id
large_stringlengths 9
16
| submitter
large_stringlengths 3
64
⌀ | authors
large_stringlengths 5
20.2k
| title
large_stringlengths 5
245
| comments
large_stringlengths 1
482
⌀ | journal-ref
large_stringlengths 4
382
⌀ | doi
large_stringlengths 9
151
⌀ | report-no
large_stringlengths 2
509
⌀ | categories
large_stringlengths 5
78
| license
large_stringclasses 9
values | abstract
large_stringlengths 83
3.55k
| versions
large listlengths 1
21
| update_date
timestamp[ms]date 2007-05-23 00:00:00
2025-05-30 00:00:00
| authors_parsed
large listlengths 1
1.35k
| prompt
large_stringlengths 166
3.64k
| label
stringclasses 2
values | prob
float64 0.5
0.73
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2304.06277 | Anand Gokul Mahalingam | Anand Gokul Mahalingam, Aayush Shah, Akshay Gulati, Royston
Mascarenhas, Rakshitha Panduranga | Optimizing Multi-Domain Performance with Active Learning-based
Improvement Strategies | 13 pages, 20 figures, draft work previously published as a medium
story | null | null | null | cs.LG cs.AI cs.CV | http://creativecommons.org/licenses/by/4.0/ | Improving performance in multiple domains is a challenging task, and often
requires significant amounts of data to train and test models. Active learning
techniques provide a promising solution by enabling models to select the most
informative samples for labeling, thus reducing the amount of labeled data
required to achieve high performance. In this paper, we present an active
learning-based framework for improving performance across multiple domains. Our
approach consists of two stages: first, we use an initial set of labeled data
to train a base model, and then we iteratively select the most informative
samples for labeling to refine the model. We evaluate our approach on several
multi-domain datasets, including image classification, sentiment analysis, and
object recognition. Our experiments demonstrate that our approach consistently
outperforms baseline methods and achieves state-of-the-art performance on
several datasets. We also show that our method is highly efficient, requiring
significantly fewer labeled samples than other active learning-based methods.
Overall, our approach provides a practical and effective solution for improving
performance across multiple domains using active learning techniques.
| [
{
"version": "v1",
"created": "Thu, 13 Apr 2023 05:27:43 GMT"
}
] | 2023-04-14T00:00:00 | [
[
"Mahalingam",
"Anand Gokul",
""
],
[
"Shah",
"Aayush",
""
],
[
"Gulati",
"Akshay",
""
],
[
"Mascarenhas",
"Royston",
""
],
[
"Panduranga",
"Rakshitha",
""
]
] | TITLE: Optimizing Multi-Domain Performance with Active Learning-based
Improvement Strategies
ABSTRACT: Improving performance in multiple domains is a challenging task, and often
requires significant amounts of data to train and test models. Active learning
techniques provide a promising solution by enabling models to select the most
informative samples for labeling, thus reducing the amount of labeled data
required to achieve high performance. In this paper, we present an active
learning-based framework for improving performance across multiple domains. Our
approach consists of two stages: first, we use an initial set of labeled data
to train a base model, and then we iteratively select the most informative
samples for labeling to refine the model. We evaluate our approach on several
multi-domain datasets, including image classification, sentiment analysis, and
object recognition. Our experiments demonstrate that our approach consistently
outperforms baseline methods and achieves state-of-the-art performance on
several datasets. We also show that our method is highly efficient, requiring
significantly fewer labeled samples than other active learning-based methods.
Overall, our approach provides a practical and effective solution for improving
performance across multiple domains using active learning techniques.
| no_new_dataset | 0.711186 |
2304.06297 | Hongchen Tan | Hongchen Tan, Baocai Yin, Kun Wei, Xiuping Liu, Xin Li | ALR-GAN: Adaptive Layout Refinement for Text-to-Image Synthesis | Accepted by TMM | null | 10.1109/TMM.2023.3238554 | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | We propose a novel Text-to-Image Generation Network, Adaptive Layout
Refinement Generative Adversarial Network (ALR-GAN), to adaptively refine the
layout of synthesized images without any auxiliary information. The ALR-GAN
includes an Adaptive Layout Refinement (ALR) module and a Layout Visual
Refinement (LVR) loss. The ALR module aligns the layout structure (which refers
to locations of objects and background) of a synthesized image with that of its
corresponding real image. In ALR module, we proposed an Adaptive Layout
Refinement (ALR) loss to balance the matching of hard and easy features, for
more efficient layout structure matching. Based on the refined layout
structure, the LVR loss further refines the visual representation within the
layout area. Experimental results on two widely-used datasets show that ALR-GAN
performs competitively at the Text-to-Image generation task.
| [
{
"version": "v1",
"created": "Thu, 13 Apr 2023 07:07:01 GMT"
}
] | 2023-04-14T00:00:00 | [
[
"Tan",
"Hongchen",
""
],
[
"Yin",
"Baocai",
""
],
[
"Wei",
"Kun",
""
],
[
"Liu",
"Xiuping",
""
],
[
"Li",
"Xin",
""
]
] | TITLE: ALR-GAN: Adaptive Layout Refinement for Text-to-Image Synthesis
ABSTRACT: We propose a novel Text-to-Image Generation Network, Adaptive Layout
Refinement Generative Adversarial Network (ALR-GAN), to adaptively refine the
layout of synthesized images without any auxiliary information. The ALR-GAN
includes an Adaptive Layout Refinement (ALR) module and a Layout Visual
Refinement (LVR) loss. The ALR module aligns the layout structure (which refers
to locations of objects and background) of a synthesized image with that of its
corresponding real image. In ALR module, we proposed an Adaptive Layout
Refinement (ALR) loss to balance the matching of hard and easy features, for
more efficient layout structure matching. Based on the refined layout
structure, the LVR loss further refines the visual representation within the
layout area. Experimental results on two widely-used datasets show that ALR-GAN
performs competitively at the Text-to-Image generation task.
| no_new_dataset | 0.711825 |
2304.06299 | Arash Fayyazi | Jingkai Hong, Arash Fayyazi, Amirhossein Esmaili, Mahdi Nazemi,
Massoud Pedram | Algorithms and Hardware for Efficient Processing of Logic-based Neural
Networks | null | null | null | null | cs.AR | http://creativecommons.org/licenses/by/4.0/ | Recent efforts to improve the performance of neural network (NN) accelerators
that meet today's application requirements have given rise to a new trend of
logic-based NN inference relying on fixed-function combinational logic (FFCL).
This paper presents an innovative optimization methodology for compiling and
mapping NNs utilizing FFCL into a logic processor. The presented method maps
FFCL blocks to a set of Boolean functions where Boolean operations in each
function are mapped to high-performance, low-latency, parallelized processing
elements. Graph partitioning and scheduling algorithms are presented to handle
FFCL blocks that cannot straightforwardly fit the logic processor. Our
experimental evaluations across several datasets and NNs demonstrate the
superior performance of our framework in terms of the inference throughput
compared to prior art NN accelerators. We achieve 25x higher throughput
compared with the XNOR-based accelerator for VGG16 model that can be amplified
5x deploying the graph partitioning and merging algorithms.
| [
{
"version": "v1",
"created": "Thu, 13 Apr 2023 07:12:40 GMT"
}
] | 2023-04-14T00:00:00 | [
[
"Hong",
"Jingkai",
""
],
[
"Fayyazi",
"Arash",
""
],
[
"Esmaili",
"Amirhossein",
""
],
[
"Nazemi",
"Mahdi",
""
],
[
"Pedram",
"Massoud",
""
]
] | TITLE: Algorithms and Hardware for Efficient Processing of Logic-based Neural
Networks
ABSTRACT: Recent efforts to improve the performance of neural network (NN) accelerators
that meet today's application requirements have given rise to a new trend of
logic-based NN inference relying on fixed-function combinational logic (FFCL).
This paper presents an innovative optimization methodology for compiling and
mapping NNs utilizing FFCL into a logic processor. The presented method maps
FFCL blocks to a set of Boolean functions where Boolean operations in each
function are mapped to high-performance, low-latency, parallelized processing
elements. Graph partitioning and scheduling algorithms are presented to handle
FFCL blocks that cannot straightforwardly fit the logic processor. Our
experimental evaluations across several datasets and NNs demonstrate the
superior performance of our framework in terms of the inference throughput
compared to prior art NN accelerators. We achieve 25x higher throughput
compared with the XNOR-based accelerator for VGG16 model that can be amplified
5x deploying the graph partitioning and merging algorithms.
| no_new_dataset | 0.707582 |
2304.06305 | Zhuo Su | Zhuo Su, Jiehua Zhang, Tianpeng Liu, Zhen Liu, Shuanghui Zhang, Matti
Pietik\"ainen, Li Liu | Boosting Convolutional Neural Networks with Middle Spectrum Grouped
Convolution | 13 pages, 11 figures, submitted to IEEEE Transactions on xxx | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper proposes a novel module called middle spectrum grouped convolution
(MSGC) for efficient deep convolutional neural networks (DCNNs) with the
mechanism of grouped convolution. It explores the broad "middle spectrum" area
between channel pruning and conventional grouped convolution. Compared with
channel pruning, MSGC can retain most of the information from the input feature
maps due to the group mechanism; compared with grouped convolution, MSGC
benefits from the learnability, the core of channel pruning, for constructing
its group topology, leading to better channel division. The middle spectrum
area is unfolded along four dimensions: group-wise, layer-wise, sample-wise,
and attention-wise, making it possible to reveal more powerful and
interpretable structures. As a result, the proposed module acts as a booster
that can reduce the computational cost of the host backbones for general image
recognition with even improved predictive accuracy. For example, in the
experiments on ImageNet dataset for image classification, MSGC can reduce the
multiply-accumulates (MACs) of ResNet-18 and ResNet-50 by half but still
increase the Top-1 accuracy by more than 1%. With 35% reduction of MACs, MSGC
can also increase the Top-1 accuracy of the MobileNetV2 backbone. Results on MS
COCO dataset for object detection show similar observations. Our code and
trained models are available at https://github.com/hellozhuo/msgc.
| [
{
"version": "v1",
"created": "Thu, 13 Apr 2023 07:31:41 GMT"
}
] | 2023-04-14T00:00:00 | [
[
"Su",
"Zhuo",
""
],
[
"Zhang",
"Jiehua",
""
],
[
"Liu",
"Tianpeng",
""
],
[
"Liu",
"Zhen",
""
],
[
"Zhang",
"Shuanghui",
""
],
[
"Pietikäinen",
"Matti",
""
],
[
"Liu",
"Li",
""
]
] | TITLE: Boosting Convolutional Neural Networks with Middle Spectrum Grouped
Convolution
ABSTRACT: This paper proposes a novel module called middle spectrum grouped convolution
(MSGC) for efficient deep convolutional neural networks (DCNNs) with the
mechanism of grouped convolution. It explores the broad "middle spectrum" area
between channel pruning and conventional grouped convolution. Compared with
channel pruning, MSGC can retain most of the information from the input feature
maps due to the group mechanism; compared with grouped convolution, MSGC
benefits from the learnability, the core of channel pruning, for constructing
its group topology, leading to better channel division. The middle spectrum
area is unfolded along four dimensions: group-wise, layer-wise, sample-wise,
and attention-wise, making it possible to reveal more powerful and
interpretable structures. As a result, the proposed module acts as a booster
that can reduce the computational cost of the host backbones for general image
recognition with even improved predictive accuracy. For example, in the
experiments on ImageNet dataset for image classification, MSGC can reduce the
multiply-accumulates (MACs) of ResNet-18 and ResNet-50 by half but still
increase the Top-1 accuracy by more than 1%. With 35% reduction of MACs, MSGC
can also increase the Top-1 accuracy of the MobileNetV2 backbone. Results on MS
COCO dataset for object detection show similar observations. Our code and
trained models are available at https://github.com/hellozhuo/msgc.
| no_new_dataset | 0.708799 |
2304.06309 | Xinyun Zhang | Xinyun Zhang and Lanqing Hong | Out-of-distribution Few-shot Learning For Edge Devices without Model
Fine-tuning | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Few-shot learning (FSL) via customization of a deep learning network with
limited data has emerged as a promising technique to achieve personalized user
experiences on edge devices. However, existing FSL methods primarily assume
independent and identically distributed (IID) data and utilize either
computational backpropagation updates for each task or a common model with
task-specific prototypes. Unfortunately, the former solution is infeasible for
edge devices that lack on-device backpropagation capabilities, while the latter
often struggles with limited generalization ability, especially for
out-of-distribution (OOD) data. This paper proposes a lightweight,
plug-and-play FSL module called Task-aware Normalization (TANO) that enables
efficient and task-aware adaptation of a deep neural network without
backpropagation. TANO covers the properties of multiple user groups by
coordinating the updates of several groups of the normalization statistics
during meta-training and automatically identifies the appropriate normalization
group for a downstream few-shot task. Consequently, TANO provides stable but
task-specific estimations of the normalization statistics to close the
distribution gaps and achieve efficient model adaptation. Results on both
intra-domain and out-of-domain generalization experiments demonstrate that TANO
outperforms recent methods in terms of accuracy, inference speed, and model
size. Moreover, TANO achieves promising results on widely-used FSL benchmarks
and data from real applications.
| [
{
"version": "v1",
"created": "Thu, 13 Apr 2023 07:33:22 GMT"
}
] | 2023-04-14T00:00:00 | [
[
"Zhang",
"Xinyun",
""
],
[
"Hong",
"Lanqing",
""
]
] | TITLE: Out-of-distribution Few-shot Learning For Edge Devices without Model
Fine-tuning
ABSTRACT: Few-shot learning (FSL) via customization of a deep learning network with
limited data has emerged as a promising technique to achieve personalized user
experiences on edge devices. However, existing FSL methods primarily assume
independent and identically distributed (IID) data and utilize either
computational backpropagation updates for each task or a common model with
task-specific prototypes. Unfortunately, the former solution is infeasible for
edge devices that lack on-device backpropagation capabilities, while the latter
often struggles with limited generalization ability, especially for
out-of-distribution (OOD) data. This paper proposes a lightweight,
plug-and-play FSL module called Task-aware Normalization (TANO) that enables
efficient and task-aware adaptation of a deep neural network without
backpropagation. TANO covers the properties of multiple user groups by
coordinating the updates of several groups of the normalization statistics
during meta-training and automatically identifies the appropriate normalization
group for a downstream few-shot task. Consequently, TANO provides stable but
task-specific estimations of the normalization statistics to close the
distribution gaps and achieve efficient model adaptation. Results on both
intra-domain and out-of-domain generalization experiments demonstrate that TANO
outperforms recent methods in terms of accuracy, inference speed, and model
size. Moreover, TANO achieves promising results on widely-used FSL benchmarks
and data from real applications.
| no_new_dataset | 0.709466 |
2304.06334 | Luigi Piccinelli | Luigi Piccinelli, Christos Sakaridis, Fisher Yu | iDisc: Internal Discretization for Monocular Depth Estimation | Accepted at CVPR 2023 | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Monocular depth estimation is fundamental for 3D scene understanding and
downstream applications. However, even under the supervised setup, it is still
challenging and ill-posed due to the lack of full geometric constraints.
Although a scene can consist of millions of pixels, there are fewer high-level
patterns. We propose iDisc to learn those patterns with internal discretized
representations. The method implicitly partitions the scene into a set of
high-level patterns. In particular, our new module, Internal Discretization
(ID), implements a continuous-discrete-continuous bottleneck to learn those
concepts without supervision. In contrast to state-of-the-art methods, the
proposed model does not enforce any explicit constraints or priors on the depth
output. The whole network with the ID module can be trained end-to-end, thanks
to the bottleneck module based on attention. Our method sets the new state of
the art with significant improvements on NYU-Depth v2 and KITTI, outperforming
all published methods on the official KITTI benchmark. iDisc can also achieve
state-of-the-art results on surface normal estimation. Further, we explore the
model generalization capability via zero-shot testing. We observe the
compelling need to promote diversification in the outdoor scenario. Hence, we
introduce splits of two autonomous driving datasets, DDAD and Argoverse. Code
is available at http://vis.xyz/pub/idisc .
| [
{
"version": "v1",
"created": "Thu, 13 Apr 2023 08:29:59 GMT"
}
] | 2023-04-14T00:00:00 | [
[
"Piccinelli",
"Luigi",
""
],
[
"Sakaridis",
"Christos",
""
],
[
"Yu",
"Fisher",
""
]
] | TITLE: iDisc: Internal Discretization for Monocular Depth Estimation
ABSTRACT: Monocular depth estimation is fundamental for 3D scene understanding and
downstream applications. However, even under the supervised setup, it is still
challenging and ill-posed due to the lack of full geometric constraints.
Although a scene can consist of millions of pixels, there are fewer high-level
patterns. We propose iDisc to learn those patterns with internal discretized
representations. The method implicitly partitions the scene into a set of
high-level patterns. In particular, our new module, Internal Discretization
(ID), implements a continuous-discrete-continuous bottleneck to learn those
concepts without supervision. In contrast to state-of-the-art methods, the
proposed model does not enforce any explicit constraints or priors on the depth
output. The whole network with the ID module can be trained end-to-end, thanks
to the bottleneck module based on attention. Our method sets the new state of
the art with significant improvements on NYU-Depth v2 and KITTI, outperforming
all published methods on the official KITTI benchmark. iDisc can also achieve
state-of-the-art results on surface normal estimation. Further, we explore the
model generalization capability via zero-shot testing. We observe the
compelling need to promote diversification in the outdoor scenario. Hence, we
introduce splits of two autonomous driving datasets, DDAD and Argoverse. Code
is available at http://vis.xyz/pub/idisc .
| no_new_dataset | 0.708168 |
2304.06335 | Chien-Pin Liu | Chien-Pin Liu, Ju-Hsuan Li, En-Ping Chu, Chia-Yeh Hsieh, Kai-Chun Liu,
Chia-Tai Chan, Yu Tsao | Deep Learning-based Fall Detection Algorithm Using Ensemble Model of
Coarse-fine CNN and GRU Networks | null | null | null | null | cs.LG eess.SP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Falls are the public health issue for the elderly all over the world since
the fall-induced injuries are associated with a large amount of healthcare
cost. Falls can cause serious injuries, even leading to death if the elderly
suffers a "long-lie". Hence, a reliable fall detection (FD) system is required
to provide an emergency alarm for first aid. Due to the advances in wearable
device technology and artificial intelligence, some fall detection systems have
been developed using machine learning and deep learning methods to analyze the
signal collected from accelerometer and gyroscopes. In order to achieve better
fall detection performance, an ensemble model that combines a coarse-fine
convolutional neural network and gated recurrent unit is proposed in this
study. The parallel structure design used in this model restores the different
grains of spatial characteristics and capture temporal dependencies for feature
representation. This study applies the FallAllD public dataset to validate the
reliability of the proposed model, which achieves a recall, precision, and
F-score of 92.54%, 96.13%, and 94.26%, respectively. The results demonstrate
the reliability of the proposed ensemble model in discriminating falls from
daily living activities and its superior performance compared to the
state-of-the-art convolutional neural network long short-term memory (CNN-LSTM)
for FD.
| [
{
"version": "v1",
"created": "Thu, 13 Apr 2023 08:30:46 GMT"
}
] | 2023-04-14T00:00:00 | [
[
"Liu",
"Chien-Pin",
""
],
[
"Li",
"Ju-Hsuan",
""
],
[
"Chu",
"En-Ping",
""
],
[
"Hsieh",
"Chia-Yeh",
""
],
[
"Liu",
"Kai-Chun",
""
],
[
"Chan",
"Chia-Tai",
""
],
[
"Tsao",
"Yu",
""
]
] | TITLE: Deep Learning-based Fall Detection Algorithm Using Ensemble Model of
Coarse-fine CNN and GRU Networks
ABSTRACT: Falls are the public health issue for the elderly all over the world since
the fall-induced injuries are associated with a large amount of healthcare
cost. Falls can cause serious injuries, even leading to death if the elderly
suffers a "long-lie". Hence, a reliable fall detection (FD) system is required
to provide an emergency alarm for first aid. Due to the advances in wearable
device technology and artificial intelligence, some fall detection systems have
been developed using machine learning and deep learning methods to analyze the
signal collected from accelerometer and gyroscopes. In order to achieve better
fall detection performance, an ensemble model that combines a coarse-fine
convolutional neural network and gated recurrent unit is proposed in this
study. The parallel structure design used in this model restores the different
grains of spatial characteristics and capture temporal dependencies for feature
representation. This study applies the FallAllD public dataset to validate the
reliability of the proposed model, which achieves a recall, precision, and
F-score of 92.54%, 96.13%, and 94.26%, respectively. The results demonstrate
the reliability of the proposed ensemble model in discriminating falls from
daily living activities and its superior performance compared to the
state-of-the-art convolutional neural network long short-term memory (CNN-LSTM)
for FD.
| no_new_dataset | 0.711036 |
2304.06344 | Sergio Gonz\'alez V\'azquez | Jonathan Hans Soeseno, Sergio Gonz\'alez, Trista Pei-Chun Chen | Streamlined Framework for Agile Forecasting Model Development towards
Efficient Inventory Management | null | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by/4.0/ | This paper proposes a framework for developing forecasting models by
streamlining the connections between core components of the developmental
process. The proposed framework enables swift and robust integration of new
datasets, experimentation on different algorithms, and selection of the best
models. We start with the datasets of different issues and apply pre-processing
steps to clean and engineer meaningful representations of time-series data. To
identify robust training configurations, we introduce a novel mechanism of
multiple cross-validation strategies. We apply different evaluation metrics to
find the best-suited models for varying applications. One of the referent
applications is our participation in the intelligent forecasting competition
held by the United States Agency of International Development (USAID). Finally,
we leverage the flexibility of the framework by applying different evaluation
metrics to assess the performance of the models in inventory management
settings.
| [
{
"version": "v1",
"created": "Thu, 13 Apr 2023 08:52:32 GMT"
}
] | 2023-04-14T00:00:00 | [
[
"Soeseno",
"Jonathan Hans",
""
],
[
"González",
"Sergio",
""
],
[
"Chen",
"Trista Pei-Chun",
""
]
] | TITLE: Streamlined Framework for Agile Forecasting Model Development towards
Efficient Inventory Management
ABSTRACT: This paper proposes a framework for developing forecasting models by
streamlining the connections between core components of the developmental
process. The proposed framework enables swift and robust integration of new
datasets, experimentation on different algorithms, and selection of the best
models. We start with the datasets of different issues and apply pre-processing
steps to clean and engineer meaningful representations of time-series data. To
identify robust training configurations, we introduce a novel mechanism of
multiple cross-validation strategies. We apply different evaluation metrics to
find the best-suited models for varying applications. One of the referent
applications is our participation in the intelligent forecasting competition
held by the United States Agency of International Development (USAID). Finally,
we leverage the flexibility of the framework by applying different evaluation
metrics to assess the performance of the models in inventory management
settings.
| no_new_dataset | 0.709818 |
2304.06351 | Lorenzo Berlincioni | Lorenzo Berlincioni, Luca Cultrera, Chiara Albisani, Lisa Cresti,
Andrea Leonardo, Sara Picchioni, Federico Becattini, Alberto Del Bimbo | Neuromorphic Event-based Facial Expression Recognition | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Recently, event cameras have shown large applicability in several computer
vision fields especially concerning tasks that require high temporal
resolution. In this work, we investigate the usage of such kind of data for
emotion recognition by presenting NEFER, a dataset for Neuromorphic Event-based
Facial Expression Recognition. NEFER is composed of paired RGB and event videos
representing human faces labeled with the respective emotions and also
annotated with face bounding boxes and facial landmarks. We detail the data
acquisition process as well as providing a baseline method for RGB and event
data. The collected data captures subtle micro-expressions, which are hard to
spot with RGB data, yet emerge in the event domain. We report a double
recognition accuracy for the event-based approach, proving the effectiveness of
a neuromorphic approach for analyzing fast and hardly detectable expressions
and the emotions they conceal.
| [
{
"version": "v1",
"created": "Thu, 13 Apr 2023 09:02:10 GMT"
}
] | 2023-04-14T00:00:00 | [
[
"Berlincioni",
"Lorenzo",
""
],
[
"Cultrera",
"Luca",
""
],
[
"Albisani",
"Chiara",
""
],
[
"Cresti",
"Lisa",
""
],
[
"Leonardo",
"Andrea",
""
],
[
"Picchioni",
"Sara",
""
],
[
"Becattini",
"Federico",
""
],
[
"Del Bimbo",
"Alberto",
""
]
] | TITLE: Neuromorphic Event-based Facial Expression Recognition
ABSTRACT: Recently, event cameras have shown large applicability in several computer
vision fields especially concerning tasks that require high temporal
resolution. In this work, we investigate the usage of such kind of data for
emotion recognition by presenting NEFER, a dataset for Neuromorphic Event-based
Facial Expression Recognition. NEFER is composed of paired RGB and event videos
representing human faces labeled with the respective emotions and also
annotated with face bounding boxes and facial landmarks. We detail the data
acquisition process as well as providing a baseline method for RGB and event
data. The collected data captures subtle micro-expressions, which are hard to
spot with RGB data, yet emerge in the event domain. We report a double
recognition accuracy for the event-based approach, proving the effectiveness of
a neuromorphic approach for analyzing fast and hardly detectable expressions
and the emotions they conceal.
| new_dataset | 0.713793 |
2304.06359 | Shun Lei | Shun Lei, Yixuan Zhou, Liyang Chen, Zhiyong Wu, Shiyin Kang, Helen
Meng | Context-aware Coherent Speaking Style Prediction with Hierarchical
Transformers for Audiobook Speech Synthesis | Accepted by ICASSP 2023 | null | null | null | cs.SD eess.AS | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recent advances in text-to-speech have significantly improved the
expressiveness of synthesized speech. However, it is still challenging to
generate speech with contextually appropriate and coherent speaking style for
multi-sentence text in audiobooks. In this paper, we propose a context-aware
coherent speaking style prediction method for audiobook speech synthesis. To
predict the style embedding of the current utterance, a hierarchical
transformer-based context-aware style predictor with a mixture attention mask
is designed, considering both text-side context information and speech-side
style information of previous speeches. Based on this, we can generate
long-form speech with coherent style and prosody sentence by sentence.
Objective and subjective evaluations on a Mandarin audiobook dataset
demonstrate that our proposed model can generate speech with more expressive
and coherent speaking style than baselines, for both single-sentence and
multi-sentence test.
| [
{
"version": "v1",
"created": "Thu, 13 Apr 2023 09:26:04 GMT"
}
] | 2023-04-14T00:00:00 | [
[
"Lei",
"Shun",
""
],
[
"Zhou",
"Yixuan",
""
],
[
"Chen",
"Liyang",
""
],
[
"Wu",
"Zhiyong",
""
],
[
"Kang",
"Shiyin",
""
],
[
"Meng",
"Helen",
""
]
] | TITLE: Context-aware Coherent Speaking Style Prediction with Hierarchical
Transformers for Audiobook Speech Synthesis
ABSTRACT: Recent advances in text-to-speech have significantly improved the
expressiveness of synthesized speech. However, it is still challenging to
generate speech with contextually appropriate and coherent speaking style for
multi-sentence text in audiobooks. In this paper, we propose a context-aware
coherent speaking style prediction method for audiobook speech synthesis. To
predict the style embedding of the current utterance, a hierarchical
transformer-based context-aware style predictor with a mixture attention mask
is designed, considering both text-side context information and speech-side
style information of previous speeches. Based on this, we can generate
long-form speech with coherent style and prosody sentence by sentence.
Objective and subjective evaluations on a Mandarin audiobook dataset
demonstrate that our proposed model can generate speech with more expressive
and coherent speaking style than baselines, for both single-sentence and
multi-sentence test.
| no_new_dataset | 0.7118 |
2304.06375 | Salvatore Citraro | Salvatore Citraro and Simon De Deyne and Massimo Stella and Giulio
Rossetti | Towards hypergraph cognitive networks as feature-rich models of
knowledge | null | null | null | null | cs.CL cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Semantic networks provide a useful tool to understand how related concepts
are retrieved from memory. However, most current network approaches use
pairwise links to represent memory recall patterns. Pairwise connections
neglect higher-order associations, i.e. relationships between more than two
concepts at a time. These higher-order interactions might covariate with (and
thus contain information about) how similar concepts are along psycholinguistic
dimensions like arousal, valence, familiarity, gender and others. We overcome
these limits by introducing feature-rich cognitive hypergraphs as quantitative
models of human memory where: (i) concepts recalled together can all engage in
hyperlinks involving also more than two concepts at once (cognitive hypergraph
aspect), and (ii) each concept is endowed with a vector of psycholinguistic
features (feature-rich aspect). We build hypergraphs from word association data
and use evaluation methods from machine learning features to predict concept
concreteness. Since concepts with similar concreteness tend to cluster together
in human memory, we expect to be able to leverage this structure. Using word
association data from the Small World of Words dataset, we compared a pairwise
network and a hypergraph with N=3586 concepts/nodes. Interpretable artificial
intelligence models trained on (1) psycholinguistic features only, (2)
pairwise-based feature aggregations, and on (3) hypergraph-based aggregations
show significant differences between pairwise and hypergraph links.
Specifically, our results show that higher-order and feature-rich hypergraph
models contain richer information than pairwise networks leading to improved
prediction of word concreteness. The relation with previous studies about
conceptual clustering and compartmentalisation in associative knowledge and
human memory are discussed.
| [
{
"version": "v1",
"created": "Thu, 13 Apr 2023 09:56:45 GMT"
}
] | 2023-04-14T00:00:00 | [
[
"Citraro",
"Salvatore",
""
],
[
"De Deyne",
"Simon",
""
],
[
"Stella",
"Massimo",
""
],
[
"Rossetti",
"Giulio",
""
]
] | TITLE: Towards hypergraph cognitive networks as feature-rich models of
knowledge
ABSTRACT: Semantic networks provide a useful tool to understand how related concepts
are retrieved from memory. However, most current network approaches use
pairwise links to represent memory recall patterns. Pairwise connections
neglect higher-order associations, i.e. relationships between more than two
concepts at a time. These higher-order interactions might covariate with (and
thus contain information about) how similar concepts are along psycholinguistic
dimensions like arousal, valence, familiarity, gender and others. We overcome
these limits by introducing feature-rich cognitive hypergraphs as quantitative
models of human memory where: (i) concepts recalled together can all engage in
hyperlinks involving also more than two concepts at once (cognitive hypergraph
aspect), and (ii) each concept is endowed with a vector of psycholinguistic
features (feature-rich aspect). We build hypergraphs from word association data
and use evaluation methods from machine learning features to predict concept
concreteness. Since concepts with similar concreteness tend to cluster together
in human memory, we expect to be able to leverage this structure. Using word
association data from the Small World of Words dataset, we compared a pairwise
network and a hypergraph with N=3586 concepts/nodes. Interpretable artificial
intelligence models trained on (1) psycholinguistic features only, (2)
pairwise-based feature aggregations, and on (3) hypergraph-based aggregations
show significant differences between pairwise and hypergraph links.
Specifically, our results show that higher-order and feature-rich hypergraph
models contain richer information than pairwise networks leading to improved
prediction of word concreteness. The relation with previous studies about
conceptual clustering and compartmentalisation in associative knowledge and
human memory are discussed.
| no_new_dataset | 0.711387 |
2304.06378 | Arun Palla | Arun Palla, Sriprabha Ramanarayanan, Keerthi Ram and Mohanasankar
Sivaprakasam | Generalizable Deep Learning Method for Suppressing Unseen and Multiple
MRI Artifacts Using Meta-learning | 5 pages, 6 figures, Accepted in EMBC 2023 | null | null | null | eess.IV cs.CV | http://creativecommons.org/licenses/by/4.0/ | Magnetic Resonance (MR) images suffer from various types of artifacts due to
motion, spatial resolution, and under-sampling. Conventional deep learning
methods deal with removing a specific type of artifact, leading to separately
trained models for each artifact type that lack the shared knowledge
generalizable across artifacts. Moreover, training a model for each type and
amount of artifact is a tedious process that consumes more training time and
storage of models. On the other hand, the shared knowledge learned by jointly
training the model on multiple artifacts might be inadequate to generalize
under deviations in the types and amounts of artifacts. Model-agnostic
meta-learning (MAML), a nested bi-level optimization framework is a promising
technique to learn common knowledge across artifacts in the outer level of
optimization, and artifact-specific restoration in the inner level. We propose
curriculum-MAML (CMAML), a learning process that integrates MAML with
curriculum learning to impart the knowledge of variable artifact complexity to
adaptively learn restoration of multiple artifacts during training. Comparative
studies against Stochastic Gradient Descent and MAML, using two cardiac
datasets reveal that CMAML exhibits (i) better generalization with improved
PSNR for 83% of unseen types and amounts of artifacts and improved SSIM in all
cases, and (ii) better artifact suppression in 4 out of 5 cases of composite
artifacts (scans with multiple artifacts).
| [
{
"version": "v1",
"created": "Thu, 13 Apr 2023 10:17:34 GMT"
}
] | 2023-04-14T00:00:00 | [
[
"Palla",
"Arun",
""
],
[
"Ramanarayanan",
"Sriprabha",
""
],
[
"Ram",
"Keerthi",
""
],
[
"Sivaprakasam",
"Mohanasankar",
""
]
] | TITLE: Generalizable Deep Learning Method for Suppressing Unseen and Multiple
MRI Artifacts Using Meta-learning
ABSTRACT: Magnetic Resonance (MR) images suffer from various types of artifacts due to
motion, spatial resolution, and under-sampling. Conventional deep learning
methods deal with removing a specific type of artifact, leading to separately
trained models for each artifact type that lack the shared knowledge
generalizable across artifacts. Moreover, training a model for each type and
amount of artifact is a tedious process that consumes more training time and
storage of models. On the other hand, the shared knowledge learned by jointly
training the model on multiple artifacts might be inadequate to generalize
under deviations in the types and amounts of artifacts. Model-agnostic
meta-learning (MAML), a nested bi-level optimization framework is a promising
technique to learn common knowledge across artifacts in the outer level of
optimization, and artifact-specific restoration in the inner level. We propose
curriculum-MAML (CMAML), a learning process that integrates MAML with
curriculum learning to impart the knowledge of variable artifact complexity to
adaptively learn restoration of multiple artifacts during training. Comparative
studies against Stochastic Gradient Descent and MAML, using two cardiac
datasets reveal that CMAML exhibits (i) better generalization with improved
PSNR for 83% of unseen types and amounts of artifacts and improved SSIM in all
cases, and (ii) better artifact suppression in 4 out of 5 cases of composite
artifacts (scans with multiple artifacts).
| no_new_dataset | 0.711838 |
2304.06401 | Martin Thi{\ss}en | Martin Thi{\ss}en and Elke Hergenr\"other | Why Existing Multimodal Crowd Counting Datasets Can Lead to Unfulfilled
Expectations in Real-World Applications | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | More information leads to better decisions and predictions, right? Confirming
this hypothesis, several studies concluded that the simultaneous use of optical
and thermal images leads to better predictions in crowd counting. However, the
way multimodal models extract enriched features from both modalities is not yet
fully understood. Since the use of multimodal data usually increases the
complexity, inference time, and memory requirements of the models, it is
relevant to examine the differences and advantages of multimodal compared to
monomodal models. In this work, all available multimodal datasets for crowd
counting are used to investigate the differences between monomodal and
multimodal models. To do so, we designed a monomodal architecture that
considers the current state of research on monomodal crowd counting. In
addition, several multimodal architectures have been developed using different
multimodal learning strategies. The key components of the monomodal
architecture are also used in the multimodal architectures to be able to answer
whether multimodal models perform better in crowd counting in general.
