for continuous action spaces, where
the gradient of the state-action value function exists. For this algorithm to
work in discrete action spaces, discrete gradient estimation must be performed.
For MADDPG, the Gumbel-Softmax (GS) estimator is used -- a reparameterisation
which relaxes a discrete distribution into a similar continuous one. This
method, however, is statistically biased, and a recent MARL benchmarking paper
suggests that this bias makes MADDPG perform poorly in grid-world situations,
where the action space is discrete. Fortunately, many alternatives to the GS
exist, boasting a wide range of properties. This paper explores several of
these alternatives and integrates them into MADDPG for discrete grid-world
scenarios. The corresponding impact on various performance metrics is then
measured and analysed. It is found that one of the proposed estimators performs
significantly better than the original GS in several tasks, achieving up to 55%
higher returns, along with faster convergence.
| [
{
"version": "v1",
"created": "Thu, 23 Feb 2023 06:13:51 GMT"
},
{
"version": "v2",
"created": "Wed, 14 Jun 2023 13:43:44 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Tilbury",
"Callum Rhys",
""
],
[
"Christianos",
"Filippos",
""
],
[
"Albrecht",
"Stefano V.",
""
]
] | TITLE: Revisiting the Gumbel-Softmax in MADDPG
ABSTRACT: MADDPG is an algorithm in multi-agent reinforcement learning (MARL) that
extends the popular single-agent method, DDPG, to multi-agent scenarios.
Importantly, DDPG is an algorithm designed for continuous action spaces, where
the gradient of the state-action value function exists. For this algorithm to
work in discrete action spaces, discrete gradient estimation must be performed.
For MADDPG, the Gumbel-Softmax (GS) estimator is used -- a reparameterisation
which relaxes a discrete distribution into a similar continuous one. This
method, however, is statistically biased, and a recent MARL benchmarking paper
suggests that this bias makes MADDPG perform poorly in grid-world situations,
where the action space is discrete. Fortunately, many alternatives to the GS
exist, boasting a wide range of properties. This paper explores several of
these alternatives and integrates them into MADDPG for discrete grid-world
scenarios. The corresponding impact on various performance metrics is then
measured and analysed. It is found that one of the proposed estimators performs
significantly better than the original GS in several tasks, achieving up to 55%
higher returns, along with faster convergence.
|
2302.14042 | Xiaoman Zhang | Xiaoman Zhang, Chaoyi Wu, Ya Zhang, Yanfeng Wang, Weidi Xie | Knowledge-enhanced Visual-Language Pre-training on Chest Radiology
Images | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | While multi-modal foundation models pre-trained on large-scale data have been
successful in natural language understanding and vision recognition, their use
in medical domains is still limited due to the fine-grained nature of medical
tasks and the high demand for domain knowledge. To address this challenge, we
propose a novel approach called Knowledge-enhanced Auto Diagnosis (KAD) which
leverages existing medical domain knowledge to guide vision-language
pre-training using paired chest X-rays and radiology reports. We evaluate KAD
on {four} external X-ray datasets and demonstrate that its zero-shot
performance is not only comparable to that of fully-supervised models, but also
superior to the average of three expert radiologists for three (out of five)
pathologies with statistical significance. Moreover, when few-shot annotation
is available, KAD outperforms all existing approaches in fine-tuning settings,
demonstrating its potential for application in different clinical scenarios.
| [
{
"version": "v1",
"created": "Mon, 27 Feb 2023 18:53:10 GMT"
},
{
"version": "v2",
"created": "Sat, 8 Apr 2023 06:33:30 GMT"
},
{
"version": "v3",
"created": "Wed, 14 Jun 2023 07:33:16 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Zhang",
"Xiaoman",
""
],
[
"Wu",
"Chaoyi",
""
],
[
"Zhang",
"Ya",
""
],
[
"Wang",
"Yanfeng",
""
],
[
"Xie",
"Weidi",
""
]
] | TITLE: Knowledge-enhanced Visual-Language Pre-training on Chest Radiology
Images
ABSTRACT: While multi-modal foundation models pre-trained on large-scale data have been
successful in natural language understanding and vision recognition, their use
in medical domains is still limited due to the fine-grained nature of medical
tasks and the high demand for domain knowledge. To address this challenge, we
propose a novel approach called Knowledge-enhanced Auto Diagnosis (KAD) which
leverages existing medical domain knowledge to guide vision-language
pre-training using paired chest X-rays and radiology reports. We evaluate KAD
on {four} external X-ray datasets and demonstrate that its zero-shot
performance is not only comparable to that of fully-supervised models, but also
superior to the average of three expert radiologists for three (out of five)
pathologies with statistical significance. Moreover, when few-shot annotation
is available, KAD outperforms all existing approaches in fine-tuning settings,
demonstrating its potential for application in different clinical scenarios.
|
2302.14192 | Sabri Mustafa Kahya | Sabri Mustafa Kahya, Muhammet Sami Yavuz, Eckehard Steinbach | Reconstruction-based Out-of-Distribution Detection for Short-Range FMCW
Radar | Accepted at EUSIPCO 2023 | null | null | null | eess.SP cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Out-of-distribution (OOD) detection recently has drawn attention due to its
critical role in the safe deployment of modern neural network architectures in
real-world applications. The OOD detectors aim to distinguish samples that lie
outside the training distribution in order to avoid the overconfident
predictions of machine learning models on OOD data. Existing detectors, which
mainly rely on the logit, intermediate feature space, softmax score, or
reconstruction loss, manage to produce promising results. However, most of
these methods are developed for the image domain. In this study, we propose a
novel reconstruction-based OOD detector to operate on the radar domain. Our
method exploits an autoencoder (AE) and its latent representation to detect the
OOD samples. We propose two scores incorporating the patch-based reconstruction
loss and the energy value calculated from the latent representations of each
patch. We achieve an AUROC of 90.72% on our dataset collected by using 60 GHz
short-range FMCW Radar. The experiments demonstrate that, in terms of AUROC and
AUPR, our method outperforms the baseline (AE) and the other state-of-the-art
methods. Also, thanks to its model size of 641 kB, our detector is suitable for
embedded usage.
| [
{
"version": "v1",
"created": "Mon, 27 Feb 2023 23:03:51 GMT"
},
{
"version": "v2",
"created": "Thu, 15 Jun 2023 09:26:42 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Kahya",
"Sabri Mustafa",
""
],
[
"Yavuz",
"Muhammet Sami",
""
],
[
"Steinbach",
"Eckehard",
""
]
] | TITLE: Reconstruction-based Out-of-Distribution Detection for Short-Range FMCW
Radar
ABSTRACT: Out-of-distribution (OOD) detection recently has drawn attention due to its
critical role in the safe deployment of modern neural network architectures in
real-world applications. The OOD detectors aim to distinguish samples that lie
outside the training distribution in order to avoid the overconfident
predictions of machine learning models on OOD data. Existing detectors, which
mainly rely on the logit, intermediate feature space, softmax score, or
reconstruction loss, manage to produce promising results. However, most of
these methods are developed for the image domain. In this study, we propose a
novel reconstruction-based OOD detector to operate on the radar domain. Our
method exploits an autoencoder (AE) and its latent representation to detect the
OOD samples. We propose two scores incorporating the patch-based reconstruction
loss and the energy value calculated from the latent representations of each
patch. We achieve an AUROC of 90.72% on our dataset collected by using 60 GHz
short-range FMCW Radar. The experiments demonstrate that, in terms of AUROC and
AUPR, our method outperforms the baseline (AE) and the other state-of-the-art
methods. Also, thanks to its model size of 641 kB, our detector is suitable for
embedded usage.
|
2302.14376 | Zhongkai Hao | Zhongkai Hao, Zhengyi Wang, Hang Su, Chengyang Ying, Yinpeng Dong,
Songming Liu, Ze Cheng, Jian Song, Jun Zhu | GNOT: A General Neural Operator Transformer for Operator Learning | null | null | null | null | cs.LG cs.NA math.NA physics.comp-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Learning partial differential equations' (PDEs) solution operators is an
essential problem in machine learning. However, there are several challenges
for learning operators in practical applications like the irregular mesh,
multiple input functions, and complexity of the PDEs' solution. To address
these challenges, we propose a general neural operator transformer (GNOT), a
scalable and effective transformer-based framework for learning operators. By
designing a novel heterogeneous normalized attention layer, our model is highly
flexible to handle multiple input functions and irregular meshes. Besides, we
introduce a geometric gating mechanism which could be viewed as a soft domain
decomposition to solve the multi-scale problems. The large model capacity of
the transformer architecture grants our model the possibility to scale to large
datasets and practical problems. We conduct extensive experiments on multiple
challenging datasets from different domains and achieve a remarkable
improvement compared with alternative methods. Our code and data are publicly
available at \url{https://github.com/thu-ml/GNOT}.
| [
{
"version": "v1",
"created": "Tue, 28 Feb 2023 07:58:49 GMT"
},
{
"version": "v2",
"created": "Wed, 31 May 2023 06:20:32 GMT"
},
{
"version": "v3",
"created": "Wed, 14 Jun 2023 12:26:03 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Hao",
"Zhongkai",
""
],
[
"Wang",
"Zhengyi",
""
],
[
"Su",
"Hang",
""
],
[
"Ying",
"Chengyang",
""
],
[
"Dong",
"Yinpeng",
""
],
[
"Liu",
"Songming",
""
],
[
"Cheng",
"Ze",
""
],
[
"Song",
"Jian",
""
],
[
"Zhu",
"Jun",
""
]
] | TITLE: GNOT: A General Neural Operator Transformer for Operator Learning
ABSTRACT: Learning partial differential equations' (PDEs) solution operators is an
essential problem in machine learning. However, there are several challenges
for learning operators in practical applications like the irregular mesh,
multiple input functions, and complexity of the PDEs' solution. To address
these challenges, we propose a general neural operator transformer (GNOT), a
scalable and effective transformer-based framework for learning operators. By
designing a novel heterogeneous normalized attention layer, our model is highly
flexible to handle multiple input functions and irregular meshes. Besides, we
introduce a geometric gating mechanism which could be viewed as a soft domain
decomposition to solve the multi-scale problems. The large model capacity of
the transformer architecture grants our model the possibility to scale to large
datasets and practical problems. We conduct extensive experiments on multiple
challenging datasets from different domains and achieve a remarkable
improvement compared with alternative methods. Our code and data are publicly
available at \url{https://github.com/thu-ml/GNOT}.
|
2303.00709 | Yuan Gao | Yuan Gao, Rasmus Kyng, Daniel A. Spielman | Robust and Practical Solution of Laplacian Equations by Approximate
Elimination | null | null | null | null | math.NA cs.DS cs.MS cs.NA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We introduce a new algorithm and software for solving linear equations in
symmetric diagonally dominant matrices with non-positive off-diagonal entries
(SDDM matrices), including Laplacian matrices. We use pre-conditioned conjugate
gradient (PCG) to solve the system of linear equations. Our preconditioner is a
variant of the Approximate Cholesky factorization of Kyng and Sachdeva (FOCS
2016). Our factorization approach is simple: we eliminate matrix rows/columns
one at a time and update the remaining matrix using sampling to approximate the
outcome of complete Cholesky factorization. Unlike earlier approaches, our
sampling always maintains a connectivity in the remaining non-zero structure.
Our algorithm comes with a tuning parameter that upper bounds the number of
samples made per original entry. We implement our algorithm in Julia, providing
two versions, AC and AC2, that respectively use 1 and 2 samples per original
entry. We compare their single-threaded performance to that of current
state-of-the-art solvers Combinatorial Multigrid (CMG),
BoomerAMG-preconditioned Krylov solvers from HyPre and PETSc, Lean Algebraic
Multigrid (LAMG), and MATLAB's with Incomplete Cholesky Factorization (ICC).
Our evaluation uses a broad class of problems, including all large SDDM
matrices from the SuiteSparse collection and diverse programmatically generated
instances. Our experiments suggest that our algorithm attains a level of
robustness and reliability not seen before in SDDM solvers, while retaining
good performance across all instances. Our code and data are public, and we
provide a tutorial on how to replicate our tests. We hope that others will
adopt this suite of tests as a benchmark, which we refer to as SDDM2023. Our
solver code is available at: https://github.com/danspielman/Laplacians.jl/ Our
benchmarking data and tutorial are available at:
https://rjkyng.github.io/SDDM2023/
| [
{
"version": "v1",
"created": "Wed, 1 Mar 2023 18:07:51 GMT"
},
{
"version": "v2",
"created": "Wed, 14 Jun 2023 08:26:46 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Gao",
"Yuan",
""
],
[
"Kyng",
"Rasmus",
""
],
[
"Spielman",
"Daniel A.",
""
]
] | TITLE: Robust and Practical Solution of Laplacian Equations by Approximate
Elimination
ABSTRACT: We introduce a new algorithm and software for solving linear equations in
symmetric diagonally dominant matrices with non-positive off-diagonal entries
(SDDM matrices), including Laplacian matrices. We use pre-conditioned conjugate
gradient (PCG) to solve the system of linear equations. Our preconditioner is a
variant of the Approximate Cholesky factorization of Kyng and Sachdeva (FOCS
2016). Our factorization approach is simple: we eliminate matrix rows/columns
one at a time and update the remaining matrix using sampling to approximate the
outcome of complete Cholesky factorization. Unlike earlier approaches, our
sampling always maintains a connectivity in the remaining non-zero structure.
Our algorithm comes with a tuning parameter that upper bounds the number of
samples made per original entry. We implement our algorithm in Julia, providing
two versions, AC and AC2, that respectively use 1 and 2 samples per original
entry. We compare their single-threaded performance to that of current
state-of-the-art solvers Combinatorial Multigrid (CMG),
BoomerAMG-preconditioned Krylov solvers from HyPre and PETSc, Lean Algebraic
Multigrid (LAMG), and MATLAB's with Incomplete Cholesky Factorization (ICC).
Our evaluation uses a broad class of problems, including all large SDDM
matrices from the SuiteSparse collection and diverse programmatically generated
instances. Our experiments suggest that our algorithm attains a level of
robustness and reliability not seen before in SDDM solvers, while retaining
good performance across all instances. Our code and data are public, and we
provide a tutorial on how to replicate our tests. We hope that others will
adopt this suite of tests as a benchmark, which we refer to as SDDM2023. Our
solver code is available at: https://github.com/danspielman/Laplacians.jl/ Our
benchmarking data and tutorial are available at:
https://rjkyng.github.io/SDDM2023/
|
2303.13952 | Xieyuanli Chen | Jiafeng Cui, Xieyuanli Chen | CCL: Continual Contrastive Learning for LiDAR Place Recognition | Accepted by IEEE RA-L in 2023 | null | 10.1109/LRA.2023.3284359 | null | cs.CV cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Place recognition is an essential and challenging task in loop closing and
global localization for robotics and autonomous driving applications.
Benefiting from the recent advances in deep learning techniques, the
performance of LiDAR place recognition (LPR) has been greatly improved.
However, current deep learning-based methods suffer from two major problems:
poor generalization ability and catastrophic forgetting. In this paper, we
propose a continual contrastive learning method, named CCL, to tackle the
catastrophic forgetting problem and generally improve the robustness of LPR
approaches. Our CCL constructs a contrastive feature pool and utilizes
contrastive loss to train more transferable representations of places. When
transferred into new environments, our CCL continuously reviews the contrastive
memory bank and applies a distribution-based knowledge distillation to maintain
the retrieval ability of the past data while continually learning to recognize
new places from the new data. We thoroughly evaluate our approach on Oxford,
MulRan, and PNV datasets using three different LPR methods. The experimental
results show that our CCL consistently improves the performance of different
methods in different environments outperforming the state-of-the-art continual
learning method. The implementation of our method has been released at
https://github.com/cloudcjf/CCL.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 12:14:54 GMT"
},
{
"version": "v2",
"created": "Wed, 14 Jun 2023 12:22:07 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Cui",
"Jiafeng",
""
],
[
"Chen",
"Xieyuanli",
""
]
] | TITLE: CCL: Continual Contrastive Learning for LiDAR Place Recognition
ABSTRACT: Place recognition is an essential and challenging task in loop closing and
global localization for robotics and autonomous driving applications.
Benefiting from the recent advances in deep learning techniques, the
performance of LiDAR place recognition (LPR) has been greatly improved.
However, current deep learning-based methods suffer from two major problems:
poor generalization ability and catastrophic forgetting. In this paper, we
propose a continual contrastive learning method, named CCL, to tackle the
catastrophic forgetting problem and generally improve the robustness of LPR
approaches. Our CCL constructs a contrastive feature pool and utilizes
contrastive loss to train more transferable representations of places. When
transferred into new environments, our CCL continuously reviews the contrastive
memory bank and applies a distribution-based knowledge distillation to maintain
the retrieval ability of the past data while continually learning to recognize
new places from the new data. We thoroughly evaluate our approach on Oxford,
MulRan, and PNV datasets using three different LPR methods. The experimental
results show that our CCL consistently improves the performance of different
methods in different environments outperforming the state-of-the-art continual
learning method. The implementation of our method has been released at
https://github.com/cloudcjf/CCL.
|
2303.16310 | Varun Mandalapu | Varun Mandalapu, Lavanya Elluri, Piyush Vyas and Nirmalya Roy | Crime Prediction Using Machine Learning and Deep Learning: A Systematic
Review and Future Directions | 35 Pages, 6 tables and 11 figures. Consists of Dataset links used for
crime prediction. Review Paper | null | 10.1109/ACCESS.2023.3286344 | null | cs.LG cs.AI cs.CV cs.CY cs.DB | http://creativecommons.org/licenses/by/4.0/ | Predicting crime using machine learning and deep learning techniques has
gained considerable attention from researchers in recent years, focusing on
identifying patterns and trends in crime occurrences. This review paper
examines over 150 articles to explore the various machine learning and deep
learning algorithms applied to predict crime. The study provides access to the
datasets used for crime prediction by researchers and analyzes prominent
approaches applied in machine learning and deep learning algorithms to predict
crime, offering insights into different trends and factors related to criminal
activities. Additionally, the paper highlights potential gaps and future
directions that can enhance the accuracy of crime prediction. Finally, the
comprehensive overview of research discussed in this paper on crime prediction
using machine learning and deep learning approaches serves as a valuable
reference for researchers in this field. By gaining a deeper understanding of
crime prediction techniques, law enforcement agencies can develop strategies to
prevent and respond to criminal activities more effectively.
| [
{
"version": "v1",
"created": "Tue, 28 Mar 2023 21:07:42 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Mandalapu",
"Varun",
""
],
[
"Elluri",
"Lavanya",
""
],
[
"Vyas",
"Piyush",
""
],
[
"Roy",
"Nirmalya",
""
]
] | TITLE: Crime Prediction Using Machine Learning and Deep Learning: A Systematic
Review and Future Directions
ABSTRACT: Predicting crime using machine learning and deep learning techniques has
gained considerable attention from researchers in recent years, focusing on
identifying patterns and trends in crime occurrences. This review paper
examines over 150 articles to explore the various machine learning and deep
learning algorithms applied to predict crime. The study provides access to the
datasets used for crime prediction by researchers and analyzes prominent
approaches applied in machine learning and deep learning algorithms to predict
crime, offering insights into different trends and factors related to criminal
activities. Additionally, the paper highlights potential gaps and future
directions that can enhance the accuracy of crime prediction. Finally, the
comprehensive overview of research discussed in this paper on crime prediction
using machine learning and deep learning approaches serves as a valuable
reference for researchers in this field. By gaining a deeper understanding of
crime prediction techniques, law enforcement agencies can develop strategies to
prevent and respond to criminal activities more effectively.
|
2303.16541 | Jiawei Liu | Jiawei Liu, Weining Wang, Sihan Chen, Xinxin Zhu, Jing Liu | Sounding Video Generator: A Unified Framework for Text-guided Sounding
Video Generation | null | IEEE Transactions on Multimedia 2023 | 10.1109/TMM.2023.3262180 | null | cs.CV cs.AI cs.MM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | As a combination of visual and audio signals, video is inherently
multi-modal. However, existing video generation methods are primarily intended
for the synthesis of visual frames, whereas audio signals in realistic videos
are disregarded. In this work, we concentrate on a rarely investigated problem
of text guided sounding video generation and propose the Sounding Video
Generator (SVG), a unified framework for generating realistic videos along with
audio signals. Specifically, we present the SVG-VQGAN to transform visual
frames and audio melspectrograms into discrete tokens. SVG-VQGAN applies a
novel hybrid contrastive learning method to model inter-modal and intra-modal
consistency and improve the quantized representations. A cross-modal attention
module is employed to extract associated features of visual frames and audio
signals for contrastive learning. Then, a Transformer-based decoder is used to
model associations between texts, visual frames, and audio signals at token
level for auto-regressive sounding video generation. AudioSetCap, a human
annotated text-video-audio paired dataset, is produced for training SVG.
Experimental results demonstrate the superiority of our method when compared
with existing textto-video generation methods as well as audio generation
methods on Kinetics and VAS datasets.
| [
{
"version": "v1",
"created": "Wed, 29 Mar 2023 09:07:31 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Liu",
"Jiawei",
""
],
[
"Wang",
"Weining",
""
],
[
"Chen",
"Sihan",
""
],
[
"Zhu",
"Xinxin",
""
],
[
"Liu",
"Jing",
""
]
] | TITLE: Sounding Video Generator: A Unified Framework for Text-guided Sounding
Video Generation
ABSTRACT: As a combination of visual and audio signals, video is inherently
multi-modal. However, existing video generation methods are primarily intended
for the synthesis of visual frames, whereas audio signals in realistic videos
are disregarded. In this work, we concentrate on a rarely investigated problem
of text guided sounding video generation and propose the Sounding Video
Generator (SVG), a unified framework for generating realistic videos along with
audio signals. Specifically, we present the SVG-VQGAN to transform visual
frames and audio melspectrograms into discrete tokens. SVG-VQGAN applies a
novel hybrid contrastive learning method to model inter-modal and intra-modal
consistency and improve the quantized representations. A cross-modal attention
module is employed to extract associated features of visual frames and audio
signals for contrastive learning. Then, a Transformer-based decoder is used to
model associations between texts, visual frames, and audio signals at token
level for auto-regressive sounding video generation. AudioSetCap, a human
annotated text-video-audio paired dataset, is produced for training SVG.
Experimental results demonstrate the superiority of our method when compared
with existing textto-video generation methods as well as audio generation
methods on Kinetics and VAS datasets.
|
2304.01329 | William Clark | Maria Oprea and Mark Walth and Robert Stephany and Gabriella Torres
Nothaft and Arnaldo Rodriguez-Gonzalez and William Clark | Learning the Delay Using Neural Delay Differential Equations | Comments welcome! | null | null | null | math.OC cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The intersection of machine learning and dynamical systems has generated
considerable interest recently. Neural Ordinary Differential Equations (NODEs)
represent a rich overlap between these fields. In this paper, we develop a
continuous time neural network approach based on Delay Differential Equations
(DDEs). Our model uses the adjoint sensitivity method to learn the model
parameters and delay directly from data. Our approach is inspired by that of
NODEs and extends earlier neural DDE models, which have assumed that the value
of the delay is known a priori. We perform a sensitivity analysis on our
proposed approach and demonstrate its ability to learn DDE parameters from
benchmark systems. We conclude our discussion with potential future directions
and applications.
| [
{
"version": "v1",
"created": "Mon, 3 Apr 2023 19:50:36 GMT"
},
{
"version": "v2",
"created": "Wed, 14 Jun 2023 16:51:05 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Oprea",
"Maria",
""
],
[
"Walth",
"Mark",
""
],
[
"Stephany",
"Robert",
""
],
[
"Nothaft",
"Gabriella Torres",
""
],
[
"Rodriguez-Gonzalez",
"Arnaldo",
""
],
[
"Clark",
"William",
""
]
] | TITLE: Learning the Delay Using Neural Delay Differential Equations
ABSTRACT: The intersection of machine learning and dynamical systems has generated
considerable interest recently. Neural Ordinary Differential Equations (NODEs)
represent a rich overlap between these fields. In this paper, we develop a
continuous time neural network approach based on Delay Differential Equations
(DDEs). Our model uses the adjoint sensitivity method to learn the model
parameters and delay directly from data. Our approach is inspired by that of
NODEs and extends earlier neural DDE models, which have assumed that the value
of the delay is known a priori. We perform a sensitivity analysis on our
proposed approach and demonstrate its ability to learn DDE parameters from
benchmark systems. We conclude our discussion with potential future directions
and applications.
|
2304.01669 | Ngoc-Bao Nguyen | Ngoc-Bao Nguyen, Keshigeyan Chandrasegaran, Milad Abdollahzadeh,
Ngai-Man Cheung | Re-thinking Model Inversion Attacks Against Deep Neural Networks | Accepted to CVPR 2023. The first two authors contributed equally | null | null | null | cs.LG cs.CR cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Model inversion (MI) attacks aim to infer and reconstruct private training
data by abusing access to a model. MI attacks have raised concerns about the
leaking of sensitive information (e.g. private face images used in training a
face recognition system). Recently, several algorithms for MI have been
proposed to improve the attack performance. In this work, we revisit MI, study
two fundamental issues pertaining to all state-of-the-art (SOTA) MI algorithms,
and propose solutions to these issues which lead to a significant boost in
attack performance for all SOTA MI. In particular, our contributions are
two-fold: 1) We analyze the optimization objective of SOTA MI algorithms, argue
that the objective is sub-optimal for achieving MI, and propose an improved
optimization objective that boosts attack performance significantly. 2) We
analyze "MI overfitting", show that it would prevent reconstructed images from
learning semantics of training data, and propose a novel "model augmentation"
idea to overcome this issue. Our proposed solutions are simple and improve all
SOTA MI attack accuracy significantly. E.g., in the standard CelebA benchmark,
our solutions improve accuracy by 11.8% and achieve for the first time over 90%
attack accuracy. Our findings demonstrate that there is a clear risk of leaking
sensitive information from deep learning models. We urge serious consideration
to be given to the privacy implications. Our code, demo, and models are
available at
https://ngoc-nguyen-0.github.io/re-thinking_model_inversion_attacks/
| [
{
"version": "v1",
"created": "Tue, 4 Apr 2023 09:58:07 GMT"
},
{
"version": "v2",
"created": "Thu, 15 Jun 2023 14:00:24 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Nguyen",
"Ngoc-Bao",
""
],
[
"Chandrasegaran",
"Keshigeyan",
""
],
[
"Abdollahzadeh",
"Milad",
""
],
[
"Cheung",
"Ngai-Man",
""
]
] | TITLE: Re-thinking Model Inversion Attacks Against Deep Neural Networks
ABSTRACT: Model inversion (MI) attacks aim to infer and reconstruct private training
data by abusing access to a model. MI attacks have raised concerns about the
leaking of sensitive information (e.g. private face images used in training a
face recognition system). Recently, several algorithms for MI have been
proposed to improve the attack performance. In this work, we revisit MI, study
two fundamental issues pertaining to all state-of-the-art (SOTA) MI algorithms,
and propose solutions to these issues which lead to a significant boost in
attack performance for all SOTA MI. In particular, our contributions are
two-fold: 1) We analyze the optimization objective of SOTA MI algorithms, argue
that the objective is sub-optimal for achieving MI, and propose an improved
optimization objective that boosts attack performance significantly. 2) We
analyze "MI overfitting", show that it would prevent reconstructed images from
learning semantics of training data, and propose a novel "model augmentation"
idea to overcome this issue. Our proposed solutions are simple and improve all
SOTA MI attack accuracy significantly. E.g., in the standard CelebA benchmark,
our solutions improve accuracy by 11.8% and achieve for the first time over 90%
attack accuracy. Our findings demonstrate that there is a clear risk of leaking
sensitive information from deep learning models. We urge serious consideration
to be given to the privacy implications. Our code, demo, and models are
available at
https://ngoc-nguyen-0.github.io/re-thinking_model_inversion_attacks/
|
2304.03129 | Liwen Hu | Liwen hu, Lei Ma, Zhaofei Yu, Boxin Shi and Tiejun Huang | Spike Stream Denoising via Spike Camera Simulation | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | As a neuromorphic sensor with high temporal resolution, the spike camera
shows enormous potential in high-speed visual tasks. However, the high-speed
sampling of light propagation processes by existing cameras brings unavoidable
noise phenomena. Eliminating the unique noise in spike stream is always a key
point for spike-based methods. No previous work has addressed the detailed
noise mechanism of the spike camera. To this end, we propose a systematic noise
model for spike camera based on its unique circuit. In addition, we carefully
constructed the noise evaluation equation and experimental scenarios to measure
noise variables. Based on our noise model, the first benchmark for spike stream
denoising is proposed which includes clear (noisy) spike stream. Further, we
design a tailored spike stream denoising framework (DnSS) where denoised spike
stream is obtained by decoding inferred inter-spike intervals. Experiments show
that DnSS has promising performance on the proposed benchmark. Eventually, DnSS
can be generalized well on real spike stream.
| [
{
"version": "v1",
"created": "Thu, 6 Apr 2023 14:59:48 GMT"
},
{
"version": "v2",
"created": "Thu, 15 Jun 2023 07:46:24 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"hu",
"Liwen",
""
],
[
"Ma",
"Lei",
""
],
[
"Yu",
"Zhaofei",
""
],
[
"Shi",
"Boxin",
""
],
[
"Huang",
"Tiejun",
""
]
] | TITLE: Spike Stream Denoising via Spike Camera Simulation
ABSTRACT: As a neuromorphic sensor with high temporal resolution, the spike camera
shows enormous potential in high-speed visual tasks. However, the high-speed
sampling of light propagation processes by existing cameras brings unavoidable
noise phenomena. Eliminating the unique noise in spike stream is always a key
point for spike-based methods. No previous work has addressed the detailed
noise mechanism of the spike camera. To this end, we propose a systematic noise
model for spike camera based on its unique circuit. In addition, we carefully
constructed the noise evaluation equation and experimental scenarios to measure
noise variables. Based on our noise model, the first benchmark for spike stream
denoising is proposed which includes clear (noisy) spike stream. Further, we
design a tailored spike stream denoising framework (DnSS) where denoised spike
stream is obtained by decoding inferred inter-spike intervals. Experiments show
that DnSS has promising performance on the proposed benchmark. Eventually, DnSS
can be generalized well on real spike stream.
|
2304.05444 | Tiffany Tseng | Tiffany Tseng, Jennifer King Chen, Mona Abdelrahman, Mary Beth Kery,
Fred Hohman, Adriana Hilliard, R. Benjamin Shapiro | Collaborative Machine Learning Model Building with Families Using Co-ML | Proceedings of the 2023 IDC Conference | null | null | null | cs.HC cs.LG | http://creativecommons.org/licenses/by/4.0/ | Existing novice-friendly machine learning (ML) modeling tools center around a
solo user experience, where a single user collects only their own data to build
a model. However, solo modeling experiences limit valuable opportunities for
encountering alternative ideas and approaches that can arise when learners work
together; consequently, it often precludes encountering critical issues in ML
around data representation and diversity that can surface when different
perspectives are manifested in a group-constructed data set. To address this
issue, we created Co-ML -- a tablet-based app for learners to collaboratively
build ML image classifiers through an end-to-end, iterative model-building
process. In this paper, we illustrate the feasibility and potential richness of
collaborative modeling by presenting an in-depth case study of a family (two
children 11 and 14-years-old working with their parents) using Co-ML in a
facilitated introductory ML activity at home. We share the Co-ML system design
and contribute a discussion of how using Co-ML in a collaborative activity
enabled beginners to collectively engage with dataset design considerations
underrepresented in prior work such as data diversity, class imbalance, and
data quality. We discuss how a distributed collaborative process, in which
individuals can take on different model-building responsibilities, provides a
rich context for children and adults to learn ML dataset design.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 18:31:07 GMT"
},
{
"version": "v2",
"created": "Tue, 13 Jun 2023 17:37:27 GMT"
},
{
"version": "v3",
"created": "Wed, 14 Jun 2023 21:03:40 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Tseng",
"Tiffany",
""
],
[
"Chen",
"Jennifer King",
""
],
[
"Abdelrahman",
"Mona",
""
],
[
"Kery",
"Mary Beth",
""
],
[
"Hohman",
"Fred",
""
],
[
"Hilliard",
"Adriana",
""
],
[
"Shapiro",
"R. Benjamin",
""
]
] | TITLE: Collaborative Machine Learning Model Building with Families Using Co-ML
ABSTRACT: Existing novice-friendly machine learning (ML) modeling tools center around a
solo user experience, where a single user collects only their own data to build
a model. However, solo modeling experiences limit valuable opportunities for
encountering alternative ideas and approaches that can arise when learners work
together; consequently, it often precludes encountering critical issues in ML
around data representation and diversity that can surface when different
perspectives are manifested in a group-constructed data set. To address this
issue, we created Co-ML -- a tablet-based app for learners to collaboratively
build ML image classifiers through an end-to-end, iterative model-building
process. In this paper, we illustrate the feasibility and potential richness of
collaborative modeling by presenting an in-depth case study of a family (two
children 11 and 14-years-old working with their parents) using Co-ML in a
facilitated introductory ML activity at home. We share the Co-ML system design
and contribute a discussion of how using Co-ML in a collaborative activity
enabled beginners to collectively engage with dataset design considerations
underrepresented in prior work such as data diversity, class imbalance, and
data quality. We discuss how a distributed collaborative process, in which
individuals can take on different model-building responsibilities, provides a
rich context for children and adults to learn ML dataset design.
|
2304.08481 | Xiong Xuan | Xuan Xiong, Yicheng Liu, Tianyuan Yuan, Yue Wang, Yilun Wang, Hang
Zhao | Neural Map Prior for Autonomous Driving | CVPR 2023 Camera Ready | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | High-definition (HD) semantic maps are crucial in enabling autonomous
vehicles to navigate urban environments. The traditional method of creating
offline HD maps involves labor-intensive manual annotation processes, which are
not only costly but also insufficient for timely updates. Recent studies have
proposed an alternative approach that generates local maps using online sensor
observations. However, this approach is limited by the sensor's perception
range and its susceptibility to occlusions. In this study, we propose Neural
Map Prior (NMP), a neural representation of global maps. This representation
automatically updates itself and improves the performance of local map
inference. Specifically, we utilize two approaches to achieve this. Firstly, to
integrate a strong map prior into local map inference, we apply
cross-attention, a mechanism that dynamically identifies correlations between
current and prior features. Secondly, to update the global neural map prior, we
utilize a learning-based fusion module that guides the network in fusing
features from previous traversals. Our experimental results, based on the
nuScenes dataset, demonstrate that our framework is highly compatible with
various map segmentation and detection architectures. It significantly improves
map prediction performance, even in challenging weather conditions and
situations with a longer perception range. To the best of our knowledge, this
is the first learning-based system for creating a global map prior.
| [
{
"version": "v1",
"created": "Mon, 17 Apr 2023 17:58:40 GMT"
},
{
"version": "v2",
"created": "Wed, 14 Jun 2023 17:10:00 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Xiong",
"Xuan",
""
],
[
"Liu",
"Yicheng",
""
],
[
"Yuan",
"Tianyuan",
""
],
[
"Wang",
"Yue",
""
],
[
"Wang",
"Yilun",
""
],
[
"Zhao",
"Hang",
""
]
] | TITLE: Neural Map Prior for Autonomous Driving
ABSTRACT: High-definition (HD) semantic maps are crucial in enabling autonomous
vehicles to navigate urban environments. The traditional method of creating
offline HD maps involves labor-intensive manual annotation processes, which are
not only costly but also insufficient for timely updates. Recent studies have
proposed an alternative approach that generates local maps using online sensor
observations. However, this approach is limited by the sensor's perception
range and its susceptibility to occlusions. In this study, we propose Neural
Map Prior (NMP), a neural representation of global maps. This representation
automatically updates itself and improves the performance of local map
inference. Specifically, we utilize two approaches to achieve this. Firstly, to
integrate a strong map prior into local map inference, we apply
cross-attention, a mechanism that dynamically identifies correlations between
current and prior features. Secondly, to update the global neural map prior, we
utilize a learning-based fusion module that guides the network in fusing
features from previous traversals. Our experimental results, based on the
nuScenes dataset, demonstrate that our framework is highly compatible with
various map segmentation and detection architectures. It significantly improves
map prediction performance, even in challenging weather conditions and
situations with a longer perception range. To the best of our knowledge, this
is the first learning-based system for creating a global map prior.
|
2304.11384 | Mingyang Geng | Mingyang Geng, Shangwen Wang, Dezun Dong, Haotian Wang, Ge Li, Zhi
Jin, Xiaoguang Mao, Xiangke Liao | Large Language Models are Few-Shot Summarizers: Multi-Intent Comment
Generation via In-Context Learning | Accepted by the 46th International Conference on Software Engineering
(ICSE 2024) | null | null | null | cs.SE | http://creativecommons.org/licenses/by/4.0/ | Code comment generation aims at generating natural language descriptions for
a code snippet to facilitate developers' program comprehension activities.