Surprisingly, no general answer to this question can be derived from the
existing datasets. We found that the existing datasets hold a bias toward
thermal images. This was determined by analyzing the relationship between the
brightness of optical images and crowd count as well as examining the
annotations made for each dataset. Since answering this question is important
for future real-world applications of crowd counting, this paper establishes
criteria for a potential dataset suitable for answering whether multimodal
models perform better in crowd counting in general.
| [
{
"version": "v1",
"created": "Thu, 13 Apr 2023 11:09:28 GMT"
}
] | 2023-04-14T00:00:00 | [
[
"Thißen",
"Martin",
""
],
[
"Hergenröther",
"Elke",
""
]
] | TITLE: Why Existing Multimodal Crowd Counting Datasets Can Lead to Unfulfilled
Expectations in Real-World Applications
ABSTRACT: More information leads to better decisions and predictions, right? Confirming
this hypothesis, several studies concluded that the simultaneous use of optical
and thermal images leads to better predictions in crowd counting. However, the
way multimodal models extract enriched features from both modalities is not yet
fully understood. Since the use of multimodal data usually increases the
complexity, inference time, and memory requirements of the models, it is
relevant to examine the differences and advantages of multimodal compared to
monomodal models. In this work, all available multimodal datasets for crowd
counting are used to investigate the differences between monomodal and
multimodal models. To do so, we designed a monomodal architecture that
considers the current state of research on monomodal crowd counting. In
addition, several multimodal architectures have been developed using different
multimodal learning strategies. The key components of the monomodal
architecture are also used in the multimodal architectures to be able to answer
whether multimodal models perform better in crowd counting in general.
Surprisingly, no general answer to this question can be derived from the
existing datasets. We found that the existing datasets hold a bias toward
thermal images. This was determined by analyzing the relationship between the
brightness of optical images and crowd count as well as examining the
annotations made for each dataset. Since answering this question is important
for future real-world applications of crowd counting, this paper establishes
criteria for a potential dataset suitable for answering whether multimodal
models perform better in crowd counting in general.
| no_new_dataset | 0.709057 |
2304.06411 | Qiongjie Cui | Qiongjie Cui, Huaijiang Sun, Jianfeng Lu, Bin Li, Weiqing Li | Meta-Auxiliary Learning for Adaptive Human Pose Prediction | 10 pages, 6 figures, AAAI 2023 accepted | null | null | null | cs.CV cs.AI cs.HC | http://creativecommons.org/licenses/by/4.0/ | Predicting high-fidelity future human poses, from a historically observed
sequence, is decisive for intelligent robots to interact with humans. Deep
end-to-end learning approaches, which typically train a generic pre-trained
model on external datasets and then directly apply it to all test samples,
emerge as the dominant solution to solve this issue. Despite encouraging
progress, they remain non-optimal, as the unique properties (e.g., motion
style, rhythm) of a specific sequence cannot be adapted. More generally, at
test-time, once encountering unseen motion categories (out-of-distribution),
the predicted poses tend to be unreliable. Motivated by this observation, we
propose a novel test-time adaptation framework that leverages two
self-supervised auxiliary tasks to help the primary forecasting network adapt
to the test sequence. In the testing phase, our model can adjust the model
parameters by several gradient updates to improve the generation quality.
However, due to catastrophic forgetting, both auxiliary tasks typically tend to
the low ability to automatically present the desired positive incentives for
the final prediction performance. For this reason, we also propose a
meta-auxiliary learning scheme for better adaptation. In terms of general
setup, our approach obtains higher accuracy, and under two new experimental
designs for out-of-distribution data (unseen subjects and categories), achieves
significant improvements.
| [
{
"version": "v1",
"created": "Thu, 13 Apr 2023 11:17:09 GMT"
}
] | 2023-04-14T00:00:00 | [
[
"Cui",
"Qiongjie",
""
],
[
"Sun",
"Huaijiang",
""
],
[
"Lu",
"Jianfeng",
""
],
[
"Li",
"Bin",
""
],
[
"Li",
"Weiqing",
""
]
] | TITLE: Meta-Auxiliary Learning for Adaptive Human Pose Prediction
ABSTRACT: Predicting high-fidelity future human poses, from a historically observed
sequence, is decisive for intelligent robots to interact with humans. Deep
end-to-end learning approaches, which typically train a generic pre-trained
model on external datasets and then directly apply it to all test samples,
emerge as the dominant solution to solve this issue. Despite encouraging
progress, they remain non-optimal, as the unique properties (e.g., motion
style, rhythm) of a specific sequence cannot be adapted. More generally, at
test-time, once encountering unseen motion categories (out-of-distribution),
the predicted poses tend to be unreliable. Motivated by this observation, we
propose a novel test-time adaptation framework that leverages two
self-supervised auxiliary tasks to help the primary forecasting network adapt
to the test sequence. In the testing phase, our model can adjust the model
parameters by several gradient updates to improve the generation quality.
However, due to catastrophic forgetting, both auxiliary tasks typically tend to
the low ability to automatically present the desired positive incentives for
the final prediction performance. For this reason, we also propose a
meta-auxiliary learning scheme for better adaptation. In terms of general
setup, our approach obtains higher accuracy, and under two new experimental
designs for out-of-distribution data (unseen subjects and categories), achieves
significant improvements.
| no_new_dataset | 0.707771 |
2304.06414 | Lauritz Thamsen | Demetris Trihinas, Lauritz Thamsen, Jossekin Beilharz, Moysis
Symeonides | Towards Energy Consumption and Carbon Footprint Testing for AI-driven
IoT Services | Presented at the 2nd International Workshop on Testing Distributed
Internet of Things Systems (TDIS 2022) | 2022 IEEE International Conference on Cloud Engineering (IC2E
2022) | 10.1109/IC2E55432.2022.00011 | null | cs.DC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Energy consumption and carbon emissions are expected to be crucial factors
for Internet of Things (IoT) applications. Both the scale and the
geo-distribution keep increasing, while Artificial Intelligence (AI) further
penetrates the "edge" in order to satisfy the need for highly-responsive and
intelligent services. To date, several edge/fog emulators are catering for IoT
testing by supporting the deployment and execution of AI-driven IoT services in
consolidated test environments. These tools enable the configuration of
infrastructures so that they closely resemble edge devices and IoT networks.
However, energy consumption and carbon emissions estimations during the testing
of AI services are still missing from the current state of IoT testing suites.
This study highlights important questions that developers of AI-driven IoT
services are in need of answers, along with a set of observations and
challenges, aiming to help researchers designing IoT testing and benchmarking
suites to cater to user needs.
| [
{
"version": "v1",
"created": "Thu, 13 Apr 2023 11:19:12 GMT"
}
] | 2023-04-14T00:00:00 | [
[
"Trihinas",
"Demetris",
""
],
[
"Thamsen",
"Lauritz",
""
],
[
"Beilharz",
"Jossekin",
""
],
[
"Symeonides",
"Moysis",
""
]
] | TITLE: Towards Energy Consumption and Carbon Footprint Testing for AI-driven
IoT Services
ABSTRACT: Energy consumption and carbon emissions are expected to be crucial factors
for Internet of Things (IoT) applications. Both the scale and the
geo-distribution keep increasing, while Artificial Intelligence (AI) further
penetrates the "edge" in order to satisfy the need for highly-responsive and
intelligent services. To date, several edge/fog emulators are catering for IoT
testing by supporting the deployment and execution of AI-driven IoT services in
consolidated test environments. These tools enable the configuration of
infrastructures so that they closely resemble edge devices and IoT networks.
However, energy consumption and carbon emissions estimations during the testing
of AI services are still missing from the current state of IoT testing suites.
This study highlights important questions that developers of AI-driven IoT
services are in need of answers, along with a set of observations and
challenges, aiming to help researchers designing IoT testing and benchmarking
suites to cater to user needs.
| no_new_dataset | 0.710396 |
2304.06419 | Denys Rozumnyi | Denys Rozumnyi, Jiri Matas, Marc Pollefeys, Vittorio Ferrari, Martin
R. Oswald | Tracking by 3D Model Estimation of Unknown Objects in Videos | null | null | null | null | cs.CV cs.GR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Most model-free visual object tracking methods formulate the tracking task as
object location estimation given by a 2D segmentation or a bounding box in each
video frame. We argue that this representation is limited and instead propose
to guide and improve 2D tracking with an explicit object representation, namely
the textured 3D shape and 6DoF pose in each video frame. Our representation
tackles a complex long-term dense correspondence problem between all 3D points
on the object for all video frames, including frames where some points are
invisible. To achieve that, the estimation is driven by re-rendering the input
video frames as well as possible through differentiable rendering, which has
not been used for tracking before. The proposed optimization minimizes a novel
loss function to estimate the best 3D shape, texture, and 6DoF pose. We improve
the state-of-the-art in 2D segmentation tracking on three different datasets
with mostly rigid objects.
| [
{
"version": "v1",
"created": "Thu, 13 Apr 2023 11:32:36 GMT"
}
] | 2023-04-14T00:00:00 | [
[
"Rozumnyi",
"Denys",
""
],
[
"Matas",
"Jiri",
""
],
[
"Pollefeys",
"Marc",
""
],
[
"Ferrari",
"Vittorio",
""
],
[
"Oswald",
"Martin R.",
""
]
] | TITLE: Tracking by 3D Model Estimation of Unknown Objects in Videos
ABSTRACT: Most model-free visual object tracking methods formulate the tracking task as
object location estimation given by a 2D segmentation or a bounding box in each
video frame. We argue that this representation is limited and instead propose
to guide and improve 2D tracking with an explicit object representation, namely
the textured 3D shape and 6DoF pose in each video frame. Our representation
tackles a complex long-term dense correspondence problem between all 3D points
on the object for all video frames, including frames where some points are
invisible. To achieve that, the estimation is driven by re-rendering the input
video frames as well as possible through differentiable rendering, which has
not been used for tracking before. The proposed optimization minimizes a novel
loss function to estimate the best 3D shape, texture, and 6DoF pose. We improve
the state-of-the-art in 2D segmentation tracking on three different datasets
with mostly rigid objects.
| no_new_dataset | 0.710007 |
2304.06440 | Kai Zhao | Kai Zhao, Kun Yuan, Ming Sun and Xing Wen | Zoom-VQA: Patches, Frames and Clips Integration for Video Quality
Assessment | Accepted by CVPR 2023 Workshop | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Video quality assessment (VQA) aims to simulate the human perception of video
quality, which is influenced by factors ranging from low-level color and
texture details to high-level semantic content. To effectively model these
complicated quality-related factors, in this paper, we decompose video into
three levels (\ie, patch level, frame level, and clip level), and propose a
novel Zoom-VQA architecture to perceive spatio-temporal features at different
levels. It integrates three components: patch attention module, frame pyramid
alignment, and clip ensemble strategy, respectively for capturing
region-of-interest in the spatial dimension, multi-level information at
different feature levels, and distortions distributed over the temporal
dimension. Owing to the comprehensive design, Zoom-VQA obtains state-of-the-art
results on four VQA benchmarks and achieves 2nd place in the NTIRE 2023 VQA
challenge. Notably, Zoom-VQA has outperformed the previous best results on two
subsets of LSVQ, achieving 0.8860 (+1.0%) and 0.7985 (+1.9%) of SRCC on the
respective subsets. Adequate ablation studies further verify the effectiveness
of each component. Codes and models are released in
https://github.com/k-zha14/Zoom-VQA.
| [
{
"version": "v1",
"created": "Thu, 13 Apr 2023 12:18:15 GMT"
}
] | 2023-04-14T00:00:00 | [
[
"Zhao",
"Kai",
""
],
[
"Yuan",
"Kun",
""
],
[
"Sun",
"Ming",
""
],
[
"Wen",
"Xing",
""
]
] | TITLE: Zoom-VQA: Patches, Frames and Clips Integration for Video Quality
Assessment
ABSTRACT: Video quality assessment (VQA) aims to simulate the human perception of video
quality, which is influenced by factors ranging from low-level color and
texture details to high-level semantic content. To effectively model these
complicated quality-related factors, in this paper, we decompose video into
three levels (\ie, patch level, frame level, and clip level), and propose a
novel Zoom-VQA architecture to perceive spatio-temporal features at different
levels. It integrates three components: patch attention module, frame pyramid
alignment, and clip ensemble strategy, respectively for capturing
region-of-interest in the spatial dimension, multi-level information at
different feature levels, and distortions distributed over the temporal
dimension. Owing to the comprehensive design, Zoom-VQA obtains state-of-the-art
results on four VQA benchmarks and achieves 2nd place in the NTIRE 2023 VQA
challenge. Notably, Zoom-VQA has outperformed the previous best results on two
subsets of LSVQ, achieving 0.8860 (+1.0%) and 0.7985 (+1.9%) of SRCC on the
respective subsets. Adequate ablation studies further verify the effectiveness
of each component. Codes and models are released in
https://github.com/k-zha14/Zoom-VQA.
| no_new_dataset | 0.709868 |
2304.06469 | Hamed Haddadi | Yuting Zhan, Hamed Haddadi, Afra Mashhadi | Analysing Fairness of Privacy-Utility Mobility Models | null | null | null | null | cs.LG cs.AI cs.CR cs.CY | http://creativecommons.org/licenses/by/4.0/ | Preserving the individuals' privacy in sharing spatial-temporal datasets is
critical to prevent re-identification attacks based on unique trajectories.
Existing privacy techniques tend to propose ideal privacy-utility tradeoffs,
however, largely ignore the fairness implications of mobility models and
whether such techniques perform equally for different groups of users. The
quantification between fairness and privacy-aware models is still unclear and
there barely exists any defined sets of metrics for measuring fairness in the
spatial-temporal context. In this work, we define a set of fairness metrics
designed explicitly for human mobility, based on structural similarity and
entropy of the trajectories. Under these definitions, we examine the fairness
of two state-of-the-art privacy-preserving models that rely on GAN and
representation learning to reduce the re-identification rate of users for data
sharing. Our results show that while both models guarantee group fairness in
terms of demographic parity, they violate individual fairness criteria,
indicating that users with highly similar trajectories receive disparate
privacy gain. We conclude that the tension between the re-identification task
and individual fairness needs to be considered for future spatial-temporal data
analysis and modelling to achieve a privacy-preserving fairness-aware setting.
| [
{
"version": "v1",
"created": "Mon, 10 Apr 2023 11:09:18 GMT"
}
] | 2023-04-14T00:00:00 | [
[
"Zhan",
"Yuting",
""
],
[
"Haddadi",
"Hamed",
""
],
[
"Mashhadi",
"Afra",
""
]
] | TITLE: Analysing Fairness of Privacy-Utility Mobility Models
ABSTRACT: Preserving the individuals' privacy in sharing spatial-temporal datasets is
critical to prevent re-identification attacks based on unique trajectories.
Existing privacy techniques tend to propose ideal privacy-utility tradeoffs,
however, largely ignore the fairness implications of mobility models and
whether such techniques perform equally for different groups of users. The
quantification between fairness and privacy-aware models is still unclear and
there barely exists any defined sets of metrics for measuring fairness in the
spatial-temporal context. In this work, we define a set of fairness metrics
designed explicitly for human mobility, based on structural similarity and
entropy of the trajectories. Under these definitions, we examine the fairness
of two state-of-the-art privacy-preserving models that rely on GAN and
representation learning to reduce the re-identification rate of users for data
sharing. Our results show that while both models guarantee group fairness in
terms of demographic parity, they violate individual fairness criteria,
indicating that users with highly similar trajectories receive disparate
privacy gain. We conclude that the tension between the re-identification task
and individual fairness needs to be considered for future spatial-temporal data
analysis and modelling to achieve a privacy-preserving fairness-aware setting.
| no_new_dataset | 0.707815 |
2304.06471 | Ruiqi Yang | Eric Modesitt, Ruiqi Yang, Qi Liu | Two Heads are Better than One: A Bio-inspired Method for Improving
Classification on EEG-ET Data | 6 pages, 3 figures, HCI International 2023 Poster | null | null | null | eess.SP cs.HC cs.LG | http://creativecommons.org/licenses/by/4.0/ | Classifying EEG data is integral to the performance of Brain Computer
Interfaces (BCI) and their applications. However, external noise often
obstructs EEG data due to its biological nature and complex data collection
process. Especially when dealing with classification tasks, standard EEG
preprocessing approaches extract relevant events and features from the entire
dataset. However, these approaches treat all relevant cognitive events equally
and overlook the dynamic nature of the brain over time. In contrast, we are
inspired by neuroscience studies to use a novel approach that integrates
feature selection and time segmentation of EEG data. When tested on the
EEGEyeNet dataset, our proposed method significantly increases the performance
of Machine Learning classifiers while reducing their respective computational
complexity.
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 23:44:39 GMT"
}
] | 2023-04-14T00:00:00 | [
[
"Modesitt",
"Eric",
""
],
[
"Yang",
"Ruiqi",
""
],
[
"Liu",
"Qi",
""
]
] | TITLE: Two Heads are Better than One: A Bio-inspired Method for Improving
Classification on EEG-ET Data
ABSTRACT: Classifying EEG data is integral to the performance of Brain Computer
Interfaces (BCI) and their applications. However, external noise often
obstructs EEG data due to its biological nature and complex data collection
process. Especially when dealing with classification tasks, standard EEG
preprocessing approaches extract relevant events and features from the entire
dataset. However, these approaches treat all relevant cognitive events equally
and overlook the dynamic nature of the brain over time. In contrast, we are
inspired by neuroscience studies to use a novel approach that integrates
feature selection and time segmentation of EEG data. When tested on the
EEGEyeNet dataset, our proposed method significantly increases the performance
of Machine Learning classifiers while reducing their respective computational
complexity.
| no_new_dataset | 0.709617 |
2304.06484 | Mark Hamilton | Howard Zhong, Mark Hamilton | Exploring Gender and Race Biases in the NFT Market | null | null | 10.1016/j.frl.2023.103651 | null | cs.CY cs.LG cs.SI econ.GN q-fin.EC stat.AP | http://creativecommons.org/licenses/by/4.0/ | Non-Fungible Tokens (NFTs) are non-interchangeable assets, usually digital
art, which are stored on the blockchain. Preliminary studies find that female
and darker-skinned NFTs are valued less than their male and lighter-skinned
counterparts. However, these studies analyze only the CryptoPunks collection.
We test the statistical significance of race and gender biases in the prices of
CryptoPunks and present the first study of gender bias in the broader NFT
market. We find evidence of racial bias but not gender bias. Our work also
introduces a dataset of gender-labeled NFT collections to advance the broader
study of social equity in this emerging market.
| [
{
"version": "v1",
"created": "Wed, 29 Mar 2023 17:38:11 GMT"
}
] | 2023-04-14T00:00:00 | [
[
"Zhong",
"Howard",
""
],
[
"Hamilton",
"Mark",
""
]
] | TITLE: Exploring Gender and Race Biases in the NFT Market
ABSTRACT: Non-Fungible Tokens (NFTs) are non-interchangeable assets, usually digital
art, which are stored on the blockchain. Preliminary studies find that female
and darker-skinned NFTs are valued less than their male and lighter-skinned
counterparts. However, these studies analyze only the CryptoPunks collection.
We test the statistical significance of race and gender biases in the prices of
CryptoPunks and present the first study of gender bias in the broader NFT
market. We find evidence of racial bias but not gender bias. Our work also
introduces a dataset of gender-labeled NFT collections to advance the broader
study of social equity in this emerging market.
| new_dataset | 0.711049 |
2304.06490 | Li-Hsiang Shen | Wen Liu, An-Hung Hsiao, Li-Hsiang Shen, Kai-Ten Feng | A New Paradigm for Device-free Indoor Localization: Deep Learning with
Error Vector Spectrum in Wi-Fi Systems | null | null | null | null | eess.SP cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The demand for device-free indoor localization using commercial Wi-Fi devices
has rapidly increased in various fields due to its convenience and versatile
applications. However, random frequency offset (RFO) in wireless channels poses
challenges to the accuracy of indoor localization when using fluctuating
channel state information (CSI). To mitigate the RFO problem, an error vector
spectrum (EVS) is conceived thanks to its higher resolution of signal and
robustness to RFO. To address these challenges, this paper proposed a novel
error vector assisted learning (EVAL) for device-free indoor localization. The
proposed EVAL scheme employs deep neural networks to classify the location of a
person in the indoor environment by extracting ample channel features from the
physical layer signals. We conducted realistic experiments based on OpenWiFi
project to extract both EVS and CSI to examine the performance of different
device-free localization techniques. Experimental results show that our
proposed EVAL scheme outperforms conventional machine learning methods and
benchmarks utilizing either CSI amplitude or phase information. Compared to
most existing CSI-based localization schemes, a new paradigm with higher
positioning accuracy by adopting EVS is revealed by our proposed EVAL system.
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 04:33:37 GMT"
}
] | 2023-04-14T00:00:00 | [
[
"Liu",
"Wen",
""
],
[
"Hsiao",
"An-Hung",
""
],
[
"Shen",
"Li-Hsiang",
""
],
[
"Feng",
"Kai-Ten",
""
]
] | TITLE: A New Paradigm for Device-free Indoor Localization: Deep Learning with
Error Vector Spectrum in Wi-Fi Systems
ABSTRACT: The demand for device-free indoor localization using commercial Wi-Fi devices
has rapidly increased in various fields due to its convenience and versatile
applications. However, random frequency offset (RFO) in wireless channels poses
challenges to the accuracy of indoor localization when using fluctuating
channel state information (CSI). To mitigate the RFO problem, an error vector
spectrum (EVS) is conceived thanks to its higher resolution of signal and
robustness to RFO. To address these challenges, this paper proposed a novel
error vector assisted learning (EVAL) for device-free indoor localization. The
proposed EVAL scheme employs deep neural networks to classify the location of a
person in the indoor environment by extracting ample channel features from the
physical layer signals. We conducted realistic experiments based on OpenWiFi
project to extract both EVS and CSI to examine the performance of different
device-free localization techniques. Experimental results show that our
proposed EVAL scheme outperforms conventional machine learning methods and
benchmarks utilizing either CSI amplitude or phase information. Compared to
most existing CSI-based localization schemes, a new paradigm with higher
positioning accuracy by adopting EVS is revealed by our proposed EVAL system.
| no_new_dataset | 0.711055 |
2304.06495 | Pierre Guetschel | Pierre Guetschel and Th\'eodore Papadopoulo and Michael Tangermann | An embedding for EEG signals learned using a triplet loss | 23 pages, 11 figures, 5 appendix pages, 6 appendix figures, work
conducted in 2020-2021 during an ARPE
(https://ens-paris-saclay.fr/en/masters/ens-paris-saclay-degree/year-pre-doctoral-research-abroad-arpe) | null | null | null | eess.SP cs.LG | http://creativecommons.org/licenses/by/4.0/ | Neurophysiological time series recordings like the electroencephalogram (EEG)
or local field potentials are obtained from multiple sensors. They can be
decoded by machine learning models in order to estimate the ongoing brain state
of a patient or healthy user. In a brain-computer interface (BCI), this decoded
brain state information can be used with minimal time delay to either control
an application, e.g., for communication or for rehabilitation after stroke, or
to passively monitor the ongoing brain state of the subject, e.g., in a
demanding work environment. A specific challenge in such decoding tasks is
posed by the small dataset sizes in BCI compared to other domains of machine
learning like computer vision or natural language processing. A possibility to
tackle classification or regression problems in BCI despite small training data
sets is through transfer learning, which utilizes data from other sessions,
subjects or even datasets to train a model. In this exploratory study, we
propose novel domain-specific embeddings for neurophysiological data. Our
approach is based on metric learning and builds upon the recently proposed
ladder loss. Using embeddings allowed us to benefit, both from the good
generalisation abilities and robustness of deep learning and from the fast
training of classical machine learning models for subject-specific calibration.
In offline analyses using EEG data of 14 subjects, we tested the embeddings'
feasibility and compared their efficiency with state-of-the-art deep learning
models and conventional machine learning pipelines. In summary, we propose the
use of metric learning to obtain pre-trained embeddings of EEG-BCI data as a
means to incorporate domain knowledge and to reach competitive performance on
novel subjects with minimal calibration requirements.
| [
{
"version": "v1",
"created": "Thu, 23 Mar 2023 09:05:20 GMT"
}
] | 2023-04-14T00:00:00 | [
[
"Guetschel",
"Pierre",
""
],
[
"Papadopoulo",
"Théodore",
""
],
[
"Tangermann",
"Michael",
""
]
] | TITLE: An embedding for EEG signals learned using a triplet loss
ABSTRACT: Neurophysiological time series recordings like the electroencephalogram (EEG)
or local field potentials are obtained from multiple sensors. They can be
decoded by machine learning models in order to estimate the ongoing brain state
of a patient or healthy user. In a brain-computer interface (BCI), this decoded
brain state information can be used with minimal time delay to either control
an application, e.g., for communication or for rehabilitation after stroke, or
to passively monitor the ongoing brain state of the subject, e.g., in a
demanding work environment. A specific challenge in such decoding tasks is
posed by the small dataset sizes in BCI compared to other domains of machine
learning like computer vision or natural language processing. A possibility to
tackle classification or regression problems in BCI despite small training data
sets is through transfer learning, which utilizes data from other sessions,
subjects or even datasets to train a model. In this exploratory study, we
propose novel domain-specific embeddings for neurophysiological data. Our
approach is based on metric learning and builds upon the recently proposed
ladder loss. Using embeddings allowed us to benefit, both from the good
generalisation abilities and robustness of deep learning and from the fast
training of classical machine learning models for subject-specific calibration.
In offline analyses using EEG data of 14 subjects, we tested the embeddings'
feasibility and compared their efficiency with state-of-the-art deep learning
models and conventional machine learning pipelines. In summary, we propose the
use of metric learning to obtain pre-trained embeddings of EEG-BCI data as a
means to incorporate domain knowledge and to reach competitive performance on
novel subjects with minimal calibration requirements.
| no_new_dataset | 0.71022 |
2304.06520 | Sabbir Ahmed | Md. Hamjajul Ashmafee, Tasnim Ahmed, Sabbir Ahmed, Md. Bakhtiar Hasan,
Mst Nura Jahan, A.B.M. Ashikur Rahman | An Efficient Transfer Learning-based Approach for Apple Leaf Disease
Classification | Accepted in ECCE 2023, 6 pages, 6 figures, 4 tables | null | null | null | cs.CV cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Correct identification and categorization of plant diseases are crucial for
ensuring the safety of the global food supply and the overall financial success
of stakeholders. In this regard, a wide range of solutions has been made
available by introducing deep learning-based classification systems for
different staple crops. Despite being one of the most important commercial
crops in many parts of the globe, research proposing a smart solution for
automatically classifying apple leaf diseases remains relatively unexplored.
This study presents a technique for identifying apple leaf diseases based on
transfer learning. The system extracts features using a pretrained
EfficientNetV2S architecture and passes to a classifier block for effective
prediction. The class imbalance issues are tackled by utilizing runtime data
augmentation. The effect of various hyperparameters, such as input resolution,
learning rate, number of epochs, etc., has been investigated carefully. The
competence of the proposed pipeline has been evaluated on the apple leaf
disease subset from the publicly available `PlantVillage' dataset, where it
achieved an accuracy of 99.21%, outperforming the existing works.
| [
{
"version": "v1",
"created": "Mon, 10 Apr 2023 08:48:36 GMT"
}
] | 2023-04-14T00:00:00 | [
[
"Ashmafee",
"Md. Hamjajul",
""
],
[
"Ahmed",
"Tasnim",
""
],
[
"Ahmed",
"Sabbir",
""
],
[
"Hasan",
"Md. Bakhtiar",
""
],
[
"Jahan",
"Mst Nura",
""
],
[
"Rahman",
"A. B. M. Ashikur",
""
]
] | TITLE: An Efficient Transfer Learning-based Approach for Apple Leaf Disease
Classification
ABSTRACT: Correct identification and categorization of plant diseases are crucial for
ensuring the safety of the global food supply and the overall financial success
of stakeholders. In this regard, a wide range of solutions has been made
available by introducing deep learning-based classification systems for
different staple crops. Despite being one of the most important commercial
crops in many parts of the globe, research proposing a smart solution for
automatically classifying apple leaf diseases remains relatively unexplored.
This study presents a technique for identifying apple leaf diseases based on
transfer learning. The system extracts features using a pretrained
EfficientNetV2S architecture and passes to a classifier block for effective
prediction. The class imbalance issues are tackled by utilizing runtime data
augmentation. The effect of various hyperparameters, such as input resolution,
learning rate, number of epochs, etc., has been investigated carefully. The
competence of the proposed pipeline has been evaluated on the apple leaf
disease subset from the publicly available `PlantVillage' dataset, where it
achieved an accuracy of 99.21%, outperforming the existing works.
| no_new_dataset | 0.708572 |
2304.06531 | Anis Kacem | Kseniya Cherenkova, Elona Dupont, Anis Kacem, Ilya Arzhannikov, Gleb
Gusev and Djamila Aouada | SepicNet: Sharp Edges Recovery by Parametric Inference of Curves in 3D
Shapes | null | null | null | null | cs.CV cs.GR | http://creativecommons.org/licenses/by-nc-sa/4.0/ | 3D scanning as a technique to digitize objects in reality and create their 3D
models, is used in many fields and areas. Though the quality of 3D scans
depends on the technical characteristics of the 3D scanner, the common drawback
is the smoothing of fine details, or the edges of an object. We introduce
SepicNet, a novel deep network for the detection and parametrization of sharp
edges in 3D shapes as primitive curves. To make the network end-to-end
trainable, we formulate the curve fitting in a differentiable manner. We
develop an adaptive point cloud sampling technique that captures the sharp
features better than uniform sampling. The experiments were conducted on a
newly introduced large-scale dataset of 50k 3D scans, where the sharp edge
annotations were extracted from their parametric CAD models, and demonstrate
significant improvement over state-of-the-art methods.
| [
{
"version": "v1",
"created": "Thu, 13 Apr 2023 13:37:21 GMT"
}
] | 2023-04-14T00:00:00 | [
[
"Cherenkova",
"Kseniya",
""
],
[
"Dupont",
"Elona",
""
],
[
"Kacem",
"Anis",
""
],
[
"Arzhannikov",
"Ilya",
""
],
[
"Gusev",
"Gleb",
""
],
[
"Aouada",
"Djamila",
""
]
] | TITLE: SepicNet: Sharp Edges Recovery by Parametric Inference of Curves in 3D
Shapes
ABSTRACT: 3D scanning as a technique to digitize objects in reality and create their 3D
models, is used in many fields and areas. Though the quality of 3D scans
depends on the technical characteristics of the 3D scanner, the common drawback
is the smoothing of fine details, or the edges of an object. We introduce
SepicNet, a novel deep network for the detection and parametrization of sharp
edges in 3D shapes as primitive curves. To make the network end-to-end
trainable, we formulate the curve fitting in a differentiable manner. We
develop an adaptive point cloud sampling technique that captures the sharp
features better than uniform sampling. The experiments were conducted on a
newly introduced large-scale dataset of 50k 3D scans, where the sharp edge
annotations were extracted from their parametric CAD models, and demonstrate
significant improvement over state-of-the-art methods.
| new_dataset | 0.713276 |
2304.06537 | Jiahao Chen | Jiahao Chen, Bing Su | Transfer Knowledge from Head to Tail: Uncertainty Calibration under
Long-tailed Distribution | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | How to estimate the uncertainty of a given model is a crucial problem.
Current calibration techniques treat different classes equally and thus
implicitly assume that the distribution of training data is balanced, but
ignore the fact that real-world data often follows a long-tailed distribution.
In this paper, we explore the problem of calibrating the model trained from a
long-tailed distribution. Due to the difference between the imbalanced training
distribution and balanced test distribution, existing calibration methods such
as temperature scaling can not generalize well to this problem. Specific
calibration methods for domain adaptation are also not applicable because they
rely on unlabeled target domain instances which are not available. Models
trained from a long-tailed distribution tend to be more overconfident to head
classes. To this end, we propose a novel knowledge-transferring-based
calibration method by estimating the importance weights for samples of tail
classes to realize long-tailed calibration. Our method models the distribution
of each class as a Gaussian distribution and views the source statistics of
head classes as a prior to calibrate the target distributions of tail classes.
We adaptively transfer knowledge from head classes to get the target
probability density of tail classes. The importance weight is estimated by the
ratio of the target probability density over the source probability density.
Extensive experiments on CIFAR-10-LT, MNIST-LT, CIFAR-100-LT, and ImageNet-LT
datasets demonstrate the effectiveness of our method.
| [
{
"version": "v1",
"created": "Thu, 13 Apr 2023 13:48:18 GMT"
}
] | 2023-04-14T00:00:00 | [
[
"Chen",
"Jiahao",
""
],
[
"Su",
"Bing",
""
]
] | TITLE: Transfer Knowledge from Head to Tail: Uncertainty Calibration under
Long-tailed Distribution
ABSTRACT: How to estimate the uncertainty of a given model is a crucial problem.
Current calibration techniques treat different classes equally and thus
implicitly assume that the distribution of training data is balanced, but
ignore the fact that real-world data often follows a long-tailed distribution.
In this paper, we explore the problem of calibrating the model trained from a
long-tailed distribution. Due to the difference between the imbalanced training
distribution and balanced test distribution, existing calibration methods such
as temperature scaling can not generalize well to this problem. Specific
calibration methods for domain adaptation are also not applicable because they
rely on unlabeled target domain instances which are not available. Models
trained from a long-tailed distribution tend to be more overconfident to head
classes. To this end, we propose a novel knowledge-transferring-based
calibration method by estimating the importance weights for samples of tail
classes to realize long-tailed calibration. Our method models the distribution
of each class as a Gaussian distribution and views the source statistics of
head classes as a prior to calibrate the target distributions of tail classes.
We adaptively transfer knowledge from head classes to get the target
probability density of tail classes. The importance weight is estimated by the
ratio of the target probability density over the source probability density.
Extensive experiments on CIFAR-10-LT, MNIST-LT, CIFAR-100-LT, and ImageNet-LT
datasets demonstrate the effectiveness of our method.
| no_new_dataset | 0.710635 |
2304.06551 | Deng Pan | Deng Pan, Mohammad Ali Khoshkholghi, Toktam Mahmoodi | Decentralized federated learning methods for reducing communication cost
and energy consumption in UAV networks | 13 pages, 7 figures | null | null | null | cs.LG cs.NI | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Unmanned aerial vehicles (UAV) or drones play many roles in a modern smart
city such as the delivery of goods, mapping real-time road traffic and
monitoring pollution. The ability of drones to perform these functions often
requires the support of machine learning technology. However, traditional
machine learning models for drones encounter data privacy problems,
communication costs and energy limitations. Federated Learning, an emerging
distributed machine learning approach, is an excellent solution to address
these issues. Federated learning (FL) allows drones to train local models
without transmitting raw data. However, existing FL requires a central server
to aggregate the trained model parameters of the UAV. A failure of the central
server can significantly impact the overall training. In this paper, we propose
two aggregation methods: Commutative FL and Alternate FL, based on the existing
architecture of decentralised Federated Learning for UAV Networks (DFL-UN) by
adding a unique aggregation method of decentralised FL. Those two methods can
effectively control energy consumption and communication cost by controlling
the number of local training epochs, local communication, and global
communication. The simulation results of the proposed training methods are also
presented to verify the feasibility and efficiency of the architecture compared
with two benchmark methods (e.g. standard machine learning training and
standard single aggregation server training). The simulation results show that
the proposed methods outperform the benchmark methods in terms of operational
stability, energy consumption and communication cost.
| [
{
"version": "v1",
"created": "Thu, 13 Apr 2023 14:00:34 GMT"
}
] | 2023-04-14T00:00:00 | [
[
"Pan",
"Deng",
""
],
[
"Khoshkholghi",
"Mohammad Ali",
""
],
[
"Mahmoodi",
"Toktam",
""
]
] | TITLE: Decentralized federated learning methods for reducing communication cost
and energy consumption in UAV networks
ABSTRACT: Unmanned aerial vehicles (UAV) or drones play many roles in a modern smart
city such as the delivery of goods, mapping real-time road traffic and
monitoring pollution. The ability of drones to perform these functions often
requires the support of machine learning technology. However, traditional
machine learning models for drones encounter data privacy problems,
communication costs and energy limitations. Federated Learning, an emerging
distributed machine learning approach, is an excellent solution to address
these issues. Federated learning (FL) allows drones to train local models
without transmitting raw data. However, existing FL requires a central server
to aggregate the trained model parameters of the UAV. A failure of the central
server can significantly impact the overall training. In this paper, we propose
two aggregation methods: Commutative FL and Alternate FL, based on the existing
architecture of decentralised Federated Learning for UAV Networks (DFL-UN) by
adding a unique aggregation method of decentralised FL. Those two methods can
effectively control energy consumption and communication cost by controlling
the number of local training epochs, local communication, and global
communication. The simulation results of the proposed training methods are also
presented to verify the feasibility and efficiency of the architecture compared
with two benchmark methods (e.g. standard machine learning training and
standard single aggregation server training). The simulation results show that
the proposed methods outperform the benchmark methods in terms of operational
stability, energy consumption and communication cost.
| no_new_dataset | 0.710653 |
2304.06575 | Benjamin Badger | Benjamin L. Badger | Adversarial Examples from Dimensional Invariance | 6 pages | null | null | null | cs.LG cs.CV cs.NA math.NA | http://creativecommons.org/licenses/by/4.0/ | Adversarial examples have been found for various deep as well as shallow
learning models, and have at various times been suggested to be either fixable
model-specific bugs, or else inherent dataset feature, or both. We present
theoretical and empirical results to show that adversarial examples are
approximate discontinuities resulting from models that specify approximately
bijective maps $f: \Bbb R^n \to \Bbb R^m; n \neq m$ over their inputs, and this
discontinuity follows from the topological invariance of dimension.
| [
{
"version": "v1",
"created": "Thu, 13 Apr 2023 14:37:45 GMT"
}
] | 2023-04-14T00:00:00 | [
[
"Badger",
"Benjamin L.",
""
]
] | TITLE: Adversarial Examples from Dimensional Invariance
ABSTRACT: Adversarial examples have been found for various deep as well as shallow
learning models, and have at various times been suggested to be either fixable
model-specific bugs, or else inherent dataset feature, or both. We present
theoretical and empirical results to show that adversarial examples are
approximate discontinuities resulting from models that specify approximately
bijective maps $f: \Bbb R^n \to \Bbb R^m; n \neq m$ over their inputs, and this
discontinuity follows from the topological invariance of dimension.
| no_new_dataset | 0.70878 |
2304.06602 | MinhDuc Vo | Duc Minh Vo, Quoc-An Luong, Akihiro Sugimoto, Hideki Nakayama | A-CAP: Anticipation Captioning with Commonsense Knowledge | Accepted to CVPR 2023 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Humans possess the capacity to reason about the future based on a sparse
collection of visual cues acquired over time. In order to emulate this ability,
we introduce a novel task called Anticipation Captioning, which generates a
caption for an unseen oracle image using a sparsely temporally-ordered set of
images. To tackle this new task, we propose a model called A-CAP, which
incorporates commonsense knowledge into a pre-trained vision-language model,
allowing it to anticipate the caption. Through both qualitative and
quantitative evaluations on a customized visual storytelling dataset, A-CAP
outperforms other image captioning methods and establishes a strong baseline
for anticipation captioning. We also address the challenges inherent in this
task.
| [
{
"version": "v1",
"created": "Thu, 13 Apr 2023 15:10:47 GMT"
}
] | 2023-04-14T00:00:00 | [
[
"Vo",
"Duc Minh",
""
],
[
"Luong",
"Quoc-An",
""
],
[
"Sugimoto",
"Akihiro",
""
],
[
"Nakayama",
"Hideki",
""
]
] | TITLE: A-CAP: Anticipation Captioning with Commonsense Knowledge
ABSTRACT: Humans possess the capacity to reason about the future based on a sparse
collection of visual cues acquired over time. In order to emulate this ability,
we introduce a novel task called Anticipation Captioning, which generates a
caption for an unseen oracle image using a sparsely temporally-ordered set of
images. To tackle this new task, we propose a model called A-CAP, which
incorporates commonsense knowledge into a pre-trained vision-language model,
allowing it to anticipate the caption. Through both qualitative and
quantitative evaluations on a customized visual storytelling dataset, A-CAP
outperforms other image captioning methods and establishes a strong baseline
for anticipation captioning. We also address the challenges inherent in this
task.
| new_dataset | 0.718403 |
2304.06603 | Erick Fredj | Erick Fredj and Yann Delorme and Sameeh Jubran and Mark Wasserman and
Zhaohui Ding and Michael Laufer | accelerating wrf i/o performance with adios2 and network-based streaming | arXiv admin note: text overlap with arXiv:2201.08228 | null | null | null | cs.DC | http://creativecommons.org/licenses/by/4.0/ | With the approach of Exascale computing power for large-scale High
Performance Computing (HPC) clusters, the gap between compute capabilities and
storage systems is growing larger. This is particularly problematic for the
Weather Research and Forecasting Model (WRF), a widely-used HPC application for
high-resolution forecasting and research that produces sizable datasets,
especially when analyzing transient weather phenomena. Despite this issue, the
I/O modules within WRF have not been updated in the past ten years, resulting
in subpar parallel I/O performance.
This research paper demonstrates the positive impact of integrating ADIOS2, a
next-generation parallel I/O framework, as a new I/O backend option in WRF. It
goes into detail about the challenges encountered during the integration
process and how they were addressed. The resulting I/O times show an over
tenfold improvement when using ADIOS2 compared to traditional MPI-I/O based
solutions. Furthermore, the study highlights the new features available to WRF
users worldwide, such as the Sustainable Staging Transport (SST) enabling
Unified Communication X (UCX) DataTransport, the node-local burst buffer write
capabilities and in-line lossless compression capabilities of ADIOS2.
Additionally, the research shows how ADIOS2's in-situ analysis capabilities
can be smoothly integrated with a simple WRF forecasting pipeline, resulting in
a significant improvement in overall time to solution. This study serves as a
reminder to legacy HPC applications that incorporating modern libraries and
tools can lead to considerable performance enhancements with minimal changes to
the core application.
| [
{
"version": "v1",
"created": "Thu, 13 Apr 2023 15:13:11 GMT"
}
] | 2023-04-14T00:00:00 | [
[
"Fredj",
"Erick",
""
],
[
"Delorme",
"Yann",
""
],
[
"Jubran",
"Sameeh",
""
],
[
"Wasserman",
"Mark",
""
],
[
"Ding",
"Zhaohui",
""
],
[
"Laufer",
"Michael",
""
]
] | TITLE: accelerating wrf i/o performance with adios2 and network-based streaming
ABSTRACT: With the approach of Exascale computing power for large-scale High
Performance Computing (HPC) clusters, the gap between compute capabilities and
storage systems is growing larger. This is particularly problematic for the
Weather Research and Forecasting Model (WRF), a widely-used HPC application for
high-resolution forecasting and research that produces sizable datasets,
especially when analyzing transient weather phenomena. Despite this issue, the
I/O modules within WRF have not been updated in the past ten years, resulting
in subpar parallel I/O performance.
This research paper demonstrates the positive impact of integrating ADIOS2, a
next-generation parallel I/O framework, as a new I/O backend option in WRF. It
goes into detail about the challenges encountered during the integration
process and how they were addressed. The resulting I/O times show an over
tenfold improvement when using ADIOS2 compared to traditional MPI-I/O based
solutions. Furthermore, the study highlights the new features available to WRF
users worldwide, such as the Sustainable Staging Transport (SST) enabling
Unified Communication X (UCX) DataTransport, the node-local burst buffer write
capabilities and in-line lossless compression capabilities of ADIOS2.
Additionally, the research shows how ADIOS2's in-situ analysis capabilities
can be smoothly integrated with a simple WRF forecasting pipeline, resulting in
a significant improvement in overall time to solution. This study serves as a
reminder to legacy HPC applications that incorporating modern libraries and
tools can lead to considerable performance enhancements with minimal changes to
the core application.
| no_new_dataset | 0.707354 |
2304.06627 | Haozhe Feng | Haozhe Feng, Zhaorui Yang, Hesun Chen, Tianyu Pang, Chao Du, Minfeng
Zhu, Wei Chen, Shuicheng Yan | CoSDA: Continual Source-Free Domain Adaptation | 15 pages, 6 figures | null | null | null | cs.LG cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Without access to the source data, source-free domain adaptation (SFDA)
transfers knowledge from a source-domain trained model to target domains.
Recently, SFDA has gained popularity due to the need to protect the data
privacy of the source domain, but it suffers from catastrophic forgetting on
the source domain due to the lack of data. To systematically investigate the
mechanism of catastrophic forgetting, we first reimplement previous SFDA
approaches within a unified framework and evaluate them on four benchmarks. We
observe that there is a trade-off between adaptation gain and forgetting loss,
which motivates us to design a consistency regularization to mitigate
forgetting. In particular, we propose a continual source-free domain adaptation
approach named CoSDA, which employs a dual-speed optimized teacher-student
model pair and is equipped with consistency learning capability. Our
experiments demonstrate that CoSDA outperforms state-of-the-art approaches in
continuous adaptation. Notably, our CoSDA can also be integrated with other
SFDA methods to alleviate forgetting.
| [
{
"version": "v1",
"created": "Thu, 13 Apr 2023 15:53:23 GMT"
}
] | 2023-04-14T00:00:00 | [
[
"Feng",
"Haozhe",
""
],
[
"Yang",
"Zhaorui",
""
],
[
"Chen",
"Hesun",
""
],
[
"Pang",
"Tianyu",
""
],
[
"Du",
"Chao",
""
],
[
"Zhu",
"Minfeng",
""
],
[
"Chen",
"Wei",
""
],
[
"Yan",
"Shuicheng",
""
]
] | TITLE: CoSDA: Continual Source-Free Domain Adaptation
ABSTRACT: Without access to the source data, source-free domain adaptation (SFDA)
transfers knowledge from a source-domain trained model to target domains.
Recently, SFDA has gained popularity due to the need to protect the data
privacy of the source domain, but it suffers from catastrophic forgetting on
the source domain due to the lack of data. To systematically investigate the
mechanism of catastrophic forgetting, we first reimplement previous SFDA
approaches within a unified framework and evaluate them on four benchmarks. We
observe that there is a trade-off between adaptation gain and forgetting loss,
which motivates us to design a consistency regularization to mitigate
forgetting. In particular, we propose a continual source-free domain adaptation
approach named CoSDA, which employs a dual-speed optimized teacher-student
model pair and is equipped with consistency learning capability. Our
experiments demonstrate that CoSDA outperforms state-of-the-art approaches in
continuous adaptation. Notably, our CoSDA can also be integrated with other
SFDA methods to alleviate forgetting.
| no_new_dataset | 0.707979 |
2304.06703 | Akshay Dudhane | Nancy Mehta, Akshay Dudhane, Subrahmanyam Murala, Syed Waqas Zamir,
Salman Khan, Fahad Shahbaz Khan | Gated Multi-Resolution Transfer Network for Burst Restoration and
Enhancement | Accepted at CVPR 2023 | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Burst image processing is becoming increasingly popular in recent years.
However, it is a challenging task since individual burst images undergo
multiple degradations and often have mutual misalignments resulting in ghosting
and zipper artifacts. Existing burst restoration methods usually do not
consider the mutual correlation and non-local contextual information among
burst frames, which tends to limit these approaches in challenging cases.
Another key challenge lies in the robust up-sampling of burst frames. The
existing up-sampling methods cannot effectively utilize the advantages of
single-stage and progressive up-sampling strategies with conventional and/or
recent up-samplers at the same time. To address these challenges, we propose a
novel Gated Multi-Resolution Transfer Network (GMTNet) to reconstruct a
spatially precise high-quality image from a burst of low-quality raw images.
GMTNet consists of three modules optimized for burst processing tasks:
Multi-scale Burst Feature Alignment (MBFA) for feature denoising and alignment,
Transposed-Attention Feature Merging (TAFM) for multi-frame feature
aggregation, and Resolution Transfer Feature Up-sampler (RTFU) to up-scale
merged features and construct a high-quality output image. Detailed
experimental analysis on five datasets validates our approach and sets a
state-of-the-art for burst super-resolution, burst denoising, and low-light
burst enhancement.
| [
{
"version": "v1",
"created": "Thu, 13 Apr 2023 17:54:00 GMT"
}
] | 2023-04-14T00:00:00 | [
[
"Mehta",
"Nancy",
""
],
[
"Dudhane",
"Akshay",
""
],
[
"Murala",
"Subrahmanyam",
""
],
[
"Zamir",
"Syed Waqas",
""
],
[
"Khan",
"Salman",
""
],
[
"Khan",
"Fahad Shahbaz",
""
]
] | TITLE: Gated Multi-Resolution Transfer Network for Burst Restoration and
Enhancement
ABSTRACT: Burst image processing is becoming increasingly popular in recent years.
However, it is a challenging task since individual burst images undergo
multiple degradations and often have mutual misalignments resulting in ghosting
and zipper artifacts. Existing burst restoration methods usually do not
consider the mutual correlation and non-local contextual information among
burst frames, which tends to limit these approaches in challenging cases.
Another key challenge lies in the robust up-sampling of burst frames. The
existing up-sampling methods cannot effectively utilize the advantages of
single-stage and progressive up-sampling strategies with conventional and/or
recent up-samplers at the same time. To address these challenges, we propose a
novel Gated Multi-Resolution Transfer Network (GMTNet) to reconstruct a
spatially precise high-quality image from a burst of low-quality raw images.
GMTNet consists of three modules optimized for burst processing tasks:
Multi-scale Burst Feature Alignment (MBFA) for feature denoising and alignment,
Transposed-Attention Feature Merging (TAFM) for multi-frame feature
aggregation, and Resolution Transfer Feature Up-sampler (RTFU) to up-scale
merged features and construct a high-quality output image. Detailed
experimental analysis on five datasets validates our approach and sets a
state-of-the-art for burst super-resolution, burst denoising, and low-light
burst enhancement.
| no_new_dataset | 0.709019 |
2304.06710 | Mustansar Fiaz | Mubashir Noman, Mustansar Fiaz, Hisham Cholakkal, Sanath Narayan, Rao
Muhammad Anwer, Salman Khan, Fahad Shahbaz Khan | Remote Sensing Change Detection With Transformers Trained from Scratch | 5 figures and 4 tables | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Current transformer-based change detection (CD) approaches either employ a
pre-trained model trained on large-scale image classification ImageNet dataset
or rely on first pre-training on another CD dataset and then fine-tuning on the
target benchmark. This current strategy is driven by the fact that transformers
typically require a large amount of training data to learn inductive biases,
which is insufficient in standard CD datasets due to their small size. We
develop an end-to-end CD approach with transformers that is trained from
scratch and yet achieves state-of-the-art performance on four public
benchmarks. Instead of using conventional self-attention that struggles to
capture inductive biases when trained from scratch, our architecture utilizes a
shuffled sparse-attention operation that focuses on selected sparse informative
regions to capture the inherent characteristics of the CD data. Moreover, we
introduce a change-enhanced feature fusion (CEFF) module to fuse the features
from input image pairs by performing a per-channel re-weighting. Our CEFF
module aids in enhancing the relevant semantic changes while suppressing the
noisy ones. Extensive experiments on four CD datasets reveal the merits of the
proposed contributions, achieving gains as high as 14.27\% in
intersection-over-union (IoU) score, compared to the best-published results in
the literature. Code is available at
\url{https://github.com/mustansarfiaz/ScratchFormer}.
| [
{
"version": "v1",
"created": "Thu, 13 Apr 2023 17:57:54 GMT"
}
] | 2023-04-14T00:00:00 | [
[
"Noman",
"Mubashir",
""
],
[
"Fiaz",
"Mustansar",
""
],
[
"Cholakkal",
"Hisham",
""
],
[
"Narayan",
"Sanath",
""
],
[
"Anwer",
"Rao Muhammad",
""
],
[
"Khan",
"Salman",
""
],
[
"Khan",
"Fahad Shahbaz",
""
]
] | TITLE: Remote Sensing Change Detection With Transformers Trained from Scratch
ABSTRACT: Current transformer-based change detection (CD) approaches either employ a
pre-trained model trained on large-scale image classification ImageNet dataset
or rely on first pre-training on another CD dataset and then fine-tuning on the
target benchmark. This current strategy is driven by the fact that transformers
typically require a large amount of training data to learn inductive biases,
which is insufficient in standard CD datasets due to their small size. We
develop an end-to-end CD approach with transformers that is trained from
scratch and yet achieves state-of-the-art performance on four public
benchmarks. Instead of using conventional self-attention that struggles to
capture inductive biases when trained from scratch, our architecture utilizes a
shuffled sparse-attention operation that focuses on selected sparse informative
regions to capture the inherent characteristics of the CD data. Moreover, we
introduce a change-enhanced feature fusion (CEFF) module to fuse the features
from input image pairs by performing a per-channel re-weighting. Our CEFF
module aids in enhancing the relevant semantic changes while suppressing the
noisy ones. Extensive experiments on four CD datasets reveal the merits of the
proposed contributions, achieving gains as high as 14.27\% in
intersection-over-union (IoU) score, compared to the best-published results in
the literature. Code is available at
\url{https://github.com/mustansarfiaz/ScratchFormer}.
| no_new_dataset | 0.711074 |
2304.06711 | Zheng Ding | Zheng Ding, Xuaner Zhang, Zhihao Xia, Lars Jebe, Zhuowen Tu, Xiuming
Zhang | DiffusionRig: Learning Personalized Priors for Facial Appearance Editing | CVPR 2023. Project website: https://diffusionrig.github.io | null | null | null | cs.CV cs.GR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We address the problem of learning person-specific facial priors from a small
number (e.g., 20) of portrait photos of the same person. This enables us to
edit this specific person's facial appearance, such as expression and lighting,
while preserving their identity and high-frequency facial details. Key to our
approach, which we dub DiffusionRig, is a diffusion model conditioned on, or
"rigged by," crude 3D face models estimated from single in-the-wild images by
an off-the-shelf estimator. On a high level, DiffusionRig learns to map
simplistic renderings of 3D face models to realistic photos of a given person.
Specifically, DiffusionRig is trained in two stages: It first learns generic
facial priors from a large-scale face dataset and then person-specific priors
from a small portrait photo collection of the person of interest. By learning
the CGI-to-photo mapping with such personalized priors, DiffusionRig can "rig"
the lighting, facial expression, head pose, etc. of a portrait photo,
conditioned only on coarse 3D models while preserving this person's identity
and other high-frequency characteristics. Qualitative and quantitative
experiments show that DiffusionRig outperforms existing approaches in both
identity preservation and photorealism. Please see the project website:
https://diffusionrig.github.io for the supplemental material, video, code, and
data.
| [
{
"version": "v1",
"created": "Thu, 13 Apr 2023 17:58:00 GMT"
}
] | 2023-04-14T00:00:00 | [
[
"Ding",
"Zheng",
""
],
[
"Zhang",
"Xuaner",
""
],
[
"Xia",
"Zhihao",
""
],
[
"Jebe",
"Lars",
""
],
[
"Tu",
"Zhuowen",
""
],
[
"Zhang",
"Xiuming",
""
]
] | TITLE: DiffusionRig: Learning Personalized Priors for Facial Appearance Editing
ABSTRACT: We address the problem of learning person-specific facial priors from a small
number (e.g., 20) of portrait photos of the same person. This enables us to
edit this specific person's facial appearance, such as expression and lighting,
while preserving their identity and high-frequency facial details. Key to our
approach, which we dub DiffusionRig, is a diffusion model conditioned on, or
"rigged by," crude 3D face models estimated from single in-the-wild images by
an off-the-shelf estimator. On a high level, DiffusionRig learns to map
simplistic renderings of 3D face models to realistic photos of a given person.
Specifically, DiffusionRig is trained in two stages: It first learns generic
facial priors from a large-scale face dataset and then person-specific priors
from a small portrait photo collection of the person of interest. By learning
the CGI-to-photo mapping with such personalized priors, DiffusionRig can "rig"
the lighting, facial expression, head pose, etc. of a portrait photo,
conditioned only on coarse 3D models while preserving this person's identity
and other high-frequency characteristics. Qualitative and quantitative
experiments show that DiffusionRig outperforms existing approaches in both
identity preservation and photorealism. Please see the project website:
https://diffusionrig.github.io for the supplemental material, video, code, and
data.
| no_new_dataset | 0.711606 |
2304.06716 | Ziyan Huang | Ziyan Huang, Haoyu Wang, Zhongying Deng, Jin Ye, Yanzhou Su, Hui Sun,
Junjun He, Yun Gu, Lixu Gu, Shaoting Zhang and Yu Qiao | STU-Net: Scalable and Transferable Medical Image Segmentation Models
Empowered by Large-Scale Supervised Pre-training | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Large-scale models pre-trained on large-scale datasets have profoundly
advanced the development of deep learning. However, the state-of-the-art models
for medical image segmentation are still small-scale, with their parameters
only in the tens of millions. Further scaling them up to higher orders of
magnitude is rarely explored. An overarching goal of exploring large-scale
models is to train them on large-scale medical segmentation datasets for better
transfer capacities. In this work, we design a series of Scalable and
Transferable U-Net (STU-Net) models, with parameter sizes ranging from 14
million to 1.4 billion. Notably, the 1.4B STU-Net is the largest medical image
segmentation model to date. Our STU-Net is based on nnU-Net framework due to
its popularity and impressive performance. We first refine the default
convolutional blocks in nnU-Net to make them scalable. Then, we empirically
evaluate different scaling combinations of network depth and width, discovering
that it is optimal to scale model depth and width together. We train our
scalable STU-Net models on a large-scale TotalSegmentator dataset and find that
increasing model size brings a stronger performance gain. This observation
reveals that a large model is promising in medical image segmentation.
Furthermore, we evaluate the transferability of our model on 14 downstream
datasets for direct inference and 3 datasets for further fine-tuning, covering
various modalities and segmentation targets. We observe good performance of our
pre-trained model in both direct inference and fine-tuning. The code and
pre-trained models are available at https://github.com/Ziyan-Huang/STU-Net.
| [
{
"version": "v1",
"created": "Thu, 13 Apr 2023 17:59:13 GMT"
}
] | 2023-04-14T00:00:00 | [
[
"Huang",
"Ziyan",
""
],
[
"Wang",
"Haoyu",
""
],
[
"Deng",
"Zhongying",
""
],
[
"Ye",
"Jin",
""
],
[
"Su",
"Yanzhou",
""
],
[
"Sun",
"Hui",
""
],
[
"He",
"Junjun",
""
],
[
"Gu",
"Yun",
""
],
[
"Gu",
"Lixu",
""
],
[
"Zhang",
"Shaoting",
""
],
[
"Qiao",
"Yu",
""
]
] | TITLE: STU-Net: Scalable and Transferable Medical Image Segmentation Models
Empowered by Large-Scale Supervised Pre-training
ABSTRACT: Large-scale models pre-trained on large-scale datasets have profoundly
advanced the development of deep learning. However, the state-of-the-art models
for medical image segmentation are still small-scale, with their parameters
only in the tens of millions. Further scaling them up to higher orders of
magnitude is rarely explored. An overarching goal of exploring large-scale
models is to train them on large-scale medical segmentation datasets for better
transfer capacities. In this work, we design a series of Scalable and
Transferable U-Net (STU-Net) models, with parameter sizes ranging from 14
million to 1.4 billion. Notably, the 1.4B STU-Net is the largest medical image
segmentation model to date. Our STU-Net is based on nnU-Net framework due to
its popularity and impressive performance. We first refine the default
convolutional blocks in nnU-Net to make them scalable. Then, we empirically
evaluate different scaling combinations of network depth and width, discovering
that it is optimal to scale model depth and width together. We train our
scalable STU-Net models on a large-scale TotalSegmentator dataset and find that
increasing model size brings a stronger performance gain. This observation
reveals that a large model is promising in medical image segmentation.
Furthermore, we evaluate the transferability of our model on 14 downstream
datasets for direct inference and 3 datasets for further fine-tuning, covering
various modalities and segmentation targets. We observe good performance of our
pre-trained model in both direct inference and fine-tuning. The code and
pre-trained models are available at https://github.com/Ziyan-Huang/STU-Net.
| no_new_dataset | 0.710434 |
2304.06717 | Sida Peng | Sida Peng, Yunzhi Yan, Qing Shuai, Hujun Bao, Xiaowei Zhou | Representing Volumetric Videos as Dynamic MLP Maps | Accepted to CVPR 2023. The first two authors contributed equally to
this paper. Project page: https://zju3dv.github.io/mlp_maps/ | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper introduces a novel representation of volumetric videos for
real-time view synthesis of dynamic scenes. Recent advances in neural scene
representations demonstrate their remarkable capability to model and render
complex static scenes, but extending them to represent dynamic scenes is not
straightforward due to their slow rendering speed or high storage cost. To
solve this problem, our key idea is to represent the radiance field of each
frame as a set of shallow MLP networks whose parameters are stored in 2D grids,
called MLP maps, and dynamically predicted by a 2D CNN decoder shared by all
frames. Representing 3D scenes with shallow MLPs significantly improves the
rendering speed, while dynamically predicting MLP parameters with a shared 2D
CNN instead of explicitly storing them leads to low storage cost. Experiments
show that the proposed approach achieves state-of-the-art rendering quality on
the NHR and ZJU-MoCap datasets, while being efficient for real-time rendering
with a speed of 41.7 fps for $512 \times 512$ images on an RTX 3090 GPU. The
code is available at https://zju3dv.github.io/mlp_maps/.
| [
{
"version": "v1",
"created": "Thu, 13 Apr 2023 17:59:33 GMT"
}
] | 2023-04-14T00:00:00 | [
[
"Peng",
"Sida",
""
],
[
"Yan",
"Yunzhi",
""
],
[
"Shuai",
"Qing",
""
],
[
"Bao",
"Hujun",
""
],
[
"Zhou",
"Xiaowei",
""
]
] | TITLE: Representing Volumetric Videos as Dynamic MLP Maps
ABSTRACT: This paper introduces a novel representation of volumetric videos for
real-time view synthesis of dynamic scenes. Recent advances in neural scene
representations demonstrate their remarkable capability to model and render
complex static scenes, but extending them to represent dynamic scenes is not
straightforward due to their slow rendering speed or high storage cost. To
solve this problem, our key idea is to represent the radiance field of each
frame as a set of shallow MLP networks whose parameters are stored in 2D grids,
called MLP maps, and dynamically predicted by a 2D CNN decoder shared by all
frames. Representing 3D scenes with shallow MLPs significantly improves the
rendering speed, while dynamically predicting MLP parameters with a shared 2D
CNN instead of explicitly storing them leads to low storage cost. Experiments
show that the proposed approach achieves state-of-the-art rendering quality on
the NHR and ZJU-MoCap datasets, while being efficient for real-time rendering
with a speed of 41.7 fps for $512 \times 512$ images on an RTX 3090 GPU. The
code is available at https://zju3dv.github.io/mlp_maps/.
| no_new_dataset | 0.710459 |
2304.06719 | Shaoyuan Xie | Shaoyuan Xie, Lingdong Kong, Wenwei Zhang, Jiawei Ren, Liang Pan, Kai
Chen, Ziwei Liu | RoboBEV: Towards Robust Bird's Eye View Perception under Corruptions | Preprint; 27 pages, 18 figures, 33 tables; Code at
https://github.com/Daniel-xsy/RoboBEV | null | null | null | cs.CV cs.RO | http://creativecommons.org/licenses/by-sa/4.0/ | The recent advances in camera-based bird's eye view (BEV) representation
exhibit great potential for in-vehicle 3D perception. Despite the substantial
progress achieved on standard benchmarks, the robustness of BEV algorithms has
not been thoroughly examined, which is critical for safe operations. To bridge
this gap, we introduce RoboBEV, a comprehensive benchmark suite that
encompasses eight distinct corruptions, including Bright, Dark, Fog, Snow,
Motion Blur, Color Quant, Camera Crash, and Frame Lost. Based on it, we
undertake extensive evaluations across a wide range of BEV-based models to
understand their resilience and reliability. Our findings indicate a strong
correlation between absolute performance on in-distribution and
out-of-distribution datasets. Nonetheless, there are considerable variations in
relative performance across different approaches. Our experiments further
demonstrate that pre-training and depth-free BEV transformation has the
potential to enhance out-of-distribution robustness. Additionally, utilizing
long and rich temporal information largely helps with robustness. Our findings
provide valuable insights for designing future BEV models that can achieve both
accuracy and robustness in real-world deployments.
| [
{
"version": "v1",
"created": "Thu, 13 Apr 2023 17:59:46 GMT"
}
] | 2023-04-14T00:00:00 | [
[
"Xie",
"Shaoyuan",
""
],
[
"Kong",
"Lingdong",
""
],
[
"Zhang",
"Wenwei",
""
],
[
"Ren",
"Jiawei",
""
],
[
"Pan",
"Liang",
""
],
[
"Chen",
"Kai",
""
],
[
"Liu",
"Ziwei",
""
]
] | TITLE: RoboBEV: Towards Robust Bird's Eye View Perception under Corruptions
ABSTRACT: The recent advances in camera-based bird's eye view (BEV) representation
exhibit great potential for in-vehicle 3D perception. Despite the substantial
progress achieved on standard benchmarks, the robustness of BEV algorithms has
not been thoroughly examined, which is critical for safe operations. To bridge
this gap, we introduce RoboBEV, a comprehensive benchmark suite that
encompasses eight distinct corruptions, including Bright, Dark, Fog, Snow,
Motion Blur, Color Quant, Camera Crash, and Frame Lost. Based on it, we
undertake extensive evaluations across a wide range of BEV-based models to
understand their resilience and reliability. Our findings indicate a strong
correlation between absolute performance on in-distribution and
out-of-distribution datasets. Nonetheless, there are considerable variations in
relative performance across different approaches. Our experiments further
demonstrate that pre-training and depth-free BEV transformation has the
potential to enhance out-of-distribution robustness. Additionally, utilizing
long and rich temporal information largely helps with robustness. Our findings
provide valuable insights for designing future BEV models that can achieve both
accuracy and robustness in real-world deployments.
| new_dataset | 0.714622 |
2003.00120 | Zhuolin Yang | Zhuolin Yang, Zhikuan Zhao, Boxin Wang, Jiawei Zhang, Linyi Li,
Hengzhi Pei, Bojan Karlas, Ji Liu, Heng Guo, Ce Zhang, and Bo Li | Improving Certified Robustness via Statistical Learning with Logical
Reasoning | Accepted by 36th Conference on Neural Information Processing Systems
(NeurIPS 2022) | null | null | null | cs.LG cs.CR stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Intensive algorithmic efforts have been made to enable the rapid improvements
of certificated robustness for complex ML models recently. However, current
robustness certification methods are only able to certify under a limited
perturbation radius. Given that existing pure data-driven statistical
approaches have reached a bottleneck, in this paper, we propose to integrate
statistical ML models with knowledge (expressed as logical rules) as a
reasoning component using Markov logic networks (MLN, so as to further improve
the overall certified robustness. This opens new research questions about
certifying the robustness of such a paradigm, especially the reasoning
component (e.g., MLN). As the first step towards understanding these questions,
we first prove that the computational complexity of certifying the robustness
of MLN is #P-hard. Guided by this hardness result, we then derive the first
certified robustness bound for MLN by carefully analyzing different model
regimes. Finally, we conduct extensive experiments on five datasets including
both high-dimensional images and natural language texts, and we show that the
certified robustness with knowledge-based logical reasoning indeed
significantly outperforms that of the state-of-the-arts.
| [
{
"version": "v1",
"created": "Fri, 28 Feb 2020 23:41:58 GMT"
},
{
"version": "v2",
"created": "Fri, 6 Mar 2020 23:30:03 GMT"
},
{
"version": "v3",
"created": "Mon, 2 Aug 2021 15:17:09 GMT"
},
{
"version": "v4",
"created": "Wed, 7 Sep 2022 12:45:44 GMT"
},
{
"version": "v5",
"created": "Fri, 9 Sep 2022 15:54:58 GMT"
},
{
"version": "v6",
"created": "Wed, 26 Oct 2022 04:43:53 GMT"
},
{
"version": "v7",
"created": "Thu, 29 Dec 2022 21:04:36 GMT"
},
{
"version": "v8",
"created": "Thu, 9 Feb 2023 18:10:39 GMT"
},
{
"version": "v9",
"created": "Wed, 12 Apr 2023 17:45:28 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Yang",
"Zhuolin",
""
],
[
"Zhao",
"Zhikuan",
""
],
[
"Wang",
"Boxin",
""
],
[
"Zhang",
"Jiawei",
""
],
[
"Li",
"Linyi",
""
],
[
"Pei",
"Hengzhi",
""
],
[
"Karlas",
"Bojan",
""
],
[
"Liu",
"Ji",
""
],
[
"Guo",
"Heng",
""
],
[
"Zhang",
"Ce",
""
],
[
"Li",
"Bo",
""
]
] | TITLE: Improving Certified Robustness via Statistical Learning with Logical
Reasoning
ABSTRACT: Intensive algorithmic efforts have been made to enable the rapid improvements
of certificated robustness for complex ML models recently. However, current
robustness certification methods are only able to certify under a limited
perturbation radius. Given that existing pure data-driven statistical
approaches have reached a bottleneck, in this paper, we propose to integrate
statistical ML models with knowledge (expressed as logical rules) as a
reasoning component using Markov logic networks (MLN, so as to further improve
the overall certified robustness. This opens new research questions about
certifying the robustness of such a paradigm, especially the reasoning
component (e.g., MLN). As the first step towards understanding these questions,
we first prove that the computational complexity of certifying the robustness
of MLN is #P-hard. Guided by this hardness result, we then derive the first
certified robustness bound for MLN by carefully analyzing different model
regimes. Finally, we conduct extensive experiments on five datasets including
both high-dimensional images and natural language texts, and we show that the
certified robustness with knowledge-based logical reasoning indeed
significantly outperforms that of the state-of-the-arts.
| no_new_dataset | 0.708616 |
2009.04131 | Linyi Li | Linyi Li, Tao Xie, Bo Li | SoK: Certified Robustness for Deep Neural Networks | To appear at 2023 IEEE Symposium on Security and Privacy (SP)
(Version 8); include recent progress till Apr 2023 in Version 9; 14 pages for
the main text; benchmark & tool website:
http://sokcertifiedrobustness.github.io/ | null | null | null | cs.LG cs.CR stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Great advances in deep neural networks (DNNs) have led to state-of-the-art
performance on a wide range of tasks. However, recent studies have shown that
DNNs are vulnerable to adversarial attacks, which have brought great concerns
when deploying these models to safety-critical applications such as autonomous
driving. Different defense approaches have been proposed against adversarial
attacks, including: a) empirical defenses, which can usually be adaptively
attacked again without providing robustness certification; and b) certifiably
robust approaches, which consist of robustness verification providing the lower
bound of robust accuracy against any attacks under certain conditions and
corresponding robust training approaches. In this paper, we systematize
certifiably robust approaches and related practical and theoretical
implications and findings. We also provide the first comprehensive benchmark on
existing robustness verification and training approaches on different datasets.