Despite being studied for a long time, a bottleneck for existing approaches is
that given a code snippet, they can only generate one comment while developers
usually need to know information from diverse perspectives such as what is the
functionality of this code snippet and how to use it. To tackle this
limitation, this study empirically investigates the feasibility of utilizing
large language models (LLMs) to generate comments that can fulfill developers'
diverse intents. Our intuition is based on the facts that (1) the code and its
pairwise comment are used during the pre-training process of LLMs to build the
semantic connection between the natural language and programming language, and
(2) comments in the real-world projects, which are collected for the
pre-training, usually contain different developers' intents. We thus postulate
that the LLMs can already understand the code from different perspectives after
the pre-training. Indeed, experiments on two large-scale datasets demonstrate
the rationale of our insights: by adopting the in-context learning paradigm and
giving adequate prompts to the LLM (e.g., providing it with ten or more
examples), the LLM can significantly outperform a state-of-the-art supervised
learning approach on generating comments with multiple intents. Results also
show that customized strategies for constructing the prompts and
post-processing strategies for reranking the results can both boost the LLM's
performances, which shed light on future research directions for using LLMs to
achieve comment generation.
| [
{
"version": "v1",
"created": "Sat, 22 Apr 2023 12:26:24 GMT"
},
{
"version": "v2",
"created": "Thu, 8 Jun 2023 15:26:21 GMT"
},
{
"version": "v3",
"created": "Wed, 14 Jun 2023 06:33:10 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Geng",
"Mingyang",
""
],
[
"Wang",
"Shangwen",
""
],
[
"Dong",
"Dezun",
""
],
[
"Wang",
"Haotian",
""
],
[
"Li",
"Ge",
""
],
[
"Jin",
"Zhi",
""
],
[
"Mao",
"Xiaoguang",
""
],
[
"Liao",
"Xiangke",
""
]
] | TITLE: Large Language Models are Few-Shot Summarizers: Multi-Intent Comment
Generation via In-Context Learning
ABSTRACT: Code comment generation aims at generating natural language descriptions for
a code snippet to facilitate developers' program comprehension activities.
Despite being studied for a long time, a bottleneck for existing approaches is
that given a code snippet, they can only generate one comment while developers
usually need to know information from diverse perspectives such as what is the
functionality of this code snippet and how to use it. To tackle this
limitation, this study empirically investigates the feasibility of utilizing
large language models (LLMs) to generate comments that can fulfill developers'
diverse intents. Our intuition is based on the facts that (1) the code and its
pairwise comment are used during the pre-training process of LLMs to build the
semantic connection between the natural language and programming language, and
(2) comments in the real-world projects, which are collected for the
pre-training, usually contain different developers' intents. We thus postulate
that the LLMs can already understand the code from different perspectives after
the pre-training. Indeed, experiments on two large-scale datasets demonstrate
the rationale of our insights: by adopting the in-context learning paradigm and
giving adequate prompts to the LLM (e.g., providing it with ten or more
examples), the LLM can significantly outperform a state-of-the-art supervised
learning approach on generating comments with multiple intents. Results also
show that customized strategies for constructing the prompts and
post-processing strategies for reranking the results can both boost the LLM's
performances, which shed light on future research directions for using LLMs to
achieve comment generation.
|
2304.11961 | Sebastian Berns | Sebastian Berns, Simon Colton, Christian Guckelsberger | Towards Mode Balancing of Generative Models via Diversity Weights | Accepted for oral presentation at the International Conference on
Computational Creativity (ICCC) 2023 | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by/4.0/ | Large data-driven image models are extensively used to support creative and
artistic work. Under the currently predominant distribution-fitting paradigm, a
dataset is treated as ground truth to be approximated as closely as possible.
Yet, many creative applications demand a diverse range of output, and creators
often strive to actively diverge from a given data distribution. We argue that
an adjustment of modelling objectives, from pure mode coverage towards mode
balancing, is necessary to accommodate the goal of higher output diversity. We
present diversity weights, a training scheme that increases a model's output
diversity by balancing the modes in the training dataset. First experiments in
a controlled setting demonstrate the potential of our method. We discuss
connections of our approach to diversity, equity, and inclusion in generative
machine learning more generally, and computational creativity specifically. An
implementation of our algorithm is available at
https://github.com/sebastianberns/diversity-weights
| [
{
"version": "v1",
"created": "Mon, 24 Apr 2023 09:55:17 GMT"
},
{
"version": "v2",
"created": "Tue, 16 May 2023 14:35:37 GMT"
},
{
"version": "v3",
"created": "Thu, 15 Jun 2023 09:21:29 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Berns",
"Sebastian",
""
],
[
"Colton",
"Simon",
""
],
[
"Guckelsberger",
"Christian",
""
]
] | TITLE: Towards Mode Balancing of Generative Models via Diversity Weights
ABSTRACT: Large data-driven image models are extensively used to support creative and
artistic work. Under the currently predominant distribution-fitting paradigm, a
dataset is treated as ground truth to be approximated as closely as possible.
Yet, many creative applications demand a diverse range of output, and creators
often strive to actively diverge from a given data distribution. We argue that
an adjustment of modelling objectives, from pure mode coverage towards mode
balancing, is necessary to accommodate the goal of higher output diversity. We
present diversity weights, a training scheme that increases a model's output
diversity by balancing the modes in the training dataset. First experiments in
a controlled setting demonstrate the potential of our method. We discuss
connections of our approach to diversity, equity, and inclusion in generative
machine learning more generally, and computational creativity specifically. An
implementation of our algorithm is available at
https://github.com/sebastianberns/diversity-weights
|
2304.14034 | Louis Tiao | Louis C. Tiao, Vincent Dutordoir, Victor Picheny | Spherical Inducing Features for Orthogonally-Decoupled Gaussian
Processes | To appear in the Proceedings of the 40th International Conference on
Machine Learning (ICML2023) | null | null | null | cs.LG stat.ML | http://creativecommons.org/licenses/by/4.0/ | Despite their many desirable properties, Gaussian processes (GPs) are often
compared unfavorably to deep neural networks (NNs) for lacking the ability to
learn representations. Recent efforts to bridge the gap between GPs and deep
NNs have yielded a new class of inter-domain variational GPs in which the
inducing variables correspond to hidden units of a feedforward NN. In this
work, we examine some practical issues associated with this approach and
propose an extension that leverages the orthogonal decomposition of GPs to
mitigate these limitations. In particular, we introduce spherical inter-domain
features to construct more flexible data-dependent basis functions for both the
principal and orthogonal components of the GP approximation and show that
incorporating NN activation features under this framework not only alleviates
these shortcomings but is more scalable than alternative strategies.
Experiments on multiple benchmark datasets demonstrate the effectiveness of our
approach.
| [
{
"version": "v1",
"created": "Thu, 27 Apr 2023 09:00:02 GMT"
},
{
"version": "v2",
"created": "Thu, 15 Jun 2023 11:51:43 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Tiao",
"Louis C.",
""
],
[
"Dutordoir",
"Vincent",
""
],
[
"Picheny",
"Victor",
""
]
] | TITLE: Spherical Inducing Features for Orthogonally-Decoupled Gaussian
Processes
ABSTRACT: Despite their many desirable properties, Gaussian processes (GPs) are often
compared unfavorably to deep neural networks (NNs) for lacking the ability to
learn representations. Recent efforts to bridge the gap between GPs and deep
NNs have yielded a new class of inter-domain variational GPs in which the
inducing variables correspond to hidden units of a feedforward NN. In this
work, we examine some practical issues associated with this approach and
propose an extension that leverages the orthogonal decomposition of GPs to
mitigate these limitations. In particular, we introduce spherical inter-domain
features to construct more flexible data-dependent basis functions for both the
principal and orthogonal components of the GP approximation and show that
incorporating NN activation features under this framework not only alleviates
these shortcomings but is more scalable than alternative strategies.
Experiments on multiple benchmark datasets demonstrate the effectiveness of our
approach.
|
2305.00562 | Yiming Qin | Yiming Qin, Huangjie Zheng, Jiangchao Yao, Mingyuan Zhou, Ya Zhang | Class-Balancing Diffusion Models | Accepted by CVPR2023 | null | null | null | cs.CV cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Diffusion-based models have shown the merits of generating high-quality
visual data while preserving better diversity in recent studies. However, such
observation is only justified with curated data distribution, where the data
samples are nicely pre-processed to be uniformly distributed in terms of their
labels. In practice, a long-tailed data distribution appears more common and
how diffusion models perform on such class-imbalanced data remains unknown. In
this work, we first investigate this problem and observe significant
degradation in both diversity and fidelity when the diffusion model is trained
on datasets with class-imbalanced distributions. Especially in tail classes,
the generations largely lose diversity and we observe severe mode-collapse
issues. To tackle this problem, we set from the hypothesis that the data
distribution is not class-balanced, and propose Class-Balancing Diffusion
Models (CBDM) that are trained with a distribution adjustment regularizer as a
solution. Experiments show that images generated by CBDM exhibit higher
diversity and quality in both quantitative and qualitative ways. Our method
benchmarked the generation results on CIFAR100/CIFAR100LT dataset and shows
outstanding performance on the downstream recognition task.
| [
{
"version": "v1",
"created": "Sun, 30 Apr 2023 20:00:14 GMT"
},
{
"version": "v2",
"created": "Wed, 14 Jun 2023 07:25:07 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Qin",
"Yiming",
""
],
[
"Zheng",
"Huangjie",
""
],
[
"Yao",
"Jiangchao",
""
],
[
"Zhou",
"Mingyuan",
""
],
[
"Zhang",
"Ya",
""
]
] | TITLE: Class-Balancing Diffusion Models
ABSTRACT: Diffusion-based models have shown the merits of generating high-quality
visual data while preserving better diversity in recent studies. However, such
observation is only justified with curated data distribution, where the data
samples are nicely pre-processed to be uniformly distributed in terms of their
labels. In practice, a long-tailed data distribution appears more common and
how diffusion models perform on such class-imbalanced data remains unknown. In
this work, we first investigate this problem and observe significant
degradation in both diversity and fidelity when the diffusion model is trained
on datasets with class-imbalanced distributions. Especially in tail classes,
the generations largely lose diversity and we observe severe mode-collapse
issues. To tackle this problem, we set from the hypothesis that the data
distribution is not class-balanced, and propose Class-Balancing Diffusion
Models (CBDM) that are trained with a distribution adjustment regularizer as a
solution. Experiments show that images generated by CBDM exhibit higher
diversity and quality in both quantitative and qualitative ways. Our method
benchmarked the generation results on CIFAR100/CIFAR100LT dataset and shows
outstanding performance on the downstream recognition task.
|
2305.05358 | Marco Peer | Marco Peer, Florian Kleber, Robert Sablatnig | Towards Writer Retrieval for Historical Datasets | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | This paper presents an unsupervised approach for writer retrieval based on
clustering SIFT descriptors detected at keypoint locations resulting in
pseudo-cluster labels. With those cluster labels, a residual network followed
by our proposed NetRVLAD, an encoding layer with reduced complexity compared to
NetVLAD, is trained on 32x32 patches at keypoint locations. Additionally, we
suggest a graph-based reranking algorithm called SGR to exploit similarities of
the page embeddings to boost the retrieval performance. Our approach is
evaluated on two historical datasets (Historical-WI and HisIR19). We include an
evaluation of different backbones and NetRVLAD. It competes with related work
on historical datasets without using explicit encodings. We set a new
State-of-the-art on both datasets by applying our reranking scheme and show
that our approach achieves comparable performance on a modern dataset as well.
| [
{
"version": "v1",
"created": "Tue, 9 May 2023 11:44:44 GMT"
},
{
"version": "v2",
"created": "Wed, 14 Jun 2023 07:04:39 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Peer",
"Marco",
""
],
[
"Kleber",
"Florian",
""
],
[
"Sablatnig",
"Robert",
""
]
] | TITLE: Towards Writer Retrieval for Historical Datasets
ABSTRACT: This paper presents an unsupervised approach for writer retrieval based on
clustering SIFT descriptors detected at keypoint locations resulting in
pseudo-cluster labels. With those cluster labels, a residual network followed
by our proposed NetRVLAD, an encoding layer with reduced complexity compared to
NetVLAD, is trained on 32x32 patches at keypoint locations. Additionally, we
suggest a graph-based reranking algorithm called SGR to exploit similarities of
the page embeddings to boost the retrieval performance. Our approach is
evaluated on two historical datasets (Historical-WI and HisIR19). We include an
evaluation of different backbones and NetRVLAD. It competes with related work
on historical datasets without using explicit encodings. We set a new
State-of-the-art on both datasets by applying our reranking scheme and show
that our approach achieves comparable performance on a modern dataset as well.
|
2305.06500 | Dongxu Li | Wenliang Dai, Junnan Li, Dongxu Li, Anthony Meng Huat Tiong, Junqi
Zhao, Weisheng Wang, Boyang Li, Pascale Fung, Steven Hoi | InstructBLIP: Towards General-purpose Vision-Language Models with
Instruction Tuning | preprint | null | null | null | cs.CV cs.LG | http://creativecommons.org/licenses/by/4.0/ | Large-scale pre-training and instruction tuning have been successful at
creating general-purpose language models with broad competence. However,
building general-purpose vision-language models is challenging due to the rich
input distributions and task diversity resulting from the additional visual
input. Although vision-language pretraining has been widely studied,
vision-language instruction tuning remains under-explored. In this paper, we
conduct a systematic and comprehensive study on vision-language instruction
tuning based on the pretrained BLIP-2 models. We gather 26 publicly available
datasets, covering a wide variety of tasks and capabilities, and transform them
into instruction tuning format. Additionally, we introduce an instruction-aware
Query Transformer, which extracts informative features tailored to the given
instruction. Trained on 13 held-in datasets, InstructBLIP attains
state-of-the-art zero-shot performance across all 13 held-out datasets,
substantially outperforming BLIP-2 and larger Flamingo models. Our models also
lead to state-of-the-art performance when finetuned on individual downstream
tasks (e.g., 90.7% accuracy on ScienceQA questions with image contexts).
Furthermore, we qualitatively demonstrate the advantages of InstructBLIP over
concurrent multimodal models. All InstructBLIP models are open-sourced at
https://github.com/salesforce/LAVIS/tree/main/projects/instructblip.
| [
{
"version": "v1",
"created": "Thu, 11 May 2023 00:38:10 GMT"
},
{
"version": "v2",
"created": "Thu, 15 Jun 2023 08:00:18 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Dai",
"Wenliang",
""
],
[
"Li",
"Junnan",
""
],
[
"Li",
"Dongxu",
""
],
[
"Tiong",
"Anthony Meng Huat",
""
],
[
"Zhao",
"Junqi",
""
],
[
"Wang",
"Weisheng",
""
],
[
"Li",
"Boyang",
""
],
[
"Fung",
"Pascale",
""
],
[
"Hoi",
"Steven",
""
]
] | TITLE: InstructBLIP: Towards General-purpose Vision-Language Models with
Instruction Tuning
ABSTRACT: Large-scale pre-training and instruction tuning have been successful at
creating general-purpose language models with broad competence. However,
building general-purpose vision-language models is challenging due to the rich
input distributions and task diversity resulting from the additional visual
input. Although vision-language pretraining has been widely studied,
vision-language instruction tuning remains under-explored. In this paper, we
conduct a systematic and comprehensive study on vision-language instruction
tuning based on the pretrained BLIP-2 models. We gather 26 publicly available
datasets, covering a wide variety of tasks and capabilities, and transform them
into instruction tuning format. Additionally, we introduce an instruction-aware
Query Transformer, which extracts informative features tailored to the given
instruction. Trained on 13 held-in datasets, InstructBLIP attains
state-of-the-art zero-shot performance across all 13 held-out datasets,
substantially outperforming BLIP-2 and larger Flamingo models. Our models also
lead to state-of-the-art performance when finetuned on individual downstream
tasks (e.g., 90.7% accuracy on ScienceQA questions with image contexts).
Furthermore, we qualitatively demonstrate the advantages of InstructBLIP over
concurrent multimodal models. All InstructBLIP models are open-sourced at
https://github.com/salesforce/LAVIS/tree/main/projects/instructblip.
|
2305.07498 | Jianfeng Kuang | Jianfeng Kuang, Wei Hua, Dingkang Liang, Mingkun Yang, Deqiang Jiang,
Bo Ren, and Xiang Bai | Visual Information Extraction in the Wild: Practical Dataset and
End-to-end Solution | 15 pages, 6 figures, ICDAR2023 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Visual information extraction (VIE), which aims to simultaneously perform OCR
and information extraction in a unified framework, has drawn increasing
attention due to its essential role in various applications like understanding
receipts, goods, and traffic signs. However, as existing benchmark datasets for
VIE mainly consist of document images without the adequate diversity of layout
structures, background disturbs, and entity categories, they cannot fully
reveal the challenges of real-world applications. In this paper, we propose a
large-scale dataset consisting of camera images for VIE, which contains not
only the larger variance of layout, backgrounds, and fonts but also much more
types of entities. Besides, we propose a novel framework for end-to-end VIE
that combines the stages of OCR and information extraction in an end-to-end
learning fashion. Different from the previous end-to-end approaches that
directly adopt OCR features as the input of an information extraction module,
we propose to use contrastive learning to narrow the semantic gap caused by the
difference between the tasks of OCR and information extraction. We evaluate the
existing end-to-end methods for VIE on the proposed dataset and observe that
the performance of these methods has a distinguishable drop from SROIE (a
widely used English dataset) to our proposed dataset due to the larger variance
of layout and entities. These results demonstrate our dataset is more practical
for promoting advanced VIE algorithms. In addition, experiments demonstrate
that the proposed VIE method consistently achieves the obvious performance
gains on the proposed and SROIE datasets.
| [
{
"version": "v1",
"created": "Fri, 12 May 2023 14:11:47 GMT"
},
{
"version": "v2",
"created": "Thu, 15 Jun 2023 03:31:12 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Kuang",
"Jianfeng",
""
],
[
"Hua",
"Wei",
""
],
[
"Liang",
"Dingkang",
""
],
[
"Yang",
"Mingkun",
""
],
[
"Jiang",
"Deqiang",
""
],
[
"Ren",
"Bo",
""
],
[
"Bai",
"Xiang",
""
]
] | TITLE: Visual Information Extraction in the Wild: Practical Dataset and
End-to-end Solution
ABSTRACT: Visual information extraction (VIE), which aims to simultaneously perform OCR
and information extraction in a unified framework, has drawn increasing
attention due to its essential role in various applications like understanding
receipts, goods, and traffic signs. However, as existing benchmark datasets for
VIE mainly consist of document images without the adequate diversity of layout
structures, background disturbs, and entity categories, they cannot fully
reveal the challenges of real-world applications. In this paper, we propose a
large-scale dataset consisting of camera images for VIE, which contains not
only the larger variance of layout, backgrounds, and fonts but also much more
types of entities. Besides, we propose a novel framework for end-to-end VIE
that combines the stages of OCR and information extraction in an end-to-end
learning fashion. Different from the previous end-to-end approaches that
directly adopt OCR features as the input of an information extraction module,
we propose to use contrastive learning to narrow the semantic gap caused by the
difference between the tasks of OCR and information extraction. We evaluate the
existing end-to-end methods for VIE on the proposed dataset and observe that
the performance of these methods has a distinguishable drop from SROIE (a
widely used English dataset) to our proposed dataset due to the larger variance
of layout and entities. These results demonstrate our dataset is more practical
for promoting advanced VIE algorithms. In addition, experiments demonstrate
that the proposed VIE method consistently achieves the obvious performance
gains on the proposed and SROIE datasets.
|
2305.07648 | Hanchen Xie | Hanchen Xie, Jiageng Zhu, Mahyar Khayatkhoei, Jiazhi Li, Mohamed E.
Hussein, Wael AbdAlmageed | A Critical View of Vision-Based Long-Term Dynamics Prediction Under
Environment Misalignment | 14 pages, 5 figures, 10 tables. Accepted to ICML 2023 | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Dynamics prediction, which is the problem of predicting future states of
scene objects based on current and prior states, is drawing increasing
attention as an instance of learning physics. To solve this problem, Region
Proposal Convolutional Interaction Network (RPCIN), a vision-based model, was
proposed and achieved state-of-the-art performance in long-term prediction.
RPCIN only takes raw images and simple object descriptions, such as the
bounding box and segmentation mask of each object, as input. However, despite
its success, the model's capability can be compromised under conditions of
environment misalignment. In this paper, we investigate two challenging
conditions for environment misalignment: Cross-Domain and Cross-Context by
proposing four datasets that are designed for these challenges: SimB-Border,
SimB-Split, BlenB-Border, and BlenB-Split. The datasets cover two domains and
two contexts. Using RPCIN as a probe, experiments conducted on the combinations
of the proposed datasets reveal potential weaknesses of the vision-based
long-term dynamics prediction model. Furthermore, we propose a promising
direction to mitigate the Cross-Domain challenge and provide concrete evidence
supporting such a direction, which provides dramatic alleviation of the
challenge on the proposed datasets.
| [
{
"version": "v1",
"created": "Fri, 12 May 2023 17:58:24 GMT"
},
{
"version": "v2",
"created": "Tue, 13 Jun 2023 19:36:17 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Xie",
"Hanchen",
""
],
[
"Zhu",
"Jiageng",
""
],
[
"Khayatkhoei",
"Mahyar",
""
],
[
"Li",
"Jiazhi",
""
],
[
"Hussein",
"Mohamed E.",
""
],
[
"AbdAlmageed",
"Wael",
""
]
] | TITLE: A Critical View of Vision-Based Long-Term Dynamics Prediction Under
Environment Misalignment
ABSTRACT: Dynamics prediction, which is the problem of predicting future states of
scene objects based on current and prior states, is drawing increasing
attention as an instance of learning physics. To solve this problem, Region
Proposal Convolutional Interaction Network (RPCIN), a vision-based model, was
proposed and achieved state-of-the-art performance in long-term prediction.
RPCIN only takes raw images and simple object descriptions, such as the
bounding box and segmentation mask of each object, as input. However, despite
its success, the model's capability can be compromised under conditions of
environment misalignment. In this paper, we investigate two challenging
conditions for environment misalignment: Cross-Domain and Cross-Context by
proposing four datasets that are designed for these challenges: SimB-Border,
SimB-Split, BlenB-Border, and BlenB-Split. The datasets cover two domains and
two contexts. Using RPCIN as a probe, experiments conducted on the combinations
of the proposed datasets reveal potential weaknesses of the vision-based
long-term dynamics prediction model. Furthermore, we propose a promising
direction to mitigate the Cross-Domain challenge and provide concrete evidence
supporting such a direction, which provides dramatic alleviation of the
challenge on the proposed datasets.
|
2305.08989 | Yang Liu | Yang Liu, Maxence Boels, Luis C. Garcia-Peraza-Herrera, Tom
Vercauteren, Prokar Dasgupta, Alejandro Granados and Sebastien Ourselin | LoViT: Long Video Transformer for Surgical Phase Recognition | Code link: https://github.com/MRUIL/LoViT | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by/4.0/ | Online surgical phase recognition plays a significant role towards building
contextual tools that could quantify performance and oversee the execution of
surgical workflows. Current approaches are limited since they train spatial
feature extractors using frame-level supervision that could lead to incorrect
predictions due to similar frames appearing at different phases, and poorly
fuse local and global features due to computational constraints which can
affect the analysis of long videos commonly encountered in surgical
interventions. In this paper, we present a two-stage method, called Long Video
Transformer (LoViT) for fusing short- and long-term temporal information that
combines a temporally-rich spatial feature extractor and a multi-scale temporal
aggregator consisting of two cascaded L-Trans modules based on self-attention,
followed by a G-Informer module based on ProbSparse self-attention for
processing global temporal information. The multi-scale temporal head then
combines local and global features and classifies surgical phases using phase
transition-aware supervision. Our approach outperforms state-of-the-art methods
on the Cholec80 and AutoLaparo datasets consistently. Compared to Trans-SVNet,
LoViT achieves a 2.4 pp (percentage point) improvement in video-level accuracy
on Cholec80 and a 3.1 pp improvement on AutoLaparo. Moreover, it achieves a 5.3
pp improvement in phase-level Jaccard on AutoLaparo and a 1.55 pp improvement
on Cholec80. Our results demonstrate the effectiveness of our approach in
achieving state-of-the-art performance of surgical phase recognition on two
datasets of different surgical procedures and temporal sequencing
characteristics whilst introducing mechanisms that cope with long videos.
| [
{
"version": "v1",
"created": "Mon, 15 May 2023 20:06:14 GMT"
},
{
"version": "v2",
"created": "Thu, 18 May 2023 12:42:44 GMT"
},
{
"version": "v3",
"created": "Wed, 14 Jun 2023 16:40:08 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Liu",
"Yang",
""
],
[
"Boels",
"Maxence",
""
],
[
"Garcia-Peraza-Herrera",
"Luis C.",
""
],
[
"Vercauteren",
"Tom",
""
],
[
"Dasgupta",
"Prokar",
""
],
[
"Granados",
"Alejandro",
""
],
[
"Ourselin",
"Sebastien",
""
]
] | TITLE: LoViT: Long Video Transformer for Surgical Phase Recognition
ABSTRACT: Online surgical phase recognition plays a significant role towards building
contextual tools that could quantify performance and oversee the execution of
surgical workflows. Current approaches are limited since they train spatial
feature extractors using frame-level supervision that could lead to incorrect
predictions due to similar frames appearing at different phases, and poorly
fuse local and global features due to computational constraints which can
affect the analysis of long videos commonly encountered in surgical
interventions. In this paper, we present a two-stage method, called Long Video
Transformer (LoViT) for fusing short- and long-term temporal information that
combines a temporally-rich spatial feature extractor and a multi-scale temporal
aggregator consisting of two cascaded L-Trans modules based on self-attention,
followed by a G-Informer module based on ProbSparse self-attention for
processing global temporal information. The multi-scale temporal head then
combines local and global features and classifies surgical phases using phase
transition-aware supervision. Our approach outperforms state-of-the-art methods
on the Cholec80 and AutoLaparo datasets consistently. Compared to Trans-SVNet,
LoViT achieves a 2.4 pp (percentage point) improvement in video-level accuracy
on Cholec80 and a 3.1 pp improvement on AutoLaparo. Moreover, it achieves a 5.3
pp improvement in phase-level Jaccard on AutoLaparo and a 1.55 pp improvement
on Cholec80. Our results demonstrate the effectiveness of our approach in
achieving state-of-the-art performance of surgical phase recognition on two
datasets of different surgical procedures and temporal sequencing
characteristics whilst introducing mechanisms that cope with long videos.
|
2305.09789 | Xiaodan Xing | Xiaodan Xing, Yang Nan, Federico Felder, Simon Walsh and Guang Yang | The Beauty or the Beast: Which Aspect of Synthetic Medical Images
Deserves Our Focus? | CBMS 2023 | null | null | null | eess.IV cs.CV | http://creativecommons.org/licenses/by/4.0/ | Training medical AI algorithms requires large volumes of accurately labeled
datasets, which are difficult to obtain in the real world. Synthetic images
generated from deep generative models can help alleviate the data scarcity
problem, but their effectiveness relies on their fidelity to real-world images.
Typically, researchers select synthesis models based on image quality
measurements, prioritizing synthetic images that appear realistic. However, our
empirical analysis shows that high-fidelity and visually appealing synthetic
images are not necessarily superior. In fact, we present a case where
low-fidelity synthetic images outperformed their high-fidelity counterparts in
downstream tasks. Our findings highlight the importance of comprehensive
analysis before incorporating synthetic data into real-world applications. We
hope our results will raise awareness among the research community of the value
of low-fidelity synthetic images in medical AI algorithm training.
| [
{
"version": "v1",
"created": "Wed, 3 May 2023 09:09:54 GMT"
},
{
"version": "v2",
"created": "Wed, 14 Jun 2023 14:39:17 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Xing",
"Xiaodan",
""
],
[
"Nan",
"Yang",
""
],
[
"Felder",
"Federico",
""
],
[
"Walsh",
"Simon",
""
],
[
"Yang",
"Guang",
""
]
] | TITLE: The Beauty or the Beast: Which Aspect of Synthetic Medical Images
Deserves Our Focus?
ABSTRACT: Training medical AI algorithms requires large volumes of accurately labeled
datasets, which are difficult to obtain in the real world. Synthetic images
generated from deep generative models can help alleviate the data scarcity
problem, but their effectiveness relies on their fidelity to real-world images.
Typically, researchers select synthesis models based on image quality
measurements, prioritizing synthetic images that appear realistic. However, our
empirical analysis shows that high-fidelity and visually appealing synthetic
images are not necessarily superior. In fact, we present a case where
low-fidelity synthetic images outperformed their high-fidelity counterparts in
downstream tasks. Our findings highlight the importance of comprehensive
analysis before incorporating synthetic data into real-world applications. We
hope our results will raise awareness among the research community of the value
of low-fidelity synthetic images in medical AI algorithm training.
|
2305.11015 | Daniel Hausmann | Oliver G\"orlitz, Daniel Hausmann, Merlin Humml, Dirk Pattinson, Simon
Prucker, Lutz Schr\"oder | COOL 2 -- A Generic Reasoner for Modal Fixpoint Logics | Final version (corrected slight mistake in Rabin-type formula series) | null | null | null | cs.LO cs.FL | http://creativecommons.org/licenses/by/4.0/ | There is a wide range of modal logics whose semantics goes beyond relational
structures, and instead involves, e.g., probabilities, multi-player games,
weights, or neighbourhood structures. Coalgebraic logic serves as a unifying
semantic and algorithmic framework for such logics. It provides uniform
reasoning algorithms that are easily instantiated to particular, concretely
given logics. The COOL 2 reasoner provides an implementation of such generic
algorithms for coalgebraic modal fixpoint logics. As concrete instances, we
obtain in particular reasoners for the aconjunctive and alternation-free
fragments of the graded $\mu$-calculus and the alternating-time $\mu$-calculus.
We evaluate the tool on standard benchmark sets for fixpoint-free graded modal
logic and alternating-time temporal logic (ATL), as well as on a dedicated set
of benchmarks for the graded $\mu$-calculus.
| [
{
"version": "v1",
"created": "Thu, 18 May 2023 14:48:38 GMT"
},
{
"version": "v2",
"created": "Wed, 24 May 2023 12:46:42 GMT"
},
{
"version": "v3",
"created": "Fri, 26 May 2023 07:12:42 GMT"
},
{
"version": "v4",
"created": "Thu, 15 Jun 2023 11:41:59 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Görlitz",
"Oliver",
""
],
[
"Hausmann",
"Daniel",
""
],
[
"Humml",
"Merlin",
""
],
[
"Pattinson",
"Dirk",
""
],
[
"Prucker",
"Simon",
""
],
[
"Schröder",
"Lutz",
""
]
] | TITLE: COOL 2 -- A Generic Reasoner for Modal Fixpoint Logics
ABSTRACT: There is a wide range of modal logics whose semantics goes beyond relational
structures, and instead involves, e.g., probabilities, multi-player games,
weights, or neighbourhood structures. Coalgebraic logic serves as a unifying
semantic and algorithmic framework for such logics. It provides uniform
reasoning algorithms that are easily instantiated to particular, concretely
given logics. The COOL 2 reasoner provides an implementation of such generic
algorithms for coalgebraic modal fixpoint logics. As concrete instances, we
obtain in particular reasoners for the aconjunctive and alternation-free
fragments of the graded $\mu$-calculus and the alternating-time $\mu$-calculus.
We evaluate the tool on standard benchmark sets for fixpoint-free graded modal
logic and alternating-time temporal logic (ATL), as well as on a dedicated set
of benchmarks for the graded $\mu$-calculus.
|
2305.13453 | Ali Owfi | Ali Owfi, ChunChih Lin, Linke Guo, Fatemeh Afghah, Jonathan Ashdown,
Kurt Turck | A Meta-learning based Generalizable Indoor Localization Model using
Channel State Information | 6 pages, 6 figures, submitted to IEEE GLOBECOM 2023 Added
Distribution Statement in first page footnote | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by/4.0/ | Indoor localization has gained significant attention in recent years due to
its various applications in smart homes, industrial automation, and healthcare,
especially since more people rely on their wireless devices for location-based
services. Deep learning-based solutions have shown promising results in
accurately estimating the position of wireless devices in indoor environments
using wireless parameters such as Channel State Information (CSI) and Received
Signal Strength Indicator (RSSI). However, despite the success of deep
learning-based approaches in achieving high localization accuracy, these models
suffer from a lack of generalizability and can not be readily-deployed to new
environments or operate in dynamic environments without retraining. In this
paper, we propose meta-learning-based localization models to address the lack
of generalizability that persists in conventionally trained DL-based
localization models. Furthermore, since meta-learning algorithms require
diverse datasets from several different scenarios, which can be hard to collect
in the context of localization, we design and propose a new meta-learning
algorithm, TB-MAML (Task Biased Model Agnostic Meta Learning), intended to
further improve generalizability when the dataset is limited. Lastly, we
evaluate the performance of TB-MAML-based localization against conventionally
trained localization models and localization done using other meta-learning
algorithms.
| [
{
"version": "v1",
"created": "Mon, 22 May 2023 19:54:59 GMT"
},
{
"version": "v2",
"created": "Tue, 13 Jun 2023 18:29:25 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Owfi",
"Ali",
""
],
[
"Lin",
"ChunChih",
""
],
[
"Guo",
"Linke",
""
],
[
"Afghah",
"Fatemeh",
""
],
[
"Ashdown",
"Jonathan",
""
],
[
"Turck",
"Kurt",
""
]
] | TITLE: A Meta-learning based Generalizable Indoor Localization Model using
Channel State Information
ABSTRACT: Indoor localization has gained significant attention in recent years due to
its various applications in smart homes, industrial automation, and healthcare,
especially since more people rely on their wireless devices for location-based
services. Deep learning-based solutions have shown promising results in
accurately estimating the position of wireless devices in indoor environments
using wireless parameters such as Channel State Information (CSI) and Received
Signal Strength Indicator (RSSI). However, despite the success of deep
learning-based approaches in achieving high localization accuracy, these models
suffer from a lack of generalizability and can not be readily-deployed to new
environments or operate in dynamic environments without retraining. In this
paper, we propose meta-learning-based localization models to address the lack
of generalizability that persists in conventionally trained DL-based
localization models. Furthermore, since meta-learning algorithms require
diverse datasets from several different scenarios, which can be hard to collect
in the context of localization, we design and propose a new meta-learning
algorithm, TB-MAML (Task Biased Model Agnostic Meta Learning), intended to
further improve generalizability when the dataset is limited. Lastly, we
evaluate the performance of TB-MAML-based localization against conventionally
trained localization models and localization done using other meta-learning
algorithms.
|
2305.14293 | Chenxi Whitehouse | Chenxi Whitehouse, Clara Vania, Alham Fikri Aji, Christos
Christodoulopoulos, Andrea Pierleoni | WebIE: Faithful and Robust Information Extraction on the Web | ACL 2023 Main Conference | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Extracting structured and grounded fact triples from raw text is a
fundamental task in Information Extraction (IE). Existing IE datasets are
typically collected from Wikipedia articles, using hyperlinks to link entities
to the Wikidata knowledge base. However, models trained only on Wikipedia have
limitations when applied to web domains, which often contain noisy text or text
that does not have any factual information. We present WebIE, the first
large-scale, entity-linked closed IE dataset consisting of 1.6M sentences
automatically collected from the English Common Crawl corpus. WebIE also
includes negative examples, i.e. sentences without fact triples, to better
reflect the data on the web. We annotate ~21K triples from WebIE through
crowdsourcing and introduce mWebIE, a translation of the annotated set in four
other languages: French, Spanish, Portuguese, and Hindi. We evaluate the
in-domain, out-of-domain, and zero-shot cross-lingual performance of generative
IE models and find models trained on WebIE show better generalisability. We
also propose three training strategies that use entity linking as an auxiliary
task. Our experiments show that adding Entity-Linking objectives improves the
faithfulness of our generative IE models.
| [
{
"version": "v1",
"created": "Tue, 23 May 2023 17:37:53 GMT"
},
{
"version": "v2",
"created": "Thu, 15 Jun 2023 13:51:36 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Whitehouse",
"Chenxi",
""
],
[
"Vania",
"Clara",
""
],
[
"Aji",
"Alham Fikri",
""
],
[
"Christodoulopoulos",
"Christos",
""
],
[
"Pierleoni",
"Andrea",
""
]
] | TITLE: WebIE: Faithful and Robust Information Extraction on the Web
ABSTRACT: Extracting structured and grounded fact triples from raw text is a
fundamental task in Information Extraction (IE). Existing IE datasets are
typically collected from Wikipedia articles, using hyperlinks to link entities
to the Wikidata knowledge base. However, models trained only on Wikipedia have
limitations when applied to web domains, which often contain noisy text or text
that does not have any factual information. We present WebIE, the first
large-scale, entity-linked closed IE dataset consisting of 1.6M sentences
automatically collected from the English Common Crawl corpus. WebIE also
includes negative examples, i.e. sentences without fact triples, to better
reflect the data on the web. We annotate ~21K triples from WebIE through
crowdsourcing and introduce mWebIE, a translation of the annotated set in four
other languages: French, Spanish, Portuguese, and Hindi. We evaluate the
in-domain, out-of-domain, and zero-shot cross-lingual performance of generative
IE models and find models trained on WebIE show better generalisability. We
also propose three training strategies that use entity linking as an auxiliary
task. Our experiments show that adding Entity-Linking objectives improves the
faithfulness of our generative IE models.
|
2305.16133 | Cheng Zhang | Zhiyu Tan, Zichao Dong, Cheng Zhang, Weikun Zhang, Hang Ji, Hao Li | OVO: Open-Vocabulary Occupancy | null | null | null | null | cs.CV cs.AI cs.LG cs.RO | http://creativecommons.org/licenses/by/4.0/ | Semantic occupancy prediction aims to infer dense geometry and semantics of
surroundings for an autonomous agent to operate safely in the 3D environment.
Existing occupancy prediction methods are almost entirely trained on
human-annotated volumetric data. Although of high quality, the generation of
such 3D annotations is laborious and costly, restricting them to a few specific
object categories in the training dataset. To address this limitation, this
paper proposes Open Vocabulary Occupancy (OVO), a novel approach that allows
semantic occupancy prediction of arbitrary classes but without the need for 3D
annotations during training. Keys to our approach are (1) knowledge
distillation from a pre-trained 2D open-vocabulary segmentation model to the 3D
occupancy network, and (2) pixel-voxel filtering for high-quality training data
generation. The resulting framework is simple, compact, and compatible with
most state-of-the-art semantic occupancy prediction models. On NYUv2 and
SemanticKITTI datasets, OVO achieves competitive performance compared to
supervised semantic occupancy prediction approaches. Furthermore, we conduct
extensive analyses and ablation studies to offer insights into the design of
the proposed framework. Our code is publicly available at
https://github.com/dzcgaara/OVO.
| [
{
"version": "v1",
"created": "Thu, 25 May 2023 15:07:25 GMT"
},
{
"version": "v2",
"created": "Wed, 14 Jun 2023 17:30:54 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Tan",
"Zhiyu",
""
],
[
"Dong",
"Zichao",
""
],
[
"Zhang",
"Cheng",
""
],
[
"Zhang",
"Weikun",
""
],
[
"Ji",
"Hang",
""
],
[
"Li",
"Hao",
""
]
] | TITLE: OVO: Open-Vocabulary Occupancy
ABSTRACT: Semantic occupancy prediction aims to infer dense geometry and semantics of
surroundings for an autonomous agent to operate safely in the 3D environment.
Existing occupancy prediction methods are almost entirely trained on
human-annotated volumetric data. Although of high quality, the generation of
such 3D annotations is laborious and costly, restricting them to a few specific
object categories in the training dataset. To address this limitation, this
paper proposes Open Vocabulary Occupancy (OVO), a novel approach that allows
semantic occupancy prediction of arbitrary classes but without the need for 3D
annotations during training. Keys to our approach are (1) knowledge
distillation from a pre-trained 2D open-vocabulary segmentation model to the 3D
occupancy network, and (2) pixel-voxel filtering for high-quality training data
generation. The resulting framework is simple, compact, and compatible with
most state-of-the-art semantic occupancy prediction models. On NYUv2 and
SemanticKITTI datasets, OVO achieves competitive performance compared to
supervised semantic occupancy prediction approaches. Furthermore, we conduct
extensive analyses and ablation studies to offer insights into the design of
the proposed framework. Our code is publicly available at
https://github.com/dzcgaara/OVO.
|
2305.16663 | Aiwei Liu | Xuming Hu, Aiwei Liu, Zeqi Tan, Xin Zhang, Chenwei Zhang, Irwin King,
Philip S. Yu | GDA: Generative Data Augmentation Techniques for Relation Extraction
Tasks | Accepted to ACL 2023 (Findings), Long Paper, 12 pages | ACL 2023 | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Relation extraction (RE) tasks show promising performance in extracting
relations from two entities mentioned in sentences, given sufficient
annotations available during training. Such annotations would be
labor-intensive to obtain in practice. Existing work adopts data augmentation
techniques to generate pseudo-annotated sentences beyond limited annotations.
These techniques neither preserve the semantic consistency of the original
sentences when rule-based augmentations are adopted, nor preserve the syntax
structure of sentences when expressing relations using seq2seq models,
resulting in less diverse augmentations. In this work, we propose a dedicated
augmentation technique for relational texts, named GDA, which uses two
complementary modules to preserve both semantic consistency and syntax
structures. We adopt a generative formulation and design a multi-tasking
solution to achieve synergies. Furthermore, GDA adopts entity hints as the
prior knowledge of the generative model to augment diverse sentences.
Experimental results in three datasets under a low-resource setting showed that
GDA could bring {\em 2.0\%} F1 improvements compared with no augmentation
technique. Source code and data are available.
| [
{
"version": "v1",
"created": "Fri, 26 May 2023 06:21:01 GMT"
},
{
"version": "v2",
"created": "Thu, 15 Jun 2023 02:43:12 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Hu",
"Xuming",
""
],
[
"Liu",
"Aiwei",
""
],
[
"Tan",
"Zeqi",
""
],
[
"Zhang",
"Xin",
""
],
[
"Zhang",
"Chenwei",
""
],
[
"King",
"Irwin",
""
],
[
"Yu",
"Philip S.",
""
]
] | TITLE: GDA: Generative Data Augmentation Techniques for Relation Extraction
Tasks
ABSTRACT: Relation extraction (RE) tasks show promising performance in extracting
relations from two entities mentioned in sentences, given sufficient
annotations available during training. Such annotations would be
labor-intensive to obtain in practice. Existing work adopts data augmentation
techniques to generate pseudo-annotated sentences beyond limited annotations.
These techniques neither preserve the semantic consistency of the original
sentences when rule-based augmentations are adopted, nor preserve the syntax
structure of sentences when expressing relations using seq2seq models,
resulting in less diverse augmentations. In this work, we propose a dedicated
augmentation technique for relational texts, named GDA, which uses two
complementary modules to preserve both semantic consistency and syntax
structures. We adopt a generative formulation and design a multi-tasking
solution to achieve synergies. Furthermore, GDA adopts entity hints as the
prior knowledge of the generative model to augment diverse sentences.
Experimental results in three datasets under a low-resource setting showed that
GDA could bring {\em 2.0\%} F1 improvements compared with no augmentation
technique. Source code and data are available.
|
2305.17940 | Qingsheng Wang | Qingsheng Wang, Lingqiao Liu, Chenchen Jing, Hao Chen, Guoqiang Liang,
Peng Wang, Chunhua Shen | Learning Conditional Attributes for Compositional Zero-Shot Learning | 10 pages, 4 figures, accepted in CVPR2023 | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Compositional Zero-Shot Learning (CZSL) aims to train models to recognize
novel compositional concepts based on learned concepts such as attribute-object
combinations. One of the challenges is to model attributes interacted with
different objects, e.g., the attribute ``wet" in ``wet apple" and ``wet cat" is
different. As a solution, we provide analysis and argue that attributes are
conditioned on the recognized object and input image and explore learning
conditional attribute embeddings by a proposed attribute learning framework
containing an attribute hyper learner and an attribute base learner. By
encoding conditional attributes, our model enables to generate flexible
attribute embeddings for generalization from seen to unseen compositions.
Experiments on CZSL benchmarks, including the more challenging C-GQA dataset,
demonstrate better performances compared with other state-of-the-art approaches
and validate the importance of learning conditional attributes. Code is
available at https://github.com/wqshmzh/CANet-CZSL
| [
{
"version": "v1",
"created": "Mon, 29 May 2023 08:04:05 GMT"
},
{
"version": "v2",
"created": "Wed, 14 Jun 2023 13:55:00 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Wang",
"Qingsheng",
""
],
[
"Liu",
"Lingqiao",
""
],
[
"Jing",
"Chenchen",
""
],
[
"Chen",
"Hao",
""
],
[
"Liang",
"Guoqiang",
""
],
[
"Wang",
"Peng",
""
],
[
"Shen",
"Chunhua",
""
]
] | TITLE: Learning Conditional Attributes for Compositional Zero-Shot Learning
ABSTRACT: Compositional Zero-Shot Learning (CZSL) aims to train models to recognize
novel compositional concepts based on learned concepts such as attribute-object
combinations. One of the challenges is to model attributes interacted with
different objects, e.g., the attribute ``wet" in ``wet apple" and ``wet cat" is
different. As a solution, we provide analysis and argue that attributes are
conditioned on the recognized object and input image and explore learning
conditional attribute embeddings by a proposed attribute learning framework
containing an attribute hyper learner and an attribute base learner. By
encoding conditional attributes, our model enables to generate flexible
attribute embeddings for generalization from seen to unseen compositions.
Experiments on CZSL benchmarks, including the more challenging C-GQA dataset,
demonstrate better performances compared with other state-of-the-art approaches
and validate the importance of learning conditional attributes. Code is
available at https://github.com/wqshmzh/CANet-CZSL
|
2305.18015 | David Jaime Tena Cucala | David Tena Cucala, Bernardo Cuenca Grau, Boris Motik, Egor V. Kostylev | On the Correspondence Between Monotonic Max-Sum GNNs and Datalog | null | null | null | null | cs.AI | http://creativecommons.org/licenses/by/4.0/ | Although there has been significant interest in applying machine learning
techniques to structured data, the expressivity (i.e., a description of what
can be learned) of such techniques is still poorly understood. In this paper,
we study data transformations based on graph neural networks (GNNs). First, we
note that the choice of how a dataset is encoded into a numeric form
processable by a GNN can obscure the characterisation of a model's
expressivity, and we argue that a canonical encoding provides an appropriate
basis. Second, we study the expressivity of monotonic max-sum GNNs, which cover
a subclass of GNNs with max and sum aggregation functions. We show that, for
each such GNN, one can compute a Datalog program such that applying the GNN to
any dataset produces the same facts as a single round of application of the
program's rules to the dataset. Monotonic max-sum GNNs can sum an unbounded
number of feature vectors which can result in arbitrarily large feature values,
whereas rule application requires only a bounded number of constants. Hence,
our result shows that the unbounded summation of monotonic max-sum GNNs does
not increase their expressive power. Third, we sharpen our result to the
subclass of monotonic max GNNs, which use only the max aggregation function,
and identify a corresponding class of Datalog programs.
| [
{
"version": "v1",
"created": "Mon, 29 May 2023 11:13:38 GMT"
},
{
"version": "v2",
"created": "Wed, 7 Jun 2023 15:06:33 GMT"
},
{
"version": "v3",
"created": "Thu, 15 Jun 2023 09:22:01 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Cucala",
"David Tena",
""
],
[
"Grau",
"Bernardo Cuenca",
""
],
[
"Motik",
"Boris",
""
],
[
"Kostylev",
"Egor V.",
""
]
] | TITLE: On the Correspondence Between Monotonic Max-Sum GNNs and Datalog
ABSTRACT: Although there has been significant interest in applying machine learning
techniques to structured data, the expressivity (i.e., a description of what
can be learned) of such techniques is still poorly understood. In this paper,
we study data transformations based on graph neural networks (GNNs). First, we
note that the choice of how a dataset is encoded into a numeric form
processable by a GNN can obscure the characterisation of a model's
expressivity, and we argue that a canonical encoding provides an appropriate
basis. Second, we study the expressivity of monotonic max-sum GNNs, which cover
a subclass of GNNs with max and sum aggregation functions. We show that, for
each such GNN, one can compute a Datalog program such that applying the GNN to
any dataset produces the same facts as a single round of application of the
program's rules to the dataset. Monotonic max-sum GNNs can sum an unbounded
number of feature vectors which can result in arbitrarily large feature values,
whereas rule application requires only a bounded number of constants. Hence,
our result shows that the unbounded summation of monotonic max-sum GNNs does
not increase their expressive power. Third, we sharpen our result to the
subclass of monotonic max GNNs, which use only the max aggregation function,
and identify a corresponding class of Datalog programs.
|
2305.18897 | Lucas Mourot | Lucas Mourot, Ludovic Hoyet, Fran\c{c}ois Le Clerc and Pierre Hellier | HuMoT: Human Motion Representation using Topology-Agnostic Transformers
for Character Animation Retargeting | 17 pages, 12 figures, 5 tables | null | null | null | cs.GR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Motion retargeting is the long-standing problem in character animation that
consists in transferring and adapting the motion of a source character to
another target character. A typical application is the creation of motion
sequences from off-the-shelf motions by transferring them onto new characters.
Motion retargeting is also promising to increase interoperability of existing
animation systems and motion databases, as they often differ in the structure
of the skeleton(s) considered. Moreover, since the goal of motion retargeting
is to abstract and transfer motion dynamics, effective solutions might provide
expressive and powerful human motion models in which operations such as
cleaning or editing are easier. In this article, we present a novel neural
network architecture for retargeting that extracts an abstract representation
of human motion agnostic to skeleton topology and morphology. Based on
transformers, our model is able to encode and decode motion sequences with
variable morphology and topology -- extending the current scope of retargeting
-- while supporting skeleton topologies not seen during the training phase.
More specifically, our model is structured as an autoencoder, and encoding and
decoding are separately conditioned on skeleton templates to extract and
control morphology and topology. Beyond motion retargeting, our model has many
applications since our abstract representation is a convenient space to embed
motion data from different sources. It may potentially be benefical to a number
of data-driven methods, allowing them to combine scarce specialised motion
datasets (e.g. with style or contact annotations) and larger general motion
datasets, for improved performance and generalisation ability. Moreover, we
show that our model can be useful for other applications beyond retargeting,
including motion denoising and joint upsampling.
| [
{
"version": "v1",
"created": "Tue, 30 May 2023 09:52:33 GMT"
},
{
"version": "v2",
"created": "Wed, 31 May 2023 13:36:26 GMT"
},
{
"version": "v3",
"created": "Thu, 15 Jun 2023 08:33:57 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Mourot",
"Lucas",
""
],
[
"Hoyet",
"Ludovic",
""
],
[
"Clerc",
"François Le",
""
],
[
"Hellier",
"Pierre",
""
]
] | TITLE: HuMoT: Human Motion Representation using Topology-Agnostic Transformers
for Character Animation Retargeting
ABSTRACT: Motion retargeting is the long-standing problem in character animation that
consists in transferring and adapting the motion of a source character to
another target character. A typical application is the creation of motion
sequences from off-the-shelf motions by transferring them onto new characters.
Motion retargeting is also promising to increase interoperability of existing
animation systems and motion databases, as they often differ in the structure
of the skeleton(s) considered. Moreover, since the goal of motion retargeting
is to abstract and transfer motion dynamics, effective solutions might provide
expressive and powerful human motion models in which operations such as
cleaning or editing are easier. In this article, we present a novel neural
network architecture for retargeting that extracts an abstract representation
of human motion agnostic to skeleton topology and morphology. Based on
transformers, our model is able to encode and decode motion sequences with
variable morphology and topology -- extending the current scope of retargeting
-- while supporting skeleton topologies not seen during the training phase.
More specifically, our model is structured as an autoencoder, and encoding and
decoding are separately conditioned on skeleton templates to extract and
control morphology and topology. Beyond motion retargeting, our model has many
applications since our abstract representation is a convenient space to embed
motion data from different sources. It may potentially be benefical to a number
of data-driven methods, allowing them to combine scarce specialised motion
datasets (e.g. with style or contact annotations) and larger general motion
datasets, for improved performance and generalisation ability. Moreover, we
show that our model can be useful for other applications beyond retargeting,
including motion denoising and joint upsampling.
|
2306.01105 | Sarah Masud | Atharva Kulkarni, Sarah Masud, Vikram Goyal, Tanmoy Chakraborty | Revisiting Hate Speech Benchmarks: From Data Curation to System
Deployment | 15 pages, 4 figures, 11 tables. Accepted at SIGKDD'23 | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Social media is awash with hateful content, much of which is often veiled
with linguistic and topical diversity. The benchmark datasets used for hate
speech detection do not account for such divagation as they are predominantly
compiled using hate lexicons. However, capturing hate signals becomes
challenging in neutrally-seeded malicious content. Thus, designing models and
datasets that mimic the real-world variability of hate warrants further
investigation.
To this end, we present GOTHate, a large-scale code-mixed crowdsourced
dataset of around 51k posts for hate speech detection from Twitter. GOTHate is
neutrally seeded, encompassing different languages and topics. We conduct
detailed comparisons of GOTHate with the existing hate speech datasets,
highlighting its novelty. We benchmark it with 10 recent baselines. Our
extensive empirical and benchmarking experiments suggest that GOTHate is hard
to classify in a text-only setup. Thus, we investigate how adding endogenous
signals enhances the hate speech detection task. We augment GOTHate with the
user's timeline information and ego network, bringing the overall data source
closer to the real-world setup for understanding hateful content. Our proposed
solution HEN-mBERT is a modular, multilingual, mixture-of-experts model that
enriches the linguistic subspace with latent endogenous signals from history,
topology, and exemplars. HEN-mBERT transcends the best baseline by 2.5% and 5%
in overall macro-F1 and hate class F1, respectively. Inspired by our
experiments, in partnership with Wipro AI, we are developing a semi-automated
pipeline to detect hateful content as a part of their mission to tackle online
harm.
| [
{
"version": "v1",
"created": "Thu, 1 Jun 2023 19:36:52 GMT"
},
{
"version": "v2",
"created": "Thu, 15 Jun 2023 12:37:34 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Kulkarni",
"Atharva",
""
],
[
"Masud",
"Sarah",
""
],
[
"Goyal",
"Vikram",
""
],
[
"Chakraborty",
"Tanmoy",
""
]
] | TITLE: Revisiting Hate Speech Benchmarks: From Data Curation to System
Deployment
ABSTRACT: Social media is awash with hateful content, much of which is often veiled
with linguistic and topical diversity. The benchmark datasets used for hate
speech detection do not account for such divagation as they are predominantly
compiled using hate lexicons. However, capturing hate signals becomes
challenging in neutrally-seeded malicious content. Thus, designing models and
datasets that mimic the real-world variability of hate warrants further
investigation.
To this end, we present GOTHate, a large-scale code-mixed crowdsourced
dataset of around 51k posts for hate speech detection from Twitter. GOTHate is
neutrally seeded, encompassing different languages and topics. We conduct
detailed comparisons of GOTHate with the existing hate speech datasets,
highlighting its novelty. We benchmark it with 10 recent baselines. Our
extensive empirical and benchmarking experiments suggest that GOTHate is hard
to classify in a text-only setup. Thus, we investigate how adding endogenous
signals enhances the hate speech detection task. We augment GOTHate with the
user's timeline information and ego network, bringing the overall data source
closer to the real-world setup for understanding hateful content. Our proposed
solution HEN-mBERT is a modular, multilingual, mixture-of-experts model that
enriches the linguistic subspace with latent endogenous signals from history,
topology, and exemplars. HEN-mBERT transcends the best baseline by 2.5% and 5%
in overall macro-F1 and hate class F1, respectively. Inspired by our
experiments, in partnership with Wipro AI, we are developing a semi-automated
pipeline to detect hateful content as a part of their mission to tackle online
harm.
|
2306.01173 | Wei Zhang | Wei Zhang, Sen Wang, Xingliang Dong, Rongwei Guo, Norbert Haala | BAMF-SLAM: Bundle Adjusted Multi-Fisheye Visual-Inertial SLAM Using
Recurrent Field Transforms | Accepted to ICRA2023 | null | null | null | cs.RO | http://creativecommons.org/licenses/by-nc-nd/4.0/ | In this paper, we present BAMF-SLAM, a novel multi-fisheye visual-inertial
SLAM system that utilizes Bundle Adjustment (BA) and recurrent field transforms
(RFT) to achieve accurate and robust state estimation in challenging scenarios.
First, our system directly operates on raw fisheye images, enabling us to fully
exploit the wide Field-of-View (FoV) of fisheye cameras. Second, to overcome
the low-texture challenge, we explore the tightly-coupled integration of
multi-camera inputs and complementary inertial measurements via a unified
factor graph and jointly optimize the poses and dense depth maps. Third, for
global consistency, the wide FoV of the fisheye camera allows the system to
find more potential loop closures, and powered by the broad convergence basin
of RFT, our system can perform very wide baseline loop closing with little
overlap. Furthermore, we introduce a semi-pose-graph BA method to avoid the
expensive full global BA. By combining relative pose factors with loop closure
factors, the global states can be adjusted efficiently with modest memory
footprint while maintaining high accuracy. Evaluations on TUM-VI, Hilti-Oxford
and Newer College datasets show the superior performance of the proposed system
over prior works. In the Hilti SLAM Challenge 2022, our VIO version achieves
second place. In a subsequent submission, our complete system, including the
global BA backend, outperforms the winning approach.
| [
{
"version": "v1",
"created": "Thu, 1 Jun 2023 22:12:06 GMT"
},
{
"version": "v2",
"created": "Wed, 14 Jun 2023 23:18:16 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Zhang",
"Wei",
""
],
[
"Wang",
"Sen",
""
],
[
"Dong",
"Xingliang",
""
],
[
"Guo",
"Rongwei",
""
],
[
"Haala",
"Norbert",
""
]
] | TITLE: BAMF-SLAM: Bundle Adjusted Multi-Fisheye Visual-Inertial SLAM Using
Recurrent Field Transforms
ABSTRACT: In this paper, we present BAMF-SLAM, a novel multi-fisheye visual-inertial
SLAM system that utilizes Bundle Adjustment (BA) and recurrent field transforms
(RFT) to achieve accurate and robust state estimation in challenging scenarios.
First, our system directly operates on raw fisheye images, enabling us to fully
exploit the wide Field-of-View (FoV) of fisheye cameras. Second, to overcome
the low-texture challenge, we explore the tightly-coupled integration of
multi-camera inputs and complementary inertial measurements via a unified
factor graph and jointly optimize the poses and dense depth maps. Third, for
global consistency, the wide FoV of the fisheye camera allows the system to
find more potential loop closures, and powered by the broad convergence basin
of RFT, our system can perform very wide baseline loop closing with little
overlap. Furthermore, we introduce a semi-pose-graph BA method to avoid the
expensive full global BA. By combining relative pose factors with loop closure
factors, the global states can be adjusted efficiently with modest memory
footprint while maintaining high accuracy. Evaluations on TUM-VI, Hilti-Oxford
and Newer College datasets show the superior performance of the proposed system
over prior works. In the Hilti SLAM Challenge 2022, our VIO version achieves
second place. In a subsequent submission, our complete system, including the
global BA backend, outperforms the winning approach.
|
2306.01584 | Semere Kiros Bitew | Semere Kiros Bitew, Johannes Deleu, A. Seza Do\u{g}ru\"oz, Chris
Develder, Thomas Demeester | Learning from Partially Annotated Data: Example-aware Creation of
Gap-filling Exercises for Language Learning | 12 pages, Accepted in the 18th Workshop on Innovative Use of NLP for
Building Educational Applications | null | null | null | cs.CL | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Since performing exercises (including, e.g., practice tests) forms a crucial
component of learning, and creating such exercises requires non-trivial effort
from the teacher, there is a great value in automatic exercise generation in
digital tools in education. In this paper, we particularly focus on automatic
creation of gapfilling exercises for language learning, specifically grammar
exercises. Since providing any annotation in this domain requires human expert
effort, we aim to avoid it entirely and explore the task of converting existing
texts into new gap-filling exercises, purely based on an example exercise,
without explicit instruction or detailed annotation of the intended grammar
topics. We contribute (i) a novel neural network architecture specifically
designed for aforementioned gap-filling exercise generation task, and (ii) a
real-world benchmark dataset for French grammar. We show that our model for
this French grammar gap-filling exercise generation outperforms a competitive
baseline classifier by 8% in F1 percentage points, achieving an average F1
score of 82%. Our model implementation and the dataset are made publicly
available to foster future research, thus offering a standardized evaluation
and baseline solution of the proposed partially annotated data prediction task
in grammar exercise creation.
| [
{
"version": "v1",
"created": "Fri, 2 Jun 2023 14:54:16 GMT"
},
{
"version": "v2",
"created": "Thu, 15 Jun 2023 10:23:53 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Bitew",
"Semere Kiros",
""
],
[
"Deleu",
"Johannes",
""
],
[
"Doğruöz",
"A. Seza",
""
],
[
"Develder",
"Chris",
""
],
[
"Demeester",
"Thomas",
""
]
] | TITLE: Learning from Partially Annotated Data: Example-aware Creation of
Gap-filling Exercises for Language Learning
ABSTRACT: Since performing exercises (including, e.g., practice tests) forms a crucial
component of learning, and creating such exercises requires non-trivial effort
from the teacher, there is a great value in automatic exercise generation in
digital tools in education. In this paper, we particularly focus on automatic
creation of gapfilling exercises for language learning, specifically grammar
exercises. Since providing any annotation in this domain requires human expert
effort, we aim to avoid it entirely and explore the task of converting existing
texts into new gap-filling exercises, purely based on an example exercise,
without explicit instruction or detailed annotation of the intended grammar
topics. We contribute (i) a novel neural network architecture specifically
designed for aforementioned gap-filling exercise generation task, and (ii) a
real-world benchmark dataset for French grammar. We show that our model for
this French grammar gap-filling exercise generation outperforms a competitive
baseline classifier by 8% in F1 percentage points, achieving an average F1
score of 82%. Our model implementation and the dataset are made publicly
available to foster future research, thus offering a standardized evaluation
and baseline solution of the proposed partially annotated data prediction task
in grammar exercise creation.
|
2306.02252 | Jianghui Wang | Jianghui Wang, Yuxuan Wang, Dongyan Zhao, Zilong Zheng | MoviePuzzle: Visual Narrative Reasoning through Multimodal Order
Learning | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We introduce MoviePuzzle, a novel challenge that targets visual narrative
reasoning and holistic movie understanding. Despite the notable progress that
has been witnessed in the realm of video understanding, most prior works fail
to present tasks and models to address holistic video understanding and the
innate visual narrative structures existing in long-form videos. To tackle this
quandary, we put forth MoviePuzzle task that amplifies the temporal feature
learning and structure learning of video models by reshuffling the shot, frame,
and clip layers of movie segments in the presence of video-dialogue
information. We start by establishing a carefully refined dataset based on
MovieNet by dissecting movies into hierarchical layers and randomly permuting
the orders. Besides benchmarking the MoviePuzzle with prior arts on movie
understanding, we devise a Hierarchical Contrastive Movie Clustering (HCMC)
model that considers the underlying structure and visual semantic orders for
movie reordering. Specifically, through a pairwise and contrastive learning
approach, we train models to predict the correct order of each layer. This
equips them with the knack for deciphering the visual narrative structure of
movies and handling the disorder lurking in video data. Experiments show that
our approach outperforms existing state-of-the-art methods on the \MoviePuzzle
benchmark, underscoring its efficacy.
| [
{
"version": "v1",
"created": "Sun, 4 Jun 2023 03:51:54 GMT"
},
{
"version": "v2",
"created": "Wed, 14 Jun 2023 10:11:38 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Wang",
"Jianghui",
""
],
[
"Wang",
"Yuxuan",
""
],
[
"Zhao",
"Dongyan",
""
],
[
"Zheng",
"Zilong",
""
]
] | TITLE: MoviePuzzle: Visual Narrative Reasoning through Multimodal Order
Learning
ABSTRACT: We introduce MoviePuzzle, a novel challenge that targets visual narrative
reasoning and holistic movie understanding. Despite the notable progress that
has been witnessed in the realm of video understanding, most prior works fail
to present tasks and models to address holistic video understanding and the
innate visual narrative structures existing in long-form videos. To tackle this
quandary, we put forth MoviePuzzle task that amplifies the temporal feature
learning and structure learning of video models by reshuffling the shot, frame,
and clip layers of movie segments in the presence of video-dialogue
information. We start by establishing a carefully refined dataset based on
MovieNet by dissecting movies into hierarchical layers and randomly permuting
the orders. Besides benchmarking the MoviePuzzle with prior arts on movie
understanding, we devise a Hierarchical Contrastive Movie Clustering (HCMC)
model that considers the underlying structure and visual semantic orders for
movie reordering. Specifically, through a pairwise and contrastive learning
approach, we train models to predict the correct order of each layer. This
equips them with the knack for deciphering the visual narrative structure of
movies and handling the disorder lurking in video data. Experiments show that
our approach outperforms existing state-of-the-art methods on the \MoviePuzzle
benchmark, underscoring its efficacy.
|
2306.03423 | Max Reuter | Max Reuter, William Schulze | I'm Afraid I Can't Do That: Predicting Prompt Refusal in Black-Box
Generative Language Models | Submitted for review to KDD 2023 via the workshop "Foundations and
Applications in Large-scale AI Models: Pre-training, Fine-tuning, and
Prompt-based Learning" | null | null | null | cs.AI | http://creativecommons.org/licenses/by/4.0/ | Since the release of OpenAI's ChatGPT, generative language models have
attracted extensive public attention. The increased usage has highlighted
generative models' broad utility, but also revealed several forms of embedded
bias. Some is induced by the pre-training corpus; but additional bias specific
to generative models arises from the use of subjective fine-tuning to avoid
generating harmful content. Fine-tuning bias may come from individual engineers
and company policies, and affects which prompts the model chooses to refuse. In
this experiment, we characterize ChatGPT's refusal behavior using a black-box
attack. We first query ChatGPT with a variety of offensive and benign prompts
(n=1,706), then manually label each response as compliance or refusal. Manual
examination of responses reveals that refusal is not cleanly binary, and lies
on a continuum; as such, we map several different kinds of responses to a
binary of compliance or refusal. The small manually-labeled dataset is used to
train a refusal classifier, which achieves an accuracy of 96%. Second, we use
this refusal classifier to bootstrap a larger (n=10,000) dataset adapted from
the Quora Insincere Questions dataset. With this machine-labeled data, we train
a prompt classifier to predict whether ChatGPT will refuse a given question,
without seeing ChatGPT's response. This prompt classifier achieves 76% accuracy
on a test set of manually labeled questions (n=985). We examine our classifiers
and the prompt n-grams that are most predictive of either compliance or
refusal. Our datasets and code are available at
https://github.com/maxwellreuter/chatgpt-refusals.