In particular, we 1) provide a taxonomy for the robustness verification and
training approaches, as well as summarize the methodologies for representative
algorithms, 2) reveal the characteristics, strengths, limitations, and
fundamental connections among these approaches, 3) discuss current research
progresses, theoretical barriers, main challenges, and future directions for
certifiably robust approaches for DNNs, and 4) provide an open-sourced unified
platform to evaluate 20+ representative certifiably robust approaches.
| [
{
"version": "v1",
"created": "Wed, 9 Sep 2020 07:00:55 GMT"
},
{
"version": "v2",
"created": "Thu, 10 Sep 2020 05:04:49 GMT"
},
{
"version": "v3",
"created": "Sat, 4 Dec 2021 08:52:27 GMT"
},
{
"version": "v4",
"created": "Tue, 7 Dec 2021 19:58:25 GMT"
},
{
"version": "v5",
"created": "Sat, 11 Dec 2021 11:33:20 GMT"
},
{
"version": "v6",
"created": "Thu, 10 Feb 2022 19:23:02 GMT"
},
{
"version": "v7",
"created": "Tue, 23 Aug 2022 07:11:04 GMT"
},
{
"version": "v8",
"created": "Fri, 9 Sep 2022 19:11:14 GMT"
},
{
"version": "v9",
"created": "Wed, 12 Apr 2023 08:51:12 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Li",
"Linyi",
""
],
[
"Xie",
"Tao",
""
],
[
"Li",
"Bo",
""
]
] | TITLE: SoK: Certified Robustness for Deep Neural Networks
ABSTRACT: Great advances in deep neural networks (DNNs) have led to state-of-the-art
performance on a wide range of tasks. However, recent studies have shown that
DNNs are vulnerable to adversarial attacks, which have brought great concerns
when deploying these models to safety-critical applications such as autonomous
driving. Different defense approaches have been proposed against adversarial
attacks, including: a) empirical defenses, which can usually be adaptively
attacked again without providing robustness certification; and b) certifiably
robust approaches, which consist of robustness verification providing the lower
bound of robust accuracy against any attacks under certain conditions and
corresponding robust training approaches. In this paper, we systematize
certifiably robust approaches and related practical and theoretical
implications and findings. We also provide the first comprehensive benchmark on
existing robustness verification and training approaches on different datasets.
In particular, we 1) provide a taxonomy for the robustness verification and
training approaches, as well as summarize the methodologies for representative
algorithms, 2) reveal the characteristics, strengths, limitations, and
fundamental connections among these approaches, 3) discuss current research
progresses, theoretical barriers, main challenges, and future directions for
certifiably robust approaches for DNNs, and 4) provide an open-sourced unified
platform to evaluate 20+ representative certifiably robust approaches.
| no_new_dataset | 0.709422 |
2106.07258 | Madelon Hulsebos | Madelon Hulsebos, \c{C}a\u{g}atay Demiralp, Paul Groth | GitTables: A Large-Scale Corpus of Relational Tables | null | null | 10.1145/3588710 | null | cs.DB cs.LG | http://creativecommons.org/licenses/by/4.0/ | The success of deep learning has sparked interest in improving relational
table tasks, like data preparation and search, with table representation models
trained on large table corpora. Existing table corpora primarily contain tables
extracted from HTML pages, limiting the capability to represent offline
database tables. To train and evaluate high-capacity models for applications
beyond the Web, we need resources with tables that resemble relational database
tables. Here we introduce GitTables, a corpus of 1M relational tables extracted
from GitHub. Our continuing curation aims at growing the corpus to at least 10M
tables. Analyses of GitTables show that its structure, content, and topical
coverage differ significantly from existing table corpora. We annotate table
columns in GitTables with semantic types, hierarchical relations and
descriptions from Schema.org and DBpedia. The evaluation of our annotation
pipeline on the T2Dv2 benchmark illustrates that our approach provides results
on par with human annotations. We present three applications of GitTables,
demonstrating its value for learned semantic type detection models, schema
completion methods, and benchmarks for table-to-KG matching, data search, and
preparation. We make the corpus and code available at
https://gittables.github.io.
| [
{
"version": "v1",
"created": "Mon, 14 Jun 2021 09:22:09 GMT"
},
{
"version": "v2",
"created": "Wed, 8 Sep 2021 11:52:20 GMT"
},
{
"version": "v3",
"created": "Thu, 9 Sep 2021 09:59:29 GMT"
},
{
"version": "v4",
"created": "Fri, 15 Apr 2022 14:45:47 GMT"
},
{
"version": "v5",
"created": "Wed, 12 Apr 2023 13:24:55 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Hulsebos",
"Madelon",
""
],
[
"Demiralp",
"Çağatay",
""
],
[
"Groth",
"Paul",
""
]
] | TITLE: GitTables: A Large-Scale Corpus of Relational Tables
ABSTRACT: The success of deep learning has sparked interest in improving relational
table tasks, like data preparation and search, with table representation models
trained on large table corpora. Existing table corpora primarily contain tables
extracted from HTML pages, limiting the capability to represent offline
database tables. To train and evaluate high-capacity models for applications
beyond the Web, we need resources with tables that resemble relational database
tables. Here we introduce GitTables, a corpus of 1M relational tables extracted
from GitHub. Our continuing curation aims at growing the corpus to at least 10M
tables. Analyses of GitTables show that its structure, content, and topical
coverage differ significantly from existing table corpora. We annotate table
columns in GitTables with semantic types, hierarchical relations and
descriptions from Schema.org and DBpedia. The evaluation of our annotation
pipeline on the T2Dv2 benchmark illustrates that our approach provides results
on par with human annotations. We present three applications of GitTables,
demonstrating its value for learned semantic type detection models, schema
completion methods, and benchmarks for table-to-KG matching, data search, and
preparation. We make the corpus and code available at
https://gittables.github.io.
| new_dataset | 0.596668 |
2110.03845 | Luciana Dalla Valle PhD | Lauren Ansell and Luciana Dalla Valle | A New Data Integration Framework for Covid-19 Social Media Information | arXiv admin note: substantial text overlap with arXiv:2104.01869 | null | null | null | cs.SI stat.AP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The Covid-19 pandemic presents a serious threat to people health, resulting
in over 250 million confirmed cases and over 5 million deaths globally. To
reduce the burden on national health care systems and to mitigate the effects
of the outbreak, accurate modelling and forecasting methods for short- and
long-term health demand are needed to inform government interventions aiming at
curbing the pandemic. Current research on Covid-19 is typically based on a
single source of information, specifically on structured historical pandemic
data. Other studies are exclusively focused on unstructured online retrieved
insights, such as data available from social media. However, the combined use
of structured and unstructured information is still uncharted. This paper aims
at filling this gap, by leveraging historical and social media information with
a novel data integration methodology. The proposed approach is based on vine
copulas, which allow us to exploit the dependencies between different sources
of information. We apply the methodology to combine structured datasets
retrieved from official sources and a big unstructured dataset of information
collected from social media. The results show that the combined use of official
and online generated information contributes to yield a more accurate
assessment of the evolution of the Covid-19 pandemic, compared to the sole use
of official data.
| [
{
"version": "v1",
"created": "Fri, 8 Oct 2021 01:15:04 GMT"
},
{
"version": "v2",
"created": "Wed, 12 Apr 2023 15:39:24 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Ansell",
"Lauren",
""
],
[
"Valle",
"Luciana Dalla",
""
]
] | TITLE: A New Data Integration Framework for Covid-19 Social Media Information
ABSTRACT: The Covid-19 pandemic presents a serious threat to people health, resulting
in over 250 million confirmed cases and over 5 million deaths globally. To
reduce the burden on national health care systems and to mitigate the effects
of the outbreak, accurate modelling and forecasting methods for short- and
long-term health demand are needed to inform government interventions aiming at
curbing the pandemic. Current research on Covid-19 is typically based on a
single source of information, specifically on structured historical pandemic
data. Other studies are exclusively focused on unstructured online retrieved
insights, such as data available from social media. However, the combined use
of structured and unstructured information is still uncharted. This paper aims
at filling this gap, by leveraging historical and social media information with
a novel data integration methodology. The proposed approach is based on vine
copulas, which allow us to exploit the dependencies between different sources
of information. We apply the methodology to combine structured datasets
retrieved from official sources and a big unstructured dataset of information
collected from social media. The results show that the combined use of official
and online generated information contributes to yield a more accurate
assessment of the evolution of the Covid-19 pandemic, compared to the sole use
of official data.
| no_new_dataset | 0.711186 |
2110.13146 | Zhengzhong Yuan | Chen Zhao, Yuqing Liu, Li Hu, Zhengzhong Yuan | Reduce the rank calculation of a high-dimensional sparse matrix based on
network controllability theory | 10 pages, 4 figures | null | 10.1155/2022/5448123 | null | math.NA cs.NA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Numerical computing of the rank of a matrix is a fundamental problem in
scientific computation. The datasets generated by the internet often correspond
to the analysis of high-dimensional sparse matrices. Notwithstanding recent
advances in the promotion of traditional singular value decomposition (SVD), an
efficient estimation algorithm for the rank of a high-dimensional sparse matrix
is still lacking. Inspired by the controllability theory of complex networks,
we converted the rank of a matrix into maximum matching computing. Then, we
established a fast rank estimation algorithm by using the cavity method, a
powerful approximate technique for computing the maximum matching, to estimate
the rank of a sparse matrix. In the merit of the natural low complexity of the
cavity method, we showed that the rank of a high-dimensional sparse matrix can
be estimated in a much faster way than SVD with high accuracy. Our method
offers an efficient pathway to quickly estimate the rank of the
high-dimensional sparse matrix when the time cost of computing the rank by SVD
is unacceptable.
| [
{
"version": "v1",
"created": "Mon, 25 Oct 2021 07:53:07 GMT"
},
{
"version": "v2",
"created": "Wed, 6 Apr 2022 01:21:59 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Zhao",
"Chen",
""
],
[
"Liu",
"Yuqing",
""
],
[
"Hu",
"Li",
""
],
[
"Yuan",
"Zhengzhong",
""
]
] | TITLE: Reduce the rank calculation of a high-dimensional sparse matrix based on
network controllability theory
ABSTRACT: Numerical computing of the rank of a matrix is a fundamental problem in
scientific computation. The datasets generated by the internet often correspond
to the analysis of high-dimensional sparse matrices. Notwithstanding recent
advances in the promotion of traditional singular value decomposition (SVD), an
efficient estimation algorithm for the rank of a high-dimensional sparse matrix
is still lacking. Inspired by the controllability theory of complex networks,
we converted the rank of a matrix into maximum matching computing. Then, we
established a fast rank estimation algorithm by using the cavity method, a
powerful approximate technique for computing the maximum matching, to estimate
the rank of a sparse matrix. In the merit of the natural low complexity of the
cavity method, we showed that the rank of a high-dimensional sparse matrix can
be estimated in a much faster way than SVD with high accuracy. Our method
offers an efficient pathway to quickly estimate the rank of the
high-dimensional sparse matrix when the time cost of computing the rank by SVD
is unacceptable.
| no_new_dataset | 0.707777 |
2111.02571 | Ping Jiang | Ping Jiang, Junji Oaki, Yoshiyuki Ishihara, Junichiro Ooga, Haifeng
Han, Atsushi Sugahara, Seiji Tokura, Haruna Eto, Kazuma Komoda, and Akihito
Ogawa | Learning suction graspability considering grasp quality and robot
reachability for bin-picking | 18 pages, 2 tables, 7 figures | null | 10.3389/fnbot.2022.806898 | null | cs.RO cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Deep learning has been widely used for inferring robust grasps. Although
human-labeled RGB-D datasets were initially used to learn grasp configurations,
preparation of this kind of large dataset is expensive. To address this
problem, images were generated by a physical simulator, and a physically
inspired model (e.g., a contact model between a suction vacuum cup and object)
was used as a grasp quality evaluation metric to annotate the synthesized
images. However, this kind of contact model is complicated and requires
parameter identification by experiments to ensure real world performance. In
addition, previous studies have not considered manipulator reachability such as
when a grasp configuration with high grasp quality is unable to reach the
target due to collisions or the physical limitations of the robot. In this
study, we propose an intuitive geometric analytic-based grasp quality
evaluation metric. We further incorporate a reachability evaluation metric. We
annotate the pixel-wise grasp quality and reachability by the proposed
evaluation metric on synthesized images in a simulator to train an
auto-encoder--decoder called suction graspability U-Net++ (SG-U-Net++).
Experiment results show that our intuitive grasp quality evaluation metric is
competitive with a physically-inspired metric. Learning the reachability helps
to reduce motion planning computation time by removing obviously unreachable
candidates. The system achieves an overall picking speed of 560 PPH (pieces per
hour).
| [
{
"version": "v1",
"created": "Thu, 4 Nov 2021 00:55:42 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Jiang",
"Ping",
""
],
[
"Oaki",
"Junji",
""
],
[
"Ishihara",
"Yoshiyuki",
""
],
[
"Ooga",
"Junichiro",
""
],
[
"Han",
"Haifeng",
""
],
[
"Sugahara",
"Atsushi",
""
],
[
"Tokura",
"Seiji",
""
],
[
"Eto",
"Haruna",
""
],
[
"Komoda",
"Kazuma",
""
],
[
"Ogawa",
"Akihito",
""
]
] | TITLE: Learning suction graspability considering grasp quality and robot
reachability for bin-picking
ABSTRACT: Deep learning has been widely used for inferring robust grasps. Although
human-labeled RGB-D datasets were initially used to learn grasp configurations,
preparation of this kind of large dataset is expensive. To address this
problem, images were generated by a physical simulator, and a physically
inspired model (e.g., a contact model between a suction vacuum cup and object)
was used as a grasp quality evaluation metric to annotate the synthesized
images. However, this kind of contact model is complicated and requires
parameter identification by experiments to ensure real world performance. In
addition, previous studies have not considered manipulator reachability such as
when a grasp configuration with high grasp quality is unable to reach the
target due to collisions or the physical limitations of the robot. In this
study, we propose an intuitive geometric analytic-based grasp quality
evaluation metric. We further incorporate a reachability evaluation metric. We
annotate the pixel-wise grasp quality and reachability by the proposed
evaluation metric on synthesized images in a simulator to train an
auto-encoder--decoder called suction graspability U-Net++ (SG-U-Net++).
Experiment results show that our intuitive grasp quality evaluation metric is
competitive with a physically-inspired metric. Learning the reachability helps
to reduce motion planning computation time by removing obviously unreachable
candidates. The system achieves an overall picking speed of 560 PPH (pieces per
hour).
| no_new_dataset | 0.711481 |
2111.10050 | Kenji Kawaguchi | Hieu Pham, Zihang Dai, Golnaz Ghiasi, Kenji Kawaguchi, Hanxiao Liu,
Adams Wei Yu, Jiahui Yu, Yi-Ting Chen, Minh-Thang Luong, Yonghui Wu, Mingxing
Tan, Quoc V. Le | Combined Scaling for Zero-shot Transfer Learning | null | null | null | null | cs.LG cs.CL cs.CV | http://creativecommons.org/licenses/by/4.0/ | We present a combined scaling method - named BASIC - that achieves 85.7%
top-1 accuracy on the ImageNet ILSVRC-2012 validation set without learning from
any labeled ImageNet example. This accuracy surpasses best published similar
models - CLIP and ALIGN - by 9.3%. Our BASIC model also shows significant
improvements in robustness benchmarks. For instance, on 5 test sets with
natural distribution shifts such as ImageNet-{A,R,V2,Sketch} and ObjectNet, our
model achieves 84.3% top-1 average accuracy, only a small drop from its
original ImageNet accuracy. To achieve these results, we scale up the
contrastive learning framework of CLIP and ALIGN in three dimensions: data
size, model size, and batch size. Our dataset has 6.6B noisy image-text pairs,
which is 4x larger than ALIGN, and 16x larger than CLIP. Our largest model has
3B weights, which is 3.75x larger in parameters and 8x larger in FLOPs than
ALIGN and CLIP. Finally, our batch size is 65536 which is 2x more than CLIP and
4x more than ALIGN. We encountered two main challenges with the scaling rules
of BASIC. First, the main challenge with implementing the combined scaling
rules of BASIC is the limited memory of accelerators, such as GPUs and TPUs. To
overcome the memory limit, we propose two simple methods which make use of
gradient checkpointing and model parallelism. Second, while increasing the
dataset size and the model size has been the defacto method to improve the
performance of deep learning models like BASIC, the effect of a large
contrastive batch size on such contrastive-trained image-text models is not
well-understood. To shed light on the benefits of large contrastive batch
sizes, we develop a theoretical framework which shows that larger contrastive
batch sizes lead to smaller generalization gaps for image-text models such as
BASIC.
| [
{
"version": "v1",
"created": "Fri, 19 Nov 2021 05:25:46 GMT"
},
{
"version": "v2",
"created": "Fri, 29 Apr 2022 20:37:00 GMT"
},
{
"version": "v3",
"created": "Wed, 12 Apr 2023 08:26:28 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Pham",
"Hieu",
""
],
[
"Dai",
"Zihang",
""
],
[
"Ghiasi",
"Golnaz",
""
],
[
"Kawaguchi",
"Kenji",
""
],
[
"Liu",
"Hanxiao",
""
],
[
"Yu",
"Adams Wei",
""
],
[
"Yu",
"Jiahui",
""
],
[
"Chen",
"Yi-Ting",
""
],
[
"Luong",
"Minh-Thang",
""
],
[
"Wu",
"Yonghui",
""
],
[
"Tan",
"Mingxing",
""
],
[
"Le",
"Quoc V.",
""
]
] | TITLE: Combined Scaling for Zero-shot Transfer Learning
ABSTRACT: We present a combined scaling method - named BASIC - that achieves 85.7%
top-1 accuracy on the ImageNet ILSVRC-2012 validation set without learning from
any labeled ImageNet example. This accuracy surpasses best published similar
models - CLIP and ALIGN - by 9.3%. Our BASIC model also shows significant
improvements in robustness benchmarks. For instance, on 5 test sets with
natural distribution shifts such as ImageNet-{A,R,V2,Sketch} and ObjectNet, our
model achieves 84.3% top-1 average accuracy, only a small drop from its
original ImageNet accuracy. To achieve these results, we scale up the
contrastive learning framework of CLIP and ALIGN in three dimensions: data
size, model size, and batch size. Our dataset has 6.6B noisy image-text pairs,
which is 4x larger than ALIGN, and 16x larger than CLIP. Our largest model has
3B weights, which is 3.75x larger in parameters and 8x larger in FLOPs than
ALIGN and CLIP. Finally, our batch size is 65536 which is 2x more than CLIP and
4x more than ALIGN. We encountered two main challenges with the scaling rules
of BASIC. First, the main challenge with implementing the combined scaling
rules of BASIC is the limited memory of accelerators, such as GPUs and TPUs. To
overcome the memory limit, we propose two simple methods which make use of
gradient checkpointing and model parallelism. Second, while increasing the
dataset size and the model size has been the defacto method to improve the
performance of deep learning models like BASIC, the effect of a large
contrastive batch size on such contrastive-trained image-text models is not
well-understood. To shed light on the benefits of large contrastive batch
sizes, we develop a theoretical framework which shows that larger contrastive
batch sizes lead to smaller generalization gaps for image-text models such as
BASIC.
| no_new_dataset | 0.707992 |
2112.02807 | Qi Pang | Qi Pang, Yuanyuan Yuan, Shuai Wang | MDPFuzz: Testing Models Solving Markov Decision Processes | null | null | null | null | cs.SE cs.LG | http://creativecommons.org/licenses/by/4.0/ | The Markov decision process (MDP) provides a mathematical framework for
modeling sequential decision-making problems, many of which are crucial to
security and safety, such as autonomous driving and robot control. The rapid
development of artificial intelligence research has created efficient methods
for solving MDPs, such as deep neural networks (DNNs), reinforcement learning
(RL), and imitation learning (IL). However, these popular models solving MDPs
are neither thoroughly tested nor rigorously reliable.
We present MDPFuzz, the first blackbox fuzz testing framework for models
solving MDPs. MDPFuzz forms testing oracles by checking whether the target
model enters abnormal and dangerous states. During fuzzing, MDPFuzz decides
which mutated state to retain by measuring if it can reduce cumulative rewards
or form a new state sequence. We design efficient techniques to quantify the
"freshness" of a state sequence using Gaussian mixture models (GMMs) and
dynamic expectation-maximization (DynEM). We also prioritize states with high
potential of revealing crashes by estimating the local sensitivity of target
models over states.
MDPFuzz is evaluated on five state-of-the-art models for solving MDPs,
including supervised DNN, RL, IL, and multi-agent RL. Our evaluation includes
scenarios of autonomous driving, aircraft collision avoidance, and two games
that are often used to benchmark RL. During a 12-hour run, we find over 80
crash-triggering state sequences on each model. We show inspiring findings that
crash-triggering states, though they look normal, induce distinct neuron
activation patterns compared with normal states. We further develop an abnormal
behavior detector to harden all the evaluated models and repair them with the
findings of MDPFuzz to significantly enhance their robustness without
sacrificing accuracy.
| [
{
"version": "v1",
"created": "Mon, 6 Dec 2021 06:35:55 GMT"
},
{
"version": "v2",
"created": "Sun, 12 Dec 2021 03:47:30 GMT"
},
{
"version": "v3",
"created": "Mon, 25 Apr 2022 11:54:47 GMT"
},
{
"version": "v4",
"created": "Tue, 11 Apr 2023 22:19:33 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Pang",
"Qi",
""
],
[
"Yuan",
"Yuanyuan",
""
],
[
"Wang",
"Shuai",
""
]
] | TITLE: MDPFuzz: Testing Models Solving Markov Decision Processes
ABSTRACT: The Markov decision process (MDP) provides a mathematical framework for
modeling sequential decision-making problems, many of which are crucial to
security and safety, such as autonomous driving and robot control. The rapid
development of artificial intelligence research has created efficient methods
for solving MDPs, such as deep neural networks (DNNs), reinforcement learning
(RL), and imitation learning (IL). However, these popular models solving MDPs
are neither thoroughly tested nor rigorously reliable.
We present MDPFuzz, the first blackbox fuzz testing framework for models
solving MDPs. MDPFuzz forms testing oracles by checking whether the target
model enters abnormal and dangerous states. During fuzzing, MDPFuzz decides
which mutated state to retain by measuring if it can reduce cumulative rewards
or form a new state sequence. We design efficient techniques to quantify the
"freshness" of a state sequence using Gaussian mixture models (GMMs) and
dynamic expectation-maximization (DynEM). We also prioritize states with high
potential of revealing crashes by estimating the local sensitivity of target
models over states.
MDPFuzz is evaluated on five state-of-the-art models for solving MDPs,
including supervised DNN, RL, IL, and multi-agent RL. Our evaluation includes
scenarios of autonomous driving, aircraft collision avoidance, and two games
that are often used to benchmark RL. During a 12-hour run, we find over 80
crash-triggering state sequences on each model. We show inspiring findings that
crash-triggering states, though they look normal, induce distinct neuron
activation patterns compared with normal states. We further develop an abnormal
behavior detector to harden all the evaluated models and repair them with the
findings of MDPFuzz to significantly enhance their robustness without
sacrificing accuracy.
| no_new_dataset | 0.707398 |
2202.03881 | Antoine Wehenkel | Antoine Wehenkel, Jens Behrmann, Hsiang Hsu, Guillermo Sapiro, Gilles
Louppe, J\"orn-Henrik Jacobsen | Robust Hybrid Learning With Expert Augmentation | null | Transaction on Machine Learning Research, 2023 | null | null | cs.LG stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Hybrid modelling reduces the misspecification of expert models by combining
them with machine learning (ML) components learned from data. Similarly to many
ML algorithms, hybrid model performance guarantees are limited to the training
distribution. Leveraging the insight that the expert model is usually valid
even outside the training domain, we overcome this limitation by introducing a
hybrid data augmentation strategy termed \textit{expert augmentation}. Based on
a probabilistic formalization of hybrid modelling, we demonstrate that expert
augmentation, which can be incorporated into existing hybrid systems, improves
generalization. We empirically validate the expert augmentation on three
controlled experiments modelling dynamical systems with ordinary and partial
differential equations. Finally, we assess the potential real-world
applicability of expert augmentation on a dataset of a real double pendulum.
| [
{
"version": "v1",
"created": "Tue, 8 Feb 2022 14:11:04 GMT"
},
{
"version": "v2",
"created": "Wed, 9 Feb 2022 10:22:33 GMT"
},
{
"version": "v3",
"created": "Tue, 11 Apr 2023 19:51:36 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Wehenkel",
"Antoine",
""
],
[
"Behrmann",
"Jens",
""
],
[
"Hsu",
"Hsiang",
""
],
[
"Sapiro",
"Guillermo",
""
],
[
"Louppe",
"Gilles",
""
],
[
"Jacobsen",
"Jörn-Henrik",
""
]
] | TITLE: Robust Hybrid Learning With Expert Augmentation
ABSTRACT: Hybrid modelling reduces the misspecification of expert models by combining
them with machine learning (ML) components learned from data. Similarly to many
ML algorithms, hybrid model performance guarantees are limited to the training
distribution. Leveraging the insight that the expert model is usually valid
even outside the training domain, we overcome this limitation by introducing a
hybrid data augmentation strategy termed \textit{expert augmentation}. Based on
a probabilistic formalization of hybrid modelling, we demonstrate that expert
augmentation, which can be incorporated into existing hybrid systems, improves
generalization. We empirically validate the expert augmentation on three
controlled experiments modelling dynamical systems with ordinary and partial
differential equations. Finally, we assess the potential real-world
applicability of expert augmentation on a dataset of a real double pendulum.
| no_new_dataset | 0.70939 |
2203.09829 | Abdul Hameed Azeemi | Abdul Hameed Azeemi, Ihsan Ayyub Qazi, Agha Ali Raza | Representative Subset Selection for Efficient Fine-Tuning in
Self-Supervised Speech Recognition | 16 pages, 8 figures | null | null | null | cs.LG cs.SD eess.AS | http://creativecommons.org/licenses/by/4.0/ | Self-supervised speech recognition models require considerable labeled
training data for learning high-fidelity representations for Automatic Speech
Recognition (ASR) which is computationally demanding and time-consuming. We
consider the task of identifying an optimal subset of data for efficient
fine-tuning in self-supervised speech models for ASR. We discover that the
dataset pruning strategies used in vision tasks for sampling the most
informative examples do not perform better than random subset selection on
fine-tuning self-supervised ASR. We then present the COWERAGE algorithm for
representative subset selection in self-supervised ASR. COWERAGE is based on
our finding that ensuring the coverage of examples based on training Word Error
Rate (WER) in the early training epochs leads to better generalization
performance. Extensive experiments with the wav2vec 2.0 and HuBERT model on
TIMIT, Librispeech, and LJSpeech datasets show the effectiveness of COWERAGE
and its transferability across models, with up to 17% relative WER improvement
over existing dataset pruning methods and random sampling. We also demonstrate
that the coverage of training instances in terms of WER values ensures the
inclusion of phonemically diverse examples, leading to better test accuracy in
self-supervised speech recognition models.
| [
{
"version": "v1",
"created": "Fri, 18 Mar 2022 10:12:24 GMT"
},
{
"version": "v2",
"created": "Fri, 24 Jun 2022 17:46:32 GMT"
},
{
"version": "v3",
"created": "Tue, 11 Apr 2023 18:13:48 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Azeemi",
"Abdul Hameed",
""
],
[
"Qazi",
"Ihsan Ayyub",
""
],
[
"Raza",
"Agha Ali",
""
]
] | TITLE: Representative Subset Selection for Efficient Fine-Tuning in
Self-Supervised Speech Recognition
ABSTRACT: Self-supervised speech recognition models require considerable labeled
training data for learning high-fidelity representations for Automatic Speech
Recognition (ASR) which is computationally demanding and time-consuming. We
consider the task of identifying an optimal subset of data for efficient
fine-tuning in self-supervised speech models for ASR. We discover that the
dataset pruning strategies used in vision tasks for sampling the most
informative examples do not perform better than random subset selection on
fine-tuning self-supervised ASR. We then present the COWERAGE algorithm for
representative subset selection in self-supervised ASR. COWERAGE is based on
our finding that ensuring the coverage of examples based on training Word Error
Rate (WER) in the early training epochs leads to better generalization
performance. Extensive experiments with the wav2vec 2.0 and HuBERT model on
TIMIT, Librispeech, and LJSpeech datasets show the effectiveness of COWERAGE
and its transferability across models, with up to 17% relative WER improvement
over existing dataset pruning methods and random sampling. We also demonstrate
that the coverage of training instances in terms of WER values ensures the
inclusion of phonemically diverse examples, leading to better test accuracy in
self-supervised speech recognition models.
| no_new_dataset | 0.712207 |
2203.09964 | Tim Keil | Tim Keil and Mario Ohlberger | A Relaxed Localized Trust-Region Reduced Basis Approach for Optimization
of Multiscale Problems | null | null | null | null | math.NA cs.NA math.OC | http://creativecommons.org/licenses/by/4.0/ | In this contribution, we are concerned with parameter optimization problems
that are constrained by multiscale PDE state equations. As an efficient
numerical solution approach for such problems, we introduce and analyze a new
relaxed and localized trust-region reduced basis method. Localization is
obtained based on a Petrov-Galerkin localized orthogonal decomposition method
and its recently introduced two-scale reduced basis approximation. We derive
efficient localizable a posteriori error estimates for the optimality system,
as well as for the two-scale reduced objective functional. While the relaxation
of the outer trust-region optimization loop still allows for a rigorous
convergence result, the resulting method converges much faster due to larger
step sizes in the initial phase of the iterative algorithms. The resulting
algorithm is parallelized in order to take advantage of the localization.
Numerical experiments are given for a multiscale thermal block benchmark
problem. The experiments demonstrate the efficiency of the approach,
particularly for large scale problems, where methods based on traditional
finite element approximation schemes are prohibitive or fail entirely.
| [
{
"version": "v1",
"created": "Fri, 18 Mar 2022 14:00:37 GMT"
},
{
"version": "v2",
"created": "Wed, 12 Apr 2023 17:04:03 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Keil",
"Tim",
""
],
[
"Ohlberger",
"Mario",
""
]
] | TITLE: A Relaxed Localized Trust-Region Reduced Basis Approach for Optimization
of Multiscale Problems
ABSTRACT: In this contribution, we are concerned with parameter optimization problems
that are constrained by multiscale PDE state equations. As an efficient
numerical solution approach for such problems, we introduce and analyze a new
relaxed and localized trust-region reduced basis method. Localization is
obtained based on a Petrov-Galerkin localized orthogonal decomposition method
and its recently introduced two-scale reduced basis approximation. We derive
efficient localizable a posteriori error estimates for the optimality system,
as well as for the two-scale reduced objective functional. While the relaxation
of the outer trust-region optimization loop still allows for a rigorous
convergence result, the resulting method converges much faster due to larger
step sizes in the initial phase of the iterative algorithms. The resulting
algorithm is parallelized in order to take advantage of the localization.
Numerical experiments are given for a multiscale thermal block benchmark
problem. The experiments demonstrate the efficiency of the approach,
particularly for large scale problems, where methods based on traditional
finite element approximation schemes are prohibitive or fail entirely.
| no_new_dataset | 0.709787 |
2203.14205 | Tianying Liu | Tianying Liu, Lu Zhang, Yang Wang, Jihong Guan, Yanwei Fu, Jiajia
Zhao, Shuigeng Zhou | Recent Few-Shot Object Detection Algorithms: A Survey with Performance
Comparison | 35 pages, 14 figures | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The generic object detection (GOD) task has been successfully tackled by
recent deep neural networks, trained by an avalanche of annotated training
samples from some common classes. However, it is still non-trivial to
generalize these object detectors to the novel long-tailed object classes,
which have only few labeled training samples. To this end, the Few-Shot Object
Detection (FSOD) has been topical recently, as it mimics the humans' ability of
learning to learn, and intelligently transfers the learned generic object
knowledge from the common heavy-tailed, to the novel long-tailed object
classes. Especially, the research in this emerging field has been flourishing
in recent years with various benchmarks, backbones, and methodologies proposed.
To review these FSOD works, there are several insightful FSOD survey articles
[58, 59, 74, 78] that systematically study and compare them as the groups of
fine-tuning/transfer learning, and meta-learning methods. In contrast, we
review the existing FSOD algorithms from a new perspective under a new taxonomy
based on their contributions, i.e., data-oriented, model-oriented, and
algorithm-oriented. Thus, a comprehensive survey with performance comparison is
conducted on recent achievements of FSOD. Furthermore, we also analyze the
technical challenges, the merits and demerits of these methods, and envision
the future directions of FSOD. Specifically, we give an overview of FSOD,
including the problem definition, common datasets, and evaluation protocols.
The taxonomy is then proposed that groups FSOD methods into three types.
Following this taxonomy, we provide a systematic review of the advances in
FSOD. Finally, further discussions on performance, challenges, and future
directions are presented.
| [
{
"version": "v1",
"created": "Sun, 27 Mar 2022 04:11:28 GMT"
},
{
"version": "v2",
"created": "Wed, 12 Apr 2023 05:26:58 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Liu",
"Tianying",
""
],
[
"Zhang",
"Lu",
""
],
[
"Wang",
"Yang",
""
],
[
"Guan",
"Jihong",
""
],
[
"Fu",
"Yanwei",
""
],
[
"Zhao",
"Jiajia",
""
],
[
"Zhou",
"Shuigeng",
""
]
] | TITLE: Recent Few-Shot Object Detection Algorithms: A Survey with Performance
Comparison
ABSTRACT: The generic object detection (GOD) task has been successfully tackled by
recent deep neural networks, trained by an avalanche of annotated training
samples from some common classes. However, it is still non-trivial to
generalize these object detectors to the novel long-tailed object classes,
which have only few labeled training samples. To this end, the Few-Shot Object
Detection (FSOD) has been topical recently, as it mimics the humans' ability of
learning to learn, and intelligently transfers the learned generic object
knowledge from the common heavy-tailed, to the novel long-tailed object
classes. Especially, the research in this emerging field has been flourishing
in recent years with various benchmarks, backbones, and methodologies proposed.
To review these FSOD works, there are several insightful FSOD survey articles
[58, 59, 74, 78] that systematically study and compare them as the groups of
fine-tuning/transfer learning, and meta-learning methods. In contrast, we
review the existing FSOD algorithms from a new perspective under a new taxonomy
based on their contributions, i.e., data-oriented, model-oriented, and
algorithm-oriented. Thus, a comprehensive survey with performance comparison is
conducted on recent achievements of FSOD. Furthermore, we also analyze the
technical challenges, the merits and demerits of these methods, and envision
the future directions of FSOD. Specifically, we give an overview of FSOD,
including the problem definition, common datasets, and evaluation protocols.
The taxonomy is then proposed that groups FSOD methods into three types.