| [
{
"version": "v1",
"created": "Tue, 6 Jun 2023 05:50:58 GMT"
},
{
"version": "v2",
"created": "Wed, 14 Jun 2023 05:13:34 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Reuter",
"Max",
""
],
[
"Schulze",
"William",
""
]
] | TITLE: I'm Afraid I Can't Do That: Predicting Prompt Refusal in Black-Box
Generative Language Models
ABSTRACT: Since the release of OpenAI's ChatGPT, generative language models have
attracted extensive public attention. The increased usage has highlighted
generative models' broad utility, but also revealed several forms of embedded
bias. Some is induced by the pre-training corpus; but additional bias specific
to generative models arises from the use of subjective fine-tuning to avoid
generating harmful content. Fine-tuning bias may come from individual engineers
and company policies, and affects which prompts the model chooses to refuse. In
this experiment, we characterize ChatGPT's refusal behavior using a black-box
attack. We first query ChatGPT with a variety of offensive and benign prompts
(n=1,706), then manually label each response as compliance or refusal. Manual
examination of responses reveals that refusal is not cleanly binary, and lies
on a continuum; as such, we map several different kinds of responses to a
binary of compliance or refusal. The small manually-labeled dataset is used to
train a refusal classifier, which achieves an accuracy of 96%. Second, we use
this refusal classifier to bootstrap a larger (n=10,000) dataset adapted from
the Quora Insincere Questions dataset. With this machine-labeled data, we train
a prompt classifier to predict whether ChatGPT will refuse a given question,
without seeing ChatGPT's response. This prompt classifier achieves 76% accuracy
on a test set of manually labeled questions (n=985). We examine our classifiers
and the prompt n-grams that are most predictive of either compliance or
refusal. Our datasets and code are available at
https://github.com/maxwellreuter/chatgpt-refusals.
|
2306.04428 | Claytone Sikasote | Claytone Sikasote, Kalinda Siaminwe, Stanly Mwape, Bangiwe Zulu, Mofya
Phiri, Martin Phiri, David Zulu, Mayumbo Nyirenda, Antonios Anastasopoulos | Zambezi Voice: A Multilingual Speech Corpus for Zambian Languages | Accepted at INTERSPEECH 2023. This pre-print version differs slightly
from the version accepted to INTERSPEECH 2023: Figure 1 is not included in
INTERSPEECH 2023! | null | null | null | cs.CL cs.SD eess.AS | http://creativecommons.org/licenses/by-nc-nd/4.0/ | This work introduces Zambezi Voice, an open-source multilingual speech
resource for Zambian languages. It contains two collections of datasets:
unlabelled audio recordings of radio news and talk shows programs (160 hours)
and labelled data (over 80 hours) consisting of read speech recorded from text
sourced from publicly available literature books. The dataset is created for
speech recognition but can be extended to multilingual speech processing
research for both supervised and unsupervised learning approaches. To our
knowledge, this is the first multilingual speech dataset created for Zambian
languages. We exploit pretraining and cross-lingual transfer learning by
finetuning the Wav2Vec2.0 large-scale multilingual pre-trained model to build
end-to-end (E2E) speech recognition models for our baseline models. The dataset
is released publicly under a Creative Commons BY-NC-ND 4.0 license and can be
accessed via https://github.com/unza-speech-lab/zambezi-voice .
| [
{
"version": "v1",
"created": "Wed, 7 Jun 2023 13:36:37 GMT"
},
{
"version": "v2",
"created": "Tue, 13 Jun 2023 20:48:02 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Sikasote",
"Claytone",
""
],
[
"Siaminwe",
"Kalinda",
""
],
[
"Mwape",
"Stanly",
""
],
[
"Zulu",
"Bangiwe",
""
],
[
"Phiri",
"Mofya",
""
],
[
"Phiri",
"Martin",
""
],
[
"Zulu",
"David",
""
],
[
"Nyirenda",
"Mayumbo",
""
],
[
"Anastasopoulos",
"Antonios",
""
]
] | TITLE: Zambezi Voice: A Multilingual Speech Corpus for Zambian Languages
ABSTRACT: This work introduces Zambezi Voice, an open-source multilingual speech
resource for Zambian languages. It contains two collections of datasets:
unlabelled audio recordings of radio news and talk shows programs (160 hours)
and labelled data (over 80 hours) consisting of read speech recorded from text
sourced from publicly available literature books. The dataset is created for
speech recognition but can be extended to multilingual speech processing
research for both supervised and unsupervised learning approaches. To our
knowledge, this is the first multilingual speech dataset created for Zambian
languages. We exploit pretraining and cross-lingual transfer learning by
finetuning the Wav2Vec2.0 large-scale multilingual pre-trained model to build
end-to-end (E2E) speech recognition models for our baseline models. The dataset
is released publicly under a Creative Commons BY-NC-ND 4.0 license and can be
accessed via https://github.com/unza-speech-lab/zambezi-voice .
|
2306.04956 | XiaoHui Zhang | Xiaohui Zhang, Jiangyan Yi, Jianhua Tao, Chenlong Wang, Le Xu and
Ruibo Fu | Adaptive Fake Audio Detection with Low-Rank Model Squeezing | null | DADA workshop on IJCAI 2023 | null | null | cs.SD cs.LG eess.AS | http://creativecommons.org/licenses/by/4.0/ | The rapid advancement of spoofing algorithms necessitates the development of
robust detection methods capable of accurately identifying emerging fake audio.
Traditional approaches, such as finetuning on new datasets containing these
novel spoofing algorithms, are computationally intensive and pose a risk of
impairing the acquired knowledge of known fake audio types. To address these
challenges, this paper proposes an innovative approach that mitigates the
limitations associated with finetuning. We introduce the concept of training
low-rank adaptation matrices tailored specifically to the newly emerging fake
audio types. During the inference stage, these adaptation matrices are combined
with the existing model to generate the final prediction output. Extensive
experimentation is conducted to evaluate the efficacy of the proposed method.
The results demonstrate that our approach effectively preserves the prediction
accuracy of the existing model for known fake audio types. Furthermore, our
approach offers several advantages, including reduced storage memory
requirements and lower equal error rates compared to conventional finetuning
methods, particularly on specific spoofing algorithms.
| [
{
"version": "v1",
"created": "Thu, 8 Jun 2023 06:06:42 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Zhang",
"Xiaohui",
""
],
[
"Yi",
"Jiangyan",
""
],
[
"Tao",
"Jianhua",
""
],
[
"Wang",
"Chenlong",
""
],
[
"Xu",
"Le",
""
],
[
"Fu",
"Ruibo",
""
]
] | TITLE: Adaptive Fake Audio Detection with Low-Rank Model Squeezing
ABSTRACT: The rapid advancement of spoofing algorithms necessitates the development of
robust detection methods capable of accurately identifying emerging fake audio.
Traditional approaches, such as finetuning on new datasets containing these
novel spoofing algorithms, are computationally intensive and pose a risk of
impairing the acquired knowledge of known fake audio types. To address these
challenges, this paper proposes an innovative approach that mitigates the
limitations associated with finetuning. We introduce the concept of training
low-rank adaptation matrices tailored specifically to the newly emerging fake
audio types. During the inference stage, these adaptation matrices are combined
with the existing model to generate the final prediction output. Extensive
experimentation is conducted to evaluate the efficacy of the proposed method.
The results demonstrate that our approach effectively preserves the prediction
accuracy of the existing model for known fake audio types. Furthermore, our
approach offers several advantages, including reduced storage memory
requirements and lower equal error rates compared to conventional finetuning
methods, particularly on specific spoofing algorithms.
|
2306.06300 | Ziyang Yan | Ali Karami, Simone Rigon, Gabriele Mazzacca, Ziyang Yan, Fabio
Remondino | NERFBK: A High-Quality Benchmark for NERF-Based 3D Reconstruction | paper result has problem | null | null | null | cs.CV cs.AI cs.GR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper introduces a new real and synthetic dataset called NeRFBK
specifically designed for testing and comparing NeRF-based 3D reconstruction
algorithms. High-quality 3D reconstruction has significant potential in various
fields, and advancements in image-based algorithms make it essential to
evaluate new advanced techniques. However, gathering diverse data with precise
ground truth is challenging and may not encompass all relevant applications.
The NeRFBK dataset addresses this issue by providing multi-scale, indoor and
outdoor datasets with high-resolution images and videos and camera parameters
for testing and comparing NeRF-based algorithms. This paper presents the design
and creation of the NeRFBK benchmark, various examples and application
scenarios, and highlights its potential for advancing the field of 3D
reconstruction.
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 23:28:33 GMT"
},
{
"version": "v2",
"created": "Thu, 15 Jun 2023 10:51:34 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Karami",
"Ali",
""
],
[
"Rigon",
"Simone",
""
],
[
"Mazzacca",
"Gabriele",
""
],
[
"Yan",
"Ziyang",
""
],
[
"Remondino",
"Fabio",
""
]
] | TITLE: NERFBK: A High-Quality Benchmark for NERF-Based 3D Reconstruction
ABSTRACT: This paper introduces a new real and synthetic dataset called NeRFBK
specifically designed for testing and comparing NeRF-based 3D reconstruction
algorithms. High-quality 3D reconstruction has significant potential in various
fields, and advancements in image-based algorithms make it essential to
evaluate new advanced techniques. However, gathering diverse data with precise
ground truth is challenging and may not encompass all relevant applications.
The NeRFBK dataset addresses this issue by providing multi-scale, indoor and
outdoor datasets with high-resolution images and videos and camera parameters
for testing and comparing NeRF-based algorithms. This paper presents the design
and creation of the NeRFBK benchmark, various examples and application
scenarios, and highlights its potential for advancing the field of 3D
reconstruction.
|
2306.06408 | Josue Page Vizcaino | Josu\'e Page Vizca\'ino, Panagiotis Symvoulidis, Zeguan Wang, Jonas
Jelten, Paolo Favaro, Edward S. Boyden, Tobias Lasser | Fast light-field 3D microscopy with out-of-distribution detection and
adaptation through Conditional Normalizing Flows | null | null | null | null | eess.IV cs.CV cs.LG q-bio.NC | http://creativecommons.org/licenses/by/4.0/ | Real-time 3D fluorescence microscopy is crucial for the spatiotemporal
analysis of live organisms, such as neural activity monitoring. The eXtended
field-of-view light field microscope (XLFM), also known as Fourier light field
microscope, is a straightforward, single snapshot solution to achieve this. The
XLFM acquires spatial-angular information in a single camera exposure. In a
subsequent step, a 3D volume can be algorithmically reconstructed, making it
exceptionally well-suited for real-time 3D acquisition and potential analysis.
Unfortunately, traditional reconstruction methods (like deconvolution) require
lengthy processing times (0.0220 Hz), hampering the speed advantages of the
XLFM. Neural network architectures can overcome the speed constraints at the
expense of lacking certainty metrics, which renders them untrustworthy for the
biomedical realm. This work proposes a novel architecture to perform fast 3D
reconstructions of live immobilized zebrafish neural activity based on a
conditional normalizing flow. It reconstructs volumes at 8 Hz spanning
512x512x96 voxels, and it can be trained in under two hours due to the small
dataset requirements (10 image-volume pairs). Furthermore, normalizing flows
allow for exact Likelihood computation, enabling distribution monitoring,
followed by out-of-distribution detection and retraining of the system when a
novel sample is detected. We evaluate the proposed method on a cross-validation
approach involving multiple in-distribution samples (genetically identical
zebrafish) and various out-of-distribution ones.
| [
{
"version": "v1",
"created": "Sat, 10 Jun 2023 10:42:49 GMT"
},
{
"version": "v2",
"created": "Wed, 14 Jun 2023 10:05:14 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Vizcaíno",
"Josué Page",
""
],
[
"Symvoulidis",
"Panagiotis",
""
],
[
"Wang",
"Zeguan",
""
],
[
"Jelten",
"Jonas",
""
],
[
"Favaro",
"Paolo",
""
],
[
"Boyden",
"Edward S.",
""
],
[
"Lasser",
"Tobias",
""
]
] | TITLE: Fast light-field 3D microscopy with out-of-distribution detection and
adaptation through Conditional Normalizing Flows
ABSTRACT: Real-time 3D fluorescence microscopy is crucial for the spatiotemporal
analysis of live organisms, such as neural activity monitoring. The eXtended
field-of-view light field microscope (XLFM), also known as Fourier light field
microscope, is a straightforward, single snapshot solution to achieve this. The
XLFM acquires spatial-angular information in a single camera exposure. In a
subsequent step, a 3D volume can be algorithmically reconstructed, making it
exceptionally well-suited for real-time 3D acquisition and potential analysis.
Unfortunately, traditional reconstruction methods (like deconvolution) require
lengthy processing times (0.0220 Hz), hampering the speed advantages of the
XLFM. Neural network architectures can overcome the speed constraints at the
expense of lacking certainty metrics, which renders them untrustworthy for the
biomedical realm. This work proposes a novel architecture to perform fast 3D
reconstructions of live immobilized zebrafish neural activity based on a
conditional normalizing flow. It reconstructs volumes at 8 Hz spanning
512x512x96 voxels, and it can be trained in under two hours due to the small
dataset requirements (10 image-volume pairs). Furthermore, normalizing flows
allow for exact Likelihood computation, enabling distribution monitoring,
followed by out-of-distribution detection and retraining of the system when a
novel sample is detected. We evaluate the proposed method on a cross-validation
approach involving multiple in-distribution samples (genetically identical
zebrafish) and various out-of-distribution ones.
|
2306.06914 | Sultan Zavrak | Sultan Zavrak | Enhancing COVID-19 Diagnosis through Vision Transformer-Based Analysis
of Chest X-ray Images | null | null | null | null | eess.IV cs.CV cs.LG | http://creativecommons.org/licenses/by-nc-sa/4.0/ | The advent of 2019 Coronavirus (COVID-19) has engendered a momentous global
health crisis, necessitating the identification of the ailment in individuals
through diverse diagnostic modalities. Radiological imaging, particularly the
deployment of X-ray imaging, has been recognized as a pivotal instrument in the
detection and characterization of COVID-19. Recent investigations have unveiled
invaluable insights pertaining to the virus within X-ray images, instigating
the exploration of methodologies aimed at augmenting diagnostic accuracy
through the utilization of artificial intelligence (AI) techniques. The current
research endeavor posits an innovative framework for the automated diagnosis of
COVID-19, harnessing raw chest X-ray images, specifically by means of
fine-tuning pre-trained Vision Transformer (ViT) models. The developed models
were appraised in terms of their binary classification performance, discerning
COVID-19 from Normal cases, as well as their ternary classification
performance, discriminating COVID-19 from Pneumonia and Normal instances, and
lastly, their quaternary classification performance, discriminating COVID-19
from Bacterial Pneumonia, Viral Pneumonia, and Normal conditions, employing
distinct datasets. The proposed model evinced extraordinary precision,
registering results of 99.92% and 99.84% for binary classification, 97.95% and
86.48% for ternary classification, and 86.81% for quaternary classification,
respectively, on the respective datasets.
| [
{
"version": "v1",
"created": "Mon, 12 Jun 2023 07:34:28 GMT"
},
{
"version": "v2",
"created": "Wed, 14 Jun 2023 09:56:59 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Zavrak",
"Sultan",
""
]
] | TITLE: Enhancing COVID-19 Diagnosis through Vision Transformer-Based Analysis
of Chest X-ray Images
ABSTRACT: The advent of 2019 Coronavirus (COVID-19) has engendered a momentous global
health crisis, necessitating the identification of the ailment in individuals
through diverse diagnostic modalities. Radiological imaging, particularly the
deployment of X-ray imaging, has been recognized as a pivotal instrument in the
detection and characterization of COVID-19. Recent investigations have unveiled
invaluable insights pertaining to the virus within X-ray images, instigating
the exploration of methodologies aimed at augmenting diagnostic accuracy
through the utilization of artificial intelligence (AI) techniques. The current
research endeavor posits an innovative framework for the automated diagnosis of
COVID-19, harnessing raw chest X-ray images, specifically by means of
fine-tuning pre-trained Vision Transformer (ViT) models. The developed models
were appraised in terms of their binary classification performance, discerning
COVID-19 from Normal cases, as well as their ternary classification
performance, discriminating COVID-19 from Pneumonia and Normal instances, and
lastly, their quaternary classification performance, discriminating COVID-19
from Bacterial Pneumonia, Viral Pneumonia, and Normal conditions, employing
distinct datasets. The proposed model evinced extraordinary precision,
registering results of 99.92% and 99.84% for binary classification, 97.95% and
86.48% for ternary classification, and 86.81% for quaternary classification,
respectively, on the respective datasets.
|
2306.06918 | Hao Peng | Hao Peng, Xiaozhi Wang, Feng Yao, Kaisheng Zeng, Lei Hou, Juanzi Li,
Zhiyuan Liu, Weixing Shen | The Devil is in the Details: On the Pitfalls of Event Extraction
Evaluation | Accepted at Findings of ACL 2023 | null | null | null | cs.CL cs.AI | http://creativecommons.org/licenses/by/4.0/ | Event extraction (EE) is a crucial task aiming at extracting events from
texts, which includes two subtasks: event detection (ED) and event argument
extraction (EAE). In this paper, we check the reliability of EE evaluations and
identify three major pitfalls: (1) The data preprocessing discrepancy makes the
evaluation results on the same dataset not directly comparable, but the data
preprocessing details are not widely noted and specified in papers. (2) The
output space discrepancy of different model paradigms makes different-paradigm
EE models lack grounds for comparison and also leads to unclear mapping issues
between predictions and annotations. (3) The absence of pipeline evaluation of
many EAE-only works makes them hard to be directly compared with EE works and
may not well reflect the model performance in real-world pipeline scenarios. We
demonstrate the significant influence of these pitfalls through comprehensive
meta-analyses of recent papers and empirical experiments. To avoid these
pitfalls, we suggest a series of remedies, including specifying data
preprocessing, standardizing outputs, and providing pipeline evaluation
results. To help implement these remedies, we develop a consistent evaluation
framework OMNIEVENT, which can be obtained from
https://github.com/THU-KEG/OmniEvent.
| [
{
"version": "v1",
"created": "Mon, 12 Jun 2023 07:38:31 GMT"
},
{
"version": "v2",
"created": "Thu, 15 Jun 2023 07:23:57 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Peng",
"Hao",
""
],
[
"Wang",
"Xiaozhi",
""
],
[
"Yao",
"Feng",
""
],
[
"Zeng",
"Kaisheng",
""
],
[
"Hou",
"Lei",
""
],
[
"Li",
"Juanzi",
""
],
[
"Liu",
"Zhiyuan",
""
],
[
"Shen",
"Weixing",
""
]
] | TITLE: The Devil is in the Details: On the Pitfalls of Event Extraction
Evaluation
ABSTRACT: Event extraction (EE) is a crucial task aiming at extracting events from
texts, which includes two subtasks: event detection (ED) and event argument
extraction (EAE). In this paper, we check the reliability of EE evaluations and
identify three major pitfalls: (1) The data preprocessing discrepancy makes the
evaluation results on the same dataset not directly comparable, but the data
preprocessing details are not widely noted and specified in papers. (2) The
output space discrepancy of different model paradigms makes different-paradigm
EE models lack grounds for comparison and also leads to unclear mapping issues
between predictions and annotations. (3) The absence of pipeline evaluation of
many EAE-only works makes them hard to be directly compared with EE works and
may not well reflect the model performance in real-world pipeline scenarios. We
demonstrate the significant influence of these pitfalls through comprehensive
meta-analyses of recent papers and empirical experiments. To avoid these
pitfalls, we suggest a series of remedies, including specifying data
preprocessing, standardizing outputs, and providing pipeline evaluation
results. To help implement these remedies, we develop a consistent evaluation
framework OMNIEVENT, which can be obtained from
https://github.com/THU-KEG/OmniEvent.
|
2306.06930 | Wenying Duan | Wenying Duan, Xiaoxi He, Zimu Zhou, Lothar Thiele, Hong Rao | Localised Adaptive Spatial-Temporal Graph Neural Network | This paper was accepted by KDD 2023 | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Spatial-temporal graph models are prevailing for abstracting and modelling
spatial and temporal dependencies. In this work, we ask the following question:
whether and to what extent can we localise spatial-temporal graph models? We
limit our scope to adaptive spatial-temporal graph neural networks (ASTGNNs),
the state-of-the-art model architecture. Our approach to localisation involves
sparsifying the spatial graph adjacency matrices. To this end, we propose
Adaptive Graph Sparsification (AGS), a graph sparsification algorithm which
successfully enables the localisation of ASTGNNs to an extreme extent (fully
localisation). We apply AGS to two distinct ASTGNN architectures and nine
spatial-temporal datasets. Intriguingly, we observe that spatial graphs in
ASTGNNs can be sparsified by over 99.5\% without any decline in test accuracy.
Furthermore, even when ASTGNNs are fully localised, becoming graph-less and
purely temporal, we record no drop in accuracy for the majority of tested
datasets, with only minor accuracy deterioration observed in the remaining
datasets. However, when the partially or fully localised ASTGNNs are
reinitialised and retrained on the same data, there is a considerable and
consistent drop in accuracy. Based on these observations, we reckon that
\textit{(i)} in the tested data, the information provided by the spatial
dependencies is primarily included in the information provided by the temporal
dependencies and, thus, can be essentially ignored for inference; and
\textit{(ii)} although the spatial dependencies provide redundant information,
it is vital for the effective training of ASTGNNs and thus cannot be ignored
during training. Furthermore, the localisation of ASTGNNs holds the potential
to reduce the heavy computation overhead required on large-scale
spatial-temporal data and further enable the distributed deployment of ASTGNNs.
| [
{
"version": "v1",
"created": "Mon, 12 Jun 2023 08:08:53 GMT"
},
{
"version": "v2",
"created": "Thu, 15 Jun 2023 13:54:24 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Duan",
"Wenying",
""
],
[
"He",
"Xiaoxi",
""
],
[
"Zhou",
"Zimu",
""
],
[
"Thiele",
"Lothar",
""
],
[
"Rao",
"Hong",
""
]
] | TITLE: Localised Adaptive Spatial-Temporal Graph Neural Network
ABSTRACT: Spatial-temporal graph models are prevailing for abstracting and modelling
spatial and temporal dependencies. In this work, we ask the following question:
whether and to what extent can we localise spatial-temporal graph models? We
limit our scope to adaptive spatial-temporal graph neural networks (ASTGNNs),
the state-of-the-art model architecture. Our approach to localisation involves
sparsifying the spatial graph adjacency matrices. To this end, we propose
Adaptive Graph Sparsification (AGS), a graph sparsification algorithm which
successfully enables the localisation of ASTGNNs to an extreme extent (fully
localisation). We apply AGS to two distinct ASTGNN architectures and nine
spatial-temporal datasets. Intriguingly, we observe that spatial graphs in
ASTGNNs can be sparsified by over 99.5\% without any decline in test accuracy.
Furthermore, even when ASTGNNs are fully localised, becoming graph-less and
purely temporal, we record no drop in accuracy for the majority of tested
datasets, with only minor accuracy deterioration observed in the remaining
datasets. However, when the partially or fully localised ASTGNNs are
reinitialised and retrained on the same data, there is a considerable and
consistent drop in accuracy. Based on these observations, we reckon that
\textit{(i)} in the tested data, the information provided by the spatial
dependencies is primarily included in the information provided by the temporal
dependencies and, thus, can be essentially ignored for inference; and
\textit{(ii)} although the spatial dependencies provide redundant information,
it is vital for the effective training of ASTGNNs and thus cannot be ignored
during training. Furthermore, the localisation of ASTGNNs holds the potential
to reduce the heavy computation overhead required on large-scale
spatial-temporal data and further enable the distributed deployment of ASTGNNs.
|
2306.07730 | Paul Streli | Valentin Bieri, Paul Streli, Berken Utku Demirel and Christian Holz | BeliefPPG: Uncertainty-aware Heart Rate Estimation from PPG signals via
Belief Propagation | Conference on Uncertainty in Artificial Intelligence (UAI) 2023. The
first two authors contributed equally | null | null | null | cs.LG cs.CV eess.SP | http://creativecommons.org/licenses/by-nc-sa/4.0/ | We present a novel learning-based method that achieves state-of-the-art
performance on several heart rate estimation benchmarks extracted from
photoplethysmography signals (PPG). We consider the evolution of the heart rate
in the context of a discrete-time stochastic process that we represent as a
hidden Markov model. We derive a distribution over possible heart rate values
for a given PPG signal window through a trained neural network. Using belief
propagation, we incorporate the statistical distribution of heart rate changes
to refine these estimates in a temporal context. From this, we obtain a
quantized probability distribution over the range of possible heart rate values
that captures a meaningful and well-calibrated estimate of the inherent
predictive uncertainty. We show the robustness of our method on eight public
datasets with three different cross-validation experiments.
| [
{
"version": "v1",
"created": "Tue, 13 Jun 2023 12:36:00 GMT"
},
{
"version": "v2",
"created": "Wed, 14 Jun 2023 16:29:53 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Bieri",
"Valentin",
""
],
[
"Streli",
"Paul",
""
],
[
"Demirel",
"Berken Utku",
""
],
[
"Holz",
"Christian",
""
]
] | TITLE: BeliefPPG: Uncertainty-aware Heart Rate Estimation from PPG signals via
Belief Propagation
ABSTRACT: We present a novel learning-based method that achieves state-of-the-art
performance on several heart rate estimation benchmarks extracted from
photoplethysmography signals (PPG). We consider the evolution of the heart rate
in the context of a discrete-time stochastic process that we represent as a
hidden Markov model. We derive a distribution over possible heart rate values
for a given PPG signal window through a trained neural network. Using belief
propagation, we incorporate the statistical distribution of heart rate changes
to refine these estimates in a temporal context. From this, we obtain a
quantized probability distribution over the range of possible heart rate values
that captures a meaningful and well-calibrated estimate of the inherent
predictive uncertainty. We show the robustness of our method on eight public
datasets with three different cross-validation experiments.
|
2306.07835 | Tobias Riedlinger | Tobias Riedlinger, Marius Schubert, Sarina Penquitt, Jan-Marcel
Kezmann, Pascal Colling, Karsten Kahl, Lutz Roese-Koerner, Michael Arnold,
Urs Zimmermann, Matthias Rottmann | LMD: Light-weight Prediction Quality Estimation for Object Detection in
Lidar Point Clouds | 19 pages, 11 figures, 11 tables | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Object detection on Lidar point cloud data is a promising technology for
autonomous driving and robotics which has seen a significant rise in
performance and accuracy during recent years. Particularly uncertainty
estimation is a crucial component for down-stream tasks and deep neural
networks remain error-prone even for predictions with high confidence.
Previously proposed methods for quantifying prediction uncertainty tend to
alter the training scheme of the detector or rely on prediction sampling which
results in vastly increased inference time. In order to address these two
issues, we propose LidarMetaDetect (LMD), a light-weight post-processing scheme
for prediction quality estimation. Our method can easily be added to any
pre-trained Lidar object detector without altering anything about the base
model and is purely based on post-processing, therefore, only leading to a
negligible computational overhead. Our experiments show a significant increase
of statistical reliability in separating true from false predictions. We
propose and evaluate an additional application of our method leading to the
detection of annotation errors. Explicit samples and a conservative count of
annotation error proposals indicates the viability of our method for
large-scale datasets like KITTI and nuScenes. On the widely-used nuScenes test
dataset, 43 out of the top 100 proposals of our method indicate, in fact,
erroneous annotations.
| [
{
"version": "v1",
"created": "Tue, 13 Jun 2023 15:13:29 GMT"
},
{
"version": "v2",
"created": "Thu, 15 Jun 2023 08:14:19 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Riedlinger",
"Tobias",
""
],
[
"Schubert",
"Marius",
""
],
[
"Penquitt",
"Sarina",
""
],
[
"Kezmann",
"Jan-Marcel",
""
],
[
"Colling",
"Pascal",
""
],
[
"Kahl",
"Karsten",
""
],
[
"Roese-Koerner",
"Lutz",
""
],
[
"Arnold",
"Michael",
""
],
[
"Zimmermann",
"Urs",
""
],
[
"Rottmann",
"Matthias",
""
]
] | TITLE: LMD: Light-weight Prediction Quality Estimation for Object Detection in
Lidar Point Clouds
ABSTRACT: Object detection on Lidar point cloud data is a promising technology for
autonomous driving and robotics which has seen a significant rise in
performance and accuracy during recent years. Particularly uncertainty
estimation is a crucial component for down-stream tasks and deep neural
networks remain error-prone even for predictions with high confidence.
Previously proposed methods for quantifying prediction uncertainty tend to
alter the training scheme of the detector or rely on prediction sampling which
results in vastly increased inference time. In order to address these two
issues, we propose LidarMetaDetect (LMD), a light-weight post-processing scheme
for prediction quality estimation. Our method can easily be added to any
pre-trained Lidar object detector without altering anything about the base
model and is purely based on post-processing, therefore, only leading to a
negligible computational overhead. Our experiments show a significant increase
of statistical reliability in separating true from false predictions. We
propose and evaluate an additional application of our method leading to the
detection of annotation errors. Explicit samples and a conservative count of
annotation error proposals indicates the viability of our method for
large-scale datasets like KITTI and nuScenes. On the widely-used nuScenes test
dataset, 43 out of the top 100 proposals of our method indicate, in fact,
erroneous annotations.
|
2306.07972 | Georgios Palaiokrassas | Georgios Palaiokrassas and Sandro Scherrers and Iason Ofeidis and
Leandros Tassiulas | Leveraging Machine Learning for Multichain DeFi Fraud Detection | null | null | null | null | q-fin.GN cs.CR cs.LG | http://creativecommons.org/licenses/by/4.0/ | Since the inception of permissionless blockchains with Bitcoin in 2008, it
became apparent that their most well-suited use case is related to making the
financial system and its advantages available to everyone seamlessly without
depending on any trusted intermediaries. Smart contracts across chains provide
an ecosystem of decentralized finance (DeFi), where users can interact with
lending pools, Automated Market Maker (AMM) exchanges, stablecoins,
derivatives, etc. with a cumulative locked value which had exceeded 160B USD.
While DeFi comes with high rewards, it also carries plenty of risks. Many
financial crimes have occurred over the years making the early detection of
malicious activity an issue of high priority. The proposed framework introduces
an effective method for extracting a set of features from different chains,
including the largest one, Ethereum and it is evaluated over an extensive
dataset we gathered with the transactions of the most widely used DeFi
protocols (23 in total, including Aave, Compound, Curve, Lido, and Yearn) based
on a novel dataset in collaboration with Covalent. Different Machine Learning
methods were employed, such as XGBoost and a Neural Network for identifying
fraud accounts detection interacting with DeFi and we demonstrate that the
introduction of novel DeFi-related features, significantly improves the
evaluation results, where Accuracy, Precision, Recall, F1-score and F2-score
where utilized.
| [
{
"version": "v1",
"created": "Wed, 17 May 2023 15:48:21 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Palaiokrassas",
"Georgios",
""
],
[
"Scherrers",
"Sandro",
""
],
[
"Ofeidis",
"Iason",
""
],
[
"Tassiulas",
"Leandros",
""
]
] | TITLE: Leveraging Machine Learning for Multichain DeFi Fraud Detection
ABSTRACT: Since the inception of permissionless blockchains with Bitcoin in 2008, it
became apparent that their most well-suited use case is related to making the
financial system and its advantages available to everyone seamlessly without
depending on any trusted intermediaries. Smart contracts across chains provide
an ecosystem of decentralized finance (DeFi), where users can interact with
lending pools, Automated Market Maker (AMM) exchanges, stablecoins,
derivatives, etc. with a cumulative locked value which had exceeded 160B USD.
While DeFi comes with high rewards, it also carries plenty of risks. Many
financial crimes have occurred over the years making the early detection of
malicious activity an issue of high priority. The proposed framework introduces
an effective method for extracting a set of features from different chains,
including the largest one, Ethereum and it is evaluated over an extensive
dataset we gathered with the transactions of the most widely used DeFi
protocols (23 in total, including Aave, Compound, Curve, Lido, and Yearn) based
on a novel dataset in collaboration with Covalent. Different Machine Learning
methods were employed, such as XGBoost and a Neural Network for identifying
fraud accounts detection interacting with DeFi and we demonstrate that the
introduction of novel DeFi-related features, significantly improves the
evaluation results, where Accuracy, Precision, Recall, F1-score and F2-score
where utilized.
|
2306.07973 | Jingwei Sun | Lin Duan, Jingwei Sun, Yiran Chen, Maria Gorlatova | PrivaScissors: Enhance the Privacy of Collaborative Inference through
the Lens of Mutual Information | null | null | null | null | cs.CR cs.LG | http://creativecommons.org/publicdomain/zero/1.0/ | Edge-cloud collaborative inference empowers resource-limited IoT devices to
support deep learning applications without disclosing their raw data to the
cloud server, thus preserving privacy. Nevertheless, prior research has shown
that collaborative inference still results in the exposure of data and
predictions from edge devices. To enhance the privacy of collaborative
inference, we introduce a defense strategy called PrivaScissors, which is
designed to reduce the mutual information between a model's intermediate
outcomes and the device's data and predictions. We evaluate PrivaScissors's
performance on several datasets in the context of diverse attacks and offer a
theoretical robustness guarantee.
| [
{
"version": "v1",
"created": "Wed, 17 May 2023 20:15:26 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Duan",
"Lin",
""
],
[
"Sun",
"Jingwei",
""
],
[
"Chen",
"Yiran",
""
],
[
"Gorlatova",
"Maria",
""
]
] | TITLE: PrivaScissors: Enhance the Privacy of Collaborative Inference through
the Lens of Mutual Information
ABSTRACT: Edge-cloud collaborative inference empowers resource-limited IoT devices to
support deep learning applications without disclosing their raw data to the
cloud server, thus preserving privacy. Nevertheless, prior research has shown
that collaborative inference still results in the exposure of data and
predictions from edge devices. To enhance the privacy of collaborative
inference, we introduce a defense strategy called PrivaScissors, which is
designed to reduce the mutual information between a model's intermediate
outcomes and the device's data and predictions. We evaluate PrivaScissors's
performance on several datasets in the context of diverse attacks and offer a
theoretical robustness guarantee.
|
2306.07981 | Mst Akter | Mst Shapna Akter, Hossain Shahriar, Juan Rodriguez Cardenas, Sheikh
Iqbal Ahamed, and Alfredo Cuzzocrea | Feature Engineering-Based Detection of Buffer Overflow Vulnerability in
Source Code Using Neural Networks | null | null | null | null | cs.CR cs.LG cs.SE | http://creativecommons.org/licenses/by/4.0/ | One of the most significant challenges in the field of software code auditing
is the presence of vulnerabilities in software source code. Every year, more
and more software flaws are discovered, either internally in proprietary code
or publicly disclosed. These flaws are highly likely to be exploited and can
lead to system compromise, data leakage, or denial of service. To create a
large-scale machine learning system for function level vulnerability
identification, we utilized a sizable dataset of C and C++ open-source code
containing millions of functions with potential buffer overflow exploits. We
have developed an efficient and scalable vulnerability detection method based
on neural network models that learn features extracted from the source codes.
The source code is first converted into an intermediate representation to
remove unnecessary components and shorten dependencies. We maintain the
semantic and syntactic information using state of the art word embedding
algorithms such as GloVe and fastText. The embedded vectors are subsequently
fed into neural networks such as LSTM, BiLSTM, LSTM Autoencoder, word2vec,
BERT, and GPT2 to classify the possible vulnerabilities. We maintain the
semantic and syntactic information using state of the art word embedding
algorithms such as GloVe and fastText. The embedded vectors are subsequently
fed into neural networks such as LSTM, BiLSTM, LSTM Autoencoder, word2vec,
BERT, and GPT2 to classify the possible vulnerabilities. Furthermore, we have
proposed a neural network model that can overcome issues associated with
traditional neural networks. We have used evaluation metrics such as F1 score,
precision, recall, accuracy, and total execution time to measure the
performance. We have conducted a comparative analysis between results derived
from features containing a minimal text representation and semantic and
syntactic information.
| [
{
"version": "v1",
"created": "Thu, 1 Jun 2023 01:44:49 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Akter",
"Mst Shapna",
""
],
[
"Shahriar",
"Hossain",
""
],
[
"Cardenas",
"Juan Rodriguez",
""
],
[
"Ahamed",
"Sheikh Iqbal",
""
],
[
"Cuzzocrea",
"Alfredo",
""
]
] | TITLE: Feature Engineering-Based Detection of Buffer Overflow Vulnerability in
Source Code Using Neural Networks
ABSTRACT: One of the most significant challenges in the field of software code auditing
is the presence of vulnerabilities in software source code. Every year, more
and more software flaws are discovered, either internally in proprietary code
or publicly disclosed. These flaws are highly likely to be exploited and can
lead to system compromise, data leakage, or denial of service. To create a
large-scale machine learning system for function level vulnerability
identification, we utilized a sizable dataset of C and C++ open-source code
containing millions of functions with potential buffer overflow exploits. We
have developed an efficient and scalable vulnerability detection method based
on neural network models that learn features extracted from the source codes.
The source code is first converted into an intermediate representation to
remove unnecessary components and shorten dependencies. We maintain the
semantic and syntactic information using state of the art word embedding
algorithms such as GloVe and fastText. The embedded vectors are subsequently
fed into neural networks such as LSTM, BiLSTM, LSTM Autoencoder, word2vec,
BERT, and GPT2 to classify the possible vulnerabilities. We maintain the
semantic and syntactic information using state of the art word embedding
algorithms such as GloVe and fastText. The embedded vectors are subsequently
fed into neural networks such as LSTM, BiLSTM, LSTM Autoencoder, word2vec,
BERT, and GPT2 to classify the possible vulnerabilities. Furthermore, we have
proposed a neural network model that can overcome issues associated with
traditional neural networks. We have used evaluation metrics such as F1 score,
precision, recall, accuracy, and total execution time to measure the
performance. We have conducted a comparative analysis between results derived
from features containing a minimal text representation and semantic and
syntactic information.
|
2306.07993 | Md. Shirajum Munir | Md. Shirajum Munir, Sachin Shetty, and Danda B. Rawat | Trustworthy Artificial Intelligence Framework for Proactive Detection
and Risk Explanation of Cyber Attacks in Smart Grid | Submitted for peer review | null | null | null | cs.CR cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The rapid growth of distributed energy resources (DERs), such as renewable
energy sources, generators, consumers, and prosumers in the smart grid
infrastructure, poses significant cybersecurity and trust challenges to the
grid controller. Consequently, it is crucial to identify adversarial tactics
and measure the strength of the attacker's DER. To enable a trustworthy smart
grid controller, this work investigates a trustworthy artificial intelligence
(AI) mechanism for proactive identification and explanation of the cyber risk
caused by the control/status message of DERs. Thus, proposing and developing a
trustworthy AI framework to facilitate the deployment of any AI algorithms for
detecting potential cyber threats and analyzing root causes based on Shapley
value interpretation while dynamically quantifying the risk of an attack based
on Ward's minimum variance formula. The experiment with a state-of-the-art
dataset establishes the proposed framework as a trustworthy AI by fulfilling
the capabilities of reliability, fairness, explainability, transparency,
reproducibility, and accountability.
| [
{
"version": "v1",
"created": "Mon, 12 Jun 2023 02:28:17 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Munir",
"Md. Shirajum",
""
],
[
"Shetty",
"Sachin",
""
],
[
"Rawat",
"Danda B.",
""
]
] | TITLE: Trustworthy Artificial Intelligence Framework for Proactive Detection
and Risk Explanation of Cyber Attacks in Smart Grid
ABSTRACT: The rapid growth of distributed energy resources (DERs), such as renewable
energy sources, generators, consumers, and prosumers in the smart grid
infrastructure, poses significant cybersecurity and trust challenges to the
grid controller. Consequently, it is crucial to identify adversarial tactics
and measure the strength of the attacker's DER. To enable a trustworthy smart
grid controller, this work investigates a trustworthy artificial intelligence
(AI) mechanism for proactive identification and explanation of the cyber risk
caused by the control/status message of DERs. Thus, proposing and developing a
trustworthy AI framework to facilitate the deployment of any AI algorithms for
detecting potential cyber threats and analyzing root causes based on Shapley
value interpretation while dynamically quantifying the risk of an attack based
on Ward's minimum variance formula. The experiment with a state-of-the-art
dataset establishes the proposed framework as a trustworthy AI by fulfilling
the capabilities of reliability, fairness, explainability, transparency,
reproducibility, and accountability.
|
2306.08000 | Aakash Mishra | Aakash Mishra, Rajat Mittal, Christy Jestin, Kostas Tingos, Pranav
Rajpurkar | Improving Zero-Shot Detection of Low Prevalence Chest Pathologies using
Domain Pre-trained Language Models | 3 pages, 1 table, Medical Imaging with Deep Learning, Short Paper | null | null | Short-Paper-120 | physics.med-ph cs.CL cs.CV cs.LG eess.IV | http://creativecommons.org/licenses/by/4.0/ | Recent advances in zero-shot learning have enabled the use of paired
image-text data to replace structured labels, replacing the need for expert
annotated datasets. Models such as CLIP-based CheXzero utilize these
advancements in the domain of chest X-ray interpretation. We hypothesize that
domain pre-trained models such as CXR-BERT, BlueBERT, and ClinicalBERT offer
the potential to improve the performance of CLIP-like models with specific
domain knowledge by replacing BERT weights at the cost of breaking the original
model's alignment. We evaluate the performance of zero-shot classification
models with domain-specific pre-training for detecting low-prevalence
pathologies. Even though replacing the weights of the original CLIP-BERT
degrades model performance on commonly found pathologies, we show that
pre-trained text towers perform exceptionally better on low-prevalence
diseases. This motivates future ensemble models with a combination of
differently trained language models for maximal performance.
| [
{
"version": "v1",
"created": "Tue, 13 Jun 2023 06:26:54 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Mishra",
"Aakash",
""
],
[
"Mittal",
"Rajat",
""
],
[
"Jestin",
"Christy",
""
],
[
"Tingos",
"Kostas",
""
],
[
"Rajpurkar",
"Pranav",
""
]
] | TITLE: Improving Zero-Shot Detection of Low Prevalence Chest Pathologies using
Domain Pre-trained Language Models
ABSTRACT: Recent advances in zero-shot learning have enabled the use of paired
image-text data to replace structured labels, replacing the need for expert
annotated datasets. Models such as CLIP-based CheXzero utilize these
advancements in the domain of chest X-ray interpretation. We hypothesize that
domain pre-trained models such as CXR-BERT, BlueBERT, and ClinicalBERT offer
the potential to improve the performance of CLIP-like models with specific
domain knowledge by replacing BERT weights at the cost of breaking the original
model's alignment. We evaluate the performance of zero-shot classification
models with domain-specific pre-training for detecting low-prevalence
pathologies. Even though replacing the weights of the original CLIP-BERT
degrades model performance on commonly found pathologies, we show that
pre-trained text towers perform exceptionally better on low-prevalence
diseases. This motivates future ensemble models with a combination of
differently trained language models for maximal performance.
|
2306.08001 | Siddarth Ijju | Sid Ijju | A Markovian Formalism for Active Querying | Active Learning, Markov, Inverse Reinforcement Learning, Query | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by/4.0/ | Active learning algorithms have been an integral part of recent advances in
artificial intelligence. However, the research in the field is widely varying
and lacks an overall organizing leans. We outline a Markovian formalism for the
field of active learning and survey the literature to demonstrate the
organizing capability of our proposed formalism. Our formalism takes a
partially observable Markovian system approach to the active learning process
as a whole. We specifically outline how querying, dataset augmentation, reward
updates, and other aspects of active learning can be viewed as a transition
between meta-states in a Markovian system, and give direction into how other
aspects of active learning can fit into our formalism.
| [
{
"version": "v1",
"created": "Tue, 13 Jun 2023 06:49:50 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Ijju",
"Sid",
""
]
] | TITLE: A Markovian Formalism for Active Querying
ABSTRACT: Active learning algorithms have been an integral part of recent advances in
artificial intelligence. However, the research in the field is widely varying
and lacks an overall organizing leans. We outline a Markovian formalism for the
field of active learning and survey the literature to demonstrate the
organizing capability of our proposed formalism. Our formalism takes a
partially observable Markovian system approach to the active learning process
as a whole. We specifically outline how querying, dataset augmentation, reward
updates, and other aspects of active learning can be viewed as a transition
between meta-states in a Markovian system, and give direction into how other
aspects of active learning can fit into our formalism.
|
2306.08009 | Gaolei Li | Zhicong Yan, Shenghong Li, Ruijie Zhao, Yuan Tian, Yuanyuan Zhao | DHBE: Data-free Holistic Backdoor Erasing in Deep Neural Networks via
Restricted Adversarial Distillation | It has been accepted by asiaccs | null | null | null | cs.LG cs.AI cs.CR | http://creativecommons.org/licenses/by/4.0/ | Backdoor attacks have emerged as an urgent threat to Deep Neural Networks
(DNNs), where victim DNNs are furtively implanted with malicious neurons that
could be triggered by the adversary. To defend against backdoor attacks, many
works establish a staged pipeline to remove backdoors from victim DNNs:
inspecting, locating, and erasing. However, in a scenario where a few clean
data can be accessible, such pipeline is fragile and cannot erase backdoors
completely without sacrificing model accuracy. To address this issue, in this
paper, we propose a novel data-free holistic backdoor erasing (DHBE) framework.
Instead of the staged pipeline, the DHBE treats the backdoor erasing task as a
unified adversarial procedure, which seeks equilibrium between two different
competing processes: distillation and backdoor regularization. In distillation,
the backdoored DNN is distilled into a proxy model, transferring its knowledge
about clean data, yet backdoors are simultaneously transferred. In backdoor
regularization, the proxy model is holistically regularized to prevent from
infecting any possible backdoor transferred from distillation. These two
processes jointly proceed with data-free adversarial optimization until a
clean, high-accuracy proxy model is obtained. With the novel adversarial
design, our framework demonstrates its superiority in three aspects: 1) minimal
detriment to model accuracy, 2) high tolerance for hyperparameters, and 3) no
demand for clean data. Extensive experiments on various backdoor attacks and
datasets are performed to verify the effectiveness of the proposed framework.
Code is available at \url{https://github.com/yanzhicong/DHBE}
| [
{
"version": "v1",
"created": "Tue, 13 Jun 2023 09:29:05 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Yan",
"Zhicong",
""
],
[
"Li",
"Shenghong",
""
],
[
"Zhao",
"Ruijie",
""
],
[
"Tian",
"Yuan",
""
],
[
"Zhao",
"Yuanyuan",
""
]
] | TITLE: DHBE: Data-free Holistic Backdoor Erasing in Deep Neural Networks via
Restricted Adversarial Distillation
ABSTRACT: Backdoor attacks have emerged as an urgent threat to Deep Neural Networks
(DNNs), where victim DNNs are furtively implanted with malicious neurons that
could be triggered by the adversary. To defend against backdoor attacks, many
works establish a staged pipeline to remove backdoors from victim DNNs:
inspecting, locating, and erasing. However, in a scenario where a few clean
data can be accessible, such pipeline is fragile and cannot erase backdoors
completely without sacrificing model accuracy. To address this issue, in this
paper, we propose a novel data-free holistic backdoor erasing (DHBE) framework.
Instead of the staged pipeline, the DHBE treats the backdoor erasing task as a
unified adversarial procedure, which seeks equilibrium between two different
competing processes: distillation and backdoor regularization. In distillation,
the backdoored DNN is distilled into a proxy model, transferring its knowledge
about clean data, yet backdoors are simultaneously transferred. In backdoor
regularization, the proxy model is holistically regularized to prevent from
infecting any possible backdoor transferred from distillation. These two
processes jointly proceed with data-free adversarial optimization until a
clean, high-accuracy proxy model is obtained. With the novel adversarial
design, our framework demonstrates its superiority in three aspects: 1) minimal
detriment to model accuracy, 2) high tolerance for hyperparameters, and 3) no
demand for clean data. Extensive experiments on various backdoor attacks and
datasets are performed to verify the effectiveness of the proposed framework.
Code is available at \url{https://github.com/yanzhicong/DHBE}
|
2306.08011 | Gaolei Li | Haochen Mei, Gaolei Li, Jun Wu, Longfei Zheng | Privacy Inference-Empowered Stealthy Backdoor Attack on Federated
Learning under Non-IID Scenarios | It can be accepted IJCNN | null | null | null | cs.LG cs.AI cs.CR | http://creativecommons.org/licenses/by/4.0/ | Federated learning (FL) naturally faces the problem of data heterogeneity in
real-world scenarios, but this is often overlooked by studies on FL security
and privacy. On the one hand, the effectiveness of backdoor attacks on FL may
drop significantly under non-IID scenarios. On the other hand, malicious
clients may steal private data through privacy inference attacks. Therefore, it
is necessary to have a comprehensive perspective of data heterogeneity,
backdoor, and privacy inference. In this paper, we propose a novel privacy
inference-empowered stealthy backdoor attack (PI-SBA) scheme for FL under
non-IID scenarios. Firstly, a diverse data reconstruction mechanism based on
generative adversarial networks (GANs) is proposed to produce a supplementary
dataset, which can improve the attacker's local data distribution and support
more sophisticated strategies for backdoor attacks. Based on this, we design a
source-specified backdoor learning (SSBL) strategy as a demonstration, allowing
the adversary to arbitrarily specify which classes are susceptible to the
backdoor trigger. Since the PI-SBA has an independent poisoned data synthesis
process, it can be integrated into existing backdoor attacks to improve their
effectiveness and stealthiness in non-IID scenarios. Extensive experiments
based on MNIST, CIFAR10 and Youtube Aligned Face datasets demonstrate that the
proposed PI-SBA scheme is effective in non-IID FL and stealthy against
state-of-the-art defense methods.
| [
{
"version": "v1",
"created": "Tue, 13 Jun 2023 11:08:30 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Mei",
"Haochen",
""
],
[
"Li",
"Gaolei",
""
],
[
"Wu",
"Jun",
""
],
[
"Zheng",
"Longfei",
""
]
] | TITLE: Privacy Inference-Empowered Stealthy Backdoor Attack on Federated
Learning under Non-IID Scenarios
ABSTRACT: Federated learning (FL) naturally faces the problem of data heterogeneity in
real-world scenarios, but this is often overlooked by studies on FL security
and privacy. On the one hand, the effectiveness of backdoor attacks on FL may
drop significantly under non-IID scenarios. On the other hand, malicious
clients may steal private data through privacy inference attacks. Therefore, it
is necessary to have a comprehensive perspective of data heterogeneity,
backdoor, and privacy inference. In this paper, we propose a novel privacy
inference-empowered stealthy backdoor attack (PI-SBA) scheme for FL under
non-IID scenarios. Firstly, a diverse data reconstruction mechanism based on
generative adversarial networks (GANs) is proposed to produce a supplementary
dataset, which can improve the attacker's local data distribution and support
more sophisticated strategies for backdoor attacks. Based on this, we design a
source-specified backdoor learning (SSBL) strategy as a demonstration, allowing
the adversary to arbitrarily specify which classes are susceptible to the
backdoor trigger. Since the PI-SBA has an independent poisoned data synthesis
process, it can be integrated into existing backdoor attacks to improve their
effectiveness and stealthiness in non-IID scenarios. Extensive experiments
based on MNIST, CIFAR10 and Youtube Aligned Face datasets demonstrate that the
proposed PI-SBA scheme is effective in non-IID FL and stealthy against
state-of-the-art defense methods.
|
2306.08021 | Ahmet Caner Y\"uz\"ug\"uler | Ahmet Caner Y\"uz\"ug\"uler and Nikolaos Dimitriadis and Pascal
Frossard | Flexible Channel Dimensions for Differentiable Architecture Search | null | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by/4.0/ | Finding optimal channel dimensions (i.e., the number of filters in DNN
layers) is essential to design DNNs that perform well under computational
resource constraints. Recent work in neural architecture search aims at
automating the optimization of the DNN model implementation. However, existing
neural architecture search methods for channel dimensions rely on fixed search
spaces, which prevents achieving an efficient and fully automated solution. In
this work, we propose a novel differentiable neural architecture search method
with an efficient dynamic channel allocation algorithm to enable a flexible
search space for channel dimensions. We show that the proposed framework is
able to find DNN architectures that are equivalent to previous methods in task
accuracy and inference latency for the CIFAR-10 dataset with an improvement of
$1.3-1.7\times$ in GPU-hours and $1.5-1.7\times$ in the memory requirements
during the architecture search stage. Moreover, the proposed frameworks do not
require a well-engineered search space a priori, which is an important step
towards fully automated design of DNN architectures.
| [
{
"version": "v1",
"created": "Tue, 13 Jun 2023 15:21:38 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Yüzügüler",
"Ahmet Caner",
""
],
[
"Dimitriadis",
"Nikolaos",
""
],
[
"Frossard",
"Pascal",
""
]
] | TITLE: Flexible Channel Dimensions for Differentiable Architecture Search
ABSTRACT: Finding optimal channel dimensions (i.e., the number of filters in DNN
layers) is essential to design DNNs that perform well under computational
resource constraints. Recent work in neural architecture search aims at
automating the optimization of the DNN model implementation. However, existing
neural architecture search methods for channel dimensions rely on fixed search
spaces, which prevents achieving an efficient and fully automated solution. In
this work, we propose a novel differentiable neural architecture search method
with an efficient dynamic channel allocation algorithm to enable a flexible
search space for channel dimensions. We show that the proposed framework is
able to find DNN architectures that are equivalent to previous methods in task
accuracy and inference latency for the CIFAR-10 dataset with an improvement of
$1.3-1.7\times$ in GPU-hours and $1.5-1.7\times$ in the memory requirements
during the architecture search stage. Moreover, the proposed frameworks do not
require a well-engineered search space a priori, which is an important step
towards fully automated design of DNN architectures.
|
2306.08055 | Ellie Kitanidis | Abraham J. Fetterman, Ellie Kitanidis, Joshua Albrecht, Zachary
Polizzi, Bryden Fogelman, Maksis Knutins, Bartosz Wr\'oblewski, James B.
Simon, Kanjun Qiu | Tune As You Scale: Hyperparameter Optimization For Compute Efficient
Training | null | null | null | null | cs.LG cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Hyperparameter tuning of deep learning models can lead to order-of-magnitude
performance gains for the same amount of compute. Despite this, systematic
tuning is uncommon, particularly for large models, which are expensive to
evaluate and tend to have many hyperparameters, necessitating difficult
judgment calls about tradeoffs, budgets, and search bounds. To address these
issues and propose a practical method for robustly tuning large models, we
present Cost-Aware Pareto Region Bayesian Search (CARBS), a Bayesian
optimization algorithm that performs local search around the performance-cost
Pareto frontier. CARBS does well even in unbounded search spaces with many
hyperparameters, learns scaling relationships so that it can tune models even
as they are scaled up, and automates much of the "black magic" of tuning. Among
our results, we effectively solve the entire ProcGen benchmark just by tuning a
simple baseline (PPO, as provided in the original ProcGen paper). We also
reproduce the model size vs. training tokens scaling result from the Chinchilla
project (Hoffmann et al. 2022), while simultaneously discovering scaling laws
for every other hyperparameter, via an easy automated process that uses
significantly less compute and is applicable to any deep learning problem (not
just language models).
| [
{
"version": "v1",
"created": "Tue, 13 Jun 2023 18:22:24 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Fetterman",
"Abraham J.",
""
],
[
"Kitanidis",
"Ellie",
""
],
[
"Albrecht",
"Joshua",
""
],
[
"Polizzi",
"Zachary",
""
],
[
"Fogelman",
"Bryden",
""
],
[
"Knutins",
"Maksis",
""
],
[
"Wróblewski",
"Bartosz",
""
],
[
"Simon",
"James B.",
""
],
[
"Qiu",
"Kanjun",
""
]
] | TITLE: Tune As You Scale: Hyperparameter Optimization For Compute Efficient
Training
ABSTRACT: Hyperparameter tuning of deep learning models can lead to order-of-magnitude
performance gains for the same amount of compute. Despite this, systematic
tuning is uncommon, particularly for large models, which are expensive to
evaluate and tend to have many hyperparameters, necessitating difficult
judgment calls about tradeoffs, budgets, and search bounds. To address these
issues and propose a practical method for robustly tuning large models, we
present Cost-Aware Pareto Region Bayesian Search (CARBS), a Bayesian
optimization algorithm that performs local search around the performance-cost
Pareto frontier. CARBS does well even in unbounded search spaces with many
hyperparameters, learns scaling relationships so that it can tune models even
as they are scaled up, and automates much of the "black magic" of tuning. Among
our results, we effectively solve the entire ProcGen benchmark just by tuning a
simple baseline (PPO, as provided in the original ProcGen paper). We also
reproduce the model size vs. training tokens scaling result from the Chinchilla
project (Hoffmann et al. 2022), while simultaneously discovering scaling laws
for every other hyperparameter, via an easy automated process that uses
significantly less compute and is applicable to any deep learning problem (not
just language models).
|
2306.08057 | Nan Jiang | Nan Jiang, Yexiang Xue | Symbolic Regression via Control Variable Genetic Programming | null | null | null | null | cs.NE cs.AI cs.LG | http://creativecommons.org/licenses/by/4.0/ | Learning symbolic expressions directly from experiment data is a vital step
in AI-driven scientific discovery. Nevertheless, state-of-the-art approaches
are limited to learning simple expressions. Regressing expressions involving
many independent variables still remain out of reach. Motivated by the control
variable experiments widely utilized in science, we propose Control Variable
Genetic Programming (CVGP) for symbolic regression over many independent
variables. CVGP expedites symbolic expression discovery via customized
experiment design, rather than learning from a fixed dataset collected a
priori. CVGP starts by fitting simple expressions involving a small set of
independent variables using genetic programming, under controlled experiments
where other variables are held as constants. It then extends expressions
learned in previous generations by adding new independent variables, using new
control variable experiments in which these variables are allowed to vary.
Theoretically, we show CVGP as an incremental building approach can yield an
exponential reduction in the search space when learning a class of expressions.
Experimentally, CVGP outperforms several baselines in learning symbolic
expressions involving multiple independent variables.
| [
{
"version": "v1",
"created": "Thu, 25 May 2023 04:11:14 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Jiang",
"Nan",
""
],
[
"Xue",
"Yexiang",
""
]
] | TITLE: Symbolic Regression via Control Variable Genetic Programming
ABSTRACT: Learning symbolic expressions directly from experiment data is a vital step
in AI-driven scientific discovery. Nevertheless, state-of-the-art approaches
are limited to learning simple expressions. Regressing expressions involving
many independent variables still remain out of reach. Motivated by the control
variable experiments widely utilized in science, we propose Control Variable
Genetic Programming (CVGP) for symbolic regression over many independent
variables. CVGP expedites symbolic expression discovery via customized
experiment design, rather than learning from a fixed dataset collected a
priori. CVGP starts by fitting simple expressions involving a small set of
independent variables using genetic programming, under controlled experiments
where other variables are held as constants. It then extends expressions
learned in previous generations by adding new independent variables, using new
control variable experiments in which these variables are allowed to vary.
Theoretically, we show CVGP as an incremental building approach can yield an
exponential reduction in the search space when learning a class of expressions.
Experimentally, CVGP outperforms several baselines in learning symbolic
expressions involving multiple independent variables.
|
2306.08060 | Mst Akter | Mst Shapna Akter, Md Jobair Hossain Faruk, Nafisa Anjum, Mohammad
Masum, Hossain Shahriar, Akond Rahman, Fan Wu, Alfredo Cuzzocrea | Software Supply Chain Vulnerabilities Detection in Source Code:
Performance Comparison between Traditional and Quantum Machine Learning
Algorithms | null | null | 10.1109/BigData55660.2022.10020813 | null | cs.CR cs.LG quant-ph | http://creativecommons.org/licenses/by/4.0/ | The software supply chain (SSC) attack has become one of the crucial issues
that are being increased rapidly with the advancement of the software
development domain. In general, SSC attacks execute during the software
development processes lead to vulnerabilities in software products targeting
downstream customers and even involved stakeholders. Machine Learning
approaches are proven in detecting and preventing software security
vulnerabilities. Besides, emerging quantum machine learning can be promising in
addressing SSC attacks. Considering the distinction between traditional and
quantum machine learning, performance could be varies based on the proportions
of the experimenting dataset. In this paper, we conduct a comparative analysis
between quantum neural networks (QNN) and conventional neural networks (NN)
with a software supply chain attack dataset known as ClaMP. Our goal is to
distinguish the performance between QNN and NN and to conduct the experiment,
we develop two different models for QNN and NN by utilizing Pennylane for
quantum and TensorFlow and Keras for traditional respectively. We evaluated the
performance of both models with different proportions of the ClaMP dataset to
identify the f1 score, recall, precision, and accuracy. We also measure the
execution time to check the efficiency of both models. The demonstration result
indicates that execution time for QNN is slower than NN with a higher
percentage of datasets. Due to recent advancements in QNN, a large level of
experiments shall be carried out to understand both models accurately in our
future research.
| [
{
"version": "v1",
"created": "Wed, 31 May 2023 06:06:28 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Akter",
"Mst Shapna",
""
],
[
"Faruk",
"Md Jobair Hossain",
""
],
[
"Anjum",
"Nafisa",
""
],
[
"Masum",
"Mohammad",
""
],
[
"Shahriar",
"Hossain",
""
],
[
"Rahman",
"Akond",
""
],
[
"Wu",
"Fan",
""
],
[
"Cuzzocrea",
"Alfredo",
""
]
] | TITLE: Software Supply Chain Vulnerabilities Detection in Source Code:
Performance Comparison between Traditional and Quantum Machine Learning
Algorithms
ABSTRACT: The software supply chain (SSC) attack has become one of the crucial issues
that are being increased rapidly with the advancement of the software
development domain. In general, SSC attacks execute during the software
development processes lead to vulnerabilities in software products targeting
downstream customers and even involved stakeholders. Machine Learning
approaches are proven in detecting and preventing software security
vulnerabilities. Besides, emerging quantum machine learning can be promising in
addressing SSC attacks. Considering the distinction between traditional and
quantum machine learning, performance could be varies based on the proportions
of the experimenting dataset. In this paper, we conduct a comparative analysis
between quantum neural networks (QNN) and conventional neural networks (NN)
with a software supply chain attack dataset known as ClaMP. Our goal is to
distinguish the performance between QNN and NN and to conduct the experiment,
we develop two different models for QNN and NN by utilizing Pennylane for
quantum and TensorFlow and Keras for traditional respectively. We evaluated the
performance of both models with different proportions of the ClaMP dataset to
identify the f1 score, recall, precision, and accuracy. We also measure the
execution time to check the efficiency of both models. The demonstration result
indicates that execution time for QNN is slower than NN with a higher
percentage of datasets. Due to recent advancements in QNN, a large level of
experiments shall be carried out to understand both models accurately in our
future research.
|
2306.08073 | Dening Lu | Dening Lu, Jun Zhou, Kyle Yilin Gao, Dilong Li, Jing Du, Linlin Xu,
Jonathan Li | Dynamic Clustering Transformer Network for Point Cloud Segmentation | 8 pages, 4 figures | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Point cloud segmentation is one of the most important tasks in computer
vision with widespread scientific, industrial, and commercial applications. The
research thereof has resulted in many breakthroughs in 3D object and scene
understanding. Previous methods typically utilized hierarchical architectures
for feature representation. However, the commonly used sampling and grouping
methods in hierarchical networks are only based on point-wise three-dimensional
coordinates, ignoring local semantic homogeneity of point clusters.
Additionally, the prevalent Farthest Point Sampling (FPS) method is often a
computational bottleneck. To address these issues, we propose a novel 3D point
cloud representation network, called Dynamic Clustering Transformer Network
(DCTNet). It has an encoder-decoder architecture, allowing for both local and
global feature learning. Specifically, we propose novel semantic feature-based
dynamic sampling and clustering methods in the encoder, which enables the model
to be aware of local semantic homogeneity for local feature aggregation.
Furthermore, in the decoder, we propose an efficient semantic feature-guided
upsampling method. Our method was evaluated on an object-based dataset
(ShapeNet), an urban navigation dataset (Toronto-3D), and a multispectral LiDAR
dataset, verifying the performance of DCTNet across a wide variety of practical
engineering applications. The inference speed of DCTNet is 3.8-16.8$\times$
faster than existing State-of-the-Art (SOTA) models on the ShapeNet dataset,
while achieving an instance-wise mIoU of $86.6\%$, the current top score. Our
method similarly outperforms previous methods on the other datasets, verifying
it as the new State-of-the-Art in point cloud segmentation.
| [
{
"version": "v1",
"created": "Tue, 30 May 2023 01:11:05 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Lu",
"Dening",
""
],
[
"Zhou",
"Jun",
""
],
[
"Gao",
"Kyle Yilin",
""
],
[
"Li",
"Dilong",
""
],
[
"Du",
"Jing",
""
],
[
"Xu",
"Linlin",
""
],
[
"Li",
"Jonathan",
""
]
] | TITLE: Dynamic Clustering Transformer Network for Point Cloud Segmentation
ABSTRACT: Point cloud segmentation is one of the most important tasks in computer
vision with widespread scientific, industrial, and commercial applications. The
research thereof has resulted in many breakthroughs in 3D object and scene
understanding. Previous methods typically utilized hierarchical architectures
for feature representation. However, the commonly used sampling and grouping
methods in hierarchical networks are only based on point-wise three-dimensional
coordinates, ignoring local semantic homogeneity of point clusters.
Additionally, the prevalent Farthest Point Sampling (FPS) method is often a
computational bottleneck. To address these issues, we propose a novel 3D point
cloud representation network, called Dynamic Clustering Transformer Network
(DCTNet). It has an encoder-decoder architecture, allowing for both local and
global feature learning. Specifically, we propose novel semantic feature-based
dynamic sampling and clustering methods in the encoder, which enables the model
to be aware of local semantic homogeneity for local feature aggregation.
Furthermore, in the decoder, we propose an efficient semantic feature-guided
upsampling method. Our method was evaluated on an object-based dataset
(ShapeNet), an urban navigation dataset (Toronto-3D), and a multispectral LiDAR
dataset, verifying the performance of DCTNet across a wide variety of practical
engineering applications. The inference speed of DCTNet is 3.8-16.8$\times$
faster than existing State-of-the-Art (SOTA) models on the ShapeNet dataset,
while achieving an instance-wise mIoU of $86.6\%$, the current top score. Our
method similarly outperforms previous methods on the other datasets, verifying
it as the new State-of-the-Art in point cloud segmentation.
|
2306.08116 | Brendan Artley Mr | Brendan Artley, Greg Mehdiyev | CipherSniffer: Classifying Cipher Types | null | null | null | null | cs.CL cs.CR cs.LG | http://creativecommons.org/licenses/by/4.0/ | Ciphers are a powerful tool for encrypting communication. There are many
different cipher types, which makes it computationally expensive to solve a
cipher using brute force. In this paper, we frame the decryption task as a
classification problem. We first create a dataset of transpositions,
substitutions, text reversals, word reversals, sentence shifts, and unencrypted
text. Then, we evaluate the performance of various tokenizer-model combinations
on this task.
| [
{
"version": "v1",
"created": "Tue, 13 Jun 2023 20:18:24 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Artley",
"Brendan",
""
],
[
"Mehdiyev",
"Greg",
""
]
] | TITLE: CipherSniffer: Classifying Cipher Types
ABSTRACT: Ciphers are a powerful tool for encrypting communication. There are many
different cipher types, which makes it computationally expensive to solve a
cipher using brute force. In this paper, we frame the decryption task as a
classification problem. We first create a dataset of transpositions,
substitutions, text reversals, word reversals, sentence shifts, and unencrypted
text. Then, we evaluate the performance of various tokenizer-model combinations
on this task.
|
2306.08131 | Nanxin Chen | Nanxin Chen, Izhak Shafran, Yu Zhang, Chung-Cheng Chiu, Hagen Soltau,
James Qin, Yonghui Wu | Efficient Adapters for Giant Speech Models | null | null | null | null | eess.AS cs.SD | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Large pre-trained speech models are widely used as the de-facto paradigm,
especially in scenarios when there is a limited amount of labeled data
available. However, finetuning all parameters from the self-supervised learned
model can be computationally expensive, and becomes infeasiable as the size of
the model and the number of downstream tasks scales. In this paper, we propose
a novel approach called Two Parallel Adapter (TPA) that is inserted into the
conformer-based model pre-trained model instead. TPA is based on systematic
studies of the residual adapter, a popular approach for finetuning a subset of
parameters. We evaluate TPA on various public benchmarks and experiment results
demonstrates its superior performance, which is close to the full finetuning on
different datasets and speech tasks. These results show that TPA is an
effective and efficient approach for serving large pre-trained speech models.