Following this taxonomy, we provide a systematic review of the advances in
FSOD. Finally, further discussions on performance, challenges, and future
directions are presented.
| no_new_dataset | 0.706823 |
2204.08358 | Hassan Eldeeb Dr. | Hassan Eldeeb, Mohamed Maher, Radwa Elshawi, and Sherif Sakr | AutoMLBench: A Comprehensive Experimental Evaluation of Automated
Machine Learning Frameworks | null | null | null | null | cs.LG cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | With the booming demand for machine learning applications, it has been
recognized that the number of knowledgeable data scientists can not scale with
the growing data volumes and application needs in our digital world. In
response to this demand, several automated machine learning (AutoML) frameworks
have been developed to fill the gap of human expertise by automating the
process of building machine learning pipelines. Each framework comes with
different heuristics-based design decisions. In this study, we present a
comprehensive evaluation and comparison of the performance characteristics of
six popular AutoML frameworks, namely, AutoWeka, AutoSKlearn, TPOT, Recipe,
ATM, and SmartML, across 100 data sets from established AutoML benchmark
suites. Our experimental evaluation considers different aspects for its
comparison, including the performance impact of several design decisions,
including time budget, size of search space, meta-learning, and ensemble
construction. The results of our study reveal various interesting insights that
can significantly guide and impact the design of AutoML frameworks.
| [
{
"version": "v1",
"created": "Mon, 18 Apr 2022 15:03:53 GMT"
},
{
"version": "v2",
"created": "Wed, 12 Apr 2023 09:29:21 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Eldeeb",
"Hassan",
""
],
[
"Maher",
"Mohamed",
""
],
[
"Elshawi",
"Radwa",
""
],
[
"Sakr",
"Sherif",
""
]
] | TITLE: AutoMLBench: A Comprehensive Experimental Evaluation of Automated
Machine Learning Frameworks
ABSTRACT: With the booming demand for machine learning applications, it has been
recognized that the number of knowledgeable data scientists can not scale with
the growing data volumes and application needs in our digital world. In
response to this demand, several automated machine learning (AutoML) frameworks
have been developed to fill the gap of human expertise by automating the
process of building machine learning pipelines. Each framework comes with
different heuristics-based design decisions. In this study, we present a
comprehensive evaluation and comparison of the performance characteristics of
six popular AutoML frameworks, namely, AutoWeka, AutoSKlearn, TPOT, Recipe,
ATM, and SmartML, across 100 data sets from established AutoML benchmark
suites. Our experimental evaluation considers different aspects for its
comparison, including the performance impact of several design decisions,
including time budget, size of search space, meta-learning, and ensemble
construction. The results of our study reveal various interesting insights that
can significantly guide and impact the design of AutoML frameworks.
| no_new_dataset | 0.707771 |
2204.14211 | Joel Jang | Joel Jang, Seonghyeon Ye, Changho Lee, Sohee Yang, Joongbo Shin,
Janghoon Han, Gyeonghun Kim, Minjoon Seo | TemporalWiki: A Lifelong Benchmark for Training and Evaluating
Ever-Evolving Language Models | published at EMNLP 2022 | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Language Models (LMs) become outdated as the world changes; they often fail
to perform tasks requiring recent factual information which was absent or
different during training, a phenomenon called temporal misalignment. This is
especially a challenging problem because the research community still lacks a
coherent dataset for assessing the adaptability of LMs to frequently-updated
knowledge corpus such as Wikipedia. To this end, we introduce TemporalWiki, a
lifelong benchmark for ever-evolving LMs that utilizes the difference between
consecutive snapshots of English Wikipedia and English Wikidata for training
and evaluation, respectively. The benchmark hence allows researchers to
periodically track an LM's ability to retain previous knowledge and acquire
updated/new knowledge at each point in time. We also find that training an LM
on the diff data through continual learning methods achieves similar or better
perplexity than on the entire snapshot in our benchmark with 12 times less
computational cost, which verifies that factual knowledge in LMs can be safely
updated with minimal training data via continual learning. The dataset and the
code are available at https://github.com/joeljang/temporalwiki.
| [
{
"version": "v1",
"created": "Fri, 29 Apr 2022 16:40:07 GMT"
},
{
"version": "v2",
"created": "Wed, 8 Feb 2023 05:15:18 GMT"
},
{
"version": "v3",
"created": "Wed, 12 Apr 2023 12:16:59 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Jang",
"Joel",
""
],
[
"Ye",
"Seonghyeon",
""
],
[
"Lee",
"Changho",
""
],
[
"Yang",
"Sohee",
""
],
[
"Shin",
"Joongbo",
""
],
[
"Han",
"Janghoon",
""
],
[
"Kim",
"Gyeonghun",
""
],
[
"Seo",
"Minjoon",
""
]
] | TITLE: TemporalWiki: A Lifelong Benchmark for Training and Evaluating
Ever-Evolving Language Models
ABSTRACT: Language Models (LMs) become outdated as the world changes; they often fail
to perform tasks requiring recent factual information which was absent or
different during training, a phenomenon called temporal misalignment. This is
especially a challenging problem because the research community still lacks a
coherent dataset for assessing the adaptability of LMs to frequently-updated
knowledge corpus such as Wikipedia. To this end, we introduce TemporalWiki, a
lifelong benchmark for ever-evolving LMs that utilizes the difference between
consecutive snapshots of English Wikipedia and English Wikidata for training
and evaluation, respectively. The benchmark hence allows researchers to
periodically track an LM's ability to retain previous knowledge and acquire
updated/new knowledge at each point in time. We also find that training an LM
on the diff data through continual learning methods achieves similar or better
perplexity than on the entire snapshot in our benchmark with 12 times less
computational cost, which verifies that factual knowledge in LMs can be safely
updated with minimal training data via continual learning. The dataset and the
code are available at https://github.com/joeljang/temporalwiki.
| new_dataset | 0.716237 |
2205.03778 | Chaoqi Liang | Chaoqi Liang, Yu Zhang, Xinyuan Li, Jinyu Zhang and Yongqi Yu | FuDFEND: Fuzzy-domain for Multi-domain Fake News Detection | null | null | 10.1007/978-3-031-17189-5_4 | null | cs.SI | http://creativecommons.org/licenses/by/4.0/ | On the Internet, fake news exists in various domain (e.g., education,
health). Since news in different domains has different features, researchers
have be-gun to use single domain label for fake news detection recently. This
emerg-ing field is called multi-domain fake news detection (MFND). Existing
works show that using single domain label can improve the accuracy of fake news
detection model. However, there are two problems in previous works. Firstly,
they ignore that a piece of news may have features from different domains. The
single domain label focuses only on the features of the news on particu-lar
domain. This may reduce the performance of the model. Secondly, their model
cannot transfer the domain knowledge to the other dataset without domain label.
In this paper, we propose a novel model, FuDFEND, which solves the limitations
above by introducing the fuzzy inference mechanism. Specifically, FuDFEND
utilizes a neural network to fit the fuzzy inference process which constructs a
fuzzy domain label for each news item. Then, the feature extraction module uses
the fuzzy domain label to extract the multi-domain features of the news and
obtain the total feature representation. Fi-nally, the discriminator module
uses the total feature representation to dis-criminate whether the news item is
fake news. The results on the Weibo21 show that our model works better than the
model using only single domain label. In addition, our model transfers domain
knowledge better to Thu da-taset which has no domain label.
| [
{
"version": "v1",
"created": "Sun, 8 May 2022 04:25:12 GMT"
},
{
"version": "v2",
"created": "Wed, 12 Apr 2023 01:50:07 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Liang",
"Chaoqi",
""
],
[
"Zhang",
"Yu",
""
],
[
"Li",
"Xinyuan",
""
],
[
"Zhang",
"Jinyu",
""
],
[
"Yu",
"Yongqi",
""
]
] | TITLE: FuDFEND: Fuzzy-domain for Multi-domain Fake News Detection
ABSTRACT: On the Internet, fake news exists in various domain (e.g., education,
health). Since news in different domains has different features, researchers
have be-gun to use single domain label for fake news detection recently. This
emerg-ing field is called multi-domain fake news detection (MFND). Existing
works show that using single domain label can improve the accuracy of fake news
detection model. However, there are two problems in previous works. Firstly,
they ignore that a piece of news may have features from different domains. The
single domain label focuses only on the features of the news on particu-lar
domain. This may reduce the performance of the model. Secondly, their model
cannot transfer the domain knowledge to the other dataset without domain label.
In this paper, we propose a novel model, FuDFEND, which solves the limitations
above by introducing the fuzzy inference mechanism. Specifically, FuDFEND
utilizes a neural network to fit the fuzzy inference process which constructs a
fuzzy domain label for each news item. Then, the feature extraction module uses
the fuzzy domain label to extract the multi-domain features of the news and
obtain the total feature representation. Fi-nally, the discriminator module
uses the total feature representation to dis-criminate whether the news item is
fake news. The results on the Weibo21 show that our model works better than the
model using only single domain label. In addition, our model transfers domain
knowledge better to Thu da-taset which has no domain label.
| no_new_dataset | 0.710226 |
2205.12676 | Sebastian Ruder | Simran Khanuja, Sebastian Ruder, Partha Talukdar | Evaluating the Diversity, Equity and Inclusion of NLP Technology: A Case
Study for Indian Languages | Accepted to EACL Findings, 2023 | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In order for NLP technology to be widely applicable, fair, and useful, it
needs to serve a diverse set of speakers across the world's languages, be
equitable, i.e., not unduly biased towards any particular language, and be
inclusive of all users, particularly in low-resource settings where compute
constraints are common. In this paper, we propose an evaluation paradigm that
assesses NLP technologies across all three dimensions. While diversity and
inclusion have received attention in recent literature, equity is currently
unexplored. We propose to address this gap using the Gini coefficient, a
well-established metric used for estimating societal wealth inequality. Using
our paradigm, we highlight the distressed state of current technologies for
Indian (IN) languages (a linguistically large and diverse set, with a varied
speaker population), across all three dimensions. To improve upon these
metrics, we demonstrate the importance of region-specific choices in model
building and dataset creation, and more importantly, propose a novel,
generalisable approach to optimal resource allocation during fine-tuning.
Finally, we discuss steps to mitigate these biases and encourage the community
to employ multi-faceted evaluation when building linguistically diverse and
equitable technologies.
| [
{
"version": "v1",
"created": "Wed, 25 May 2022 11:38:04 GMT"
},
{
"version": "v2",
"created": "Tue, 21 Mar 2023 03:34:36 GMT"
},
{
"version": "v3",
"created": "Wed, 12 Apr 2023 14:49:46 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Khanuja",
"Simran",
""
],
[
"Ruder",
"Sebastian",
""
],
[
"Talukdar",
"Partha",
""
]
] | TITLE: Evaluating the Diversity, Equity and Inclusion of NLP Technology: A Case
Study for Indian Languages
ABSTRACT: In order for NLP technology to be widely applicable, fair, and useful, it
needs to serve a diverse set of speakers across the world's languages, be
equitable, i.e., not unduly biased towards any particular language, and be
inclusive of all users, particularly in low-resource settings where compute
constraints are common. In this paper, we propose an evaluation paradigm that
assesses NLP technologies across all three dimensions. While diversity and
inclusion have received attention in recent literature, equity is currently
unexplored. We propose to address this gap using the Gini coefficient, a
well-established metric used for estimating societal wealth inequality. Using
our paradigm, we highlight the distressed state of current technologies for
Indian (IN) languages (a linguistically large and diverse set, with a varied
speaker population), across all three dimensions. To improve upon these
metrics, we demonstrate the importance of region-specific choices in model
building and dataset creation, and more importantly, propose a novel,
generalisable approach to optimal resource allocation during fine-tuning.
Finally, we discuss steps to mitigate these biases and encourage the community
to employ multi-faceted evaluation when building linguistically diverse and
equitable technologies.
| no_new_dataset | 0.710415 |
2205.15960 | Genta Indra Winata | Genta Indra Winata, Alham Fikri Aji, Samuel Cahyawijaya, Rahmad
Mahendra, Fajri Koto, Ade Romadhony, Kemal Kurniawan, David Moeljadi, Radityo
Eko Prasojo, Pascale Fung, Timothy Baldwin, Jey Han Lau, Rico Sennrich,
Sebastian Ruder | NusaX: Multilingual Parallel Sentiment Dataset for 10 Indonesian Local
Languages | EACL 2023 | null | null | null | cs.CL | http://creativecommons.org/licenses/by-sa/4.0/ | Natural language processing (NLP) has a significant impact on society via
technologies such as machine translation and search engines. Despite its
success, NLP technology is only widely available for high-resource languages
such as English and Chinese, while it remains inaccessible to many languages
due to the unavailability of data resources and benchmarks. In this work, we
focus on developing resources for languages in Indonesia. Despite being the
second most linguistically diverse country, most languages in Indonesia are
categorized as endangered and some are even extinct. We develop the first-ever
parallel resource for 10 low-resource languages in Indonesia. Our resource
includes datasets, a multi-task benchmark, and lexicons, as well as a parallel
Indonesian-English dataset. We provide extensive analyses and describe the
challenges when creating such resources. We hope that our work can spark NLP
research on Indonesian and other underrepresented languages.
| [
{
"version": "v1",
"created": "Tue, 31 May 2022 17:03:50 GMT"
},
{
"version": "v2",
"created": "Wed, 12 Apr 2023 16:42:53 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Winata",
"Genta Indra",
""
],
[
"Aji",
"Alham Fikri",
""
],
[
"Cahyawijaya",
"Samuel",
""
],
[
"Mahendra",
"Rahmad",
""
],
[
"Koto",
"Fajri",
""
],
[
"Romadhony",
"Ade",
""
],
[
"Kurniawan",
"Kemal",
""
],
[
"Moeljadi",
"David",
""
],
[
"Prasojo",
"Radityo Eko",
""
],
[
"Fung",
"Pascale",
""
],
[
"Baldwin",
"Timothy",
""
],
[
"Lau",
"Jey Han",
""
],
[
"Sennrich",
"Rico",
""
],
[
"Ruder",
"Sebastian",
""
]
] | TITLE: NusaX: Multilingual Parallel Sentiment Dataset for 10 Indonesian Local
Languages
ABSTRACT: Natural language processing (NLP) has a significant impact on society via
technologies such as machine translation and search engines. Despite its
success, NLP technology is only widely available for high-resource languages
such as English and Chinese, while it remains inaccessible to many languages
due to the unavailability of data resources and benchmarks. In this work, we
focus on developing resources for languages in Indonesia. Despite being the
second most linguistically diverse country, most languages in Indonesia are
categorized as endangered and some are even extinct. We develop the first-ever
parallel resource for 10 low-resource languages in Indonesia. Our resource
includes datasets, a multi-task benchmark, and lexicons, as well as a parallel
Indonesian-English dataset. We provide extensive analyses and describe the
challenges when creating such resources. We hope that our work can spark NLP
research on Indonesian and other underrepresented languages.
| new_dataset | 0.71408 |
2206.12126 | Cheng Tan | Cheng Tan, Zhangyang Gao, Lirong Wu, Yongjie Xu, Jun Xia, Siyuan Li,
Stan Z. Li | Temporal Attention Unit: Towards Efficient Spatiotemporal Predictive
Learning | Accepted by CVPR 2023 | null | null | null | cs.CV cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Spatiotemporal predictive learning aims to generate future frames by learning
from historical frames. In this paper, we investigate existing methods and
present a general framework of spatiotemporal predictive learning, in which the
spatial encoder and decoder capture intra-frame features and the middle
temporal module catches inter-frame correlations. While the mainstream methods
employ recurrent units to capture long-term temporal dependencies, they suffer
from low computational efficiency due to their unparallelizable architectures.
To parallelize the temporal module, we propose the Temporal Attention Unit
(TAU), which decomposes the temporal attention into intra-frame statical
attention and inter-frame dynamical attention. Moreover, while the mean squared
error loss focuses on intra-frame errors, we introduce a novel differential
divergence regularization to take inter-frame variations into account.
Extensive experiments demonstrate that the proposed method enables the derived
model to achieve competitive performance on various spatiotemporal prediction
benchmarks.
| [
{
"version": "v1",
"created": "Fri, 24 Jun 2022 07:43:50 GMT"
},
{
"version": "v2",
"created": "Fri, 17 Mar 2023 13:25:42 GMT"
},
{
"version": "v3",
"created": "Wed, 12 Apr 2023 08:08:50 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Tan",
"Cheng",
""
],
[
"Gao",
"Zhangyang",
""
],
[
"Wu",
"Lirong",
""
],
[
"Xu",
"Yongjie",
""
],
[
"Xia",
"Jun",
""
],
[
"Li",
"Siyuan",
""
],
[
"Li",
"Stan Z.",
""
]
] | TITLE: Temporal Attention Unit: Towards Efficient Spatiotemporal Predictive
Learning
ABSTRACT: Spatiotemporal predictive learning aims to generate future frames by learning
from historical frames. In this paper, we investigate existing methods and
present a general framework of spatiotemporal predictive learning, in which the
spatial encoder and decoder capture intra-frame features and the middle
temporal module catches inter-frame correlations. While the mainstream methods
employ recurrent units to capture long-term temporal dependencies, they suffer
from low computational efficiency due to their unparallelizable architectures.
To parallelize the temporal module, we propose the Temporal Attention Unit
(TAU), which decomposes the temporal attention into intra-frame statical
attention and inter-frame dynamical attention. Moreover, while the mean squared
error loss focuses on intra-frame errors, we introduce a novel differential
divergence regularization to take inter-frame variations into account.
Extensive experiments demonstrate that the proposed method enables the derived
model to achieve competitive performance on various spatiotemporal prediction
benchmarks.
| no_new_dataset | 0.708231 |
2209.10913 | Shengcai Liu | Shengcai Liu, Yu Zhang, Ke Tang, Xin Yao | How Good Is Neural Combinatorial Optimization? A Systematic Evaluation
on the Traveling Salesman Problem | Accepted by IEEE CIM | null | null | null | cs.NE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Traditional solvers for tackling combinatorial optimization (CO) problems are
usually designed by human experts. Recently, there has been a surge of interest
in utilizing deep learning, especially deep reinforcement learning, to
automatically learn effective solvers for CO. The resultant new paradigm is
termed neural combinatorial optimization (NCO). However, the advantages and
disadvantages of NCO relative to other approaches have not been empirically or
theoretically well studied. This work presents a comprehensive comparative
study of NCO solvers and alternative solvers. Specifically, taking the
traveling salesman problem as the testbed problem, the performance of the
solvers is assessed in five aspects, i.e., effectiveness, efficiency,
stability, scalability, and generalization ability. Our results show that the
solvers learned by NCO approaches, in general, still fall short of traditional
solvers in nearly all these aspects. A potential benefit of NCO solvers would
be their superior time and energy efficiency for small-size problem instances
when sufficient training instances are available. Hopefully, this work would
help with a better understanding of the strengths and weaknesses of NCO and
provide a comprehensive evaluation protocol for further benchmarking NCO
approaches in comparison to other approaches.
| [
{
"version": "v1",
"created": "Thu, 22 Sep 2022 10:50:36 GMT"
},
{
"version": "v2",
"created": "Wed, 12 Apr 2023 09:12:26 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Liu",
"Shengcai",
""
],
[
"Zhang",
"Yu",
""
],
[
"Tang",
"Ke",
""
],
[
"Yao",
"Xin",
""
]
] | TITLE: How Good Is Neural Combinatorial Optimization? A Systematic Evaluation
on the Traveling Salesman Problem
ABSTRACT: Traditional solvers for tackling combinatorial optimization (CO) problems are
usually designed by human experts. Recently, there has been a surge of interest
in utilizing deep learning, especially deep reinforcement learning, to
automatically learn effective solvers for CO. The resultant new paradigm is
termed neural combinatorial optimization (NCO). However, the advantages and
disadvantages of NCO relative to other approaches have not been empirically or
theoretically well studied. This work presents a comprehensive comparative
study of NCO solvers and alternative solvers. Specifically, taking the
traveling salesman problem as the testbed problem, the performance of the
solvers is assessed in five aspects, i.e., effectiveness, efficiency,
stability, scalability, and generalization ability. Our results show that the
solvers learned by NCO approaches, in general, still fall short of traditional
solvers in nearly all these aspects. A potential benefit of NCO solvers would
be their superior time and energy efficiency for small-size problem instances
when sufficient training instances are available. Hopefully, this work would
help with a better understanding of the strengths and weaknesses of NCO and
provide a comprehensive evaluation protocol for further benchmarking NCO
approaches in comparison to other approaches.
| no_new_dataset | 0.707209 |
2209.14072 | Zirui Wu | Yongliang Shi, Runyi Yang, Pengfei Li, Zirui Wu, Hao Zhao, Guyue Zhou | City-scale Incremental Neural Mapping with Three-layer Sampling and
Panoptic Representation | 8 pages, 9 figures | null | null | null | cs.CV cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Neural implicit representations are drawing a lot of attention from the
robotics community recently, as they are expressive, continuous and compact.
However, city-scale continual implicit dense mapping based on sparse LiDAR
input is still an under-explored challenge. To this end, we successfully build
a city-scale continual neural mapping system with a panoptic representation
that consists of environment-level and instance-level modelling. Given a stream
of sparse LiDAR point cloud, it maintains a dynamic generative model that maps
3D coordinates to signed distance field (SDF) values. To address the difficulty
of representing geometric information at different levels in city-scale space,
we propose a tailored three-layer sampling strategy to dynamically sample the
global, local and near-surface domains. Meanwhile, to realize high fidelity
mapping of instance under incomplete observation, category-specific prior is
introduced to better model the geometric details. We evaluate on the public
SemanticKITTI dataset and demonstrate the significance of the newly proposed
three-layer sampling strategy and panoptic representation, using both
quantitative and qualitative results. Codes and model will be publicly
available.
| [
{
"version": "v1",
"created": "Wed, 28 Sep 2022 13:14:40 GMT"
},
{
"version": "v2",
"created": "Wed, 12 Apr 2023 12:06:09 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Shi",
"Yongliang",
""
],
[
"Yang",
"Runyi",
""
],
[
"Li",
"Pengfei",
""
],
[
"Wu",
"Zirui",
""
],
[
"Zhao",
"Hao",
""
],
[
"Zhou",
"Guyue",
""
]
] | TITLE: City-scale Incremental Neural Mapping with Three-layer Sampling and
Panoptic Representation
ABSTRACT: Neural implicit representations are drawing a lot of attention from the
robotics community recently, as they are expressive, continuous and compact.
However, city-scale continual implicit dense mapping based on sparse LiDAR
input is still an under-explored challenge. To this end, we successfully build
a city-scale continual neural mapping system with a panoptic representation
that consists of environment-level and instance-level modelling. Given a stream
of sparse LiDAR point cloud, it maintains a dynamic generative model that maps
3D coordinates to signed distance field (SDF) values. To address the difficulty
of representing geometric information at different levels in city-scale space,
we propose a tailored three-layer sampling strategy to dynamically sample the
global, local and near-surface domains. Meanwhile, to realize high fidelity
mapping of instance under incomplete observation, category-specific prior is
introduced to better model the geometric details. We evaluate on the public
SemanticKITTI dataset and demonstrate the significance of the newly proposed
three-layer sampling strategy and panoptic representation, using both
quantitative and qualitative results. Codes and model will be publicly
available.
| no_new_dataset | 0.708843 |
2210.07713 | Leonardos Pantiskas | Leonardos Pantiskas, Kees Verstoep, Mark Hoogendoorn, Henri Bal | An Empirical Evaluation of Multivariate Time Series Classification with
Input Transformation across Different Dimensions | IEEE ICMLA 2022 | null | 10.1109/ICMLA55696.2022.00012 | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In current research, machine and deep learning solutions for the
classification of temporal data are shifting from single-channel datasets
(univariate) to problems with multiple channels of information (multivariate).
The majority of these works are focused on the method novelty and architecture,
and the format of the input data is often treated implicitly. Particularly,
multivariate datasets are often treated as a stack of univariate time series in
terms of input preprocessing, with scaling methods applied across each channel
separately. In this evaluation, we aim to demonstrate that the additional
channel dimension is far from trivial and different approaches to scaling can
lead to significantly different results in the accuracy of a solution. To that
end, we test seven different data transformation methods on four different
temporal dimensions and study their effect on the classification accuracy of
five recent methods. We show that, for the large majority of tested datasets,
the best transformation-dimension configuration leads to an increase in the
accuracy compared to the result of each model with the same hyperparameters and
no scaling, ranging from 0.16 to 76.79 percentage points. We also show that if
we keep the transformation method constant, there is a statistically
significant difference in accuracy results when applying it across different
dimensions, with accuracy differences ranging from 0.23 to 47.79 percentage
points. Finally, we explore the relation of the transformation methods and
dimensions to the classifiers, and we conclude that there is no prominent
general trend, and the optimal configuration is dataset- and
classifier-specific.
| [
{
"version": "v1",
"created": "Fri, 14 Oct 2022 11:27:24 GMT"
},
{
"version": "v2",
"created": "Wed, 12 Apr 2023 15:15:17 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Pantiskas",
"Leonardos",
""
],
[
"Verstoep",
"Kees",
""
],
[
"Hoogendoorn",
"Mark",
""
],
[
"Bal",
"Henri",
""
]
] | TITLE: An Empirical Evaluation of Multivariate Time Series Classification with
Input Transformation across Different Dimensions
ABSTRACT: In current research, machine and deep learning solutions for the
classification of temporal data are shifting from single-channel datasets
(univariate) to problems with multiple channels of information (multivariate).
The majority of these works are focused on the method novelty and architecture,
and the format of the input data is often treated implicitly. Particularly,
multivariate datasets are often treated as a stack of univariate time series in
terms of input preprocessing, with scaling methods applied across each channel
separately. In this evaluation, we aim to demonstrate that the additional
channel dimension is far from trivial and different approaches to scaling can
lead to significantly different results in the accuracy of a solution. To that
end, we test seven different data transformation methods on four different
temporal dimensions and study their effect on the classification accuracy of
five recent methods. We show that, for the large majority of tested datasets,
the best transformation-dimension configuration leads to an increase in the
accuracy compared to the result of each model with the same hyperparameters and
no scaling, ranging from 0.16 to 76.79 percentage points. We also show that if
we keep the transformation method constant, there is a statistically
significant difference in accuracy results when applying it across different
dimensions, with accuracy differences ranging from 0.23 to 47.79 percentage
points. Finally, we explore the relation of the transformation methods and
dimensions to the classifiers, and we conclude that there is no prominent
general trend, and the optimal configuration is dataset- and
classifier-specific.
| no_new_dataset | 0.709416 |
2210.10462 | Yaming Yang | Yaming Yang, Ziyu Guan, Zhe Wang, Wei Zhao, Cai Xu, Weigang Lu,
Jianbin Huang | Self-supervised Heterogeneous Graph Pre-training Based on Structural
Clustering | Accepted by NeurIPS 2022 | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recent self-supervised pre-training methods on Heterogeneous Information
Networks (HINs) have shown promising competitiveness over traditional
semi-supervised Heterogeneous Graph Neural Networks (HGNNs). Unfortunately,
their performance heavily depends on careful customization of various
strategies for generating high-quality positive examples and negative examples,
which notably limits their flexibility and generalization ability. In this
work, we present SHGP, a novel Self-supervised Heterogeneous Graph Pre-training
approach, which does not need to generate any positive examples or negative
examples. It consists of two modules that share the same attention-aggregation
scheme. In each iteration, the Att-LPA module produces pseudo-labels through
structural clustering, which serve as the self-supervision signals to guide the
Att-HGNN module to learn object embeddings and attention coefficients. The two
modules can effectively utilize and enhance each other, promoting the model to
learn discriminative embeddings. Extensive experiments on four real-world
datasets demonstrate the superior effectiveness of SHGP against
state-of-the-art unsupervised baselines and even semi-supervised baselines. We
release our source code at: https://github.com/kepsail/SHGP.
| [
{
"version": "v1",
"created": "Wed, 19 Oct 2022 10:55:48 GMT"
},
{
"version": "v2",
"created": "Wed, 12 Apr 2023 12:20:55 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Yang",
"Yaming",
""
],
[
"Guan",
"Ziyu",
""
],
[
"Wang",
"Zhe",
""
],
[
"Zhao",
"Wei",
""
],
[
"Xu",
"Cai",
""
],
[
"Lu",
"Weigang",
""
],
[
"Huang",
"Jianbin",
""
]
] | TITLE: Self-supervised Heterogeneous Graph Pre-training Based on Structural
Clustering
ABSTRACT: Recent self-supervised pre-training methods on Heterogeneous Information
Networks (HINs) have shown promising competitiveness over traditional
semi-supervised Heterogeneous Graph Neural Networks (HGNNs). Unfortunately,
their performance heavily depends on careful customization of various
strategies for generating high-quality positive examples and negative examples,
which notably limits their flexibility and generalization ability. In this
work, we present SHGP, a novel Self-supervised Heterogeneous Graph Pre-training
approach, which does not need to generate any positive examples or negative
examples. It consists of two modules that share the same attention-aggregation
scheme. In each iteration, the Att-LPA module produces pseudo-labels through
structural clustering, which serve as the self-supervision signals to guide the
Att-HGNN module to learn object embeddings and attention coefficients. The two
modules can effectively utilize and enhance each other, promoting the model to
learn discriminative embeddings. Extensive experiments on four real-world
datasets demonstrate the superior effectiveness of SHGP against
state-of-the-art unsupervised baselines and even semi-supervised baselines. We
release our source code at: https://github.com/kepsail/SHGP.
| no_new_dataset | 0.71039 |
2211.05853 | Subhabrata Majumdar | Harsh Raj, Domenic Rosati, Subhabrata Majumdar | Measuring Reliability of Large Language Models through Semantic
Consistency | NeurIPS 2022 ML Safety Workshop, https://neurips2022.mlsafety.org | null | null | null | cs.CL cs.AI cs.CY | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | While large pretrained language models (PLMs) demonstrate incredible fluency
and performance on many natural language tasks, recent work has shown that
well-performing PLMs are very sensitive to what prompts are feed into them.
Even when prompts are semantically identical, language models may give very
different answers. When considering safe and trustworthy deployments of PLMs we
would like their outputs to be consistent under prompts that mean the same
thing or convey the same intent. While some work has looked into how
state-of-the-art PLMs address this need, they have been limited to only
evaluating lexical equality of single- or multi-word answers and do not address
consistency of generative text sequences. In order to understand consistency of
PLMs under text generation settings, we develop a measure of semantic
consistency that allows the comparison of open-ended text outputs. We implement
several versions of this consistency metric to evaluate the performance of a
number of PLMs on paraphrased versions of questions in the TruthfulQA dataset,
we find that our proposed metrics are considerably more consistent than
traditional metrics embodying lexical consistency, and also correlate with
human evaluation of output consistency to a higher degree.
| [
{
"version": "v1",
"created": "Thu, 10 Nov 2022 20:21:07 GMT"
},
{
"version": "v2",
"created": "Tue, 11 Apr 2023 18:53:23 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Raj",
"Harsh",
""
],
[
"Rosati",
"Domenic",
""
],
[
"Majumdar",
"Subhabrata",
""
]
] | TITLE: Measuring Reliability of Large Language Models through Semantic
Consistency
ABSTRACT: While large pretrained language models (PLMs) demonstrate incredible fluency
and performance on many natural language tasks, recent work has shown that
well-performing PLMs are very sensitive to what prompts are feed into them.
Even when prompts are semantically identical, language models may give very
different answers. When considering safe and trustworthy deployments of PLMs we
would like their outputs to be consistent under prompts that mean the same
thing or convey the same intent. While some work has looked into how
state-of-the-art PLMs address this need, they have been limited to only
evaluating lexical equality of single- or multi-word answers and do not address
consistency of generative text sequences. In order to understand consistency of
PLMs under text generation settings, we develop a measure of semantic
consistency that allows the comparison of open-ended text outputs. We implement
several versions of this consistency metric to evaluate the performance of a
number of PLMs on paraphrased versions of questions in the TruthfulQA dataset,
we find that our proposed metrics are considerably more consistent than
traditional metrics embodying lexical consistency, and also correlate with
human evaluation of output consistency to a higher degree.
| no_new_dataset | 0.70133 |
2211.09224 | Alessandro Flaborea | Alessandro Flaborea, Bardh Prenkaj, Bharti Munjal, Marco Aurelio
Sterpa, Dario Aragona, Luca Podo, Fabio Galasso | Are we certain it's anomalous? | Accepted at CVPR '23 Visual Anomaly and Novelty Detection (VAND)
Workshop | null | null | null | cs.LG cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The progress in modelling time series and, more generally, sequences of
structured data has recently revamped research in anomaly detection. The task
stands for identifying abnormal behaviors in financial series, IT systems,
aerospace measurements, and the medical domain, where anomaly detection may aid
in isolating cases of depression and attend the elderly. Anomaly detection in
time series is a complex task since anomalies are rare due to highly non-linear
temporal correlations and since the definition of anomalous is sometimes
subjective. Here we propose the novel use of Hyperbolic uncertainty for Anomaly
Detection (HypAD). HypAD learns self-supervisedly to reconstruct the input
signal. We adopt best practices from the state-of-the-art to encode the
sequence by an LSTM, jointly learned with a decoder to reconstruct the signal,
with the aid of GAN critics. Uncertainty is estimated end-to-end by means of a
hyperbolic neural network. By using uncertainty, HypAD may assess whether it is
certain about the input signal but it fails to reconstruct it because this is
anomalous; or whether the reconstruction error does not necessarily imply
anomaly, as the model is uncertain, e.g. a complex but regular input signal.
The novel key idea is that a detectable anomaly is one where the model is
certain but it predicts wrongly. HypAD outperforms the current state-of-the-art
for univariate anomaly detection on established benchmarks based on data from
NASA, Yahoo, Numenta, Amazon, and Twitter. It also yields state-of-the-art
performance on a multivariate dataset of anomaly activities in elderly home
residences, and it outperforms the baseline on SWaT. Overall, HypAD yields the
lowest false alarms at the best performance rate, thanks to successfully
identifying detectable anomalies.
| [
{
"version": "v1",
"created": "Wed, 16 Nov 2022 21:31:39 GMT"
},
{
"version": "v2",
"created": "Tue, 21 Feb 2023 08:53:19 GMT"
},
{
"version": "v3",
"created": "Tue, 11 Apr 2023 12:06:22 GMT"
},
{
"version": "v4",
"created": "Wed, 12 Apr 2023 07:20:21 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Flaborea",
"Alessandro",
""
],
[
"Prenkaj",
"Bardh",
""
],
[
"Munjal",
"Bharti",
""
],
[
"Sterpa",
"Marco Aurelio",
""
],
[
"Aragona",
"Dario",
""
],
[
"Podo",
"Luca",
""
],
[
"Galasso",
"Fabio",
""
]
] | TITLE: Are we certain it's anomalous?
ABSTRACT: The progress in modelling time series and, more generally, sequences of
structured data has recently revamped research in anomaly detection. The task
stands for identifying abnormal behaviors in financial series, IT systems,
aerospace measurements, and the medical domain, where anomaly detection may aid
in isolating cases of depression and attend the elderly. Anomaly detection in
time series is a complex task since anomalies are rare due to highly non-linear
temporal correlations and since the definition of anomalous is sometimes
subjective. Here we propose the novel use of Hyperbolic uncertainty for Anomaly
Detection (HypAD). HypAD learns self-supervisedly to reconstruct the input
signal. We adopt best practices from the state-of-the-art to encode the
sequence by an LSTM, jointly learned with a decoder to reconstruct the signal,
with the aid of GAN critics. Uncertainty is estimated end-to-end by means of a
hyperbolic neural network. By using uncertainty, HypAD may assess whether it is
certain about the input signal but it fails to reconstruct it because this is
anomalous; or whether the reconstruction error does not necessarily imply
anomaly, as the model is uncertain, e.g. a complex but regular input signal.
The novel key idea is that a detectable anomaly is one where the model is
certain but it predicts wrongly. HypAD outperforms the current state-of-the-art
for univariate anomaly detection on established benchmarks based on data from
NASA, Yahoo, Numenta, Amazon, and Twitter. It also yields state-of-the-art
performance on a multivariate dataset of anomaly activities in elderly home
residences, and it outperforms the baseline on SWaT. Overall, HypAD yields the
lowest false alarms at the best performance rate, thanks to successfully
identifying detectable anomalies.
| no_new_dataset | 0.708389 |
2211.12005 | Sizhe Chen | Sizhe Chen, Geng Yuan, Xinwen Cheng, Yifan Gong, Minghai Qin, Yanzhi
Wang, Xiaolin Huang | Self-Ensemble Protection: Training Checkpoints Are Good Data Protectors | ICLR 2023 | null | null | null | cs.LG cs.CR stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | As data becomes increasingly vital, a company would be very cautious about
releasing data, because the competitors could use it to train high-performance
models, thereby posing a tremendous threat to the company's commercial
competence. To prevent training good models on the data, we could add
imperceptible perturbations to it. Since such perturbations aim at hurting the
entire training process, they should reflect the vulnerability of DNN training,
rather than that of a single model. Based on this new idea, we seek perturbed
examples that are always unrecognized (never correctly classified) in training.
In this paper, we uncover them by model checkpoints' gradients, forming the
proposed self-ensemble protection (SEP), which is very effective because (1)
learning on examples ignored during normal training tends to yield DNNs
ignoring normal examples; (2) checkpoints' cross-model gradients are close to
orthogonal, meaning that they are as diverse as DNNs with different
architectures. That is, our amazing performance of ensemble only requires the
computation of training one model. By extensive experiments with 9 baselines on
3 datasets and 5 architectures, SEP is verified to be a new state-of-the-art,
e.g., our small $\ell_\infty=2/255$ perturbations reduce the accuracy of a
CIFAR-10 ResNet18 from 94.56% to 14.68%, compared to 41.35% by the best-known
method. Code is available at https://github.com/Sizhe-Chen/SEP.
| [
{
"version": "v1",
"created": "Tue, 22 Nov 2022 04:54:20 GMT"
},
{
"version": "v2",
"created": "Tue, 28 Feb 2023 13:11:00 GMT"
},
{
"version": "v3",
"created": "Wed, 12 Apr 2023 11:04:04 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Chen",
"Sizhe",
""
],
[
"Yuan",
"Geng",
""
],
[
"Cheng",
"Xinwen",
""
],
[
"Gong",
"Yifan",
""
],
[
"Qin",
"Minghai",
""
],
[
"Wang",
"Yanzhi",
""
],
[
"Huang",
"Xiaolin",
""
]
] | TITLE: Self-Ensemble Protection: Training Checkpoints Are Good Data Protectors
ABSTRACT: As data becomes increasingly vital, a company would be very cautious about
releasing data, because the competitors could use it to train high-performance
models, thereby posing a tremendous threat to the company's commercial
competence. To prevent training good models on the data, we could add
imperceptible perturbations to it. Since such perturbations aim at hurting the
entire training process, they should reflect the vulnerability of DNN training,
rather than that of a single model. Based on this new idea, we seek perturbed
examples that are always unrecognized (never correctly classified) in training.
In this paper, we uncover them by model checkpoints' gradients, forming the
proposed self-ensemble protection (SEP), which is very effective because (1)
learning on examples ignored during normal training tends to yield DNNs
ignoring normal examples; (2) checkpoints' cross-model gradients are close to
orthogonal, meaning that they are as diverse as DNNs with different
architectures. That is, our amazing performance of ensemble only requires the
computation of training one model. By extensive experiments with 9 baselines on
3 datasets and 5 architectures, SEP is verified to be a new state-of-the-art,
e.g., our small $\ell_\infty=2/255$ perturbations reduce the accuracy of a
CIFAR-10 ResNet18 from 94.56% to 14.68%, compared to 41.35% by the best-known
method. Code is available at https://github.com/Sizhe-Chen/SEP.
| no_new_dataset | 0.708036 |
2211.15088 | Bingyuan Liu | Bingyuan Liu, J\'er\^ome Rony, Adrian Galdran, Jose Dolz, Ismail Ben
Ayed | Class Adaptive Network Calibration | CVPR 2023. Code: https://github.com/by-liu/CALS | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recent studies have revealed that, beyond conventional accuracy, calibration
should also be considered for training modern deep neural networks. To address
miscalibration during learning, some methods have explored different penalty
functions as part of the learning objective, alongside a standard
classification loss, with a hyper-parameter controlling the relative
contribution of each term. Nevertheless, these methods share two major
drawbacks: 1) the scalar balancing weight is the same for all classes,
hindering the ability to address different intrinsic difficulties or imbalance
among classes; and 2) the balancing weight is usually fixed without an adaptive
strategy, which may prevent from reaching the best compromise between accuracy
and calibration, and requires hyper-parameter search for each application. We
propose Class Adaptive Label Smoothing (CALS) for calibrating deep networks,
which allows to learn class-wise multipliers during training, yielding a
powerful alternative to common label smoothing penalties. Our method builds on
a general Augmented Lagrangian approach, a well-established technique in
constrained optimization, but we introduce several modifications to tailor it
for large-scale, class-adaptive training. Comprehensive evaluation and multiple
comparisons on a variety of benchmarks, including standard and long-tailed
image classification, semantic segmentation, and text classification,
demonstrate the superiority of the proposed method. The code is available at
https://github.com/by-liu/CALS.
| [
{
"version": "v1",
"created": "Mon, 28 Nov 2022 06:05:31 GMT"
},
{
"version": "v2",
"created": "Wed, 12 Apr 2023 06:51:01 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Liu",
"Bingyuan",
""
],
[
"Rony",
"Jérôme",
""
],
[
"Galdran",
"Adrian",
""
],
[
"Dolz",
"Jose",
""
],
[
"Ayed",
"Ismail Ben",
""
]
] | TITLE: Class Adaptive Network Calibration
ABSTRACT: Recent studies have revealed that, beyond conventional accuracy, calibration
should also be considered for training modern deep neural networks. To address
miscalibration during learning, some methods have explored different penalty
functions as part of the learning objective, alongside a standard
classification loss, with a hyper-parameter controlling the relative
contribution of each term. Nevertheless, these methods share two major
drawbacks: 1) the scalar balancing weight is the same for all classes,
hindering the ability to address different intrinsic difficulties or imbalance
among classes; and 2) the balancing weight is usually fixed without an adaptive
strategy, which may prevent from reaching the best compromise between accuracy
and calibration, and requires hyper-parameter search for each application. We
propose Class Adaptive Label Smoothing (CALS) for calibrating deep networks,
which allows to learn class-wise multipliers during training, yielding a
powerful alternative to common label smoothing penalties. Our method builds on
a general Augmented Lagrangian approach, a well-established technique in
constrained optimization, but we introduce several modifications to tailor it
for large-scale, class-adaptive training. Comprehensive evaluation and multiple
comparisons on a variety of benchmarks, including standard and long-tailed
image classification, semantic segmentation, and text classification,
demonstrate the superiority of the proposed method. The code is available at
https://github.com/by-liu/CALS.
| no_new_dataset | 0.709988 |
2212.05699 | Ziwang Zhao | Linmei Hu, Ziwang Zhao, Weijian Qi, Xuemeng Song, Liqiang Nie | Multimodal Matching-aware Co-attention Networks with Mutual Knowledge
Distillation for Fake News Detection | null | null | null | null | cs.IR cs.MM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Fake news often involves multimedia information such as text and image to
mislead readers, proliferating and expanding its influence. Most existing fake
news detection methods apply the co-attention mechanism to fuse multimodal
features while ignoring the consistency of image and text in co-attention. In
this paper, we propose multimodal matching-aware co-attention networks with
mutual knowledge distillation for improving fake news detection. Specifically,
we design an image-text matching-aware co-attention mechanism which captures
the alignment of image and text for better multimodal fusion. The image-text
matching representation can be obtained via a vision-language pre-trained
model. Additionally, based on the designed image-text matching-aware
co-attention mechanism, we propose to build two co-attention networks
respectively centered on text and image for mutual knowledge distillation to
improve fake news detection. Extensive experiments on three benchmark datasets
demonstrate that our proposed model achieves state-of-the-art performance on
multimodal fake news detection.
| [
{
"version": "v1",
"created": "Mon, 12 Dec 2022 04:36:06 GMT"
},
{
"version": "v2",
"created": "Wed, 12 Apr 2023 05:31:36 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Hu",
"Linmei",
""
],
[
"Zhao",
"Ziwang",
""
],
[
"Qi",
"Weijian",
""
],
[
"Song",
"Xuemeng",
""
],
[
"Nie",
"Liqiang",
""
]
] | TITLE: Multimodal Matching-aware Co-attention Networks with Mutual Knowledge
Distillation for Fake News Detection
ABSTRACT: Fake news often involves multimedia information such as text and image to
mislead readers, proliferating and expanding its influence. Most existing fake
news detection methods apply the co-attention mechanism to fuse multimodal
features while ignoring the consistency of image and text in co-attention. In
this paper, we propose multimodal matching-aware co-attention networks with
mutual knowledge distillation for improving fake news detection. Specifically,
we design an image-text matching-aware co-attention mechanism which captures
the alignment of image and text for better multimodal fusion. The image-text
matching representation can be obtained via a vision-language pre-trained
model. Additionally, based on the designed image-text matching-aware
co-attention mechanism, we propose to build two co-attention networks
respectively centered on text and image for mutual knowledge distillation to
improve fake news detection. Extensive experiments on three benchmark datasets
demonstrate that our proposed model achieves state-of-the-art performance on
multimodal fake news detection.
| no_new_dataset | 0.710804 |
2301.03410 | Xudong Lin | Andrew Lu, Xudong Lin, Yulei Niu, Shih-Fu Chang | In Defense of Structural Symbolic Representation for Video
Event-Relation Prediction | CVPRW 23, Learning with Limited Labelled Data | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Understanding event relationships in videos requires a model to understand
the underlying structures of events (i.e. the event type, the associated
argument roles, and corresponding entities) and factual knowledge for
reasoning. Structural symbolic representation (SSR) based methods directly take
event types and associated argument roles/entities as inputs to perform
reasoning. However, the state-of-the-art video event-relation prediction system
shows the necessity of using continuous feature vectors from input videos;
existing methods based solely on SSR inputs fail completely, even when given
oracle event types and argument roles. In this paper, we conduct an extensive
empirical analysis to answer the following questions: 1) why SSR-based method
failed; 2) how to understand the evaluation setting of video event relation
prediction properly; 3) how to uncover the potential of SSR-based methods. We
first identify suboptimal training settings as causing the failure of previous
SSR-based video event prediction models. Then through qualitative and
quantitative analysis, we show how evaluation that takes only video as inputs
is currently unfeasible, as well as the reliance on oracle event information to
obtain an accurate evaluation. Based on these findings, we propose to further
contextualize the SSR-based model to an Event-Sequence Model and equip it with
more factual knowledge through a simple yet effective way of reformulating
external visual commonsense knowledge bases into an event-relation prediction
pretraining dataset. The resultant new state-of-the-art model eventually
establishes a 25% Macro-accuracy performance boost.
| [
{
"version": "v1",
"created": "Fri, 6 Jan 2023 14:43:19 GMT"
},
{
"version": "v2",
"created": "Wed, 12 Apr 2023 15:19:16 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Lu",
"Andrew",
""
],
[
"Lin",
"Xudong",
""
],
[
"Niu",
"Yulei",
""
],
[
"Chang",
"Shih-Fu",
""
]
] | TITLE: In Defense of Structural Symbolic Representation for Video
Event-Relation Prediction
ABSTRACT: Understanding event relationships in videos requires a model to understand
the underlying structures of events (i.e. the event type, the associated
argument roles, and corresponding entities) and factual knowledge for
reasoning. Structural symbolic representation (SSR) based methods directly take
event types and associated argument roles/entities as inputs to perform
reasoning. However, the state-of-the-art video event-relation prediction system
shows the necessity of using continuous feature vectors from input videos;
existing methods based solely on SSR inputs fail completely, even when given
oracle event types and argument roles. In this paper, we conduct an extensive
empirical analysis to answer the following questions: 1) why SSR-based method
failed; 2) how to understand the evaluation setting of video event relation
prediction properly; 3) how to uncover the potential of SSR-based methods. We
first identify suboptimal training settings as causing the failure of previous
SSR-based video event prediction models. Then through qualitative and
quantitative analysis, we show how evaluation that takes only video as inputs
is currently unfeasible, as well as the reliance on oracle event information to
obtain an accurate evaluation. Based on these findings, we propose to further
contextualize the SSR-based model to an Event-Sequence Model and equip it with
more factual knowledge through a simple yet effective way of reformulating
external visual commonsense knowledge bases into an event-relation prediction
pretraining dataset. The resultant new state-of-the-art model eventually
establishes a 25% Macro-accuracy performance boost.
| no_new_dataset | 0.71227 |
2301.04122 | Mingyu Liang | Mingyu Liang, Wenyin Fu, Louis Feng, Zhongyi Lin, Pavani Panakanti,
Shengbao Zheng, Srinivas Sridharan, Christina Delimitrou | Mystique: Enabling Accurate and Scalable Generation of Production AI
Benchmarks | Accepted to ISCA 2023 | null | null | null | cs.DC cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Building large AI fleets to support the rapidly growing DL workloads is an
active research topic for modern cloud providers. Generating accurate
benchmarks plays an essential role in designing the fast-paced software and
hardware solutions in this space. Two fundamental challenges to make this
scalable are (i) workload representativeness and (ii) the ability to quickly
incorporate changes to the fleet into the benchmarks.
To overcome these issues, we propose Mystique, an accurate and scalable
framework for production AI benchmark generation. It leverages the PyTorch
execution trace (ET), a new feature that captures the runtime information of AI
models at the granularity of operators, in a graph format, together with their
metadata. By sourcing fleet ETs, we can build AI benchmarks that are portable
and representative. Mystique is scalable, due to its lightweight data
collection, in terms of runtime overhead and instrumentation effort. It is also
adaptive because ET composability allows flexible control on benchmark
creation.
We evaluate our methodology on several production AI models, and show that
benchmarks generated with Mystique closely resemble original AI models, both in
execution time and system-level metrics. We also showcase the portability of
the generated benchmarks across platforms, and demonstrate several use cases
enabled by the fine-grained composability of the execution trace.
| [
{
"version": "v1",
"created": "Fri, 16 Dec 2022 18:46:37 GMT"
},
{
"version": "v2",
"created": "Fri, 3 Mar 2023 21:08:50 GMT"
},
{
"version": "v3",
"created": "Tue, 11 Apr 2023 18:16:38 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Liang",
"Mingyu",
""
],
[
"Fu",
"Wenyin",
""
],
[
"Feng",
"Louis",
""
],
[
"Lin",
"Zhongyi",
""
],
[
"Panakanti",
"Pavani",
""
],
[
"Zheng",
"Shengbao",
""
],
[
"Sridharan",
"Srinivas",
""
],
[
"Delimitrou",
"Christina",
""
]
] | TITLE: Mystique: Enabling Accurate and Scalable Generation of Production AI
Benchmarks
ABSTRACT: Building large AI fleets to support the rapidly growing DL workloads is an
active research topic for modern cloud providers. Generating accurate
benchmarks plays an essential role in designing the fast-paced software and
hardware solutions in this space. Two fundamental challenges to make this
scalable are (i) workload representativeness and (ii) the ability to quickly
incorporate changes to the fleet into the benchmarks.
To overcome these issues, we propose Mystique, an accurate and scalable
framework for production AI benchmark generation. It leverages the PyTorch
execution trace (ET), a new feature that captures the runtime information of AI
models at the granularity of operators, in a graph format, together with their
metadata. By sourcing fleet ETs, we can build AI benchmarks that are portable
and representative. Mystique is scalable, due to its lightweight data
collection, in terms of runtime overhead and instrumentation effort. It is also
adaptive because ET composability allows flexible control on benchmark
creation.
We evaluate our methodology on several production AI models, and show that
benchmarks generated with Mystique closely resemble original AI models, both in
execution time and system-level metrics. We also showcase the portability of
the generated benchmarks across platforms, and demonstrate several use cases
enabled by the fine-grained composability of the execution trace.
| no_new_dataset | 0.707657 |
2301.13112 | Fei Lu | Zehong Zhang, Fei Lu, Esther Xu Fei, Terry Lyons, Yannis Kevrekidis,
and Tom Woolf | Benchmarking optimality of time series classification methods in
distinguishing diffusions | 23 pages, 8 figures | null | null | null | stat.ML cs.LG | http://creativecommons.org/licenses/by/4.0/ | Statistical optimality benchmarking is crucial for analyzing and designing
time series classification (TSC) algorithms. This study proposes to benchmark
the optimality of TSC algorithms in distinguishing diffusion processes by the
likelihood ratio test (LRT). The LRT is an optimal classifier by the
Neyman-Pearson lemma. The LRT benchmarks are computationally efficient because
the LRT does not need training, and the diffusion processes can be efficiently
simulated and are flexible to reflect the specific features of real-world
applications. We demonstrate the benchmarking with three widely-used TSC
algorithms: random forest, ResNet, and ROCKET. These algorithms can achieve the
LRT optimality for univariate time series and multivariate Gaussian processes.
However, these model-agnostic algorithms are suboptimal in classifying
high-dimensional nonlinear multivariate time series. Additionally, the LRT
benchmark provides tools to analyze the dependence of classification accuracy
on the time length, dimension, temporal sampling frequency, and randomness of
the time series.
| [
{
"version": "v1",
"created": "Mon, 30 Jan 2023 17:49:12 GMT"
},
{
"version": "v2",
"created": "Sun, 5 Feb 2023 00:53:07 GMT"
},
{
"version": "v3",
"created": "Wed, 12 Apr 2023 03:49:48 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Zhang",
"Zehong",
""
],
[
"Lu",
"Fei",
""
],
[
"Fei",
"Esther Xu",
""
],
[
"Lyons",
"Terry",
""
],
[
"Kevrekidis",
"Yannis",
""
],
[
"Woolf",
"Tom",
""
]
] | TITLE: Benchmarking optimality of time series classification methods in
distinguishing diffusions
ABSTRACT: Statistical optimality benchmarking is crucial for analyzing and designing
time series classification (TSC) algorithms. This study proposes to benchmark
the optimality of TSC algorithms in distinguishing diffusion processes by the
likelihood ratio test (LRT). The LRT is an optimal classifier by the
Neyman-Pearson lemma. The LRT benchmarks are computationally efficient because
the LRT does not need training, and the diffusion processes can be efficiently
simulated and are flexible to reflect the specific features of real-world
applications. We demonstrate the benchmarking with three widely-used TSC
algorithms: random forest, ResNet, and ROCKET. These algorithms can achieve the
LRT optimality for univariate time series and multivariate Gaussian processes.
However, these model-agnostic algorithms are suboptimal in classifying
high-dimensional nonlinear multivariate time series. Additionally, the LRT
benchmark provides tools to analyze the dependence of classification accuracy
on the time length, dimension, temporal sampling frequency, and randomness of
the time series.
| no_new_dataset | 0.709007 |
2302.05991 | Zhihao Zhao | Weiyu Feng, Seth Z. Zhao, Chuanyu Pan, Adam Chang, Yichen Chen, Zekun
Wang, Allen Y. Yang | Digital Twin Tracking Dataset (DTTD): A New RGB+Depth 3D Dataset for
Longer-Range Object Tracking Applications | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by-sa/4.0/ | Digital twin is a problem of augmenting real objects with their digital
counterparts. It can underpin a wide range of applications in augmented reality
(AR), autonomy, and UI/UX. A critical component in a good digital-twin system
is real-time, accurate 3D object tracking. Most existing works solve 3D object
tracking through the lens of robotic grasping, employ older generations of
depth sensors, and measure performance metrics that may not apply to other
digital-twin applications such as in AR. In this work, we create a novel RGB-D
dataset, called Digital Twin Tracking Dataset (DTTD), to enable further
research of the problem and extend potential solutions towards longer ranges
and mm localization accuracy. To reduce point cloud noise from the input
source, we select the latest Microsoft Azure Kinect as the state-of-the-art
time-of-flight (ToF) camera. In total, 103 scenes of 10 common off-the-shelf
objects with rich textures are recorded, with each frame annotated with a
per-pixel semantic segmentation and ground-truth object poses provided by a
commercial motion capturing system. Through extensive experiments with
model-level and dataset-level analysis, we demonstrate that DTTD can help
researchers develop future object tracking methods and analyze new challenges.
The dataset, data generation, annotation, and model evaluation pipeline are
made publicly available as open source code at:
https://github.com/augcog/DTTDv1.
| [
{
"version": "v1",
"created": "Sun, 12 Feb 2023 20:06:07 GMT"
},
{
"version": "v2",
"created": "Tue, 11 Apr 2023 20:31:38 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Feng",
"Weiyu",
""
],
[
"Zhao",
"Seth Z.",
""
],
[
"Pan",
"Chuanyu",
""
],
[
"Chang",
"Adam",
""
],
[
"Chen",
"Yichen",
""
],
[
"Wang",
"Zekun",
""
],
[
"Yang",
"Allen Y.",
""
]
] | TITLE: Digital Twin Tracking Dataset (DTTD): A New RGB+Depth 3D Dataset for
Longer-Range Object Tracking Applications
ABSTRACT: Digital twin is a problem of augmenting real objects with their digital
counterparts. It can underpin a wide range of applications in augmented reality
(AR), autonomy, and UI/UX. A critical component in a good digital-twin system
is real-time, accurate 3D object tracking. Most existing works solve 3D object
tracking through the lens of robotic grasping, employ older generations of
depth sensors, and measure performance metrics that may not apply to other
digital-twin applications such as in AR. In this work, we create a novel RGB-D
dataset, called Digital Twin Tracking Dataset (DTTD), to enable further
research of the problem and extend potential solutions towards longer ranges
and mm localization accuracy. To reduce point cloud noise from the input
source, we select the latest Microsoft Azure Kinect as the state-of-the-art
time-of-flight (ToF) camera. In total, 103 scenes of 10 common off-the-shelf
objects with rich textures are recorded, with each frame annotated with a
per-pixel semantic segmentation and ground-truth object poses provided by a
commercial motion capturing system. Through extensive experiments with
model-level and dataset-level analysis, we demonstrate that DTTD can help
researchers develop future object tracking methods and analyze new challenges.
The dataset, data generation, annotation, and model evaluation pipeline are
made publicly available as open source code at:
https://github.com/augcog/DTTDv1.
| new_dataset | 0.718832 |
2302.08788 | Seunghyeon Seo | Seunghyeon Seo, Donghoon Han, Yeonjin Chang, Nojun Kwak | MixNeRF: Modeling a Ray with Mixture Density for Novel View Synthesis
from Sparse Inputs | CVPR 2023. Project Page: https://shawn615.github.io/mixnerf/ | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Neural Radiance Field (NeRF) has broken new ground in the novel view
synthesis due to its simple concept and state-of-the-art quality. However, it
suffers from severe performance degradation unless trained with a dense set of
images with different camera poses, which hinders its practical applications.
Although previous methods addressing this problem achieved promising results,
they relied heavily on the additional training resources, which goes against
the philosophy of sparse-input novel-view synthesis pursuing the training
efficiency. In this work, we propose MixNeRF, an effective training strategy
for novel view synthesis from sparse inputs by modeling a ray with a mixture
density model. Our MixNeRF estimates the joint distribution of RGB colors along
the ray samples by modeling it with mixture of distributions. We also propose a
new task of ray depth estimation as a useful training objective, which is
highly correlated with 3D scene geometry. Moreover, we remodel the colors with
regenerated blending weights based on the estimated ray depth and further
improves the robustness for colors and viewpoints. Our MixNeRF outperforms
other state-of-the-art methods in various standard benchmarks with superior
efficiency of training and inference.
| [
{
"version": "v1",
"created": "Fri, 17 Feb 2023 10:07:35 GMT"
},
{
"version": "v2",
"created": "Wed, 12 Apr 2023 05:15:11 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Seo",
"Seunghyeon",
""
],
[
"Han",
"Donghoon",
""
],
[
"Chang",
"Yeonjin",
""
],
[
"Kwak",
"Nojun",
""
]
] | TITLE: MixNeRF: Modeling a Ray with Mixture Density for Novel View Synthesis
from Sparse Inputs
ABSTRACT: Neural Radiance Field (NeRF) has broken new ground in the novel view
synthesis due to its simple concept and state-of-the-art quality. However, it
suffers from severe performance degradation unless trained with a dense set of
images with different camera poses, which hinders its practical applications.
Although previous methods addressing this problem achieved promising results,
they relied heavily on the additional training resources, which goes against
the philosophy of sparse-input novel-view synthesis pursuing the training
efficiency. In this work, we propose MixNeRF, an effective training strategy
for novel view synthesis from sparse inputs by modeling a ray with a mixture
density model. Our MixNeRF estimates the joint distribution of RGB colors along
the ray samples by modeling it with mixture of distributions. We also propose a
new task of ray depth estimation as a useful training objective, which is
highly correlated with 3D scene geometry. Moreover, we remodel the colors with
regenerated blending weights based on the estimated ray depth and further
improves the robustness for colors and viewpoints. Our MixNeRF outperforms
other state-of-the-art methods in various standard benchmarks with superior
efficiency of training and inference.
| no_new_dataset | 0.709837 |
2302.09755 | YoonHyuk Choi | Yoonhyuk Choi, Jiho Choi, Taewook Ko, Chong-Kwon Kim | Finding Heterophilic Neighbors via Confidence-based Subgraph Matching
for Semi-supervised Node Classification | Proceedings of the 31st ACM International Conference on Information &
Knowledge Management | null | 10.1145/3511808.3557324 | null | cs.LG cs.SI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Graph Neural Networks (GNNs) have proven to be powerful in many graph-based
applications. However, they fail to generalize well under heterophilic setups,
where neighbor nodes have different labels. To address this challenge, we
employ a confidence ratio as a hyper-parameter, assuming that some of the edges
are disassortative (heterophilic). Here, we propose a two-phased algorithm.
Firstly, we determine edge coefficients through subgraph matching using a
supplementary module. Then, we apply GNNs with a modified label propagation
mechanism to utilize the edge coefficients effectively. Specifically, our
supplementary module identifies a certain proportion of task-irrelevant edges
based on a given confidence ratio. Using the remaining edges, we employ the
widely used optimal transport to measure the similarity between two nodes with
their subgraphs. Finally, using the coefficients as supplementary information
on GNNs, we improve the label propagation mechanism which can prevent two nodes
with smaller weights from being closer. The experiments on benchmark datasets
show that our model alleviates over-smoothing and improves performance.
| [
{
"version": "v1",
"created": "Mon, 20 Feb 2023 04:23:34 GMT"
},
{
"version": "v2",
"created": "Wed, 12 Apr 2023 08:57:05 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Choi",
"Yoonhyuk",
""
],
[
"Choi",
"Jiho",
""
],
[
"Ko",
"Taewook",
""
],
[
"Kim",
"Chong-Kwon",
""
]
] | TITLE: Finding Heterophilic Neighbors via Confidence-based Subgraph Matching
for Semi-supervised Node Classification
ABSTRACT: Graph Neural Networks (GNNs) have proven to be powerful in many graph-based
applications. However, they fail to generalize well under heterophilic setups,
where neighbor nodes have different labels. To address this challenge, we
employ a confidence ratio as a hyper-parameter, assuming that some of the edges
are disassortative (heterophilic). Here, we propose a two-phased algorithm.
Firstly, we determine edge coefficients through subgraph matching using a
supplementary module. Then, we apply GNNs with a modified label propagation
mechanism to utilize the edge coefficients effectively. Specifically, our
supplementary module identifies a certain proportion of task-irrelevant edges
based on a given confidence ratio. Using the remaining edges, we employ the
widely used optimal transport to measure the similarity between two nodes with
their subgraphs. Finally, using the coefficients as supplementary information
on GNNs, we improve the label propagation mechanism which can prevent two nodes
with smaller weights from being closer. The experiments on benchmark datasets
show that our model alleviates over-smoothing and improves performance.
| no_new_dataset | 0.709435 |
2302.11044 | Edith Cohen | Edith Cohen and Xin Lyu | The Target-Charging Technique for Privacy Accounting across Interactive
Computations | null | null | null | null | cs.DS | http://creativecommons.org/licenses/by/4.0/ | We propose the \emph{Target Charging Technique} (TCT), a unified privacy
analysis framework for interactive settings where a sensitive dataset is
accessed multiple times using differentially private algorithms. Unlike
traditional composition, where privacy guarantees deteriorate quickly with the
number of accesses, TCT allows computations that don't hit a specified
\emph{target}, often the vast majority, to be essentially free (while incurring
instead a small overhead on those that do hit their targets). TCT generalizes
tools such as the sparse vector technique and top-$k$ selection from private
candidates and extends their remarkable privacy enhancement benefits from noisy
Lipschitz functions to general private algorithms.
| [
{
"version": "v1",
"created": "Tue, 21 Feb 2023 22:57:14 GMT"
},
{
"version": "v2",
"created": "Tue, 11 Apr 2023 23:59:07 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Cohen",
"Edith",
""
],
[
"Lyu",
"Xin",
""
]
] | TITLE: The Target-Charging Technique for Privacy Accounting across Interactive
Computations
ABSTRACT: We propose the \emph{Target Charging Technique} (TCT), a unified privacy
analysis framework for interactive settings where a sensitive dataset is
accessed multiple times using differentially private algorithms. Unlike
traditional composition, where privacy guarantees deteriorate quickly with the
number of accesses, TCT allows computations that don't hit a specified
\emph{target}, often the vast majority, to be essentially free (while incurring
instead a small overhead on those that do hit their targets). TCT generalizes
tools such as the sparse vector technique and top-$k$ selection from private
candidates and extends their remarkable privacy enhancement benefits from noisy
Lipschitz functions to general private algorithms.
| no_new_dataset | 0.712188 |
2302.13610 | Daichi Guo | Guanting Dong and Zechen Wang and Liwen Wang and Daichi Guo and Dayuan
Fu and Yuxiang Wu and Chen Zeng and Xuefeng Li and Tingfeng Hui and Keqing He
and Xinyue Cui and Qixiang Gao and Weiran Xu | A Prototypical Semantic Decoupling Method via Joint Contrastive Learning
for Few-Shot Name Entity Recognition | we want to revise our paper and upload this article in few days | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Few-shot named entity recognition (NER) aims at identifying named entities
based on only few labeled instances. Most existing prototype-based sequence
labeling models tend to memorize entity mentions which would be easily confused
by close prototypes. In this paper, we proposed a Prototypical Semantic
Decoupling method via joint Contrastive learning (PSDC) for few-shot NER.
Specifically, we decouple class-specific prototypes and contextual semantic
prototypes by two masking strategies to lead the model to focus on two
different semantic information for inference. Besides, we further introduce
joint contrastive learning objectives to better integrate two kinds of
decoupling information and prevent semantic collapse. Experimental results on
two few-shot NER benchmarks demonstrate that PSDC consistently outperforms the
previous SOTA methods in terms of overall performance. Extensive analysis
further validates the effectiveness and generalization of PSDC.
| [
{
"version": "v1",
"created": "Mon, 27 Feb 2023 09:20:00 GMT"
},
{
"version": "v2",
"created": "Wed, 12 Apr 2023 06:10:11 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Dong",
"Guanting",
""
],
[
"Wang",
"Zechen",
""
],
[
"Wang",
"Liwen",
""
],
[
"Guo",
"Daichi",
""
],
[
"Fu",
"Dayuan",
""
],
[
"Wu",
"Yuxiang",
""
],
[
"Zeng",
"Chen",
""
],
[
"Li",
"Xuefeng",
""
],
[
"Hui",
"Tingfeng",
""
],
[
"He",
"Keqing",
""
],
[
"Cui",
"Xinyue",
""
],
[
"Gao",
"Qixiang",
""
],
[
"Xu",
"Weiran",
""
]
] | TITLE: A Prototypical Semantic Decoupling Method via Joint Contrastive Learning
for Few-Shot Name Entity Recognition
ABSTRACT: Few-shot named entity recognition (NER) aims at identifying named entities
based on only few labeled instances. Most existing prototype-based sequence
labeling models tend to memorize entity mentions which would be easily confused
by close prototypes. In this paper, we proposed a Prototypical Semantic
Decoupling method via joint Contrastive learning (PSDC) for few-shot NER.