Ablation studies show that TPA can also be pruned, especially for lower blocks.
| [
{
"version": "v1",
"created": "Tue, 13 Jun 2023 20:51:00 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Chen",
"Nanxin",
""
],
[
"Shafran",
"Izhak",
""
],
[
"Zhang",
"Yu",
""
],
[
"Chiu",
"Chung-Cheng",
""
],
[
"Soltau",
"Hagen",
""
],
[
"Qin",
"James",
""
],
[
"Wu",
"Yonghui",
""
]
] | TITLE: Efficient Adapters for Giant Speech Models
ABSTRACT: Large pre-trained speech models are widely used as the de-facto paradigm,
especially in scenarios when there is a limited amount of labeled data
available. However, finetuning all parameters from the self-supervised learned
model can be computationally expensive, and becomes infeasiable as the size of
the model and the number of downstream tasks scales. In this paper, we propose
a novel approach called Two Parallel Adapter (TPA) that is inserted into the
conformer-based model pre-trained model instead. TPA is based on systematic
studies of the residual adapter, a popular approach for finetuning a subset of
parameters. We evaluate TPA on various public benchmarks and experiment results
demonstrates its superior performance, which is close to the full finetuning on
different datasets and speech tasks. These results show that TPA is an
effective and efficient approach for serving large pre-trained speech models.
Ablation studies show that TPA can also be pruned, especially for lower blocks.
|
2306.08132 | Dylan Turpin | Dylan Turpin, Tao Zhong, Shutong Zhang, Guanglei Zhu, Jingzhou Liu,
Ritvik Singh, Eric Heiden, Miles Macklin, Stavros Tsogkas, Sven Dickinson,
Animesh Garg | Fast-Grasp'D: Dexterous Multi-finger Grasp Generation Through
Differentiable Simulation | null | null | null | null | cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Multi-finger grasping relies on high quality training data, which is hard to
obtain: human data is hard to transfer and synthetic data relies on simplifying
assumptions that reduce grasp quality. By making grasp simulation
differentiable, and contact dynamics amenable to gradient-based optimization,
we accelerate the search for high-quality grasps with fewer limiting
assumptions. We present Grasp'D-1M: a large-scale dataset for multi-finger
robotic grasping, synthesized with Fast- Grasp'D, a novel differentiable
grasping simulator. Grasp'D- 1M contains one million training examples for
three robotic hands (three, four and five-fingered), each with multimodal
visual inputs (RGB+depth+segmentation, available in mono and stereo). Grasp
synthesis with Fast-Grasp'D is 10x faster than GraspIt! and 20x faster than the
prior Grasp'D differentiable simulator. Generated grasps are more stable and
contact-rich than GraspIt! grasps, regardless of the distance threshold used
for contact generation. We validate the usefulness of our dataset by retraining
an existing vision-based grasping pipeline on Grasp'D-1M, and showing a
dramatic increase in model performance, predicting grasps with 30% more
contact, a 33% higher epsilon metric, and 35% lower simulated displacement.
Additional details at https://dexgrasp.github.io.
| [
{
"version": "v1",
"created": "Tue, 13 Jun 2023 20:54:07 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Turpin",
"Dylan",
""
],
[
"Zhong",
"Tao",
""
],
[
"Zhang",
"Shutong",
""
],
[
"Zhu",
"Guanglei",
""
],
[
"Liu",
"Jingzhou",
""
],
[
"Singh",
"Ritvik",
""
],
[
"Heiden",
"Eric",
""
],
[
"Macklin",
"Miles",
""
],
[
"Tsogkas",
"Stavros",
""
],
[
"Dickinson",
"Sven",
""
],
[
"Garg",
"Animesh",
""
]
] | TITLE: Fast-Grasp'D: Dexterous Multi-finger Grasp Generation Through
Differentiable Simulation
ABSTRACT: Multi-finger grasping relies on high quality training data, which is hard to
obtain: human data is hard to transfer and synthetic data relies on simplifying
assumptions that reduce grasp quality. By making grasp simulation
differentiable, and contact dynamics amenable to gradient-based optimization,
we accelerate the search for high-quality grasps with fewer limiting
assumptions. We present Grasp'D-1M: a large-scale dataset for multi-finger
robotic grasping, synthesized with Fast- Grasp'D, a novel differentiable
grasping simulator. Grasp'D- 1M contains one million training examples for
three robotic hands (three, four and five-fingered), each with multimodal
visual inputs (RGB+depth+segmentation, available in mono and stereo). Grasp
synthesis with Fast-Grasp'D is 10x faster than GraspIt! and 20x faster than the
prior Grasp'D differentiable simulator. Generated grasps are more stable and
contact-rich than GraspIt! grasps, regardless of the distance threshold used
for contact generation. We validate the usefulness of our dataset by retraining
an existing vision-based grasping pipeline on Grasp'D-1M, and showing a
dramatic increase in model performance, predicting grasps with 30% more
contact, a 33% higher epsilon metric, and 35% lower simulated displacement.
Additional details at https://dexgrasp.github.io.
|
2306.08189 | Thinh Hung Truong | Thinh Hung Truong, Timothy Baldwin, Karin Verspoor, Trevor Cohn | Language models are not naysayers: An analysis of language models on
negation benchmarks | null | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Negation has been shown to be a major bottleneck for masked language models,
such as BERT. However, whether this finding still holds for larger-sized
auto-regressive language models (``LLMs'') has not been studied
comprehensively. With the ever-increasing volume of research and applications
of LLMs, we take a step back to evaluate the ability of current-generation LLMs
to handle negation, a fundamental linguistic phenomenon that is central to
language understanding. We evaluate different LLMs -- including the open-source
GPT-neo, GPT-3, and InstructGPT -- against a wide range of negation benchmarks.
Through systematic experimentation with varying model sizes and prompts, we
show that LLMs have several limitations including insensitivity to the presence
of negation, an inability to capture the lexical semantics of negation, and a
failure to reason under negation.
| [
{
"version": "v1",
"created": "Wed, 14 Jun 2023 01:16:37 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Truong",
"Thinh Hung",
""
],
[
"Baldwin",
"Timothy",
""
],
[
"Verspoor",
"Karin",
""
],
[
"Cohn",
"Trevor",
""
]
] | TITLE: Language models are not naysayers: An analysis of language models on
negation benchmarks
ABSTRACT: Negation has been shown to be a major bottleneck for masked language models,
such as BERT. However, whether this finding still holds for larger-sized
auto-regressive language models (``LLMs'') has not been studied
comprehensively. With the ever-increasing volume of research and applications
of LLMs, we take a step back to evaluate the ability of current-generation LLMs
to handle negation, a fundamental linguistic phenomenon that is central to
language understanding. We evaluate different LLMs -- including the open-source
GPT-neo, GPT-3, and InstructGPT -- against a wide range of negation benchmarks.
Through systematic experimentation with varying model sizes and prompts, we
show that LLMs have several limitations including insensitivity to the presence
of negation, an inability to capture the lexical semantics of negation, and a
failure to reason under negation.
|
2306.08190 | Mars Gokturk Buchholz | Mars Gokturk Buchholz | Assessing the Effectiveness of GPT-3 in Detecting False Political
Statements: A Case Study on the LIAR Dataset | 7 pages, 1 figure | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | The detection of political fake statements is crucial for maintaining
information integrity and preventing the spread of misinformation in society.
Historically, state-of-the-art machine learning models employed various methods
for detecting deceptive statements. These methods include the use of metadata
(W. Wang et al., 2018), n-grams analysis (Singh et al., 2021), and linguistic
(Wu et al., 2022) and stylometric (Islam et al., 2020) features. Recent
advancements in large language models, such as GPT-3 (Brown et al., 2020) have
achieved state-of-the-art performance on a wide range of tasks. In this study,
we conducted experiments with GPT-3 on the LIAR dataset (W. Wang et al., 2018)
and achieved higher accuracy than state-of-the-art models without using any
additional meta or linguistic features. Additionally, we experimented with
zero-shot learning using a carefully designed prompt and achieved near
state-of-the-art performance. An advantage of this approach is that the model
provided evidence for its decision, which adds transparency to the model's
decision-making and offers a chance for users to verify the validity of the
evidence provided.
| [
{
"version": "v1",
"created": "Wed, 14 Jun 2023 01:16:49 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Buchholz",
"Mars Gokturk",
""
]
] | TITLE: Assessing the Effectiveness of GPT-3 in Detecting False Political
Statements: A Case Study on the LIAR Dataset
ABSTRACT: The detection of political fake statements is crucial for maintaining
information integrity and preventing the spread of misinformation in society.
Historically, state-of-the-art machine learning models employed various methods
for detecting deceptive statements. These methods include the use of metadata
(W. Wang et al., 2018), n-grams analysis (Singh et al., 2021), and linguistic
(Wu et al., 2022) and stylometric (Islam et al., 2020) features. Recent
advancements in large language models, such as GPT-3 (Brown et al., 2020) have
achieved state-of-the-art performance on a wide range of tasks. In this study,
we conducted experiments with GPT-3 on the LIAR dataset (W. Wang et al., 2018)
and achieved higher accuracy than state-of-the-art models without using any
additional meta or linguistic features. Additionally, we experimented with
zero-shot learning using a carefully designed prompt and achieved near
state-of-the-art performance. An advantage of this approach is that the model
provided evidence for its decision, which adds transparency to the model's
decision-making and offers a chance for users to verify the validity of the
evidence provided.
|
2306.08194 | Wei Ju | Jingyang Yuan, Xiao Luo, Yifang Qin, Yusheng Zhao, Wei Ju, Ming Zhang | Learning on Graphs under Label Noise | Accepted by IEEE International Conference on Acoustics, Speech and
Signal Processing (ICASSP 2023) | null | null | null | cs.LG cs.AI cs.IR cs.SI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Node classification on graphs is a significant task with a wide range of
applications, including social analysis and anomaly detection. Even though
graph neural networks (GNNs) have produced promising results on this task,
current techniques often presume that label information of nodes is accurate,
which may not be the case in real-world applications. To tackle this issue, we
investigate the problem of learning on graphs with label noise and develop a
novel approach dubbed Consistent Graph Neural Network (CGNN) to solve it.
Specifically, we employ graph contrastive learning as a regularization term,
which promotes two views of augmented nodes to have consistent representations.
Since this regularization term cannot utilize label information, it can enhance
the robustness of node representations to label noise. Moreover, to detect
noisy labels on the graph, we present a sample selection technique based on the
homophily assumption, which identifies noisy nodes by measuring the consistency
between the labels with their neighbors. Finally, we purify these confident
noisy labels to permit efficient semantic graph learning. Extensive experiments
on three well-known benchmark datasets demonstrate the superiority of our CGNN
over competing approaches.
| [
{
"version": "v1",
"created": "Wed, 14 Jun 2023 01:38:01 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Yuan",
"Jingyang",
""
],
[
"Luo",
"Xiao",
""
],
[
"Qin",
"Yifang",
""
],
[
"Zhao",
"Yusheng",
""
],
[
"Ju",
"Wei",
""
],
[
"Zhang",
"Ming",
""
]
] | TITLE: Learning on Graphs under Label Noise
ABSTRACT: Node classification on graphs is a significant task with a wide range of
applications, including social analysis and anomaly detection. Even though
graph neural networks (GNNs) have produced promising results on this task,
current techniques often presume that label information of nodes is accurate,
which may not be the case in real-world applications. To tackle this issue, we
investigate the problem of learning on graphs with label noise and develop a
novel approach dubbed Consistent Graph Neural Network (CGNN) to solve it.
Specifically, we employ graph contrastive learning as a regularization term,
which promotes two views of augmented nodes to have consistent representations.
Since this regularization term cannot utilize label information, it can enhance
the robustness of node representations to label noise. Moreover, to detect
noisy labels on the graph, we present a sample selection technique based on the
homophily assumption, which identifies noisy nodes by measuring the consistency
between the labels with their neighbors. Finally, we purify these confident
noisy labels to permit efficient semantic graph learning. Extensive experiments
on three well-known benchmark datasets demonstrate the superiority of our CGNN
over competing approaches.
|
2306.08198 | Milan Aryal | Milan Aryal and Nasim Yahyasoltani | Explainable and Position-Aware Learning in Digital Pathology | null | null | null | null | eess.IV cs.CV cs.LG | http://creativecommons.org/licenses/by/4.0/ | Encoding whole slide images (WSI) as graphs is well motivated since it makes
it possible for the gigapixel resolution WSI to be represented in its entirety
for the purpose of graph learning. To this end, WSIs can be broken into smaller
patches that represent the nodes of the graph. Then, graph-based learning
methods can be utilized for the grading and classification of cancer. Message
passing among neighboring nodes is the foundation of graph-based learning
methods. However, they do not take into consideration any positional
information for any of the patches, and if two patches are found in
topologically isomorphic neighborhoods, their embeddings are nearly similar to
one another. In this work, classification of cancer from WSIs is performed with
positional embedding and graph attention. In order to represent the positional
embedding of the nodes in graph classification, the proposed method makes use
of spline convolutional neural networks (CNN). The algorithm is then tested
with the WSI dataset for grading prostate cancer and kidney cancer. A
comparison of the proposed method with leading approaches in cancer diagnosis
and grading verify improved performance. The identification of cancerous
regions in WSIs is another critical task in cancer diagnosis. In this work, the
explainability of the proposed model is also addressed. A gradient-based
explainbility approach is used to generate the saliency mapping for the WSIs.
This can be used to look into regions of WSI that are responsible for cancer
diagnosis thus rendering the proposed model explainable.
| [
{
"version": "v1",
"created": "Wed, 14 Jun 2023 01:53:17 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Aryal",
"Milan",
""
],
[
"Yahyasoltani",
"Nasim",
""
]
] | TITLE: Explainable and Position-Aware Learning in Digital Pathology
ABSTRACT: Encoding whole slide images (WSI) as graphs is well motivated since it makes
it possible for the gigapixel resolution WSI to be represented in its entirety
for the purpose of graph learning. To this end, WSIs can be broken into smaller
patches that represent the nodes of the graph. Then, graph-based learning
methods can be utilized for the grading and classification of cancer. Message
passing among neighboring nodes is the foundation of graph-based learning
methods. However, they do not take into consideration any positional
information for any of the patches, and if two patches are found in
topologically isomorphic neighborhoods, their embeddings are nearly similar to
one another. In this work, classification of cancer from WSIs is performed with
positional embedding and graph attention. In order to represent the positional
embedding of the nodes in graph classification, the proposed method makes use
of spline convolutional neural networks (CNN). The algorithm is then tested
with the WSI dataset for grading prostate cancer and kidney cancer. A
comparison of the proposed method with leading approaches in cancer diagnosis
and grading verify improved performance. The identification of cancerous
regions in WSIs is another critical task in cancer diagnosis. In this work, the
explainability of the proposed model is also addressed. A gradient-based
explainbility approach is used to generate the saliency mapping for the WSIs.
This can be used to look into regions of WSI that are responsible for cancer
diagnosis thus rendering the proposed model explainable.
|
2306.08200 | Zhiyuan Hu | Zhiyuan Hu, Jiancheng Lyu, Dashan Gao, Nuno Vasconcelos | POP: Prompt Of Prompts for Continual Learning | null | null | null | null | cs.CV cs.LG | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Continual learning (CL) has attracted increasing attention in the recent
past. It aims to mimic the human ability to learn new concepts without
catastrophic forgetting. While existing CL methods accomplish this to some
extent, they are still prone to semantic drift of the learned feature space.
Foundation models, which are endowed with a robust feature representation,
learned from very large datasets, provide an interesting substrate for the
solution of the CL problem. Recent work has also shown that they can be adapted
to specific tasks by prompt tuning techniques that leave the generality of the
representation mostly unscathed. An open question is, however, how to learn
both prompts that are task specific and prompts that are global, i.e. capture
cross-task information. In this work, we propose the Prompt Of Prompts (POP)
model, which addresses this goal by progressively learning a group of
task-specified prompts and a group of global prompts, denoted as POP, to
integrate information from the former. We show that a foundation model equipped
with POP learning is able to outperform classic CL methods by a significant
margin. Moreover, as prompt tuning only requires a small set of training
samples, POP is able to perform CL in the few-shot setting, while still
outperforming competing methods trained on the entire dataset.
| [
{
"version": "v1",
"created": "Wed, 14 Jun 2023 02:09:26 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Hu",
"Zhiyuan",
""
],
[
"Lyu",
"Jiancheng",
""
],
[
"Gao",
"Dashan",
""
],
[
"Vasconcelos",
"Nuno",
""
]
] | TITLE: POP: Prompt Of Prompts for Continual Learning
ABSTRACT: Continual learning (CL) has attracted increasing attention in the recent
past. It aims to mimic the human ability to learn new concepts without
catastrophic forgetting. While existing CL methods accomplish this to some
extent, they are still prone to semantic drift of the learned feature space.
Foundation models, which are endowed with a robust feature representation,
learned from very large datasets, provide an interesting substrate for the
solution of the CL problem. Recent work has also shown that they can be adapted
to specific tasks by prompt tuning techniques that leave the generality of the
representation mostly unscathed. An open question is, however, how to learn
both prompts that are task specific and prompts that are global, i.e. capture
cross-task information. In this work, we propose the Prompt Of Prompts (POP)
model, which addresses this goal by progressively learning a group of
task-specified prompts and a group of global prompts, denoted as POP, to
integrate information from the former. We show that a foundation model equipped
with POP learning is able to outperform classic CL methods by a significant
margin. Moreover, as prompt tuning only requires a small set of training
samples, POP is able to perform CL in the few-shot setting, while still
outperforming competing methods trained on the entire dataset.
|
2306.08204 | Sundong Kim | Jaehyun Park, Jaegyun Im, Sanha Hwang, Mintaek Lim, Sabina Ualibekova,
Sejin Kim, Sundong Kim | Unraveling the ARC Puzzle: Mimicking Human Solutions with Object-Centric
Decision Transformer | null | null | null | null | cs.AI cs.LG | http://creativecommons.org/licenses/by/4.0/ | In the pursuit of artificial general intelligence (AGI), we tackle
Abstraction and Reasoning Corpus (ARC) tasks using a novel two-pronged
approach. We employ the Decision Transformer in an imitation learning paradigm
to model human problem-solving, and introduce an object detection algorithm,
the Push and Pull clustering method. This dual strategy enhances AI's ARC
problem-solving skills and provides insights for AGI progression. Yet, our work
reveals the need for advanced data collection tools, robust training datasets,
and refined model structures. This study highlights potential improvements for
Decision Transformers and propels future AGI research.
| [
{
"version": "v1",
"created": "Wed, 14 Jun 2023 02:12:49 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Park",
"Jaehyun",
""
],
[
"Im",
"Jaegyun",
""
],
[
"Hwang",
"Sanha",
""
],
[
"Lim",
"Mintaek",
""
],
[
"Ualibekova",
"Sabina",
""
],
[
"Kim",
"Sejin",
""
],
[
"Kim",
"Sundong",
""
]
] | TITLE: Unraveling the ARC Puzzle: Mimicking Human Solutions with Object-Centric
Decision Transformer
ABSTRACT: In the pursuit of artificial general intelligence (AGI), we tackle
Abstraction and Reasoning Corpus (ARC) tasks using a novel two-pronged
approach. We employ the Decision Transformer in an imitation learning paradigm
to model human problem-solving, and introduce an object detection algorithm,
the Push and Pull clustering method. This dual strategy enhances AI's ARC
problem-solving skills and provides insights for AGI progression. Yet, our work
reveals the need for advanced data collection tools, robust training datasets,
and refined model structures. This study highlights potential improvements for
Decision Transformers and propels future AGI research.
|
2306.08213 | Zhusi Zhong | Zhusi Zhong, Jie Li, Lulu Bi, Li Yang, Ihab Kamel, Rama Chellappa,
Xinbo Gao, Harrison Bai, Zhicheng Jiao | SMC-UDA: Structure-Modal Constraint for Unsupervised Cross-Domain Renal
Segmentation | conference | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by/4.0/ | Medical image segmentation based on deep learning often fails when deployed
on images from a different domain. The domain adaptation methods aim to solve
domain-shift challenges, but still face some problems. The transfer learning
methods require annotation on the target domain, and the generative
unsupervised domain adaptation (UDA) models ignore domain-specific
representations, whose generated quality highly restricts segmentation
performance. In this study, we propose a novel Structure-Modal Constrained
(SMC) UDA framework based on a discriminative paradigm and introduce edge
structure as a bridge between domains. The proposed multi-modal learning
backbone distills structure information from image texture to distinguish
domain-invariant edge structure. With the structure-constrained self-learning
and progressive ROI, our methods segment the kidney by locating the 3D spatial
structure of the edge. We evaluated SMC-UDA on public renal segmentation
datasets, adapting from the labeled source domain (CT) to the unlabeled target
domain (CT/MRI). The experiments show that our proposed SMC-UDA has a strong
generalization and outperforms generative UDA methods.
| [
{
"version": "v1",
"created": "Wed, 14 Jun 2023 02:57:23 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Zhong",
"Zhusi",
""
],
[
"Li",
"Jie",
""
],
[
"Bi",
"Lulu",
""
],
[
"Yang",
"Li",
""
],
[
"Kamel",
"Ihab",
""
],
[
"Chellappa",
"Rama",
""
],
[
"Gao",
"Xinbo",
""
],
[
"Bai",
"Harrison",
""
],
[
"Jiao",
"Zhicheng",
""
]
] | TITLE: SMC-UDA: Structure-Modal Constraint for Unsupervised Cross-Domain Renal
Segmentation
ABSTRACT: Medical image segmentation based on deep learning often fails when deployed
on images from a different domain. The domain adaptation methods aim to solve
domain-shift challenges, but still face some problems. The transfer learning
methods require annotation on the target domain, and the generative
unsupervised domain adaptation (UDA) models ignore domain-specific
representations, whose generated quality highly restricts segmentation
performance. In this study, we propose a novel Structure-Modal Constrained
(SMC) UDA framework based on a discriminative paradigm and introduce edge
structure as a bridge between domains. The proposed multi-modal learning
backbone distills structure information from image texture to distinguish
domain-invariant edge structure. With the structure-constrained self-learning
and progressive ROI, our methods segment the kidney by locating the 3D spatial
structure of the edge. We evaluated SMC-UDA on public renal segmentation
datasets, adapting from the labeled source domain (CT) to the unlabeled target
domain (CT/MRI). The experiments show that our proposed SMC-UDA has a strong
generalization and outperforms generative UDA methods.
|
2306.08219 | Hongyang Liu | Xinghua Qu, Hongyang Liu, Zhu Sun, Xiang Yin, Yew Soon Ong, Lu Lu,
Zejun Ma | Towards Building Voice-based Conversational Recommender Systems:
Datasets, Potential Solutions, and Prospects | Accepted by SIGIR 2023 Resource Track | null | 10.1145/3539618.3591876 | null | cs.IR cs.SD eess.AS | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Conversational recommender systems (CRSs) have become crucial emerging
research topics in the field of RSs, thanks to their natural advantages of
explicitly acquiring user preferences via interactive conversations and
revealing the reasons behind recommendations. However, the majority of current
CRSs are text-based, which is less user-friendly and may pose challenges for
certain users, such as those with visual impairments or limited writing and
reading abilities. Therefore, for the first time, this paper investigates the
potential of voice-based CRS (VCRSs) to revolutionize the way users interact
with RSs in a natural, intuitive, convenient, and accessible fashion. To
support such studies, we create two VCRSs benchmark datasets in the e-commerce
and movie domains, after realizing the lack of such datasets through an
exhaustive literature review. Specifically, we first empirically verify the
benefits and necessity of creating such datasets. Thereafter, we convert the
user-item interactions to text-based conversations through the ChatGPT-driven
prompts for generating diverse and natural templates, and then synthesize the
corresponding audios via the text-to-speech model. Meanwhile, a number of
strategies are delicately designed to ensure the naturalness and high quality
of voice conversations. On this basis, we further explore the potential
solutions and point out possible directions to build end-to-end VCRSs by
seamlessly extracting and integrating voice-based inputs, thus delivering
performance-enhanced, self-explainable, and user-friendly VCRSs. Our study aims
to establish the foundation and motivate further pioneering research in the
emerging field of VCRSs. This aligns with the principles of explainable AI and
AI for social good, viz., utilizing technology's potential to create a fair,
sustainable, and just world.
| [
{
"version": "v1",
"created": "Wed, 14 Jun 2023 03:17:02 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Qu",
"Xinghua",
""
],
[
"Liu",
"Hongyang",
""
],
[
"Sun",
"Zhu",
""
],
[
"Yin",
"Xiang",
""
],
[
"Ong",
"Yew Soon",
""
],
[
"Lu",
"Lu",
""
],
[
"Ma",
"Zejun",
""
]
] | TITLE: Towards Building Voice-based Conversational Recommender Systems:
Datasets, Potential Solutions, and Prospects
ABSTRACT: Conversational recommender systems (CRSs) have become crucial emerging
research topics in the field of RSs, thanks to their natural advantages of
explicitly acquiring user preferences via interactive conversations and
revealing the reasons behind recommendations. However, the majority of current
CRSs are text-based, which is less user-friendly and may pose challenges for
certain users, such as those with visual impairments or limited writing and
reading abilities. Therefore, for the first time, this paper investigates the
potential of voice-based CRS (VCRSs) to revolutionize the way users interact
with RSs in a natural, intuitive, convenient, and accessible fashion. To
support such studies, we create two VCRSs benchmark datasets in the e-commerce
and movie domains, after realizing the lack of such datasets through an
exhaustive literature review. Specifically, we first empirically verify the
benefits and necessity of creating such datasets. Thereafter, we convert the
user-item interactions to text-based conversations through the ChatGPT-driven
prompts for generating diverse and natural templates, and then synthesize the
corresponding audios via the text-to-speech model. Meanwhile, a number of
strategies are delicately designed to ensure the naturalness and high quality
of voice conversations. On this basis, we further explore the potential
solutions and point out possible directions to build end-to-end VCRSs by
seamlessly extracting and integrating voice-based inputs, thus delivering
performance-enhanced, self-explainable, and user-friendly VCRSs. Our study aims
to establish the foundation and motivate further pioneering research in the
emerging field of VCRSs. This aligns with the principles of explainable AI and
AI for social good, viz., utilizing technology's potential to create a fair,
sustainable, and just world.
|
2306.08223 | Zhigang Kan | Zhigang Kan, Linbo Qiao, Hao Yu, Liwen Peng, Yifu Gao, Dongsheng Li | Protecting User Privacy in Remote Conversational Systems: A
Privacy-Preserving framework based on text sanitization | 9 pages, 2 figures | null | null | null | cs.CR cs.HC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Large Language Models (LLMs) are gaining increasing attention due to their
exceptional performance across numerous tasks. As a result, the general public
utilize them as an influential tool for boosting their productivity while
natural language processing researchers endeavor to employ them in solving
existing or new research problems. Unfortunately, individuals can only access
such powerful AIs through APIs, which ultimately leads to the transmission of
raw data to the models' providers and increases the possibility of privacy data
leakage. Current privacy-preserving methods for cloud-deployed language models
aim to protect privacy information in the pre-training dataset or during the
model training phase. However, they do not meet the specific challenges
presented by the remote access approach of new large-scale language models.
This paper introduces a novel task, "User Privacy Protection for Dialogue
Models," which aims to safeguard sensitive user information from any possible
disclosure while conversing with chatbots. We also present an evaluation scheme
for this task, which covers evaluation metrics for privacy protection, data
availability, and resistance to simulation attacks. Moreover, we propose the
first framework for this task, namely privacy protection through text
sanitization. Before sending the input to remote large models, it filters out
the sensitive information, using several rounds of text sanitization based on
privacy types that users define. Upon receiving responses from the larger
model, our framework automatically restores privacy to ensure that the
conversation goes smoothly, without intervention from the privacy filter.
Experiments based on real-world datasets demonstrate the efficacy of our
privacy-preserving approach against eavesdropping from potential attackers.
| [
{
"version": "v1",
"created": "Wed, 14 Jun 2023 03:28:51 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Kan",
"Zhigang",
""
],
[
"Qiao",
"Linbo",
""
],
[
"Yu",
"Hao",
""
],
[
"Peng",
"Liwen",
""
],
[
"Gao",
"Yifu",
""
],
[
"Li",
"Dongsheng",
""
]
] | TITLE: Protecting User Privacy in Remote Conversational Systems: A
Privacy-Preserving framework based on text sanitization
ABSTRACT: Large Language Models (LLMs) are gaining increasing attention due to their
exceptional performance across numerous tasks. As a result, the general public
utilize them as an influential tool for boosting their productivity while
natural language processing researchers endeavor to employ them in solving
existing or new research problems. Unfortunately, individuals can only access
such powerful AIs through APIs, which ultimately leads to the transmission of
raw data to the models' providers and increases the possibility of privacy data
leakage. Current privacy-preserving methods for cloud-deployed language models
aim to protect privacy information in the pre-training dataset or during the
model training phase. However, they do not meet the specific challenges
presented by the remote access approach of new large-scale language models.
This paper introduces a novel task, "User Privacy Protection for Dialogue
Models," which aims to safeguard sensitive user information from any possible
disclosure while conversing with chatbots. We also present an evaluation scheme
for this task, which covers evaluation metrics for privacy protection, data
availability, and resistance to simulation attacks. Moreover, we propose the
first framework for this task, namely privacy protection through text
sanitization. Before sending the input to remote large models, it filters out
the sensitive information, using several rounds of text sanitization based on
privacy types that users define. Upon receiving responses from the larger
model, our framework automatically restores privacy to ensure that the
conversation goes smoothly, without intervention from the privacy filter.
Experiments based on real-world datasets demonstrate the efficacy of our
privacy-preserving approach against eavesdropping from potential attackers.
|
2306.08232 | Shunyu Liu | Shunyu Liu, Yunpeng Qing, Shuqi Xu, Hongyan Wu, Jiangtao Zhang,
Jingyuan Cong, Tianhao Chen, Yunfu Liu, Mingli Song | Curricular Subgoals for Inverse Reinforcement Learning | null | null | null | null | cs.LG cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Inverse Reinforcement Learning (IRL) aims to reconstruct the reward function
from expert demonstrations to facilitate policy learning, and has demonstrated
its remarkable success in imitation learning. To promote expert-like behavior,
existing IRL methods mainly focus on learning global reward functions to
minimize the trajectory difference between the imitator and the expert.
However, these global designs are still limited by the redundant noise and
error propagation problems, leading to the unsuitable reward assignment and
thus downgrading the agent capability in complex multi-stage tasks. In this
paper, we propose a novel Curricular Subgoal-based Inverse Reinforcement
Learning (CSIRL) framework, that explicitly disentangles one task with several
local subgoals to guide agent imitation. Specifically, CSIRL firstly introduces
decision uncertainty of the trained agent over expert trajectories to
dynamically select subgoals, which directly determines the exploration boundary
of different task stages. To further acquire local reward functions for each
stage, we customize a meta-imitation objective based on these curricular
subgoals to train an intrinsic reward generator. Experiments on the D4RL and
autonomous driving benchmarks demonstrate that the proposed methods yields
results superior to the state-of-the-art counterparts, as well as better
interpretability. Our code is available at https://github.com/Plankson/CSIRL.
| [
{
"version": "v1",
"created": "Wed, 14 Jun 2023 04:06:41 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Liu",
"Shunyu",
""
],
[
"Qing",
"Yunpeng",
""
],
[
"Xu",
"Shuqi",
""
],
[
"Wu",
"Hongyan",
""
],
[
"Zhang",
"Jiangtao",
""
],
[
"Cong",
"Jingyuan",
""
],
[
"Chen",
"Tianhao",
""
],
[
"Liu",
"Yunfu",
""
],
[
"Song",
"Mingli",
""
]
] | TITLE: Curricular Subgoals for Inverse Reinforcement Learning
ABSTRACT: Inverse Reinforcement Learning (IRL) aims to reconstruct the reward function
from expert demonstrations to facilitate policy learning, and has demonstrated
its remarkable success in imitation learning. To promote expert-like behavior,
existing IRL methods mainly focus on learning global reward functions to
minimize the trajectory difference between the imitator and the expert.
However, these global designs are still limited by the redundant noise and
error propagation problems, leading to the unsuitable reward assignment and
thus downgrading the agent capability in complex multi-stage tasks. In this
paper, we propose a novel Curricular Subgoal-based Inverse Reinforcement
Learning (CSIRL) framework, that explicitly disentangles one task with several
local subgoals to guide agent imitation. Specifically, CSIRL firstly introduces
decision uncertainty of the trained agent over expert trajectories to
dynamically select subgoals, which directly determines the exploration boundary
of different task stages. To further acquire local reward functions for each
stage, we customize a meta-imitation objective based on these curricular
subgoals to train an intrinsic reward generator. Experiments on the D4RL and
autonomous driving benchmarks demonstrate that the proposed methods yields
results superior to the state-of-the-art counterparts, as well as better
interpretability. Our code is available at https://github.com/Plankson/CSIRL.
|
2306.08240 | Zhongyi Shui | Zhongyi Shui, Yizhi Zhao, Sunyi Zheng, Yunlong Zhang, Honglin Li,
Shichuan Zhang, Xiaoxuan Yu, Chenglu Zhu, Lin Yang | Semi-supervised Cell Recognition under Point Supervision | null | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Cell recognition is a fundamental task in digital histopathology image
analysis. Point-based cell recognition (PCR) methods normally require a vast
number of annotations, which is extremely costly, time-consuming and
labor-intensive. Semi-supervised learning (SSL) can provide a shortcut to make
full use of cell information in gigapixel whole slide images without exhaustive
labeling. However, research into semi-supervised point-based cell recognition
(SSPCR) remains largely overlooked. Previous SSPCR works are all built on
density map-based PCR models, which suffer from unsatisfactory accuracy, slow
inference speed and high sensitivity to hyper-parameters. To address these
issues, end-to-end PCR models are proposed recently. In this paper, we develop
a SSPCR framework suitable for the end-to-end PCR models for the first time.
Overall, we use the current models to generate pseudo labels for unlabeled
images, which are in turn utilized to supervise the models training. Besides,
we introduce a co-teaching strategy to overcome the confirmation bias problem
that generally exists in self-training. A distribution alignment technique is
also incorporated to produce high-quality, unbiased pseudo labels for unlabeled
data. Experimental results on four histopathology datasets concerning different
types of staining styles show the effectiveness and versatility of the proposed
framework. Code is available at
\textcolor{magenta}{\url{https://github.com/windygooo/SSPCR}
| [
{
"version": "v1",
"created": "Wed, 14 Jun 2023 04:56:31 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Shui",
"Zhongyi",
""
],
[
"Zhao",
"Yizhi",
""
],
[
"Zheng",
"Sunyi",
""
],
[
"Zhang",
"Yunlong",
""
],
[
"Li",
"Honglin",
""
],
[
"Zhang",
"Shichuan",
""
],
[
"Yu",
"Xiaoxuan",
""
],
[
"Zhu",
"Chenglu",
""
],
[
"Yang",
"Lin",
""
]
] | TITLE: Semi-supervised Cell Recognition under Point Supervision
ABSTRACT: Cell recognition is a fundamental task in digital histopathology image
analysis. Point-based cell recognition (PCR) methods normally require a vast
number of annotations, which is extremely costly, time-consuming and
labor-intensive. Semi-supervised learning (SSL) can provide a shortcut to make
full use of cell information in gigapixel whole slide images without exhaustive
labeling. However, research into semi-supervised point-based cell recognition
(SSPCR) remains largely overlooked. Previous SSPCR works are all built on
density map-based PCR models, which suffer from unsatisfactory accuracy, slow
inference speed and high sensitivity to hyper-parameters. To address these
issues, end-to-end PCR models are proposed recently. In this paper, we develop
a SSPCR framework suitable for the end-to-end PCR models for the first time.
Overall, we use the current models to generate pseudo labels for unlabeled
images, which are in turn utilized to supervise the models training. Besides,
we introduce a co-teaching strategy to overcome the confirmation bias problem
that generally exists in self-training. A distribution alignment technique is
also incorporated to produce high-quality, unbiased pseudo labels for unlabeled
data. Experimental results on four histopathology datasets concerning different
types of staining styles show the effectiveness and versatility of the proposed
framework. Code is available at
\textcolor{magenta}{\url{https://github.com/windygooo/SSPCR}
|
2306.08257 | Jianping Zhang | Jianping Zhang, Zhuoer Xu, Shiwen Cui, Changhua Meng, Weibin Wu,
Michael R. Lyu | On the Robustness of Latent Diffusion Models | null | null | null | null | cs.CV cs.CR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Latent diffusion models achieve state-of-the-art performance on a variety of
generative tasks, such as image synthesis and image editing. However, the
robustness of latent diffusion models is not well studied. Previous works only
focus on the adversarial attacks against the encoder or the output image under
white-box settings, regardless of the denoising process. Therefore, in this
paper, we aim to analyze the robustness of latent diffusion models more
thoroughly. We first study the influence of the components inside latent
diffusion models on their white-box robustness. In addition to white-box
scenarios, we evaluate the black-box robustness of latent diffusion models via
transfer attacks, where we consider both prompt-transfer and model-transfer
settings and possible defense mechanisms. However, all these explorations need
a comprehensive benchmark dataset, which is missing in the literature.
Therefore, to facilitate the research of the robustness of latent diffusion
models, we propose two automatic dataset construction pipelines for two kinds
of image editing models and release the whole dataset. Our code and dataset are
available at \url{https://github.com/jpzhang1810/LDM-Robustness}.
| [
{
"version": "v1",
"created": "Wed, 14 Jun 2023 05:47:15 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Zhang",
"Jianping",
""
],
[
"Xu",
"Zhuoer",
""
],
[
"Cui",
"Shiwen",
""
],
[
"Meng",
"Changhua",
""
],
[
"Wu",
"Weibin",
""
],
[
"Lyu",
"Michael R.",
""
]
] | TITLE: On the Robustness of Latent Diffusion Models
ABSTRACT: Latent diffusion models achieve state-of-the-art performance on a variety of
generative tasks, such as image synthesis and image editing. However, the
robustness of latent diffusion models is not well studied. Previous works only
focus on the adversarial attacks against the encoder or the output image under
white-box settings, regardless of the denoising process. Therefore, in this
paper, we aim to analyze the robustness of latent diffusion models more
thoroughly. We first study the influence of the components inside latent
diffusion models on their white-box robustness. In addition to white-box
scenarios, we evaluate the black-box robustness of latent diffusion models via
transfer attacks, where we consider both prompt-transfer and model-transfer
settings and possible defense mechanisms. However, all these explorations need
a comprehensive benchmark dataset, which is missing in the literature.
Therefore, to facilitate the research of the robustness of latent diffusion
models, we propose two automatic dataset construction pipelines for two kinds
of image editing models and release the whole dataset. Our code and dataset are
available at \url{https://github.com/jpzhang1810/LDM-Robustness}.
|
2306.08271 | Bimsara Pathiraja | Bimsara Pathiraja, Malitha Gunawardhana, Muhammad Haris Khan | Multiclass Confidence and Localization Calibration for Object Detection | Project page - https://bimsarapathiraja.github.io/mccl-project-page/ | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Albeit achieving high predictive accuracy across many challenging computer
vision problems, recent studies suggest that deep neural networks (DNNs) tend
to make overconfident predictions, rendering them poorly calibrated. Most of
the existing attempts for improving DNN calibration are limited to
classification tasks and restricted to calibrating in-domain predictions.
Surprisingly, very little to no attempts have been made in studying the
calibration of object detection methods, which occupy a pivotal space in
vision-based security-sensitive, and safety-critical applications. In this
paper, we propose a new train-time technique for calibrating modern object
detection methods. It is capable of jointly calibrating multiclass confidence
and box localization by leveraging their predictive uncertainties. We perform
extensive experiments on several in-domain and out-of-domain detection
benchmarks. Results demonstrate that our proposed train-time calibration method
consistently outperforms several baselines in reducing calibration error for
both in-domain and out-of-domain predictions. Our code and models are available
at https://github.com/bimsarapathiraja/MCCL.
| [
{
"version": "v1",
"created": "Wed, 14 Jun 2023 06:14:16 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Pathiraja",
"Bimsara",
""
],
[
"Gunawardhana",
"Malitha",
""
],
[
"Khan",
"Muhammad Haris",
""
]
] | TITLE: Multiclass Confidence and Localization Calibration for Object Detection
ABSTRACT: Albeit achieving high predictive accuracy across many challenging computer
vision problems, recent studies suggest that deep neural networks (DNNs) tend
to make overconfident predictions, rendering them poorly calibrated. Most of
the existing attempts for improving DNN calibration are limited to
classification tasks and restricted to calibrating in-domain predictions.
Surprisingly, very little to no attempts have been made in studying the
calibration of object detection methods, which occupy a pivotal space in
vision-based security-sensitive, and safety-critical applications. In this
paper, we propose a new train-time technique for calibrating modern object
detection methods. It is capable of jointly calibrating multiclass confidence
and box localization by leveraging their predictive uncertainties. We perform
extensive experiments on several in-domain and out-of-domain detection
benchmarks. Results demonstrate that our proposed train-time calibration method
consistently outperforms several baselines in reducing calibration error for
both in-domain and out-of-domain predictions. Our code and models are available
at https://github.com/bimsarapathiraja/MCCL.
|
2306.08290 | Sanjana Sankar | Sanjana Sankar (GIPSA-CRISSP), Denis Beautemps (GIPSA-CRISSP),
Fr\'ed\'eric Elisei (ICP), Olivier Perrotin (GIPSA-CRISSP), Thomas Hueber
(GIPSA-CRISSP) | Investigating the dynamics of hand and lips in French Cued Speech using
attention mechanisms and CTC-based decoding | null | Interspeech, ISCA, Aug 2023, Dublin, Ireland | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Hard of hearing or profoundly deaf people make use of cued speech (CS) as a
communication tool to understand spoken language. By delivering cues that are
relevant to the phonetic information, CS offers a way to enhance lipreading. In
literature, there have been several studies on the dynamics between the hand
and the lips in the context of human production. This article proposes a way to
investigate how a neural network learns this relation for a single speaker
while performing a recognition task using attention mechanisms. Further, an
analysis of the learnt dynamics is utilized to establish the relationship
between the two modalities and extract automatic segments. For the purpose of
this study, a new dataset has been recorded for French CS. Along with the
release of this dataset, a benchmark will be reported for word-level
recognition, a novelty in the automatic recognition of French CS.
| [
{
"version": "v1",
"created": "Wed, 14 Jun 2023 06:58:07 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Sankar",
"Sanjana",
"",
"GIPSA-CRISSP"
],
[
"Beautemps",
"Denis",
"",
"GIPSA-CRISSP"
],
[
"Elisei",
"Frédéric",
"",
"ICP"
],
[
"Perrotin",
"Olivier",
"",
"GIPSA-CRISSP"
],
[
"Hueber",
"Thomas",
"",
"GIPSA-CRISSP"
]
] | TITLE: Investigating the dynamics of hand and lips in French Cued Speech using
attention mechanisms and CTC-based decoding
ABSTRACT: Hard of hearing or profoundly deaf people make use of cued speech (CS) as a
communication tool to understand spoken language. By delivering cues that are
relevant to the phonetic information, CS offers a way to enhance lipreading. In
literature, there have been several studies on the dynamics between the hand
and the lips in the context of human production. This article proposes a way to
investigate how a neural network learns this relation for a single speaker
while performing a recognition task using attention mechanisms. Further, an
analysis of the learnt dynamics is utilized to establish the relationship
between the two modalities and extract automatic segments. For the purpose of
this study, a new dataset has been recorded for French CS. Along with the
release of this dataset, a benchmark will be reported for word-level
recognition, a novelty in the automatic recognition of French CS.
|
2306.08293 | Shikhar Nilabh | Shikhar Nilabh and Fidel Grandia | Efficient Training of Physics-Informed Neural Networks with Direct Grid
Refinement Algorithm | Submitted to ICML 2023 workshop: The Synergy of Scientific and
Machine Learning Modelling | null | null | null | cs.LG cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This research presents the development of an innovative algorithm tailored
for the adaptive sampling of residual points within the framework of
Physics-Informed Neural Networks (PINNs). By addressing the limitations
inherent in existing adaptive sampling techniques, our proposed methodology
introduces a direct mesh refinement approach that effectively ensures both
computational efficiency and adaptive point placement. Verification studies
were conducted to evaluate the performance of our algorithm, showcasing
reasonable agreement between the model based on our novel approach and
benchmark model results. Comparative analyses with established adaptive
resampling techniques demonstrated the superior performance of our approach,
particularly when implemented with higher refinement factor. Overall, our
findings highlight the enhancement of simulation accuracy achievable through
the application of our adaptive sampling algorithm for Physics-Informed Neural
Networks.
| [
{
"version": "v1",
"created": "Wed, 14 Jun 2023 07:04:02 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Nilabh",
"Shikhar",
""
],
[
"Grandia",
"Fidel",
""
]
] | TITLE: Efficient Training of Physics-Informed Neural Networks with Direct Grid
Refinement Algorithm
ABSTRACT: This research presents the development of an innovative algorithm tailored
for the adaptive sampling of residual points within the framework of
Physics-Informed Neural Networks (PINNs). By addressing the limitations
inherent in existing adaptive sampling techniques, our proposed methodology
introduces a direct mesh refinement approach that effectively ensures both
computational efficiency and adaptive point placement. Verification studies
were conducted to evaluate the performance of our algorithm, showcasing
reasonable agreement between the model based on our novel approach and
benchmark model results. Comparative analyses with established adaptive
resampling techniques demonstrated the superior performance of our approach,
particularly when implemented with higher refinement factor. Overall, our
findings highlight the enhancement of simulation accuracy achievable through
the application of our adaptive sampling algorithm for Physics-Informed Neural
Networks.
|
2306.08299 | Ziqing Ma | Hengbo Liu, Ziqing Ma, Linxiao Yang, Tian Zhou, Rui Xia, Yi Wang,
Qingsong Wen, Liang Sun | SaDI: A Self-adaptive Decomposed Interpretable Framework for Electric
Load Forecasting under Extreme Events | null | null | null | null | cs.LG | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Accurate prediction of electric load is crucial in power grid planning and
management. In this paper, we solve the electric load forecasting problem under
extreme events such as scorching heats. One challenge for accurate forecasting
is the lack of training samples under extreme conditions. Also load usually
changes dramatically in these extreme conditions, which calls for interpretable
model to make better decisions. In this paper, we propose a novel forecasting
framework, named Self-adaptive Decomposed Interpretable framework~(SaDI), which
ensembles long-term trend, short-term trend, and period modelings to capture
temporal characteristics in different components. The external variable
triggered loss is proposed for the imbalanced learning under extreme events.
Furthermore, Generalized Additive Model (GAM) is employed in the framework for
desirable interpretability. The experiments on both Central China electric load
and public energy meters from buildings show that the proposed SaDI framework
achieves average 22.14% improvement compared with the current state-of-the-art
algorithms in forecasting under extreme events in terms of daily mean of
normalized RMSE. Code, Public datasets, and Appendix are available at:
https://doi.org/10.24433/CO.9696980.v1 .
| [
{
"version": "v1",
"created": "Wed, 14 Jun 2023 07:11:30 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Liu",
"Hengbo",
""
],
[
"Ma",
"Ziqing",
""
],
[
"Yang",
"Linxiao",
""
],
[
"Zhou",
"Tian",
""
],
[
"Xia",
"Rui",
""
],
[
"Wang",
"Yi",
""
],
[
"Wen",
"Qingsong",
""
],
[
"Sun",
"Liang",
""
]
] | TITLE: SaDI: A Self-adaptive Decomposed Interpretable Framework for Electric
Load Forecasting under Extreme Events
ABSTRACT: Accurate prediction of electric load is crucial in power grid planning and
management. In this paper, we solve the electric load forecasting problem under
extreme events such as scorching heats. One challenge for accurate forecasting
is the lack of training samples under extreme conditions. Also load usually
changes dramatically in these extreme conditions, which calls for interpretable
model to make better decisions. In this paper, we propose a novel forecasting
framework, named Self-adaptive Decomposed Interpretable framework~(SaDI), which
ensembles long-term trend, short-term trend, and period modelings to capture
temporal characteristics in different components. The external variable
triggered loss is proposed for the imbalanced learning under extreme events.
Furthermore, Generalized Additive Model (GAM) is employed in the framework for
desirable interpretability. The experiments on both Central China electric load
and public energy meters from buildings show that the proposed SaDI framework
achieves average 22.14% improvement compared with the current state-of-the-art
algorithms in forecasting under extreme events in terms of daily mean of
normalized RMSE. Code, Public datasets, and Appendix are available at:
https://doi.org/10.24433/CO.9696980.v1 .
|
2306.08303 | Haoming Li | Haoming Li, Yu Xiang, Haodong Xu, Wenyong Wang | Pedestrian Recognition with Radar Data-Enhanced Deep Learning Approach
Based on Micro-Doppler Signatures | 6 pages,17 figures | null | null | null | eess.SP cs.CV cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | As a hot topic in recent years, the ability of pedestrians identification
based on radar micro-Doppler signatures is limited by the lack of adequate
training data. In this paper, we propose a data-enhanced multi-characteristic
learning (DEMCL) model with data enhancement (DE) module and
multi-characteristic learning (MCL) module to learn more complementary
pedestrian micro-Doppler (m-D) signatures. In DE module, a range-Doppler
generative adversarial network (RDGAN) is proposed to enhance free walking
datasets, and MCL module with multi-scale convolution neural network (MCNN) and
radial basis function neural network (RBFNN) is trained to learn m-D signatures
extracted from enhanced datasets. Experimental results show that our model is
3.33% to 10.24% more accurate than other studies and has a short run time of
0.9324 seconds on a 25-minute walking dataset.
| [
{
"version": "v1",
"created": "Wed, 14 Jun 2023 07:19:13 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Li",
"Haoming",
""
],
[
"Xiang",
"Yu",
""
],
[
"Xu",
"Haodong",
""
],
[
"Wang",
"Wenyong",
""
]
] | TITLE: Pedestrian Recognition with Radar Data-Enhanced Deep Learning Approach
Based on Micro-Doppler Signatures
ABSTRACT: As a hot topic in recent years, the ability of pedestrians identification
based on radar micro-Doppler signatures is limited by the lack of adequate
training data. In this paper, we propose a data-enhanced multi-characteristic
learning (DEMCL) model with data enhancement (DE) module and
multi-characteristic learning (MCL) module to learn more complementary
pedestrian micro-Doppler (m-D) signatures. In DE module, a range-Doppler
generative adversarial network (RDGAN) is proposed to enhance free walking
datasets, and MCL module with multi-scale convolution neural network (MCNN) and
radial basis function neural network (RBFNN) is trained to learn m-D signatures
extracted from enhanced datasets. Experimental results show that our model is
3.33% to 10.24% more accurate than other studies and has a short run time of
0.9324 seconds on a 25-minute walking dataset.
|
2306.08315 | Weidong Ji | Weidong Ji, Yousheng Zhang, Guohui Zhou, Xu Wang | Research on Named Entity Recognition in Improved transformer with R-Drop
structure | 16 pages,7 figure | null | null | null | cs.CL cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | To enhance the generalization ability of the model and improve the
effectiveness of the transformer for named entity recognition tasks, the
XLNet-Transformer-R model is proposed in this paper. The XLNet pre-trained
model and the Transformer encoder with relative positional encodings are
combined to enhance the model's ability to process long text and learn
contextual information to improve robustness. To prevent overfitting, the
R-Drop structure is used to improve the generalization capability and enhance
the accuracy of the model in named entity recognition tasks. The model in this
paper performs ablation experiments on the MSRA dataset and comparison
experiments with other models on four datasets with excellent performance,
demonstrating the strategic effectiveness of the XLNet-Transformer-R model.
| [
{
"version": "v1",
"created": "Wed, 14 Jun 2023 07:34:27 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Ji",
"Weidong",
""
],
[
"Zhang",
"Yousheng",
""
],
[
"Zhou",
"Guohui",
""
],
[
"Wang",
"Xu",
""
]
] | TITLE: Research on Named Entity Recognition in Improved transformer with R-Drop
structure
ABSTRACT: To enhance the generalization ability of the model and improve the
effectiveness of the transformer for named entity recognition tasks, the
XLNet-Transformer-R model is proposed in this paper. The XLNet pre-trained
model and the Transformer encoder with relative positional encodings are
combined to enhance the model's ability to process long text and learn
contextual information to improve robustness. To prevent overfitting, the
R-Drop structure is used to improve the generalization capability and enhance
the accuracy of the model in named entity recognition tasks. The model in this
paper performs ablation experiments on the MSRA dataset and comparison
experiments with other models on four datasets with excellent performance,
demonstrating the strategic effectiveness of the XLNet-Transformer-R model.
|
2306.08328 | Runpeng Yu | Runpeng Yu, Songhua Liu, Xingyi Yang, Xinchao Wang | Distribution Shift Inversion for Out-of-Distribution Prediction | null | null | null | null | cs.LG cs.CV | http://creativecommons.org/licenses/by/4.0/ | Machine learning society has witnessed the emergence of a myriad of
Out-of-Distribution (OoD) algorithms, which address the distribution shift
between the training and the testing distribution by searching for a unified
predictor or invariant feature representation. However, the task of directly
mitigating the distribution shift in the unseen testing set is rarely
investigated, due to the unavailability of the testing distribution during the
training phase and thus the impossibility of training a distribution translator
mapping between the training and testing distribution. In this paper, we
explore how to bypass the requirement of testing distribution for distribution
translator training and make the distribution translation useful for OoD
prediction. We propose a portable Distribution Shift Inversion algorithm, in
which, before being fed into the prediction model, the OoD testing samples are
first linearly combined with additional Gaussian noise and then transferred
back towards the training distribution using a diffusion model trained only on
the source distribution. Theoretical analysis reveals the feasibility of our
method. Experimental results, on both multiple-domain generalization datasets
and single-domain generalization datasets, show that our method provides a
general performance gain when plugged into a wide range of commonly used OoD
algorithms.
| [
{
"version": "v1",
"created": "Wed, 14 Jun 2023 08:00:49 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Yu",
"Runpeng",
""
],
[
"Liu",
"Songhua",
""
],
[
"Yang",
"Xingyi",
""
],
[
"Wang",
"Xinchao",
""
]
] | TITLE: Distribution Shift Inversion for Out-of-Distribution Prediction
ABSTRACT: Machine learning society has witnessed the emergence of a myriad of
Out-of-Distribution (OoD) algorithms, which address the distribution shift
between the training and the testing distribution by searching for a unified
predictor or invariant feature representation. However, the task of directly
mitigating the distribution shift in the unseen testing set is rarely
investigated, due to the unavailability of the testing distribution during the
training phase and thus the impossibility of training a distribution translator
mapping between the training and testing distribution. In this paper, we
explore how to bypass the requirement of testing distribution for distribution
translator training and make the distribution translation useful for OoD
prediction. We propose a portable Distribution Shift Inversion algorithm, in
which, before being fed into the prediction model, the OoD testing samples are
first linearly combined with additional Gaussian noise and then transferred
back towards the training distribution using a diffusion model trained only on
the source distribution. Theoretical analysis reveals the feasibility of our
method. Experimental results, on both multiple-domain generalization datasets
and single-domain generalization datasets, show that our method provides a
general performance gain when plugged into a wide range of commonly used OoD
algorithms.
|
2306.08329 | Weidong Ji | Weidong Ji, Shijie Zan, Guohui Zhou, and Xu Wang | Research on an improved Conformer end-to-end Speech Recognition Model
with R-Drop Structure | 15 pages, 9 figures | null | null | null | cs.SD cs.CL eess.AS | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | To address the issue of poor generalization ability in end-to-end speech
recognition models within deep learning, this study proposes a new
Conformer-based speech recognition model called "Conformer-R" that incorporates
the R-drop structure. This model combines the Conformer model, which has shown
promising results in speech recognition, with the R-drop structure. By doing
so, the model is able to effectively model both local and global speech
information while also reducing overfitting through the use of the R-drop
structure. This enhances the model's ability to generalize and improves overall
recognition efficiency. The model was first pre-trained on the Aishell1 and
Wenetspeech datasets for general domain adaptation, and subsequently fine-tuned
on computer-related audio data. Comparison tests with classic models such as
LAS and Wenet were performed on the same test set, demonstrating the
Conformer-R model's ability to effectively improve generalization.
| [
{
"version": "v1",
"created": "Wed, 14 Jun 2023 08:01:23 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Ji",
"Weidong",
""
],
[
"Zan",
"Shijie",
""
],
[
"Zhou",
"Guohui",
""
],
[
"Wang",
"Xu",
""
]
] | TITLE: Research on an improved Conformer end-to-end Speech Recognition Model
with R-Drop Structure
ABSTRACT: To address the issue of poor generalization ability in end-to-end speech
recognition models within deep learning, this study proposes a new
Conformer-based speech recognition model called "Conformer-R" that incorporates
the R-drop structure. This model combines the Conformer model, which has shown
promising results in speech recognition, with the R-drop structure. By doing
so, the model is able to effectively model both local and global speech
information while also reducing overfitting through the use of the R-drop
structure. This enhances the model's ability to generalize and improves overall
recognition efficiency. The model was first pre-trained on the Aishell1 and
Wenetspeech datasets for general domain adaptation, and subsequently fine-tuned
on computer-related audio data. Comparison tests with classic models such as
LAS and Wenet were performed on the same test set, demonstrating the
Conformer-R model's ability to effectively improve generalization.
|
2306.08350 | Yanzhou Li | Yanzhou Li, Shangqing Liu, Kangjie Chen, Xiaofei Xie, Tianwei Zhang
and Yang Liu | Multi-target Backdoor Attacks for Code Pre-trained Models | ACL 2023 main conference | null | null | null | cs.CR cs.AI cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Backdoor attacks for neural code models have gained considerable attention
due to the advancement of code intelligence. However, most existing works
insert triggers into task-specific data for code-related downstream tasks,
thereby limiting the scope of attacks. Moreover, the majority of attacks for
pre-trained models are designed for understanding tasks. In this paper, we
propose task-agnostic backdoor attacks for code pre-trained models. Our
backdoored model is pre-trained with two learning strategies (i.e., Poisoned
Seq2Seq learning and token representation learning) to support the multi-target
attack of downstream code understanding and generation tasks. During the
deployment phase, the implanted backdoors in the victim models can be activated
by the designed triggers to achieve the targeted attack. We evaluate our
approach on two code understanding tasks and three code generation tasks over
seven datasets. Extensive experiments demonstrate that our approach can
effectively and stealthily attack code-related downstream tasks.
| [
{
"version": "v1",
"created": "Wed, 14 Jun 2023 08:38:51 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Li",
"Yanzhou",
""
],
[
"Liu",
"Shangqing",
""
],
[
"Chen",
"Kangjie",
""
],
[
"Xie",
"Xiaofei",
""
],
[
"Zhang",
"Tianwei",
""
],
[
"Liu",
"Yang",
""
]
] | TITLE: Multi-target Backdoor Attacks for Code Pre-trained Models
ABSTRACT: Backdoor attacks for neural code models have gained considerable attention
due to the advancement of code intelligence. However, most existing works
insert triggers into task-specific data for code-related downstream tasks,
thereby limiting the scope of attacks. Moreover, the majority of attacks for
pre-trained models are designed for understanding tasks. In this paper, we
propose task-agnostic backdoor attacks for code pre-trained models. Our
backdoored model is pre-trained with two learning strategies (i.e., Poisoned
Seq2Seq learning and token representation learning) to support the multi-target
attack of downstream code understanding and generation tasks. During the
deployment phase, the implanted backdoors in the victim models can be activated
by the designed triggers to achieve the targeted attack. We evaluate our
approach on two code understanding tasks and three code generation tasks over
seven datasets. Extensive experiments demonstrate that our approach can
effectively and stealthily attack code-related downstream tasks.
|
2306.08359 | Xuechen Mu | Xuechen Mu, Hankz Hankui Zhuo, Chen Chen, Kai Zhang, Chao Yu and
Jianye Hao | Hierarchical Task Network Planning for Facilitating Cooperative
Multi-Agent Reinforcement Learning | null | null | null | null | cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Exploring sparse reward multi-agent reinforcement learning (MARL)
environments with traps in a collaborative manner is a complex task. Agents
typically fail to reach the goal state and fall into traps, which affects the
overall performance of the system. To overcome this issue, we present SOMARL, a
framework that uses prior knowledge to reduce the exploration space and assist
learning. In SOMARL, agents are treated as part of the MARL environment, and
symbolic knowledge is embedded using a tree structure to build a knowledge
hierarchy. The framework has a two-layer hierarchical structure, comprising a
hybrid module with a Hierarchical Task Network (HTN) planning and
meta-controller at the higher level, and a MARL-based interactive module at the
lower level. The HTN module and meta-controller use Hierarchical Domain
Definition Language (HDDL) and the option framework to formalize symbolic
knowledge and obtain domain knowledge and a symbolic option set, respectively.
Moreover, the HTN module leverages domain knowledge to guide low-level agent
exploration by assisting the meta-controller in selecting symbolic options. The
meta-controller further computes intrinsic rewards of symbolic options to limit
exploration behavior and adjust HTN planning solutions as needed. We evaluate
SOMARL on two benchmarks, FindTreasure and MoveBox, and report superior
performance over state-of-the-art MARL and subgoal-based baselines for MARL
environments significantly.
| [
{
"version": "v1",
"created": "Wed, 14 Jun 2023 08:51:43 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Mu",
"Xuechen",
""
],
[
"Zhuo",
"Hankz Hankui",
""
],
[
"Chen",
"Chen",
""
],
[
"Zhang",
"Kai",
""
],
[
"Yu",
"Chao",
""
],
[
"Hao",
"Jianye",
""
]
] | TITLE: Hierarchical Task Network Planning for Facilitating Cooperative
Multi-Agent Reinforcement Learning
ABSTRACT: Exploring sparse reward multi-agent reinforcement learning (MARL)
environments with traps in a collaborative manner is a complex task. Agents
typically fail to reach the goal state and fall into traps, which affects the
overall performance of the system. To overcome this issue, we present SOMARL, a
framework that uses prior knowledge to reduce the exploration space and assist
learning. In SOMARL, agents are treated as part of the MARL environment, and
symbolic knowledge is embedded using a tree structure to build a knowledge
hierarchy. The framework has a two-layer hierarchical structure, comprising a
hybrid module with a Hierarchical Task Network (HTN) planning and
meta-controller at the higher level, and a MARL-based interactive module at the
lower level. The HTN module and meta-controller use Hierarchical Domain
Definition Language (HDDL) and the option framework to formalize symbolic
knowledge and obtain domain knowledge and a symbolic option set, respectively.
Moreover, the HTN module leverages domain knowledge to guide low-level agent
exploration by assisting the meta-controller in selecting symbolic options. The
meta-controller further computes intrinsic rewards of symbolic options to limit
exploration behavior and adjust HTN planning solutions as needed. We evaluate
SOMARL on two benchmarks, FindTreasure and MoveBox, and report superior
performance over state-of-the-art MARL and subgoal-based baselines for MARL
environments significantly.
|
2306.08364 | Chengshuai Shi | Chengshuai Shi, Wei Xiong, Cong Shen, Jing Yang | Provably Efficient Offline Reinforcement Learning with Perturbed Data
Sources | ICML 2023 | null | null | null | stat.ML cs.IT cs.LG math.IT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Existing theoretical studies on offline reinforcement learning (RL) mostly
consider a dataset sampled directly from the target task. In practice, however,
data often come from several heterogeneous but related sources. Motivated by
this gap, this work aims at rigorously understanding offline RL with multiple
datasets that are collected from randomly perturbed versions of the target task
instead of from itself. An information-theoretic lower bound is derived, which
reveals a necessary requirement on the number of involved sources in addition
to that on the number of data samples. Then, a novel HetPEVI algorithm is
proposed, which simultaneously considers the sample uncertainties from a finite
number of data samples per data source and the source uncertainties due to a
finite number of available data sources. Theoretical analyses demonstrate that
HetPEVI can solve the target task as long as the data sources collectively
provide a good data coverage. Moreover, HetPEVI is demonstrated to be optimal
up to a polynomial factor of the horizon length. Finally, the study is extended
to offline Markov games and offline robust RL, which demonstrates the
generality of the proposed designs and theoretical analyses.
| [
{
"version": "v1",
"created": "Wed, 14 Jun 2023 08:53:20 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Shi",
"Chengshuai",
""
],
[
"Xiong",
"Wei",
""
],
[
"Shen",
"Cong",
""
],
[
"Yang",
"Jing",
""
]
] | TITLE: Provably Efficient Offline Reinforcement Learning with Perturbed Data
Sources
ABSTRACT: Existing theoretical studies on offline reinforcement learning (RL) mostly
consider a dataset sampled directly from the target task. In practice, however,
data often come from several heterogeneous but related sources. Motivated by
this gap, this work aims at rigorously understanding offline RL with multiple
datasets that are collected from randomly perturbed versions of the target task
instead of from itself. An information-theoretic lower bound is derived, which
reveals a necessary requirement on the number of involved sources in addition
to that on the number of data samples. Then, a novel HetPEVI algorithm is
proposed, which simultaneously considers the sample uncertainties from a finite
number of data samples per data source and the source uncertainties due to a
finite number of available data sources. Theoretical analyses demonstrate that
HetPEVI can solve the target task as long as the data sources collectively
provide a good data coverage. Moreover, HetPEVI is demonstrated to be optimal
up to a polynomial factor of the horizon length. Finally, the study is extended
to offline Markov games and offline robust RL, which demonstrates the
generality of the proposed designs and theoretical analyses.
|
2306.08368 | Zhenpeng Su | Yuntao Li and Zhenpeng Su and Yutian Li and Hanchu Zhang and Sirui
Wang and Wei Wu and Yan Zhang | T5-SR: A Unified Seq-to-Seq Decoding Strategy for Semantic Parsing | This paper has been accepted by ICASSP2023 | null | null | null | cs.CL cs.AI | http://creativecommons.org/licenses/by/4.0/ | Translating natural language queries into SQLs in a seq2seq manner has
attracted much attention recently. However, compared with
abstract-syntactic-tree-based SQL generation, seq2seq semantic parsers face
much more challenges, including poor quality on schematical information
prediction and poor semantic coherence between natural language queries and
SQLs. This paper analyses the above difficulties and proposes a
seq2seq-oriented decoding strategy called SR, which includes a new intermediate
representation SSQL and a reranking method with score re-estimator to solve the
above obstacles respectively. Experimental results demonstrate the
effectiveness of our proposed techniques and T5-SR-3b achieves new
state-of-the-art results on the Spider dataset.
| [
{
"version": "v1",
"created": "Wed, 14 Jun 2023 08:57:13 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Li",
"Yuntao",
""
],
[
"Su",
"Zhenpeng",
""
],
[
"Li",
"Yutian",
""
],
[
"Zhang",
"Hanchu",
""
],
[
"Wang",
"Sirui",
""
],
[
"Wu",
"Wei",
""
],
[
"Zhang",
"Yan",
""
]
] | TITLE: T5-SR: A Unified Seq-to-Seq Decoding Strategy for Semantic Parsing
ABSTRACT: Translating natural language queries into SQLs in a seq2seq manner has
attracted much attention recently. However, compared with
abstract-syntactic-tree-based SQL generation, seq2seq semantic parsers face
much more challenges, including poor quality on schematical information
prediction and poor semantic coherence between natural language queries and
SQLs. This paper analyses the above difficulties and proposes a
seq2seq-oriented decoding strategy called SR, which includes a new intermediate
representation SSQL and a reranking method with score re-estimator to solve the
above obstacles respectively. Experimental results demonstrate the
effectiveness of our proposed techniques and T5-SR-3b achieves new
state-of-the-art results on the Spider dataset.
|
2306.08397 | Arseny Skryagin | Arseny Skryagin and Daniel Ochs and Devendra Singh Dhami and Kristian
Kersting | Scalable Neural-Probabilistic Answer Set Programming | 37 pages, 14 figures | null | null | null | cs.AI | http://creativecommons.org/licenses/by/4.0/ | The goal of combining the robustness of neural networks and the
expressiveness of symbolic methods has rekindled the interest in Neuro-Symbolic
AI. Deep Probabilistic Programming Languages (DPPLs) have been developed for
probabilistic logic programming to be carried out via the probability
estimations of deep neural networks. However, recent SOTA DPPL approaches allow
only for limited conditional probabilistic queries and do not offer the power
of true joint probability estimation. In our work, we propose an easy
integration of tractable probabilistic inference within a DPPL. To this end, we
introduce SLASH, a novel DPPL that consists of Neural-Probabilistic Predicates
(NPPs) and a logic program, united via answer set programming (ASP). NPPs are a
novel design principle allowing for combining all deep model types and
combinations thereof to be represented as a single probabilistic predicate. In
this context, we introduce a novel $+/-$ notation for answering various types
of probabilistic queries by adjusting the atom notations of a predicate. To
scale well, we show how to prune the stochastically insignificant parts of the
(ground) program, speeding up reasoning without sacrificing the predictive
performance. We evaluate SLASH on a variety of different tasks, including the
benchmark task of MNIST addition and Visual Question Answering (VQA).
| [
{
"version": "v1",
"created": "Wed, 14 Jun 2023 09:45:29 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Skryagin",
"Arseny",
""
],
[
"Ochs",
"Daniel",
""
],
[
"Dhami",
"Devendra Singh",
""
],
[
"Kersting",
"Kristian",
""
]
] | TITLE: Scalable Neural-Probabilistic Answer Set Programming
ABSTRACT: The goal of combining the robustness of neural networks and the
expressiveness of symbolic methods has rekindled the interest in Neuro-Symbolic
AI. Deep Probabilistic Programming Languages (DPPLs) have been developed for
probabilistic logic programming to be carried out via the probability
estimations of deep neural networks. However, recent SOTA DPPL approaches allow
only for limited conditional probabilistic queries and do not offer the power
of true joint probability estimation. In our work, we propose an easy
integration of tractable probabilistic inference within a DPPL. To this end, we
introduce SLASH, a novel DPPL that consists of Neural-Probabilistic Predicates
(NPPs) and a logic program, united via answer set programming (ASP). NPPs are a
novel design principle allowing for combining all deep model types and
combinations thereof to be represented as a single probabilistic predicate. In
this context, we introduce a novel $+/-$ notation for answering various types
of probabilistic queries by adjusting the atom notations of a predicate. To
scale well, we show how to prune the stochastically insignificant parts of the
(ground) program, speeding up reasoning without sacrificing the predictive
performance. We evaluate SLASH on a variety of different tasks, including the
benchmark task of MNIST addition and Visual Question Answering (VQA).
|
2306.08401 | Jingsheng Gao | Jingsheng Gao, Yixin Lian, Ziyi Zhou, Yuzhuo Fu, Baoyuan Wang | LiveChat: A Large-Scale Personalized Dialogue Dataset Automatically
Constructed from Live Streaming | ACL 2023 Main Conference | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Open-domain dialogue systems have made promising progress in recent years.