Specifically, we decouple class-specific prototypes and contextual semantic
prototypes by two masking strategies to lead the model to focus on two
different semantic information for inference. Besides, we further introduce
joint contrastive learning objectives to better integrate two kinds of
decoupling information and prevent semantic collapse. Experimental results on
two few-shot NER benchmarks demonstrate that PSDC consistently outperforms the
previous SOTA methods in terms of overall performance. Extensive analysis
further validates the effectiveness and generalization of PSDC.
| no_new_dataset | 0.710597 |
2303.05725 | Jiangbin Zheng | Jiangbin Zheng, Yile Wang, Cheng Tan, Siyuan Li, Ge Wang, Jun Xia,
Yidong Chen, Stan Z. Li | CVT-SLR: Contrastive Visual-Textual Transformation for Sign Language
Recognition with Variational Alignment | Accepted to CVPR 2023 (Highlight paper, 2.5% acceptance rate); Open
source | null | null | null | cs.CV cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Sign language recognition (SLR) is a weakly supervised task that annotates
sign videos as textual glosses. Recent studies show that insufficient training
caused by the lack of large-scale available sign datasets becomes the main
bottleneck for SLR. Most SLR works thereby adopt pretrained visual modules and
develop two mainstream solutions. The multi-stream architectures extend
multi-cue visual features, yielding the current SOTA performances but requiring
complex designs and might introduce potential noise. Alternatively, the
advanced single-cue SLR frameworks using explicit cross-modal alignment between
visual and textual modalities are simple and effective, potentially competitive
with the multi-cue framework. In this work, we propose a novel contrastive
visual-textual transformation for SLR, CVT-SLR, to fully explore the pretrained
knowledge of both the visual and language modalities. Based on the single-cue
cross-modal alignment framework, we propose a variational autoencoder (VAE) for
pretrained contextual knowledge while introducing the complete pretrained
language module. The VAE implicitly aligns visual and textual modalities while
benefiting from pretrained contextual knowledge as the traditional contextual
module. Meanwhile, a contrastive cross-modal alignment algorithm is designed to
explicitly enhance the consistency constraints. Extensive experiments on public
datasets (PHOENIX-2014 and PHOENIX-2014T) demonstrate that our proposed CVT-SLR
consistently outperforms existing single-cue methods and even outperforms SOTA
multi-cue methods.
| [
{
"version": "v1",
"created": "Fri, 10 Mar 2023 06:12:36 GMT"
},
{
"version": "v2",
"created": "Tue, 21 Mar 2023 13:28:49 GMT"
},
{
"version": "v3",
"created": "Thu, 23 Mar 2023 12:00:33 GMT"
},
{
"version": "v4",
"created": "Wed, 12 Apr 2023 10:07:11 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Zheng",
"Jiangbin",
""
],
[
"Wang",
"Yile",
""
],
[
"Tan",
"Cheng",
""
],
[
"Li",
"Siyuan",
""
],
[
"Wang",
"Ge",
""
],
[
"Xia",
"Jun",
""
],
[
"Chen",
"Yidong",
""
],
[
"Li",
"Stan Z.",
""
]
] | TITLE: CVT-SLR: Contrastive Visual-Textual Transformation for Sign Language
Recognition with Variational Alignment
ABSTRACT: Sign language recognition (SLR) is a weakly supervised task that annotates
sign videos as textual glosses. Recent studies show that insufficient training
caused by the lack of large-scale available sign datasets becomes the main
bottleneck for SLR. Most SLR works thereby adopt pretrained visual modules and
develop two mainstream solutions. The multi-stream architectures extend
multi-cue visual features, yielding the current SOTA performances but requiring
complex designs and might introduce potential noise. Alternatively, the
advanced single-cue SLR frameworks using explicit cross-modal alignment between
visual and textual modalities are simple and effective, potentially competitive
with the multi-cue framework. In this work, we propose a novel contrastive
visual-textual transformation for SLR, CVT-SLR, to fully explore the pretrained
knowledge of both the visual and language modalities. Based on the single-cue
cross-modal alignment framework, we propose a variational autoencoder (VAE) for
pretrained contextual knowledge while introducing the complete pretrained
language module. The VAE implicitly aligns visual and textual modalities while
benefiting from pretrained contextual knowledge as the traditional contextual
module. Meanwhile, a contrastive cross-modal alignment algorithm is designed to
explicitly enhance the consistency constraints. Extensive experiments on public
datasets (PHOENIX-2014 and PHOENIX-2014T) demonstrate that our proposed CVT-SLR
consistently outperforms existing single-cue methods and even outperforms SOTA
multi-cue methods.
| no_new_dataset | 0.711036 |
2303.11436 | Manmeet Singh | Sifatkaur Dhingra, Manmeet Singh, Vaisakh SB, Neetiraj Malviya,
Sukhpal Singh Gill | Mind meets machine: Unravelling GPT-4's cognitive psychology | null | null | null | null | cs.CL cs.AI | http://creativecommons.org/licenses/by/4.0/ | Cognitive psychology delves on understanding perception, attention, memory,
language, problem-solving, decision-making, and reasoning. Large language
models (LLMs) are emerging as potent tools increasingly capable of performing
human-level tasks. The recent development in the form of GPT-4 and its
demonstrated success in tasks complex to humans exam and complex problems has
led to an increased confidence in the LLMs to become perfect instruments of
intelligence. Although GPT-4 report has shown performance on some cognitive
psychology tasks, a comprehensive assessment of GPT-4, via the existing
well-established datasets is required. In this study, we focus on the
evaluation of GPT-4's performance on a set of cognitive psychology datasets
such as CommonsenseQA, SuperGLUE, MATH and HANS. In doing so, we understand how
GPT-4 processes and integrates cognitive psychology with contextual
information, providing insight into the underlying cognitive processes that
enable its ability to generate the responses. We show that GPT-4 exhibits a
high level of accuracy in cognitive psychology tasks relative to the prior
state-of-the-art models. Our results strengthen the already available
assessments and confidence on GPT-4's cognitive psychology abilities. It has
significant potential to revolutionize the field of AI, by enabling machines to
bridge the gap between human and machine reasoning.
| [
{
"version": "v1",
"created": "Mon, 20 Mar 2023 20:28:26 GMT"
},
{
"version": "v2",
"created": "Wed, 12 Apr 2023 15:46:20 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Dhingra",
"Sifatkaur",
""
],
[
"Singh",
"Manmeet",
""
],
[
"SB",
"Vaisakh",
""
],
[
"Malviya",
"Neetiraj",
""
],
[
"Gill",
"Sukhpal Singh",
""
]
] | TITLE: Mind meets machine: Unravelling GPT-4's cognitive psychology
ABSTRACT: Cognitive psychology delves on understanding perception, attention, memory,
language, problem-solving, decision-making, and reasoning. Large language
models (LLMs) are emerging as potent tools increasingly capable of performing
human-level tasks. The recent development in the form of GPT-4 and its
demonstrated success in tasks complex to humans exam and complex problems has
led to an increased confidence in the LLMs to become perfect instruments of
intelligence. Although GPT-4 report has shown performance on some cognitive
psychology tasks, a comprehensive assessment of GPT-4, via the existing
well-established datasets is required. In this study, we focus on the
evaluation of GPT-4's performance on a set of cognitive psychology datasets
such as CommonsenseQA, SuperGLUE, MATH and HANS. In doing so, we understand how
GPT-4 processes and integrates cognitive psychology with contextual
information, providing insight into the underlying cognitive processes that
enable its ability to generate the responses. We show that GPT-4 exhibits a
high level of accuracy in cognitive psychology tasks relative to the prior
state-of-the-art models. Our results strengthen the already available
assessments and confidence on GPT-4's cognitive psychology abilities. It has
significant potential to revolutionize the field of AI, by enabling machines to
bridge the gap between human and machine reasoning.
| no_new_dataset | 0.708023 |
2303.13375 | Harsha Nori | Harsha Nori, Nicholas King, Scott Mayer McKinney, Dean Carignan, Eric
Horvitz | Capabilities of GPT-4 on Medical Challenge Problems | 35 pages, 15 figures; added GPT-4-base model results and discussion | null | null | null | cs.CL cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Large language models (LLMs) have demonstrated remarkable capabilities in
natural language understanding and generation across various domains, including
medicine. We present a comprehensive evaluation of GPT-4, a state-of-the-art
LLM, on medical competency examinations and benchmark datasets. GPT-4 is a
general-purpose model that is not specialized for medical problems through
training or engineered to solve clinical tasks. Our analysis covers two sets of
official practice materials for the USMLE, a three-step examination program
used to assess clinical competency and grant licensure in the United States. We
also evaluate performance on the MultiMedQA suite of benchmark datasets. Beyond
measuring model performance, experiments were conducted to investigate the
influence of test questions containing both text and images on model
performance, probe for memorization of content during training, and study
probability calibration, which is of critical importance in high-stakes
applications like medicine. Our results show that GPT-4, without any
specialized prompt crafting, exceeds the passing score on USMLE by over 20
points and outperforms earlier general-purpose models (GPT-3.5) as well as
models specifically fine-tuned on medical knowledge (Med-PaLM, a prompt-tuned
version of Flan-PaLM 540B). In addition, GPT-4 is significantly better
calibrated than GPT-3.5, demonstrating a much-improved ability to predict the
likelihood that its answers are correct. We also explore the behavior of the
model qualitatively through a case study that shows the ability of GPT-4 to
explain medical reasoning, personalize explanations to students, and
interactively craft new counterfactual scenarios around a medical case.
Implications of the findings are discussed for potential uses of GPT-4 in
medical education, assessment, and clinical practice, with appropriate
attention to challenges of accuracy and safety.
| [
{
"version": "v1",
"created": "Mon, 20 Mar 2023 16:18:38 GMT"
},
{
"version": "v2",
"created": "Wed, 12 Apr 2023 16:48:39 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Nori",
"Harsha",
""
],
[
"King",
"Nicholas",
""
],
[
"McKinney",
"Scott Mayer",
""
],
[
"Carignan",
"Dean",
""
],
[
"Horvitz",
"Eric",
""
]
] | TITLE: Capabilities of GPT-4 on Medical Challenge Problems
ABSTRACT: Large language models (LLMs) have demonstrated remarkable capabilities in
natural language understanding and generation across various domains, including
medicine. We present a comprehensive evaluation of GPT-4, a state-of-the-art
LLM, on medical competency examinations and benchmark datasets. GPT-4 is a
general-purpose model that is not specialized for medical problems through
training or engineered to solve clinical tasks. Our analysis covers two sets of
official practice materials for the USMLE, a three-step examination program
used to assess clinical competency and grant licensure in the United States. We
also evaluate performance on the MultiMedQA suite of benchmark datasets. Beyond
measuring model performance, experiments were conducted to investigate the
influence of test questions containing both text and images on model
performance, probe for memorization of content during training, and study
probability calibration, which is of critical importance in high-stakes
applications like medicine. Our results show that GPT-4, without any
specialized prompt crafting, exceeds the passing score on USMLE by over 20
points and outperforms earlier general-purpose models (GPT-3.5) as well as
models specifically fine-tuned on medical knowledge (Med-PaLM, a prompt-tuned
version of Flan-PaLM 540B). In addition, GPT-4 is significantly better
calibrated than GPT-3.5, demonstrating a much-improved ability to predict the
likelihood that its answers are correct. We also explore the behavior of the
model qualitatively through a case study that shows the ability of GPT-4 to
explain medical reasoning, personalize explanations to students, and
interactively craft new counterfactual scenarios around a medical case.
Implications of the findings are discussed for potential uses of GPT-4 in
medical education, assessment, and clinical practice, with appropriate
attention to challenges of accuracy and safety.
| no_new_dataset | 0.706437 |
2303.16869 | Sandipp Krishnan Ravi | Lele Luan, Nesar Ramachandra, Sandipp Krishnan Ravi, Anindya Bhaduri,
Piyush Pandita, Prasanna Balaprakash, Mihai Anitescu, Changjie Sun, Liping
Wang | Application of probabilistic modeling and automated machine learning
framework for high-dimensional stress field | 17 pages, 16 figures, IDETC Conference Submission | null | null | null | cs.CE cs.LG cs.NA math.NA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Modern computational methods, involving highly sophisticated mathematical
formulations, enable several tasks like modeling complex physical phenomenon,
predicting key properties and design optimization. The higher fidelity in these
computer models makes it computationally intensive to query them hundreds of
times for optimization and one usually relies on a simplified model albeit at
the cost of losing predictive accuracy and precision. Towards this, data-driven
surrogate modeling methods have shown a lot of promise in emulating the
behavior of the expensive computer models. However, a major bottleneck in such
methods is the inability to deal with high input dimensionality and the need
for relatively large datasets. With such problems, the input and output
quantity of interest are tensors of high dimensionality. Commonly used
surrogate modeling methods for such problems, suffer from requirements like
high number of computational evaluations that precludes one from performing
other numerical tasks like uncertainty quantification and statistical analysis.
In this work, we propose an end-to-end approach that maps a high-dimensional
image like input to an output of high dimensionality or its key statistics. Our
approach uses two main framework that perform three steps: a) reduce the input
and output from a high-dimensional space to a reduced or low-dimensional space,
b) model the input-output relationship in the low-dimensional space, and c)
enable the incorporation of domain-specific physical constraints as masks. In
order to accomplish the task of reducing input dimensionality we leverage
principal component analysis, that is coupled with two surrogate modeling
methods namely: a) Bayesian hybrid modeling, and b) DeepHyper's deep neural
networks. We demonstrate the applicability of the approach on a problem of a
linear elastic stress field data.
| [
{
"version": "v1",
"created": "Wed, 15 Mar 2023 13:10:58 GMT"
},
{
"version": "v2",
"created": "Tue, 11 Apr 2023 19:50:48 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Luan",
"Lele",
""
],
[
"Ramachandra",
"Nesar",
""
],
[
"Ravi",
"Sandipp Krishnan",
""
],
[
"Bhaduri",
"Anindya",
""
],
[
"Pandita",
"Piyush",
""
],
[
"Balaprakash",
"Prasanna",
""
],
[
"Anitescu",
"Mihai",
""
],
[
"Sun",
"Changjie",
""
],
[
"Wang",
"Liping",
""
]
] | TITLE: Application of probabilistic modeling and automated machine learning
framework for high-dimensional stress field
ABSTRACT: Modern computational methods, involving highly sophisticated mathematical
formulations, enable several tasks like modeling complex physical phenomenon,
predicting key properties and design optimization. The higher fidelity in these
computer models makes it computationally intensive to query them hundreds of
times for optimization and one usually relies on a simplified model albeit at
the cost of losing predictive accuracy and precision. Towards this, data-driven
surrogate modeling methods have shown a lot of promise in emulating the
behavior of the expensive computer models. However, a major bottleneck in such
methods is the inability to deal with high input dimensionality and the need
for relatively large datasets. With such problems, the input and output
quantity of interest are tensors of high dimensionality. Commonly used
surrogate modeling methods for such problems, suffer from requirements like
high number of computational evaluations that precludes one from performing
other numerical tasks like uncertainty quantification and statistical analysis.
In this work, we propose an end-to-end approach that maps a high-dimensional
image like input to an output of high dimensionality or its key statistics. Our
approach uses two main framework that perform three steps: a) reduce the input
and output from a high-dimensional space to a reduced or low-dimensional space,
b) model the input-output relationship in the low-dimensional space, and c)
enable the incorporation of domain-specific physical constraints as masks. In
order to accomplish the task of reducing input dimensionality we leverage
principal component analysis, that is coupled with two surrogate modeling
methods namely: a) Bayesian hybrid modeling, and b) DeepHyper's deep neural
networks. We demonstrate the applicability of the approach on a problem of a
linear elastic stress field data.
| no_new_dataset | 0.708591 |
2304.00668 | Weijie Li | Weijie Li, Wei Yang, Li Liu, Wenpeng Zhang, Yongxiang Liu | Discovering and Explaining the Non-Causality of Deep Learning in SAR ATR | null | null | 10.1109/LGRS.2023.3266493 | null | cs.CV cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In recent years, deep learning has been widely used in SAR ATR and achieved
excellent performance on the MSTAR dataset. However, due to constrained imaging
conditions, MSTAR has data biases such as background correlation, i.e.,
background clutter properties have a spurious correlation with target classes.
Deep learning can overfit clutter to reduce training errors. Therefore, the
degree of overfitting for clutter reflects the non-causality of deep learning
in SAR ATR. Existing methods only qualitatively analyze this phenomenon. In
this paper, we quantify the contributions of different regions to target
recognition based on the Shapley value. The Shapley value of clutter measures
the degree of overfitting. Moreover, we explain how data bias and model bias
contribute to non-causality. Concisely, data bias leads to comparable
signal-to-clutter ratios and clutter textures in training and test sets. And
various model structures have different degrees of overfitting for these
biases. The experimental results of various models under standard operating
conditions on the MSTAR dataset support our conclusions. Our code is available
at https://github.com/waterdisappear/Data-Bias-in-MSTAR.
| [
{
"version": "v1",
"created": "Mon, 3 Apr 2023 00:45:11 GMT"
},
{
"version": "v2",
"created": "Wed, 5 Apr 2023 00:40:41 GMT"
},
{
"version": "v3",
"created": "Mon, 10 Apr 2023 14:58:53 GMT"
},
{
"version": "v4",
"created": "Wed, 12 Apr 2023 04:42:21 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Li",
"Weijie",
""
],
[
"Yang",
"Wei",
""
],
[
"Liu",
"Li",
""
],
[
"Zhang",
"Wenpeng",
""
],
[
"Liu",
"Yongxiang",
""
]
] | TITLE: Discovering and Explaining the Non-Causality of Deep Learning in SAR ATR
ABSTRACT: In recent years, deep learning has been widely used in SAR ATR and achieved
excellent performance on the MSTAR dataset. However, due to constrained imaging
conditions, MSTAR has data biases such as background correlation, i.e.,
background clutter properties have a spurious correlation with target classes.
Deep learning can overfit clutter to reduce training errors. Therefore, the
degree of overfitting for clutter reflects the non-causality of deep learning
in SAR ATR. Existing methods only qualitatively analyze this phenomenon. In
this paper, we quantify the contributions of different regions to target
recognition based on the Shapley value. The Shapley value of clutter measures
the degree of overfitting. Moreover, we explain how data bias and model bias
contribute to non-causality. Concisely, data bias leads to comparable
signal-to-clutter ratios and clutter textures in training and test sets. And
various model structures have different degrees of overfitting for these
biases. The experimental results of various models under standard operating
conditions on the MSTAR dataset support our conclusions. Our code is available
at https://github.com/waterdisappear/Data-Bias-in-MSTAR.
| no_new_dataset | 0.711863 |
2304.02213 | Yuwei Wan | Tong Xie, Yuwei Wan, Wei Huang, Yufei Zhou, Yixuan Liu, Qingyuan
Linghu, Shaozhou Wang, Chunyu Kit, Clara Grazian, Wenjie Zhang and Bram Hoex | Large Language Models as Master Key: Unlocking the Secrets of Materials
Science with GPT | null | null | null | null | cs.CL cs.AI | http://creativecommons.org/licenses/by-nc-sa/4.0/ | The amount of data has growing significance in exploring cutting-edge
materials and a number of datasets have been generated either by hand or
automated approaches. However, the materials science field struggles to
effectively utilize the abundance of data, especially in applied disciplines
where materials are evaluated based on device performance rather than their
properties. This article presents a new natural language processing (NLP) task
called structured information inference (SII) to address the complexities of
information extraction at the device level in materials science. We
accomplished this task by tuning GPT-3 on an existing perovskite solar cell
FAIR (Findable, Accessible, Interoperable, Reusable) dataset with 91.8%
F1-score and extended the dataset with data published since its release. The
produced data is formatted and normalized, enabling its direct utilization as
input in subsequent data analysis. This feature empowers materials scientists
to develop models by selecting high-quality review articles within their
domain. Additionally, we designed experiments to predict the electrical
performance of solar cells and design materials or devices with targeted
parameters using large language models (LLMs). Our results demonstrate
comparable performance to traditional machine learning methods without feature
selection, highlighting the potential of LLMs to acquire scientific knowledge
and design new materials akin to materials scientists.
| [
{
"version": "v1",
"created": "Wed, 5 Apr 2023 04:01:52 GMT"
},
{
"version": "v2",
"created": "Thu, 6 Apr 2023 12:08:53 GMT"
},
{
"version": "v3",
"created": "Mon, 10 Apr 2023 15:53:26 GMT"
},
{
"version": "v4",
"created": "Tue, 11 Apr 2023 15:44:30 GMT"
},
{
"version": "v5",
"created": "Wed, 12 Apr 2023 14:06:02 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Xie",
"Tong",
""
],
[
"Wan",
"Yuwei",
""
],
[
"Huang",
"Wei",
""
],
[
"Zhou",
"Yufei",
""
],
[
"Liu",
"Yixuan",
""
],
[
"Linghu",
"Qingyuan",
""
],
[
"Wang",
"Shaozhou",
""
],
[
"Kit",
"Chunyu",
""
],
[
"Grazian",
"Clara",
""
],
[
"Zhang",
"Wenjie",
""
],
[
"Hoex",
"Bram",
""
]
] | TITLE: Large Language Models as Master Key: Unlocking the Secrets of Materials
Science with GPT
ABSTRACT: The amount of data has growing significance in exploring cutting-edge
materials and a number of datasets have been generated either by hand or
automated approaches. However, the materials science field struggles to
effectively utilize the abundance of data, especially in applied disciplines
where materials are evaluated based on device performance rather than their
properties. This article presents a new natural language processing (NLP) task
called structured information inference (SII) to address the complexities of
information extraction at the device level in materials science. We
accomplished this task by tuning GPT-3 on an existing perovskite solar cell
FAIR (Findable, Accessible, Interoperable, Reusable) dataset with 91.8%
F1-score and extended the dataset with data published since its release. The
produced data is formatted and normalized, enabling its direct utilization as
input in subsequent data analysis. This feature empowers materials scientists
to develop models by selecting high-quality review articles within their
domain. Additionally, we designed experiments to predict the electrical
performance of solar cells and design materials or devices with targeted
parameters using large language models (LLMs). Our results demonstrate
comparable performance to traditional machine learning methods without feature
selection, highlighting the potential of LLMs to acquire scientific knowledge
and design new materials akin to materials scientists.
| no_new_dataset | 0.708648 |
2304.02798 | Laya Rafiee Sevyeri | Laya Rafiee Sevyeri, Ivaxi Sheth, Farhood Farahnak, Alexandre See,
Samira Ebrahimi Kahou, Thomas Fevens, Mohammad Havaei | Source-free Domain Adaptation Requires Penalized Diversity | null | null | null | null | cs.LG cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | While neural networks are capable of achieving human-like performance in many
tasks such as image classification, the impressive performance of each model is
limited to its own dataset. Source-free domain adaptation (SFDA) was introduced
to address knowledge transfer between different domains in the absence of
source data, thus, increasing data privacy. Diversity in representation space
can be vital to a model`s adaptability in varied and difficult domains. In
unsupervised SFDA, the diversity is limited to learning a single hypothesis on
the source or learning multiple hypotheses with a shared feature extractor.
Motivated by the improved predictive performance of ensembles, we propose a
novel unsupervised SFDA algorithm that promotes representational diversity
through the use of separate feature extractors with Distinct Backbone
Architectures (DBA). Although diversity in feature space is increased, the
unconstrained mutual information (MI) maximization may potentially introduce
amplification of weak hypotheses. Thus we introduce the Weak Hypothesis
Penalization (WHP) regularizer as a mitigation strategy. Our work proposes
Penalized Diversity (PD) where the synergy of DBA and WHP is applied to
unsupervised source-free domain adaptation for covariate shift. In addition, PD
is augmented with a weighted MI maximization objective for label distribution
shift. Empirical results on natural, synthetic, and medical domains demonstrate
the effectiveness of PD under different distributional shifts.
| [
{
"version": "v1",
"created": "Thu, 6 Apr 2023 00:20:19 GMT"
},
{
"version": "v2",
"created": "Wed, 12 Apr 2023 15:50:35 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Sevyeri",
"Laya Rafiee",
""
],
[
"Sheth",
"Ivaxi",
""
],
[
"Farahnak",
"Farhood",
""
],
[
"See",
"Alexandre",
""
],
[
"Kahou",
"Samira Ebrahimi",
""
],
[
"Fevens",
"Thomas",
""
],
[
"Havaei",
"Mohammad",
""
]
] | TITLE: Source-free Domain Adaptation Requires Penalized Diversity
ABSTRACT: While neural networks are capable of achieving human-like performance in many
tasks such as image classification, the impressive performance of each model is
limited to its own dataset. Source-free domain adaptation (SFDA) was introduced
to address knowledge transfer between different domains in the absence of
source data, thus, increasing data privacy. Diversity in representation space
can be vital to a model`s adaptability in varied and difficult domains. In
unsupervised SFDA, the diversity is limited to learning a single hypothesis on
the source or learning multiple hypotheses with a shared feature extractor.
Motivated by the improved predictive performance of ensembles, we propose a
novel unsupervised SFDA algorithm that promotes representational diversity
through the use of separate feature extractors with Distinct Backbone
Architectures (DBA). Although diversity in feature space is increased, the
unconstrained mutual information (MI) maximization may potentially introduce
amplification of weak hypotheses. Thus we introduce the Weak Hypothesis
Penalization (WHP) regularizer as a mitigation strategy. Our work proposes
Penalized Diversity (PD) where the synergy of DBA and WHP is applied to
unsupervised source-free domain adaptation for covariate shift. In addition, PD
is augmented with a weighted MI maximization objective for label distribution
shift. Empirical results on natural, synthetic, and medical domains demonstrate
the effectiveness of PD under different distributional shifts.
| no_new_dataset | 0.708395 |
2304.04175 | Lin Geng Foo | Tianjiao Li, Lin Geng Foo, Ping Hu, Xindi Shang, Hossein Rahmani,
Zehuan Yuan, Jun Liu | Token Boosting for Robust Self-Supervised Visual Transformer
Pre-training | Accepted to CVPR 2023 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Learning with large-scale unlabeled data has become a powerful tool for
pre-training Visual Transformers (VTs). However, prior works tend to overlook
that, in real-world scenarios, the input data may be corrupted and unreliable.
Pre-training VTs on such corrupted data can be challenging, especially when we
pre-train via the masked autoencoding approach, where both the inputs and
masked ``ground truth" targets can potentially be unreliable in this case. To
address this limitation, we introduce the Token Boosting Module (TBM) as a
plug-and-play component for VTs that effectively allows the VT to learn to
extract clean and robust features during masked autoencoding pre-training. We
provide theoretical analysis to show how TBM improves model pre-training with
more robust and generalizable representations, thus benefiting downstream
tasks. We conduct extensive experiments to analyze TBM's effectiveness, and
results on four corrupted datasets demonstrate that TBM consistently improves
performance on downstream tasks.
| [
{
"version": "v1",
"created": "Sun, 9 Apr 2023 06:42:33 GMT"
},
{
"version": "v2",
"created": "Wed, 12 Apr 2023 04:35:47 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Li",
"Tianjiao",
""
],
[
"Foo",
"Lin Geng",
""
],
[
"Hu",
"Ping",
""
],
[
"Shang",
"Xindi",
""
],
[
"Rahmani",
"Hossein",
""
],
[
"Yuan",
"Zehuan",
""
],
[
"Liu",
"Jun",
""
]
] | TITLE: Token Boosting for Robust Self-Supervised Visual Transformer
Pre-training
ABSTRACT: Learning with large-scale unlabeled data has become a powerful tool for
pre-training Visual Transformers (VTs). However, prior works tend to overlook
that, in real-world scenarios, the input data may be corrupted and unreliable.
Pre-training VTs on such corrupted data can be challenging, especially when we
pre-train via the masked autoencoding approach, where both the inputs and
masked ``ground truth" targets can potentially be unreliable in this case. To
address this limitation, we introduce the Token Boosting Module (TBM) as a
plug-and-play component for VTs that effectively allows the VT to learn to
extract clean and robust features during masked autoencoding pre-training. We
provide theoretical analysis to show how TBM improves model pre-training with
more robust and generalizable representations, thus benefiting downstream
tasks. We conduct extensive experiments to analyze TBM's effectiveness, and
results on four corrupted datasets demonstrate that TBM consistently improves
performance on downstream tasks.
| no_new_dataset | 0.708042 |
2304.04415 | Li Xu | Li Xu, Mark He Huang, Xindi Shang, Zehuan Yuan, Ying Sun, Jun Liu | Meta Compositional Referring Expression Segmentation | Accepted by CVPR 2023 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Referring expression segmentation aims to segment an object described by a
language expression from an image. Despite the recent progress on this task,
existing models tackling this task may not be able to fully capture semantics
and visual representations of individual concepts, which limits their
generalization capability, especially when handling novel compositions of
learned concepts. In this work, through the lens of meta learning, we propose a
Meta Compositional Referring Expression Segmentation (MCRES) framework to
enhance model compositional generalization performance. Specifically, to handle
various levels of novel compositions, our framework first uses training data to
construct a virtual training set and multiple virtual testing sets, where data
samples in each virtual testing set contain a level of novel compositions
w.r.t. the virtual training set. Then, following a novel meta optimization
scheme to optimize the model to obtain good testing performance on the virtual
testing sets after training on the virtual training set, our framework can
effectively drive the model to better capture semantics and visual
representations of individual concepts, and thus obtain robust generalization
performance even when handling novel compositions. Extensive experiments on
three benchmark datasets demonstrate the effectiveness of our framework.
| [
{
"version": "v1",
"created": "Mon, 10 Apr 2023 06:55:25 GMT"
},
{
"version": "v2",
"created": "Tue, 11 Apr 2023 04:01:24 GMT"
},
{
"version": "v3",
"created": "Wed, 12 Apr 2023 07:19:48 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Xu",
"Li",
""
],
[
"Huang",
"Mark He",
""
],
[
"Shang",
"Xindi",
""
],
[
"Yuan",
"Zehuan",
""
],
[
"Sun",
"Ying",
""
],
[
"Liu",
"Jun",
""
]
] | TITLE: Meta Compositional Referring Expression Segmentation
ABSTRACT: Referring expression segmentation aims to segment an object described by a
language expression from an image. Despite the recent progress on this task,
existing models tackling this task may not be able to fully capture semantics
and visual representations of individual concepts, which limits their
generalization capability, especially when handling novel compositions of
learned concepts. In this work, through the lens of meta learning, we propose a
Meta Compositional Referring Expression Segmentation (MCRES) framework to
enhance model compositional generalization performance. Specifically, to handle
various levels of novel compositions, our framework first uses training data to
construct a virtual training set and multiple virtual testing sets, where data
samples in each virtual testing set contain a level of novel compositions
w.r.t. the virtual training set. Then, following a novel meta optimization
scheme to optimize the model to obtain good testing performance on the virtual
testing sets after training on the virtual training set, our framework can
effectively drive the model to better capture semantics and visual
representations of individual concepts, and thus obtain robust generalization
performance even when handling novel compositions. Extensive experiments on
three benchmark datasets demonstrate the effectiveness of our framework.
| no_new_dataset | 0.711212 |
2304.04468 | Changshuo Liu | Changshuo Liu, Wenqiao Zhang, Beng Chin Ooi, James Wei Luen Yip,
Lingze Zeng, Kaiping Zheng | Toward Cohort Intelligence: A Universal Cohort Representation Learning
Framework for Electronic Health Record Analysis | 10 pages | null | null | null | cs.LG cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Electronic Health Records (EHR) are generated from clinical routine care
recording valuable information of broad patient populations, which provide
plentiful opportunities for improving patient management and intervention
strategies in clinical practice. To exploit the enormous potential of EHR data,
a popular EHR data analysis paradigm in machine learning is EHR representation
learning, which first leverages the individual patient's EHR data to learn
informative representations by a backbone, and supports diverse health-care
downstream tasks grounded on the representations. Unfortunately, such a
paradigm fails to access the in-depth analysis of patients' relevance, which is
generally known as cohort studies in clinical practice. Specifically, patients
in the same cohort tend to share similar characteristics, implying their
resemblance in medical conditions such as symptoms or diseases. In this paper,
we propose a universal COhort Representation lEarning (CORE) framework to
augment EHR utilization by leveraging the fine-grained cohort information among
patients. In particular, CORE first develops an explicit patient modeling task
based on the prior knowledge of patients' diagnosis codes, which measures the
latent relevance among patients to adaptively divide the cohorts for each
patient. Based on the constructed cohorts, CORE recodes the pre-extracted EHR
data representation from intra- and inter-cohort perspectives, yielding
augmented EHR data representation learning. CORE is readily applicable to
diverse backbone models, serving as a universal plug-in framework to infuse
cohort information into healthcare methods for boosted performance. We conduct
an extensive experimental evaluation on two real-world datasets, and the
experimental results demonstrate the effectiveness and generalizability of
CORE.
| [
{
"version": "v1",
"created": "Mon, 10 Apr 2023 09:12:37 GMT"
},
{
"version": "v2",
"created": "Tue, 11 Apr 2023 15:15:05 GMT"
},
{
"version": "v3",
"created": "Wed, 12 Apr 2023 07:12:42 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Liu",
"Changshuo",
""
],
[
"Zhang",
"Wenqiao",
""
],
[
"Ooi",
"Beng Chin",
""
],
[
"Yip",
"James Wei Luen",
""
],
[
"Zeng",
"Lingze",
""
],
[
"Zheng",
"Kaiping",
""
]
] | TITLE: Toward Cohort Intelligence: A Universal Cohort Representation Learning
Framework for Electronic Health Record Analysis
ABSTRACT: Electronic Health Records (EHR) are generated from clinical routine care
recording valuable information of broad patient populations, which provide
plentiful opportunities for improving patient management and intervention
strategies in clinical practice. To exploit the enormous potential of EHR data,
a popular EHR data analysis paradigm in machine learning is EHR representation
learning, which first leverages the individual patient's EHR data to learn
informative representations by a backbone, and supports diverse health-care
downstream tasks grounded on the representations. Unfortunately, such a
paradigm fails to access the in-depth analysis of patients' relevance, which is
generally known as cohort studies in clinical practice. Specifically, patients
in the same cohort tend to share similar characteristics, implying their
resemblance in medical conditions such as symptoms or diseases. In this paper,
we propose a universal COhort Representation lEarning (CORE) framework to
augment EHR utilization by leveraging the fine-grained cohort information among
patients. In particular, CORE first develops an explicit patient modeling task
based on the prior knowledge of patients' diagnosis codes, which measures the
latent relevance among patients to adaptively divide the cohorts for each
patient. Based on the constructed cohorts, CORE recodes the pre-extracted EHR
data representation from intra- and inter-cohort perspectives, yielding
augmented EHR data representation learning. CORE is readily applicable to
diverse backbone models, serving as a universal plug-in framework to infuse
cohort information into healthcare methods for boosted performance. We conduct
an extensive experimental evaluation on two real-world datasets, and the
experimental results demonstrate the effectiveness and generalizability of
CORE.
| no_new_dataset | 0.709988 |
2304.05395 | ChuXin Wang | Jiacheng Deng, Chuxin Wang, Jiahao Lu, Jianfeng He, Tianzhu Zhang,
Jiyang Yu, Zhe Zhang | SE-ORNet: Self-Ensembling Orientation-aware Network for Unsupervised
Point Cloud Shape Correspondence | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Unsupervised point cloud shape correspondence aims to obtain dense
point-to-point correspondences between point clouds without manually annotated
pairs. However, humans and some animals have bilateral symmetry and various
orientations, which lead to severe mispredictions of symmetrical parts.
Besides, point cloud noise disrupts consistent representations for point cloud
and thus degrades the shape correspondence accuracy. To address the above
issues, we propose a Self-Ensembling ORientation-aware Network termed SE-ORNet.
The key of our approach is to exploit an orientation estimation module with a
domain adaptive discriminator to align the orientations of point cloud pairs,
which significantly alleviates the mispredictions of symmetrical parts.
Additionally, we design a selfensembling framework for unsupervised point cloud
shape correspondence. In this framework, the disturbances of point cloud noise
are overcome by perturbing the inputs of the student and teacher networks with
different data augmentations and constraining the consistency of predictions.
Extensive experiments on both human and animal datasets show that our SE-ORNet
can surpass state-of-the-art unsupervised point cloud shape correspondence
methods.
| [
{
"version": "v1",
"created": "Mon, 10 Apr 2023 18:00:02 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Deng",
"Jiacheng",
""
],
[
"Wang",
"Chuxin",
""
],
[
"Lu",
"Jiahao",
""
],
[
"He",
"Jianfeng",
""
],
[
"Zhang",
"Tianzhu",
""
],
[
"Yu",
"Jiyang",
""
],
[
"Zhang",
"Zhe",
""
]
] | TITLE: SE-ORNet: Self-Ensembling Orientation-aware Network for Unsupervised
Point Cloud Shape Correspondence
ABSTRACT: Unsupervised point cloud shape correspondence aims to obtain dense
point-to-point correspondences between point clouds without manually annotated
pairs. However, humans and some animals have bilateral symmetry and various
orientations, which lead to severe mispredictions of symmetrical parts.
Besides, point cloud noise disrupts consistent representations for point cloud
and thus degrades the shape correspondence accuracy. To address the above
issues, we propose a Self-Ensembling ORientation-aware Network termed SE-ORNet.
The key of our approach is to exploit an orientation estimation module with a
domain adaptive discriminator to align the orientations of point cloud pairs,
which significantly alleviates the mispredictions of symmetrical parts.
Additionally, we design a selfensembling framework for unsupervised point cloud
shape correspondence. In this framework, the disturbances of point cloud noise
are overcome by perturbing the inputs of the student and teacher networks with
different data augmentations and constraining the consistency of predictions.