While the state-of-the-art dialogue agents are built upon large-scale
text-based social media data and large pre-trained models, there is no
guarantee these agents could also perform well in fast-growing scenarios, such
as live streaming, due to the bounded transferability of pre-trained models and
biased distributions of public datasets from Reddit and Weibo, etc. To improve
the essential capability of responding and establish a benchmark in the live
open-domain scenario, we introduce the LiveChat dataset, composed of 1.33
million real-life Chinese dialogues with almost 3800 average sessions across
351 personas and fine-grained profiles for each persona. LiveChat is
automatically constructed by processing numerous live videos on the Internet
and naturally falls within the scope of multi-party conversations, where the
issues of Who says What to Whom should be considered. Therefore, we target two
critical tasks of response modeling and addressee recognition and propose
retrieval-based baselines grounded on advanced techniques. Experimental results
have validated the positive effects of leveraging persona profiles and larger
average sessions per persona. In addition, we also benchmark the
transferability of advanced generation-based models on LiveChat and pose some
future directions for current challenges.
| [
{
"version": "v1",
"created": "Wed, 14 Jun 2023 09:50:06 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Gao",
"Jingsheng",
""
],
[
"Lian",
"Yixin",
""
],
[
"Zhou",
"Ziyi",
""
],
[
"Fu",
"Yuzhuo",
""
],
[
"Wang",
"Baoyuan",
""
]
] | TITLE: LiveChat: A Large-Scale Personalized Dialogue Dataset Automatically
Constructed from Live Streaming
ABSTRACT: Open-domain dialogue systems have made promising progress in recent years.
While the state-of-the-art dialogue agents are built upon large-scale
text-based social media data and large pre-trained models, there is no
guarantee these agents could also perform well in fast-growing scenarios, such
as live streaming, due to the bounded transferability of pre-trained models and
biased distributions of public datasets from Reddit and Weibo, etc. To improve
the essential capability of responding and establish a benchmark in the live
open-domain scenario, we introduce the LiveChat dataset, composed of 1.33
million real-life Chinese dialogues with almost 3800 average sessions across
351 personas and fine-grained profiles for each persona. LiveChat is
automatically constructed by processing numerous live videos on the Internet
and naturally falls within the scope of multi-party conversations, where the
issues of Who says What to Whom should be considered. Therefore, we target two
critical tasks of response modeling and addressee recognition and propose
retrieval-based baselines grounded on advanced techniques. Experimental results
have validated the positive effects of leveraging persona profiles and larger
average sessions per persona. In addition, we also benchmark the
transferability of advanced generation-based models on LiveChat and pose some
future directions for current challenges.
|
2306.08424 | Matthew Barker | Matthew Barker, Katherine M. Collins, Krishnamurthy Dvijotham, Adrian
Weller, Umang Bhatt | Selective Concept Models: Permitting Stakeholder Customisation at
Test-Time | null | null | null | null | cs.HC cs.AI cs.LG | http://creativecommons.org/licenses/by/4.0/ | Concept-based models perform prediction using a set of concepts that are
interpretable to stakeholders. However, such models often involve a fixed,
large number of concepts, which may place a substantial cognitive load on
stakeholders. We propose Selective COncept Models (SCOMs) which make
predictions using only a subset of concepts and can be customised by
stakeholders at test-time according to their preferences. We show that SCOMs
only require a fraction of the total concepts to achieve optimal accuracy on
multiple real-world datasets. Further, we collect and release a new dataset,
CUB-Sel, consisting of human concept set selections for 900 bird images from
the popular CUB dataset. Using CUB-Sel, we show that humans have unique
individual preferences for the choice of concepts they prefer to reason about,
and struggle to identify the most theoretically informative concepts. The
customisation and concept selection provided by SCOM improves the efficiency of
interpretation and intervention for stakeholders.
| [
{
"version": "v1",
"created": "Wed, 14 Jun 2023 10:37:13 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Barker",
"Matthew",
""
],
[
"Collins",
"Katherine M.",
""
],
[
"Dvijotham",
"Krishnamurthy",
""
],
[
"Weller",
"Adrian",
""
],
[
"Bhatt",
"Umang",
""
]
] | TITLE: Selective Concept Models: Permitting Stakeholder Customisation at
Test-Time
ABSTRACT: Concept-based models perform prediction using a set of concepts that are
interpretable to stakeholders. However, such models often involve a fixed,
large number of concepts, which may place a substantial cognitive load on
stakeholders. We propose Selective COncept Models (SCOMs) which make
predictions using only a subset of concepts and can be customised by
stakeholders at test-time according to their preferences. We show that SCOMs
only require a fraction of the total concepts to achieve optimal accuracy on
multiple real-world datasets. Further, we collect and release a new dataset,
CUB-Sel, consisting of human concept set selections for 900 bird images from
the popular CUB dataset. Using CUB-Sel, we show that humans have unique
individual preferences for the choice of concepts they prefer to reason about,
and struggle to identify the most theoretically informative concepts. The
customisation and concept selection provided by SCOM improves the efficiency of
interpretation and intervention for stakeholders.
|
2306.08458 | Xiaolin He | Xiaolin He, Riender Happee, Meng Wang | A new computational perceived risk model for automated vehicles based on
potential collision avoidance difficulty (PCAD) | null | null | null | null | cs.HC | http://creativecommons.org/licenses/by/4.0/ | Perceived risk is crucial in designing trustworthy and acceptable vehicle
automation systems. However, our understanding of its dynamics is limited, and
models for perceived risk dynamics are scarce in the literature. This study
formulates a new computational perceived risk model based on potential
collision avoidance difficulty (PCAD) for drivers of SAE level 2 driving
automation. PCAD uses the 2D safe velocity gap as the potential collision
avoidance difficulty, and takes into account collision severity. The safe
velocity gap is defined as the 2D gap between the current velocity and the safe
velocity region, and represents the amount of braking and steering needed,
considering behavioural uncertainty of neighbouring vehicles and imprecise
control of the subject vehicle. The PCAD predicts perceived risk both in
continuous time and per event. We compare the PCAD model with three
state-of-the-art models and analyse the models both theoretically and
empirically with two unique datasets: Dataset Merging and Dataset Obstacle
Avoidance. The PCAD model generally outperforms the other models in terms of
model error, detection rate, and the ability to accurately capture the
tendencies of human drivers' perceived risk, albeit at a longer computation
time. Additionally, the study shows that the perceived risk is not static and
varies with the surrounding traffic conditions. This research advances our
understanding of perceived risk in automated driving and paves the way for
improved safety and acceptance of driving automation systems.
| [
{
"version": "v1",
"created": "Wed, 14 Jun 2023 11:59:11 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"He",
"Xiaolin",
""
],
[
"Happee",
"Riender",
""
],
[
"Wang",
"Meng",
""
]
] | TITLE: A new computational perceived risk model for automated vehicles based on
potential collision avoidance difficulty (PCAD)
ABSTRACT: Perceived risk is crucial in designing trustworthy and acceptable vehicle
automation systems. However, our understanding of its dynamics is limited, and
models for perceived risk dynamics are scarce in the literature. This study
formulates a new computational perceived risk model based on potential
collision avoidance difficulty (PCAD) for drivers of SAE level 2 driving
automation. PCAD uses the 2D safe velocity gap as the potential collision
avoidance difficulty, and takes into account collision severity. The safe
velocity gap is defined as the 2D gap between the current velocity and the safe
velocity region, and represents the amount of braking and steering needed,
considering behavioural uncertainty of neighbouring vehicles and imprecise
control of the subject vehicle. The PCAD predicts perceived risk both in
continuous time and per event. We compare the PCAD model with three
state-of-the-art models and analyse the models both theoretically and
empirically with two unique datasets: Dataset Merging and Dataset Obstacle
Avoidance. The PCAD model generally outperforms the other models in terms of
model error, detection rate, and the ability to accurately capture the
tendencies of human drivers' perceived risk, albeit at a longer computation
time. Additionally, the study shows that the perceived risk is not static and
varies with the surrounding traffic conditions. This research advances our
understanding of perceived risk in automated driving and paves the way for
improved safety and acceptance of driving automation systems.
|
2306.08460 | Ren Wang | Ren Wang, Haoliang Sun, Qi Wei, Xiushan Nie, Yuling Ma, Yilong Yin | Improving Generalization in Meta-Learning via Meta-Gradient Augmentation | null | null | null | null | cs.LG cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Meta-learning methods typically follow a two-loop framework, where each loop
potentially suffers from notorious overfitting, hindering rapid adaptation and
generalization to new tasks. Existing schemes solve it by enhancing the
mutual-exclusivity or diversity of training samples, but these data
manipulation strategies are data-dependent and insufficiently flexible. This
work alleviates overfitting in meta-learning from the perspective of gradient
regularization and proposes a data-independent \textbf{M}eta-\textbf{G}radient
\textbf{Aug}mentation (\textbf{MGAug}) method. The key idea is to first break
the rote memories by network pruning to address memorization overfitting in the
inner loop, and then the gradients of pruned sub-networks naturally form the
high-quality augmentation of the meta-gradient to alleviate learner overfitting
in the outer loop. Specifically, we explore three pruning strategies, including
\textit{random width pruning}, \textit{random parameter pruning}, and a newly
proposed \textit{catfish pruning} that measures a Meta-Memorization Carrying
Amount (MMCA) score for each parameter and prunes high-score ones to break rote
memories as much as possible. The proposed MGAug is theoretically guaranteed by
the generalization bound from the PAC-Bayes framework. In addition, we extend a
lightweight version, called MGAug-MaxUp, as a trade-off between performance
gains and resource overhead. Extensive experiments on multiple few-shot
learning benchmarks validate MGAug's effectiveness and significant improvement
over various meta-baselines. The code is publicly available at
\url{https://github.com/xxLifeLover/Meta-Gradient-Augmentation}.
| [
{
"version": "v1",
"created": "Wed, 14 Jun 2023 12:04:28 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Wang",
"Ren",
""
],
[
"Sun",
"Haoliang",
""
],
[
"Wei",
"Qi",
""
],
[
"Nie",
"Xiushan",
""
],
[
"Ma",
"Yuling",
""
],
[
"Yin",
"Yilong",
""
]
] | TITLE: Improving Generalization in Meta-Learning via Meta-Gradient Augmentation
ABSTRACT: Meta-learning methods typically follow a two-loop framework, where each loop
potentially suffers from notorious overfitting, hindering rapid adaptation and
generalization to new tasks. Existing schemes solve it by enhancing the
mutual-exclusivity or diversity of training samples, but these data
manipulation strategies are data-dependent and insufficiently flexible. This
work alleviates overfitting in meta-learning from the perspective of gradient
regularization and proposes a data-independent \textbf{M}eta-\textbf{G}radient
\textbf{Aug}mentation (\textbf{MGAug}) method. The key idea is to first break
the rote memories by network pruning to address memorization overfitting in the
inner loop, and then the gradients of pruned sub-networks naturally form the
high-quality augmentation of the meta-gradient to alleviate learner overfitting
in the outer loop. Specifically, we explore three pruning strategies, including
\textit{random width pruning}, \textit{random parameter pruning}, and a newly
proposed \textit{catfish pruning} that measures a Meta-Memorization Carrying
Amount (MMCA) score for each parameter and prunes high-score ones to break rote
memories as much as possible. The proposed MGAug is theoretically guaranteed by
the generalization bound from the PAC-Bayes framework. In addition, we extend a
lightweight version, called MGAug-MaxUp, as a trade-off between performance
gains and resource overhead. Extensive experiments on multiple few-shot
learning benchmarks validate MGAug's effectiveness and significant improvement
over various meta-baselines. The code is publicly available at
\url{https://github.com/xxLifeLover/Meta-Gradient-Augmentation}.
|
2306.08522 | Pushyami Kaveti | Pushyami Kaveti, Aniket Gupta, Dennis Giaya, Madeline Karp, Colin
Keil, Jagatpreet Nir, Zhiyong Zhang, Hanumant Singh | Challenges of Indoor SLAM: A multi-modal multi-floor dataset for SLAM
evaluation | null | null | null | null | cs.RO | http://creativecommons.org/licenses/by/4.0/ | Robustness in Simultaneous Localization and Mapping (SLAM) remains one of the
key challenges for the real-world deployment of autonomous systems. SLAM
research has seen significant progress in the last two and a half decades, yet
many state-of-the-art (SOTA) algorithms still struggle to perform reliably in
real-world environments. There is a general consensus in the research community
that we need challenging real-world scenarios which bring out different failure
modes in sensing modalities. In this paper, we present a novel multi-modal
indoor SLAM dataset covering challenging common scenarios that a robot will
encounter and should be robust to. Our data was collected with a mobile
robotics platform across multiple floors at Northeastern University's ISEC
building. Such a multi-floor sequence is typical of commercial office spaces
characterized by symmetry across floors and, thus, is prone to perceptual
aliasing due to similar floor layouts. The sensor suite comprises seven global
shutter cameras, a high-grade MEMS inertial measurement unit (IMU), a ZED
stereo camera, and a 128-channel high-resolution lidar. Along with the dataset,
we benchmark several SLAM algorithms and highlight the problems faced during
the runs, such as perceptual aliasing, visual degradation, and trajectory
drift. The benchmarking results indicate that parts of the dataset work well
with some algorithms, while other data sections are challenging for even the
best SOTA algorithms. The dataset is available at
https://github.com/neufieldrobotics/NUFR-M3F.
| [
{
"version": "v1",
"created": "Wed, 14 Jun 2023 14:12:57 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Kaveti",
"Pushyami",
""
],
[
"Gupta",
"Aniket",
""
],
[
"Giaya",
"Dennis",
""
],
[
"Karp",
"Madeline",
""
],
[
"Keil",
"Colin",
""
],
[
"Nir",
"Jagatpreet",
""
],
[
"Zhang",
"Zhiyong",
""
],
[
"Singh",
"Hanumant",
""
]
] | TITLE: Challenges of Indoor SLAM: A multi-modal multi-floor dataset for SLAM
evaluation
ABSTRACT: Robustness in Simultaneous Localization and Mapping (SLAM) remains one of the
key challenges for the real-world deployment of autonomous systems. SLAM
research has seen significant progress in the last two and a half decades, yet
many state-of-the-art (SOTA) algorithms still struggle to perform reliably in
real-world environments. There is a general consensus in the research community
that we need challenging real-world scenarios which bring out different failure
modes in sensing modalities. In this paper, we present a novel multi-modal
indoor SLAM dataset covering challenging common scenarios that a robot will
encounter and should be robust to. Our data was collected with a mobile
robotics platform across multiple floors at Northeastern University's ISEC
building. Such a multi-floor sequence is typical of commercial office spaces
characterized by symmetry across floors and, thus, is prone to perceptual
aliasing due to similar floor layouts. The sensor suite comprises seven global
shutter cameras, a high-grade MEMS inertial measurement unit (IMU), a ZED
stereo camera, and a 128-channel high-resolution lidar. Along with the dataset,
we benchmark several SLAM algorithms and highlight the problems faced during
the runs, such as perceptual aliasing, visual degradation, and trajectory
drift. The benchmarking results indicate that parts of the dataset work well
with some algorithms, while other data sections are challenging for even the
best SOTA algorithms. The dataset is available at
https://github.com/neufieldrobotics/NUFR-M3F.
|
2306.08531 | Fernando Amodeo | Fernando Amodeo, No\'e P\'erez-Higueras, Luis Merino, Fernando
Caballero | FROG: A new people detection dataset for knee-high 2D range finders | Code and data are publicly available at:
https://github.com/robotics-upo/2DLaserPeopleBenchmark | null | null | null | cs.RO | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Mobile robots require knowledge of the environment, especially of humans
located in its vicinity. While the most common approaches for detecting humans
involve computer vision, an often overlooked hardware feature of robots for
people detection are their 2D range finders. These were originally intended for
obstacle avoidance and mapping/SLAM tasks. In most robots, they are
conveniently located at a height approximately between the ankle and the knee,
so they can be used for detecting people too, and with a larger field of view
and depth resolution compared to cameras.
In this paper, we present a new dataset for people detection using knee-high
2D range finders called FROG. This dataset has greater laser resolution,
scanning frequency, and more complete annotation data compared to existing
datasets such as DROW. Particularly, the FROG dataset contains annotations for
100% of its laser scans (unlike DROW which only annotates 5%), 17x more
annotated scans, 100x more people annotations, and over twice the distance
traveled by the robot. We propose a benchmark based on the FROG dataset, and
analyze a collection of state-of-the-art people detectors based on 2D range
finder data.
We also propose and evaluate a new end-to-end deep learning approach for
people detection. Our solution works with the raw sensor data directly (not
needing hand-crafted input data features), thus avoiding CPU preprocessing and
releasing the developer of understanding specific domain heuristics.
Experimental results show how the proposed people detector attains results
comparable to the state of the art, while an optimized implementation for ROS
can operate at more than 500 Hz.
| [
{
"version": "v1",
"created": "Wed, 14 Jun 2023 14:24:10 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Amodeo",
"Fernando",
""
],
[
"Pérez-Higueras",
"Noé",
""
],
[
"Merino",
"Luis",
""
],
[
"Caballero",
"Fernando",
""
]
] | TITLE: FROG: A new people detection dataset for knee-high 2D range finders
ABSTRACT: Mobile robots require knowledge of the environment, especially of humans
located in its vicinity. While the most common approaches for detecting humans
involve computer vision, an often overlooked hardware feature of robots for
people detection are their 2D range finders. These were originally intended for
obstacle avoidance and mapping/SLAM tasks. In most robots, they are
conveniently located at a height approximately between the ankle and the knee,
so they can be used for detecting people too, and with a larger field of view
and depth resolution compared to cameras.
In this paper, we present a new dataset for people detection using knee-high
2D range finders called FROG. This dataset has greater laser resolution,
scanning frequency, and more complete annotation data compared to existing
datasets such as DROW. Particularly, the FROG dataset contains annotations for
100% of its laser scans (unlike DROW which only annotates 5%), 17x more
annotated scans, 100x more people annotations, and over twice the distance
traveled by the robot. We propose a benchmark based on the FROG dataset, and
analyze a collection of state-of-the-art people detectors based on 2D range
finder data.
We also propose and evaluate a new end-to-end deep learning approach for
people detection. Our solution works with the raw sensor data directly (not
needing hand-crafted input data features), thus avoiding CPU preprocessing and
releasing the developer of understanding specific domain heuristics.
Experimental results show how the proposed people detector attains results
comparable to the state of the art, while an optimized implementation for ROS
can operate at more than 500 Hz.
|
2306.08549 | Mustafa Atay | Megh Pudyel and Mustafa Atay | An Exploratory Study of Masked Face Recognition with Machine Learning
Algorithms | 6 pages published in IEEE SoutheastCon 2023 | SoutheastCon 2023 (2023) 877-882 | 10.1109/SoutheastCon51012.2023.10115205 | null | cs.CV cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Automated face recognition is a widely adopted machine learning technology
for contactless identification of people in various processes such as automated
border control, secure login to electronic devices, community surveillance,
tracking school attendance, workplace clock in and clock out. Using face masks
have become crucial in our daily life with the recent world-wide COVID-19
pandemic. The use of face masks causes the performance of conventional face
recognition technologies to degrade considerably. The effect of mask-wearing in
face recognition is yet an understudied issue. In this paper, we address this
issue by evaluating the performance of a number of face recognition models
which are tested by identifying masked and unmasked face images. We use six
conventional machine learning algorithms, which are SVC, KNN, LDA, DT, LR and
NB, to find out the ones which perform best, besides the ones which poorly
perform, in the presence of masked face images. Local Binary Pattern (LBP) is
utilized as the feature extraction operator. We generated and used synthesized
masked face images. We prepared unmasked, masked, and half-masked training
datasets and evaluated the face recognition performance against both masked and
unmasked images to present a broad view of this crucial problem. We believe
that our study is unique in elaborating the mask-aware facial recognition with
almost all possible scenarios including half_masked-to-masked and
half_masked-to-unmasked besides evaluating a larger number of conventional
machine learning algorithms compared the other studies in the literature.
| [
{
"version": "v1",
"created": "Wed, 14 Jun 2023 14:50:23 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Pudyel",
"Megh",
""
],
[
"Atay",
"Mustafa",
""
]
] | TITLE: An Exploratory Study of Masked Face Recognition with Machine Learning
Algorithms
ABSTRACT: Automated face recognition is a widely adopted machine learning technology
for contactless identification of people in various processes such as automated
border control, secure login to electronic devices, community surveillance,
tracking school attendance, workplace clock in and clock out. Using face masks
have become crucial in our daily life with the recent world-wide COVID-19
pandemic. The use of face masks causes the performance of conventional face
recognition technologies to degrade considerably. The effect of mask-wearing in
face recognition is yet an understudied issue. In this paper, we address this
issue by evaluating the performance of a number of face recognition models
which are tested by identifying masked and unmasked face images. We use six
conventional machine learning algorithms, which are SVC, KNN, LDA, DT, LR and
NB, to find out the ones which perform best, besides the ones which poorly
perform, in the presence of masked face images. Local Binary Pattern (LBP) is
utilized as the feature extraction operator. We generated and used synthesized
masked face images. We prepared unmasked, masked, and half-masked training
datasets and evaluated the face recognition performance against both masked and
unmasked images to present a broad view of this crucial problem. We believe
that our study is unique in elaborating the mask-aware facial recognition with
almost all possible scenarios including half_masked-to-masked and
half_masked-to-unmasked besides evaluating a larger number of conventional
machine learning algorithms compared the other studies in the literature.
|
2306.08564 | Raja Marjieh | Raja Marjieh, Nori Jacoby, Joshua C. Peterson, Thomas L. Griffiths | The Universal Law of Generalization Holds for Naturalistic Stimuli | 36 pages, 6 figures | null | null | null | q-bio.NC cs.AI stat.AP | http://creativecommons.org/licenses/by/4.0/ | Shepard's universal law of generalization is a remarkable hypothesis about
how intelligent organisms should perceive similarity. In its broadest form, the
universal law states that the level of perceived similarity between a pair of
stimuli should decay as a concave function of their distance when embedded in
an appropriate psychological space. While extensively studied, evidence in
support of the universal law has relied on low-dimensional stimuli and small
stimulus sets that are very different from their real-world counterparts. This
is largely because pairwise comparisons -- as required for similarity judgments
-- scale quadratically in the number of stimuli. We provide direct evidence for
the universal law in a naturalistic high-dimensional regime by analyzing an
existing dataset of 214,200 human similarity judgments and a newly collected
dataset of 390,819 human generalization judgments (N=2406 US participants)
across three sets of natural images.
| [
{
"version": "v1",
"created": "Wed, 14 Jun 2023 15:17:48 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Marjieh",
"Raja",
""
],
[
"Jacoby",
"Nori",
""
],
[
"Peterson",
"Joshua C.",
""
],
[
"Griffiths",
"Thomas L.",
""
]
] | TITLE: The Universal Law of Generalization Holds for Naturalistic Stimuli
ABSTRACT: Shepard's universal law of generalization is a remarkable hypothesis about
how intelligent organisms should perceive similarity. In its broadest form, the
universal law states that the level of perceived similarity between a pair of
stimuli should decay as a concave function of their distance when embedded in
an appropriate psychological space. While extensively studied, evidence in
support of the universal law has relied on low-dimensional stimuli and small
stimulus sets that are very different from their real-world counterparts. This
is largely because pairwise comparisons -- as required for similarity judgments
-- scale quadratically in the number of stimuli. We provide direct evidence for
the universal law in a naturalistic high-dimensional regime by analyzing an
existing dataset of 214,200 human similarity judgments and a newly collected
dataset of 390,819 human generalization judgments (N=2406 US participants)
across three sets of natural images.
|
2306.08565 | Wenqian Yu | Wenqian Yu and Jindong Gu and Zhijiang Li and Philip Torr | Reliable Evaluation of Adversarial Transferability | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Adversarial examples (AEs) with small adversarial perturbations can mislead
deep neural networks (DNNs) into wrong predictions. The AEs created on one DNN
can also fool another DNN. Over the last few years, the transferability of AEs
has garnered significant attention as it is a crucial property for facilitating
black-box attacks. Many approaches have been proposed to improve adversarial
transferability. However, they are mainly verified across different
convolutional neural network (CNN) architectures, which is not a reliable
evaluation since all CNNs share some similar architectural biases. In this
work, we re-evaluate 12 representative transferability-enhancing attack methods
where we test on 18 popular models from 4 types of neural networks. Our
reevaluation revealed that the adversarial transferability is often
overestimated, and there is no single AE that can be transferred to all popular
models. The transferability rank of previous attacking methods changes when
under our comprehensive evaluation. Based on our analysis, we propose a
reliable benchmark including three evaluation protocols. Adversarial
transferability on our new benchmark is extremely low, which further confirms
the overestimation of adversarial transferability. We release our benchmark at
https://adv-trans-eval.github.io to facilitate future research, which includes
code, model checkpoints, and evaluation protocols.
| [
{
"version": "v1",
"created": "Wed, 14 Jun 2023 15:17:51 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Yu",
"Wenqian",
""
],
[
"Gu",
"Jindong",
""
],
[
"Li",
"Zhijiang",
""
],
[
"Torr",
"Philip",
""
]
] | TITLE: Reliable Evaluation of Adversarial Transferability
ABSTRACT: Adversarial examples (AEs) with small adversarial perturbations can mislead
deep neural networks (DNNs) into wrong predictions. The AEs created on one DNN
can also fool another DNN. Over the last few years, the transferability of AEs
has garnered significant attention as it is a crucial property for facilitating
black-box attacks. Many approaches have been proposed to improve adversarial
transferability. However, they are mainly verified across different
convolutional neural network (CNN) architectures, which is not a reliable
evaluation since all CNNs share some similar architectural biases. In this
work, we re-evaluate 12 representative transferability-enhancing attack methods
where we test on 18 popular models from 4 types of neural networks. Our
reevaluation revealed that the adversarial transferability is often
overestimated, and there is no single AE that can be transferred to all popular
models. The transferability rank of previous attacking methods changes when
under our comprehensive evaluation. Based on our analysis, we propose a
reliable benchmark including three evaluation protocols. Adversarial
transferability on our new benchmark is extremely low, which further confirms
the overestimation of adversarial transferability. We release our benchmark at
https://adv-trans-eval.github.io to facilitate future research, which includes
code, model checkpoints, and evaluation protocols.
|
2306.08593 | Divyam Madaan | Divyam Madaan, Hongxu Yin, Wonmin Byeon, Jan Kautz, Pavlo Molchanov | Heterogeneous Continual Learning | Accepted to CVPR 2023 | null | null | null | cs.CV cs.LG | http://creativecommons.org/licenses/by-sa/4.0/ | We propose a novel framework and a solution to tackle the continual learning
(CL) problem with changing network architectures. Most CL methods focus on
adapting a single architecture to a new task/class by modifying its weights.
However, with rapid progress in architecture design, the problem of adapting
existing solutions to novel architectures becomes relevant. To address this
limitation, we propose Heterogeneous Continual Learning (HCL), where a wide
range of evolving network architectures emerge continually together with novel
data/tasks. As a solution, we build on top of the distillation family of
techniques and modify it to a new setting where a weaker model takes the role
of a teacher; meanwhile, a new stronger architecture acts as a student.
Furthermore, we consider a setup of limited access to previous data and propose
Quick Deep Inversion (QDI) to recover prior task visual features to support
knowledge transfer. QDI significantly reduces computational costs compared to
previous solutions and improves overall performance. In summary, we propose a
new setup for CL with a modified knowledge distillation paradigm and design a
quick data inversion method to enhance distillation. Our evaluation of various
benchmarks shows a significant improvement on accuracy in comparison to
state-of-the-art methods over various networks architectures.
| [
{
"version": "v1",
"created": "Wed, 14 Jun 2023 15:54:42 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Madaan",
"Divyam",
""
],
[
"Yin",
"Hongxu",
""
],
[
"Byeon",
"Wonmin",
""
],
[
"Kautz",
"Jan",
""
],
[
"Molchanov",
"Pavlo",
""
]
] | TITLE: Heterogeneous Continual Learning
ABSTRACT: We propose a novel framework and a solution to tackle the continual learning
(CL) problem with changing network architectures. Most CL methods focus on
adapting a single architecture to a new task/class by modifying its weights.
However, with rapid progress in architecture design, the problem of adapting
existing solutions to novel architectures becomes relevant. To address this
limitation, we propose Heterogeneous Continual Learning (HCL), where a wide
range of evolving network architectures emerge continually together with novel
data/tasks. As a solution, we build on top of the distillation family of
techniques and modify it to a new setting where a weaker model takes the role
of a teacher; meanwhile, a new stronger architecture acts as a student.
Furthermore, we consider a setup of limited access to previous data and propose
Quick Deep Inversion (QDI) to recover prior task visual features to support
knowledge transfer. QDI significantly reduces computational costs compared to
previous solutions and improves overall performance. In summary, we propose a
new setup for CL with a modified knowledge distillation paradigm and design a
quick data inversion method to enhance distillation. Our evaluation of various
benchmarks shows a significant improvement on accuracy in comparison to
state-of-the-art methods over various networks architectures.
|
2306.08604 | Enyan Dai | Enyan Dai, Limeng Cui, Zhengyang Wang, Xianfeng Tang, Yinghan Wang,
Monica Cheng, Bing Yin, Suhang Wang | A Unified Framework of Graph Information Bottleneck for Robustness and
Membership Privacy | null | null | null | null | cs.LG cs.AI cs.CR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Graph Neural Networks (GNNs) have achieved great success in modeling
graph-structured data. However, recent works show that GNNs are vulnerable to
adversarial attacks which can fool the GNN model to make desired predictions of
the attacker. In addition, training data of GNNs can be leaked under membership
inference attacks. This largely hinders the adoption of GNNs in high-stake
domains such as e-commerce, finance and bioinformatics. Though investigations
have been made in conducting robust predictions and protecting membership
privacy, they generally fail to simultaneously consider the robustness and
membership privacy. Therefore, in this work, we study a novel problem of
developing robust and membership privacy-preserving GNNs. Our analysis shows
that Information Bottleneck (IB) can help filter out noisy information and
regularize the predictions on labeled samples, which can benefit robustness and
membership privacy. However, structural noises and lack of labels in node
classification challenge the deployment of IB on graph-structured data. To
mitigate these issues, we propose a novel graph information bottleneck
framework that can alleviate structural noises with neighbor bottleneck. Pseudo
labels are also incorporated in the optimization to minimize the gap between
the predictions on the labeled set and unlabeled set for membership privacy.
Extensive experiments on real-world datasets demonstrate that our method can
give robust predictions and simultaneously preserve membership privacy.
| [
{
"version": "v1",
"created": "Wed, 14 Jun 2023 16:11:00 GMT"
}
] | 2023-06-16T00:00:00 | [
[
"Dai",
"Enyan",
""
],
[
"Cui",
"Limeng",
""
],
[
"Wang",
"Zhengyang",
""
],
[
"Tang",
"Xianfeng",
""
],
[
"Wang",
"Yinghan",
""
],
[
"Cheng",
"Monica",
""
],
[
"Yin",
"Bing",
""
],
[
"Wang",
"Suhang",
""
]
] | TITLE: A Unified Framework of Graph Information Bottleneck for Robustness and
Membership Privacy
ABSTRACT: Graph Neural Networks (GNNs) have achieved great success in modeling
graph-structured data. However, recent works show that GNNs are vulnerable to
adversarial attacks which can fool the GNN model to make desired predictions of
the attacker. In addition, training data of GNNs can be leaked under membership
inference attacks. This largely hinders the adoption of GNNs in high-stake
domains such as e-commerce, finance and bioinformatics. Though investigations
have been made in conducting robust predictions and protecting membership
privacy, they generally fail to simultaneously consider the robustness and
membership privacy. Therefore, in this work, we study a novel problem of
developing robust and membership privacy-preserving GNNs. Our analysis shows
that Information Bottleneck (IB) can help filter out noisy information and
regularize the predictions on labeled samples, which can benefit robustness and
membership privacy. However, structural noises and lack of labels in node
classification challenge the deployment of IB on graph-structured data. To
mitigate these issues, we propose a novel graph information bottleneck
framework that can alleviate structural noises with neighbor bottleneck. Pseudo
labels are also incorporated in the optimization to minimize the gap between
the predictions on the labeled set and unlabeled set for membership privacy.
Extensive experiments on real-world datasets demonstrate that our method can
give robust predictions and simultaneously preserve membership privacy.
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.