Extensive experiments on both human and animal datasets show that our SE-ORNet
can surpass state-of-the-art unsupervised point cloud shape correspondence
methods.
| no_new_dataset | 0.710867 |
2304.05403 | Gilberto Ochoa-Ruiz | David Laines, Gissella Bejarano, Miguel Gonzalez-Mendoza, Gilberto
Ochoa-Ruiz | Isolated Sign Language Recognition based on Tree Structure Skeleton
Images | This paper has been accepted the the LatinX in Computer Vision
Research Workshop at CVPR 2023 | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Sign Language Recognition (SLR) systems aim to be embedded in video stream
platforms to recognize the sign performed in front of a camera. SLR research
has taken advantage of recent advances in pose estimation models to use
skeleton sequences estimated from videos instead of RGB information to predict
signs. This approach can make HAR-related tasks less complex and more robust to
diverse backgrounds, lightning conditions, and physical appearances. In this
work, we explore the use of a spatio-temporal skeleton representation such as
Tree Structure Skeleton Image (TSSI) as an alternative input to improve the
accuracy of skeleton-based models for SLR. TSSI converts a skeleton sequence
into an RGB image where the columns represent the joints of the skeleton in a
depth-first tree traversal order, the rows represent the temporal evolution of
the joints, and the three channels represent the (x, y, z) coordinates of the
joints. We trained a DenseNet-121 using this type of input and compared it with
other skeleton-based deep learning methods using a large-scale American Sign
Language (ASL) dataset, WLASL. Our model (SL-TSSI-DenseNet) overcomes the
state-of-the-art of other skeleton-based models. Moreover, when including data
augmentation our proposal achieves better results than both skeleton-based and
RGB-based models. We evaluated the effectiveness of our model on the Ankara
University Turkish Sign Language (TSL) dataset, AUTSL, and a Mexican Sign
Language (LSM) dataset. On the AUTSL dataset, the model achieves similar
results to the state-of-the-art of other skeleton-based models. On the LSM
dataset, the model achieves higher results than the baseline. Code has been
made available at: https://github.com/davidlainesv/SL-TSSI-DenseNet.
| [
{
"version": "v1",
"created": "Mon, 10 Apr 2023 01:58:50 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Laines",
"David",
""
],
[
"Bejarano",
"Gissella",
""
],
[
"Gonzalez-Mendoza",
"Miguel",
""
],
[
"Ochoa-Ruiz",
"Gilberto",
""
]
] | TITLE: Isolated Sign Language Recognition based on Tree Structure Skeleton
Images
ABSTRACT: Sign Language Recognition (SLR) systems aim to be embedded in video stream
platforms to recognize the sign performed in front of a camera. SLR research
has taken advantage of recent advances in pose estimation models to use
skeleton sequences estimated from videos instead of RGB information to predict
signs. This approach can make HAR-related tasks less complex and more robust to
diverse backgrounds, lightning conditions, and physical appearances. In this
work, we explore the use of a spatio-temporal skeleton representation such as
Tree Structure Skeleton Image (TSSI) as an alternative input to improve the
accuracy of skeleton-based models for SLR. TSSI converts a skeleton sequence
into an RGB image where the columns represent the joints of the skeleton in a
depth-first tree traversal order, the rows represent the temporal evolution of
the joints, and the three channels represent the (x, y, z) coordinates of the
joints. We trained a DenseNet-121 using this type of input and compared it with
other skeleton-based deep learning methods using a large-scale American Sign
Language (ASL) dataset, WLASL. Our model (SL-TSSI-DenseNet) overcomes the
state-of-the-art of other skeleton-based models. Moreover, when including data
augmentation our proposal achieves better results than both skeleton-based and
RGB-based models. We evaluated the effectiveness of our model on the Ankara
University Turkish Sign Language (TSL) dataset, AUTSL, and a Mexican Sign
Language (LSM) dataset. On the AUTSL dataset, the model achieves similar
results to the state-of-the-art of other skeleton-based models. On the LSM
dataset, the model achieves higher results than the baseline. Code has been
made available at: https://github.com/davidlainesv/SL-TSSI-DenseNet.
| no_new_dataset | 0.70557 |
2304.05442 | Shubhang Pandey | Murali Dadi, Shubhang Pandey, Aparna Behera and T G Venkatesh | Performance Study of Partitioned Caches in Asymmetric Multi-Core
Processors | null | null | null | null | cs.AR | http://creativecommons.org/licenses/by-nc-sa/4.0/ | The current workloads and applications are highly diversified, facing
critical challenges such as the Power Wall and the Memory Wall Problem.
Different strategies over the multiple levels of Caches have evolved to
mitigate these problems. Also, to work with such diversified applications, the
Asymmetric Multi-Core Processor (AMP) presents itself as a viable solution. In
this paper, we study the performance of L2 and Last Level Cache for different
cache partitions against various AMP configurations. In addition, this study
investigates the optimal cache partitioning for a collection of Multi-threaded
benchmarks from PARSEC and SPLASH2 benchmark suites under medium-sized inputs.
We have studied the effect of block replacement strategies and their impact on
the key metrics such as total on-chip power consumption and L2 \& LLC Miss
rates. Our study presents an intermediate cache design for AMPs between the two
extremities of fully shared and fully private L2 \& LLC level Cache, which
helps achieve the desired power values and optimal cache miss penalties.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 18:30:06 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Dadi",
"Murali",
""
],
[
"Pandey",
"Shubhang",
""
],
[
"Behera",
"Aparna",
""
],
[
"Venkatesh",
"T G",
""
]
] | TITLE: Performance Study of Partitioned Caches in Asymmetric Multi-Core
Processors
ABSTRACT: The current workloads and applications are highly diversified, facing
critical challenges such as the Power Wall and the Memory Wall Problem.
Different strategies over the multiple levels of Caches have evolved to
mitigate these problems. Also, to work with such diversified applications, the
Asymmetric Multi-Core Processor (AMP) presents itself as a viable solution. In
this paper, we study the performance of L2 and Last Level Cache for different
cache partitions against various AMP configurations. In addition, this study
investigates the optimal cache partitioning for a collection of Multi-threaded
benchmarks from PARSEC and SPLASH2 benchmark suites under medium-sized inputs.
We have studied the effect of block replacement strategies and their impact on
the key metrics such as total on-chip power consumption and L2 \& LLC Miss
rates. Our study presents an intermediate cache design for AMPs between the two
extremities of fully shared and fully private L2 \& LLC level Cache, which
helps achieve the desired power values and optimal cache miss penalties.
| no_new_dataset | 0.70844 |
2304.05464 | Patrick Ebel | Patrick Ebel, Vivien Sainte Fare Garnot, Michael Schmitt, Jan Dirk
Wegner, Xiao Xiang Zhu | UnCRtainTS: Uncertainty Quantification for Cloud Removal in Optical
Satellite Time Series | null | null | null | null | cs.CV eess.IV | http://creativecommons.org/licenses/by/4.0/ | Clouds and haze often occlude optical satellite images, hindering continuous,
dense monitoring of the Earth's surface. Although modern deep learning methods
can implicitly learn to ignore such occlusions, explicit cloud removal as
pre-processing enables manual interpretation and allows training models when
only few annotations are available. Cloud removal is challenging due to the
wide range of occlusion scenarios -- from scenes partially visible through
haze, to completely opaque cloud coverage. Furthermore, integrating
reconstructed images in downstream applications would greatly benefit from
trustworthy quality assessment. In this paper, we introduce UnCRtainTS, a
method for multi-temporal cloud removal combining a novel attention-based
architecture, and a formulation for multivariate uncertainty prediction. These
two components combined set a new state-of-the-art performance in terms of
image reconstruction on two public cloud removal datasets. Additionally, we
show how the well-calibrated predicted uncertainties enable a precise control
of the reconstruction quality.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 19:27:18 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Ebel",
"Patrick",
""
],
[
"Garnot",
"Vivien Sainte Fare",
""
],
[
"Schmitt",
"Michael",
""
],
[
"Wegner",
"Jan Dirk",
""
],
[
"Zhu",
"Xiao Xiang",
""
]
] | TITLE: UnCRtainTS: Uncertainty Quantification for Cloud Removal in Optical
Satellite Time Series
ABSTRACT: Clouds and haze often occlude optical satellite images, hindering continuous,
dense monitoring of the Earth's surface. Although modern deep learning methods
can implicitly learn to ignore such occlusions, explicit cloud removal as
pre-processing enables manual interpretation and allows training models when
only few annotations are available. Cloud removal is challenging due to the
wide range of occlusion scenarios -- from scenes partially visible through
haze, to completely opaque cloud coverage. Furthermore, integrating
reconstructed images in downstream applications would greatly benefit from
trustworthy quality assessment. In this paper, we introduce UnCRtainTS, a
method for multi-temporal cloud removal combining a novel attention-based
architecture, and a formulation for multivariate uncertainty prediction. These
two components combined set a new state-of-the-art performance in terms of
image reconstruction on two public cloud removal datasets. Additionally, we
show how the well-calibrated predicted uncertainties enable a precise control
of the reconstruction quality.
| no_new_dataset | 0.709038 |
2304.05469 | Zongwei Wu | Xue-Jing Luo, Shuo Wang, Zongwei Wu, Christos Sakaridis, Yun Cheng,
Deng-Ping Fan, Luc Van Gool | CamDiff: Camouflage Image Augmentation via Diffusion Model | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | The burgeoning field of camouflaged object detection (COD) seeks to identify
objects that blend into their surroundings. Despite the impressive performance
of recent models, we have identified a limitation in their robustness, where
existing methods may misclassify salient objects as camouflaged ones, despite
these two characteristics being contradictory. This limitation may stem from
lacking multi-pattern training images, leading to less saliency robustness. To
address this issue, we introduce CamDiff, a novel approach inspired by
AI-Generated Content (AIGC) that overcomes the scarcity of multi-pattern
training images. Specifically, we leverage the latent diffusion model to
synthesize salient objects in camouflaged scenes, while using the zero-shot
image classification ability of the Contrastive Language-Image Pre-training
(CLIP) model to prevent synthesis failures and ensure the synthesized object
aligns with the input prompt. Consequently, the synthesized image retains its
original camouflage label while incorporating salient objects, yielding
camouflage samples with richer characteristics. The results of user studies
show that the salient objects in the scenes synthesized by our framework
attract the user's attention more; thus, such samples pose a greater challenge
to the existing COD models. Our approach enables flexible editing and efficient
large-scale dataset generation at a low cost. It significantly enhances COD
baselines' training and testing phases, emphasizing robustness across diverse
domains. Our newly-generated datasets and source code are available at
https://github.com/drlxj/CamDiff.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 19:37:47 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Luo",
"Xue-Jing",
""
],
[
"Wang",
"Shuo",
""
],
[
"Wu",
"Zongwei",
""
],
[
"Sakaridis",
"Christos",
""
],
[
"Cheng",
"Yun",
""
],
[
"Fan",
"Deng-Ping",
""
],
[
"Van Gool",
"Luc",
""
]
] | TITLE: CamDiff: Camouflage Image Augmentation via Diffusion Model
ABSTRACT: The burgeoning field of camouflaged object detection (COD) seeks to identify
objects that blend into their surroundings. Despite the impressive performance
of recent models, we have identified a limitation in their robustness, where
existing methods may misclassify salient objects as camouflaged ones, despite
these two characteristics being contradictory. This limitation may stem from
lacking multi-pattern training images, leading to less saliency robustness. To
address this issue, we introduce CamDiff, a novel approach inspired by
AI-Generated Content (AIGC) that overcomes the scarcity of multi-pattern
training images. Specifically, we leverage the latent diffusion model to
synthesize salient objects in camouflaged scenes, while using the zero-shot
image classification ability of the Contrastive Language-Image Pre-training
(CLIP) model to prevent synthesis failures and ensure the synthesized object
aligns with the input prompt. Consequently, the synthesized image retains its
original camouflage label while incorporating salient objects, yielding
camouflage samples with richer characteristics. The results of user studies
show that the salient objects in the scenes synthesized by our framework
attract the user's attention more; thus, such samples pose a greater challenge
to the existing COD models. Our approach enables flexible editing and efficient
large-scale dataset generation at a low cost. It significantly enhances COD
baselines' training and testing phases, emphasizing robustness across diverse
domains. Our newly-generated datasets and source code are available at
https://github.com/drlxj/CamDiff.
| no_new_dataset | 0.641254 |
2304.05474 | Shaofeng Wang | Kai Song, Shaofeng Wang, Ziwei Xie, Shanyu Wang, Jiahong Li, Yongqiang
Yang | Audience Expansion for Multi-show Release Based on an Edge-prompted
Heterogeneous Graph Network | null | null | null | null | cs.SI cs.AI cs.IR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In the user targeting and expanding of new shows on a video platform, the key
point is how their embeddings are generated. It's supposed to be personalized
from the perspective of both users and shows. Furthermore, the pursue of both
instant (click) and long-time (view time) rewards, and the cold-start problem
for new shows bring additional challenges. Such a problem is suitable for
processing by heterogeneous graph models, because of the natural graph
structure of data. But real-world networks usually have billions of nodes and
various types of edges. Few existing methods focus on handling large-scale data
and exploiting different types of edges, especially the latter. In this paper,
we propose a two-stage audience expansion scheme based on an edge-prompted
heterogeneous graph network which can take different double-sided interactions
and features into account. In the offline stage, to construct the graph, user
IDs and specific side information combinations of the shows are chosen to be
the nodes, and click/co-click relations and view time are used to build the
edges. Embeddings and clustered user groups are then calculated. When new shows
arrive, their embeddings and subsequent matching users can be produced within a
consistent space. In the online stage, posterior data including click/view
users are employed as seeds to look for similar users. The results on the
public datasets and our billion-scale data demonstrate the accuracy and
efficiency of our approach.
| [
{
"version": "v1",
"created": "Sat, 8 Apr 2023 15:14:32 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Song",
"Kai",
""
],
[
"Wang",
"Shaofeng",
""
],
[
"Xie",
"Ziwei",
""
],
[
"Wang",
"Shanyu",
""
],
[
"Li",
"Jiahong",
""
],
[
"Yang",
"Yongqiang",
""
]
] | TITLE: Audience Expansion for Multi-show Release Based on an Edge-prompted
Heterogeneous Graph Network
ABSTRACT: In the user targeting and expanding of new shows on a video platform, the key
point is how their embeddings are generated. It's supposed to be personalized
from the perspective of both users and shows. Furthermore, the pursue of both
instant (click) and long-time (view time) rewards, and the cold-start problem
for new shows bring additional challenges. Such a problem is suitable for
processing by heterogeneous graph models, because of the natural graph
structure of data. But real-world networks usually have billions of nodes and
various types of edges. Few existing methods focus on handling large-scale data
and exploiting different types of edges, especially the latter. In this paper,
we propose a two-stage audience expansion scheme based on an edge-prompted
heterogeneous graph network which can take different double-sided interactions
and features into account. In the offline stage, to construct the graph, user
IDs and specific side information combinations of the shows are chosen to be
the nodes, and click/co-click relations and view time are used to build the
edges. Embeddings and clustered user groups are then calculated. When new shows
arrive, their embeddings and subsequent matching users can be produced within a
consistent space. In the online stage, posterior data including click/view
users are employed as seeds to look for similar users. The results on the
public datasets and our billion-scale data demonstrate the accuracy and
efficiency of our approach.
| no_new_dataset | 0.711406 |
2304.05523 | Rakesh Chada | Rakesh Chada, Zhaoheng Zheng, Pradeep Natarajan | MoMo: A shared encoder Model for text, image and multi-Modal
representations | null | null | null | null | cs.CV cs.AI cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We propose a self-supervised shared encoder model that achieves strong
results on several visual, language and multimodal benchmarks while being data,
memory and run-time efficient. We make three key contributions. First, in
contrast to most existing works, we use a single transformer with all the
encoder layers processing both the text and the image modalities. Second, we
propose a stage-wise training strategy where the model is first trained on
images, then jointly with unimodal text and image datasets and finally jointly
with text and text-image datasets. Third, to preserve information across both
the modalities, we propose a training pipeline that learns simultaneously from
gradient updates of different modalities at each training update step. The
results on downstream text-only, image-only and multimodal tasks show that our
model is competitive with several strong models while using fewer parameters
and lesser pre-training data. For example, MoMo performs competitively with
FLAVA on multimodal (+3.1), image-only (+1.1) and text-only (-0.1) tasks
despite having 2/5th the number of parameters and using 1/3rd the image-text
training pairs. Finally, we ablate various design choices and further show that
increasing model size produces significant performance gains indicating
potential for substantial improvements with larger models using our approach.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 22:26:10 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Chada",
"Rakesh",
""
],
[
"Zheng",
"Zhaoheng",
""
],
[
"Natarajan",
"Pradeep",
""
]
] | TITLE: MoMo: A shared encoder Model for text, image and multi-Modal
representations
ABSTRACT: We propose a self-supervised shared encoder model that achieves strong
results on several visual, language and multimodal benchmarks while being data,
memory and run-time efficient. We make three key contributions. First, in
contrast to most existing works, we use a single transformer with all the
encoder layers processing both the text and the image modalities. Second, we
propose a stage-wise training strategy where the model is first trained on
images, then jointly with unimodal text and image datasets and finally jointly
with text and text-image datasets. Third, to preserve information across both
the modalities, we propose a training pipeline that learns simultaneously from
gradient updates of different modalities at each training update step. The
results on downstream text-only, image-only and multimodal tasks show that our
model is competitive with several strong models while using fewer parameters
and lesser pre-training data. For example, MoMo performs competitively with
FLAVA on multimodal (+3.1), image-only (+1.1) and text-only (-0.1) tasks
despite having 2/5th the number of parameters and using 1/3rd the image-text
training pairs. Finally, we ablate various design choices and further show that
increasing model size produces significant performance gains indicating
potential for substantial improvements with larger models using our approach.
| no_new_dataset | 0.710848 |
2304.05542 | Weihua Zhou | Chen Zhao, Anqi Liu, Xiao Zhang, Xuewei Cao, Zhengming Ding, Qiuying
Sha, Hui Shen, Hong-Wen Deng, Weihua Zhou | CLCLSA: Cross-omics Linked embedding with Contrastive Learning and Self
Attention for multi-omics integration with incomplete multi-omics data | 21 pages; 5 figures | null | null | null | cs.LG cs.AI q-bio.GN | http://creativecommons.org/licenses/by/4.0/ | Integration of heterogeneous and high-dimensional multi-omics data is
becoming increasingly important in understanding genetic data. Each omics
technique only provides a limited view of the underlying biological process and
integrating heterogeneous omics layers simultaneously would lead to a more
comprehensive and detailed understanding of diseases and phenotypes. However,
one obstacle faced when performing multi-omics data integration is the
existence of unpaired multi-omics data due to instrument sensitivity and cost.
Studies may fail if certain aspects of the subjects are missing or incomplete.
In this paper, we propose a deep learning method for multi-omics integration
with incomplete data by Cross-omics Linked unified embedding with Contrastive
Learning and Self Attention (CLCLSA). Utilizing complete multi-omics data as
supervision, the model employs cross-omics autoencoders to learn the feature
representation across different types of biological data. The multi-omics
contrastive learning, which is used to maximize the mutual information between
different types of omics, is employed before latent feature concatenation. In
addition, the feature-level self-attention and omics-level self-attention are
employed to dynamically identify the most informative features for multi-omics
data integration. Extensive experiments were conducted on four public
multi-omics datasets. The experimental results indicated that the proposed
CLCLSA outperformed the state-of-the-art approaches for multi-omics data
classification using incomplete multi-omics data.
| [
{
"version": "v1",
"created": "Wed, 12 Apr 2023 00:22:18 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Zhao",
"Chen",
""
],
[
"Liu",
"Anqi",
""
],
[
"Zhang",
"Xiao",
""
],
[
"Cao",
"Xuewei",
""
],
[
"Ding",
"Zhengming",
""
],
[
"Sha",
"Qiuying",
""
],
[
"Shen",
"Hui",
""
],
[
"Deng",
"Hong-Wen",
""
],
[
"Zhou",
"Weihua",
""
]
] | TITLE: CLCLSA: Cross-omics Linked embedding with Contrastive Learning and Self
Attention for multi-omics integration with incomplete multi-omics data
ABSTRACT: Integration of heterogeneous and high-dimensional multi-omics data is
becoming increasingly important in understanding genetic data. Each omics
technique only provides a limited view of the underlying biological process and
integrating heterogeneous omics layers simultaneously would lead to a more
comprehensive and detailed understanding of diseases and phenotypes. However,
one obstacle faced when performing multi-omics data integration is the
existence of unpaired multi-omics data due to instrument sensitivity and cost.
Studies may fail if certain aspects of the subjects are missing or incomplete.
In this paper, we propose a deep learning method for multi-omics integration
with incomplete data by Cross-omics Linked unified embedding with Contrastive
Learning and Self Attention (CLCLSA). Utilizing complete multi-omics data as
supervision, the model employs cross-omics autoencoders to learn the feature
representation across different types of biological data. The multi-omics
contrastive learning, which is used to maximize the mutual information between
different types of omics, is employed before latent feature concatenation. In
addition, the feature-level self-attention and omics-level self-attention are
employed to dynamically identify the most informative features for multi-omics
data integration. Extensive experiments were conducted on four public
multi-omics datasets. The experimental results indicated that the proposed
CLCLSA outperformed the state-of-the-art approaches for multi-omics data
classification using incomplete multi-omics data.
| no_new_dataset | 0.710879 |
2304.05544 | Vikas Natesh | Andrew Sabot, Vikas Natesh, H.T. Kung, Wei-Te Ting | MEMA Runtime Framework: Minimizing External Memory Accesses for TinyML
on Microcontrollers | Accepted as a full paper by the TinyML Research Symposium 2023 | null | null | null | cs.LG cs.AR cs.PF cs.PL | http://creativecommons.org/licenses/by-nc-nd/4.0/ | We present the MEMA framework for the easy and quick derivation of efficient
inference runtimes that minimize external memory accesses for matrix
multiplication on TinyML systems. The framework accounts for hardware resource
constraints and problem sizes in analytically determining optimized schedules
and kernels that minimize memory accesses. MEMA provides a solution to a
well-known problem in the current practice, that is, optimal schedules tend to
be found only through a time consuming and heuristic search of a large
scheduling space. We compare the performance of runtimes derived from MEMA to
existing state-of-the-art libraries on ARM-based TinyML systems. For example,
for neural network benchmarks on the ARM Cortex-M4, we achieve up to a 1.8x
speedup and 44% energy reduction over CMSIS-NN.
| [
{
"version": "v1",
"created": "Wed, 12 Apr 2023 00:27:11 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Sabot",
"Andrew",
""
],
[
"Natesh",
"Vikas",
""
],
[
"Kung",
"H. T.",
""
],
[
"Ting",
"Wei-Te",
""
]
] | TITLE: MEMA Runtime Framework: Minimizing External Memory Accesses for TinyML
on Microcontrollers
ABSTRACT: We present the MEMA framework for the easy and quick derivation of efficient
inference runtimes that minimize external memory accesses for matrix
multiplication on TinyML systems. The framework accounts for hardware resource
constraints and problem sizes in analytically determining optimized schedules
and kernels that minimize memory accesses. MEMA provides a solution to a
well-known problem in the current practice, that is, optimal schedules tend to
be found only through a time consuming and heuristic search of a large
scheduling space. We compare the performance of runtimes derived from MEMA to
existing state-of-the-art libraries on ARM-based TinyML systems. For example,
for neural network benchmarks on the ARM Cortex-M4, we achieve up to a 1.8x
speedup and 44% energy reduction over CMSIS-NN.
| no_new_dataset | 0.710189 |
2304.05548 | Feixiang Ren | Feixiang Ren | Distilling Token-Pruned Pose Transformer for 2D Human Pose Estimation | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Human pose estimation has seen widespread use of transformer models in recent
years. Pose transformers benefit from the self-attention map, which captures
the correlation between human joint tokens and the image. However, training
such models is computationally expensive. The recent token-Pruned Pose
Transformer (PPT) solves this problem by pruning the background tokens of the
image, which are usually less informative. However, although it improves
efficiency, PPT inevitably leads to worse performance than TokenPose due to the
pruning of tokens. To overcome this problem, we present a novel method called
Distilling Pruned-Token Transformer for human pose estimation (DPPT). Our
method leverages the output of a pre-trained TokenPose to supervise the
learning process of PPT. We also establish connections between the internal
structure of pose transformers and PPT, such as attention maps and joint
features. Our experimental results on the MPII datasets show that our DPPT can
significantly improve PCK compared to previous PPT models while still reducing
computational complexity.
| [
{
"version": "v1",
"created": "Wed, 12 Apr 2023 00:46:41 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Ren",
"Feixiang",
""
]
] | TITLE: Distilling Token-Pruned Pose Transformer for 2D Human Pose Estimation
ABSTRACT: Human pose estimation has seen widespread use of transformer models in recent
years. Pose transformers benefit from the self-attention map, which captures
the correlation between human joint tokens and the image. However, training
such models is computationally expensive. The recent token-Pruned Pose
Transformer (PPT) solves this problem by pruning the background tokens of the
image, which are usually less informative. However, although it improves
efficiency, PPT inevitably leads to worse performance than TokenPose due to the
pruning of tokens. To overcome this problem, we present a novel method called
Distilling Pruned-Token Transformer for human pose estimation (DPPT). Our
method leverages the output of a pre-trained TokenPose to supervise the
learning process of PPT. We also establish connections between the internal
structure of pose transformers and PPT, such as attention maps and joint
features. Our experimental results on the MPII datasets show that our DPPT can
significantly improve PCK compared to previous PPT models while still reducing
computational complexity.
| no_new_dataset | 0.709038 |
2304.05552 | Zhihao Lin | Zhihao Lin, Yongtao Wang, Jinhe Zhang, Xiaojie Chu | DynamicDet: A Unified Dynamic Architecture for Object Detection | Accepted by CVPR 2023 | null | null | null | cs.CV cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Dynamic neural network is an emerging research topic in deep learning. With
adaptive inference, dynamic models can achieve remarkable accuracy and
computational efficiency. However, it is challenging to design a powerful
dynamic detector, because of no suitable dynamic architecture and exiting
criterion for object detection. To tackle these difficulties, we propose a
dynamic framework for object detection, named DynamicDet. Firstly, we carefully
design a dynamic architecture based on the nature of the object detection task.
Then, we propose an adaptive router to analyze the multi-scale information and
to decide the inference route automatically. We also present a novel
optimization strategy with an exiting criterion based on the detection losses
for our dynamic detectors. Last, we present a variable-speed inference
strategy, which helps to realize a wide range of accuracy-speed trade-offs with
only one dynamic detector. Extensive experiments conducted on the COCO
benchmark demonstrate that the proposed DynamicDet achieves new
state-of-the-art accuracy-speed trade-offs. For instance, with comparable
accuracy, the inference speed of our dynamic detector Dy-YOLOv7-W6 surpasses
YOLOv7-E6 by 12%, YOLOv7-D6 by 17%, and YOLOv7-E6E by 39%. The code is
available at https://github.com/VDIGPKU/DynamicDet.
| [
{
"version": "v1",
"created": "Wed, 12 Apr 2023 01:16:53 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Lin",
"Zhihao",
""
],
[
"Wang",
"Yongtao",
""
],
[
"Zhang",
"Jinhe",
""
],
[
"Chu",
"Xiaojie",
""
]
] | TITLE: DynamicDet: A Unified Dynamic Architecture for Object Detection
ABSTRACT: Dynamic neural network is an emerging research topic in deep learning. With
adaptive inference, dynamic models can achieve remarkable accuracy and
computational efficiency. However, it is challenging to design a powerful
dynamic detector, because of no suitable dynamic architecture and exiting
criterion for object detection. To tackle these difficulties, we propose a
dynamic framework for object detection, named DynamicDet. Firstly, we carefully
design a dynamic architecture based on the nature of the object detection task.
Then, we propose an adaptive router to analyze the multi-scale information and
to decide the inference route automatically. We also present a novel
optimization strategy with an exiting criterion based on the detection losses
for our dynamic detectors. Last, we present a variable-speed inference
strategy, which helps to realize a wide range of accuracy-speed trade-offs with
only one dynamic detector. Extensive experiments conducted on the COCO
benchmark demonstrate that the proposed DynamicDet achieves new
state-of-the-art accuracy-speed trade-offs. For instance, with comparable
accuracy, the inference speed of our dynamic detector Dy-YOLOv7-W6 surpasses
YOLOv7-E6 by 12%, YOLOv7-D6 by 17%, and YOLOv7-E6E by 39%. The code is
available at https://github.com/VDIGPKU/DynamicDet.
| no_new_dataset | 0.705734 |
2304.05554 | Liping Bao | Liping Bao, Longhui Wei, Xiaoyu Qiu, Wengang Zhou, Houqiang Li, Qi
Tian | Learning Transferable Pedestrian Representation from Multimodal
Information Supervision | null | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by/4.0/ | Recent researches on unsupervised person re-identification~(reID) have
demonstrated that pre-training on unlabeled person images achieves superior
performance on downstream reID tasks than pre-training on ImageNet. However,
those pre-trained methods are specifically designed for reID and suffer
flexible adaption to other pedestrian analysis tasks. In this paper, we propose
VAL-PAT, a novel framework that learns transferable representations to enhance
various pedestrian analysis tasks with multimodal information. To train our
framework, we introduce three learning objectives, \emph{i.e.,} self-supervised
contrastive learning, image-text contrastive learning and multi-attribute
classification. The self-supervised contrastive learning facilitates the
learning of the intrinsic pedestrian properties, while the image-text
contrastive learning guides the model to focus on the appearance information of
pedestrians.Meanwhile, multi-attribute classification encourages the model to
recognize attributes to excavate fine-grained pedestrian information. We first
perform pre-training on LUPerson-TA dataset, where each image contains text and
attribute annotations, and then transfer the learned representations to various
downstream tasks, including person reID, person attribute recognition and
text-based person search. Extensive experiments demonstrate that our framework
facilitates the learning of general pedestrian representations and thus leads
to promising results on various pedestrian analysis tasks.
| [
{
"version": "v1",
"created": "Wed, 12 Apr 2023 01:20:58 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Bao",
"Liping",
""
],
[
"Wei",
"Longhui",
""
],
[
"Qiu",
"Xiaoyu",
""
],
[
"Zhou",
"Wengang",
""
],
[
"Li",
"Houqiang",
""
],
[
"Tian",
"Qi",
""
]
] | TITLE: Learning Transferable Pedestrian Representation from Multimodal
Information Supervision
ABSTRACT: Recent researches on unsupervised person re-identification~(reID) have
demonstrated that pre-training on unlabeled person images achieves superior
performance on downstream reID tasks than pre-training on ImageNet. However,
those pre-trained methods are specifically designed for reID and suffer
flexible adaption to other pedestrian analysis tasks. In this paper, we propose
VAL-PAT, a novel framework that learns transferable representations to enhance
various pedestrian analysis tasks with multimodal information. To train our
framework, we introduce three learning objectives, \emph{i.e.,} self-supervised
contrastive learning, image-text contrastive learning and multi-attribute
classification. The self-supervised contrastive learning facilitates the
learning of the intrinsic pedestrian properties, while the image-text
contrastive learning guides the model to focus on the appearance information of
pedestrians.Meanwhile, multi-attribute classification encourages the model to
recognize attributes to excavate fine-grained pedestrian information. We first
perform pre-training on LUPerson-TA dataset, where each image contains text and
attribute annotations, and then transfer the learned representations to various
downstream tasks, including person reID, person attribute recognition and
text-based person search. Extensive experiments demonstrate that our framework
facilitates the learning of general pedestrian representations and thus leads
to promising results on various pedestrian analysis tasks.
| no_new_dataset | 0.708377 |
2304.05561 | Hassan Jameel Asghar | Gioacchino Tangari and Shreesh Keskar and Hassan Jameel Asghar and
Dali Kaafar | On the Adversarial Inversion of Deep Biometric Representations | null | null | null | null | cs.CV cs.CR cs.LG | http://creativecommons.org/licenses/by/4.0/ | Biometric authentication service providers often claim that it is not
possible to reverse-engineer a user's raw biometric sample, such as a
fingerprint or a face image, from its mathematical (feature-space)
representation. In this paper, we investigate this claim on the specific
example of deep neural network (DNN) embeddings. Inversion of DNN embeddings
has been investigated for explaining deep image representations or synthesizing
normalized images. Existing studies leverage full access to all layers of the
original model, as well as all possible information on the original dataset.
For the biometric authentication use case, we need to investigate this under
adversarial settings where an attacker has access to a feature-space
representation but no direct access to the exact original dataset nor the
original learned model. Instead, we assume varying degree of attacker's
background knowledge about the distribution of the dataset as well as the
original learned model (architecture and training process). In these cases, we
show that the attacker can exploit off-the-shelf DNN models and public
datasets, to mimic the behaviour of the original learned model to varying
degrees of success, based only on the obtained representation and attacker's
prior knowledge. We propose a two-pronged attack that first infers the original
DNN by exploiting the model footprint on the embedding, and then reconstructs
the raw data by using the inferred model. We show the practicality of the
attack on popular DNNs trained for two prominent biometric modalities, face and
fingerprint recognition. The attack can effectively infer the original
recognition model (mean accuracy 83\% for faces, 86\% for fingerprints), and
can craft effective biometric reconstructions that are successfully
authenticated with 1-vs-1 authentication accuracy of up to 92\% for some
models.
| [
{
"version": "v1",
"created": "Wed, 12 Apr 2023 01:47:11 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Tangari",
"Gioacchino",
""
],
[
"Keskar",
"Shreesh",
""
],
[
"Asghar",
"Hassan Jameel",
""
],
[
"Kaafar",
"Dali",
""
]
] | TITLE: On the Adversarial Inversion of Deep Biometric Representations
ABSTRACT: Biometric authentication service providers often claim that it is not
possible to reverse-engineer a user's raw biometric sample, such as a
fingerprint or a face image, from its mathematical (feature-space)
representation. In this paper, we investigate this claim on the specific
example of deep neural network (DNN) embeddings. Inversion of DNN embeddings
has been investigated for explaining deep image representations or synthesizing
normalized images. Existing studies leverage full access to all layers of the
original model, as well as all possible information on the original dataset.
For the biometric authentication use case, we need to investigate this under
adversarial settings where an attacker has access to a feature-space
representation but no direct access to the exact original dataset nor the
original learned model. Instead, we assume varying degree of attacker's
background knowledge about the distribution of the dataset as well as the
original learned model (architecture and training process). In these cases, we
show that the attacker can exploit off-the-shelf DNN models and public
datasets, to mimic the behaviour of the original learned model to varying
degrees of success, based only on the obtained representation and attacker's
prior knowledge. We propose a two-pronged attack that first infers the original
DNN by exploiting the model footprint on the embedding, and then reconstructs
the raw data by using the inferred model. We show the practicality of the
attack on popular DNNs trained for two prominent biometric modalities, face and
fingerprint recognition. The attack can effectively infer the original
recognition model (mean accuracy 83\% for faces, 86\% for fingerprints), and
can craft effective biometric reconstructions that are successfully
authenticated with 1-vs-1 authentication accuracy of up to 92\% for some
models.
| no_new_dataset | 0.711418 |
2304.05564 | Shuang Cui | Shuang Cui, Bingnan Wang, Quan Zheng | Neural Invertible Variable-degree Optical Aberrations Correction | null | Optics Express 2023 | 10.1364/OE.485258 | null | cs.CV eess.IV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Optical aberrations of optical systems cause significant degradation of
imaging quality. Aberration correction by sophisticated lens designs and
special glass materials generally incurs high cost of manufacturing and the
increase in the weight of optical systems, thus recent work has shifted to
aberration correction with deep learning-based post-processing. Though
real-world optical aberrations vary in degree, existing methods cannot
eliminate variable-degree aberrations well, especially for the severe degrees
of degradation. Also, previous methods use a single feed-forward neural network
and suffer from information loss in the output. To address the issues, we
propose a novel aberration correction method with an invertible architecture by
leveraging its information-lossless property. Within the architecture, we
develop conditional invertible blocks to allow the processing of aberrations
with variable degrees. Our method is evaluated on both a synthetic dataset from
physics-based imaging simulation and a real captured dataset. Quantitative and
qualitative experimental results demonstrate that our method outperforms
compared methods in correcting variable-degree optical aberrations.
| [
{
"version": "v1",
"created": "Wed, 12 Apr 2023 01:56:42 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Cui",
"Shuang",
""
],
[
"Wang",
"Bingnan",
""
],
[
"Zheng",
"Quan",
""
]
] | TITLE: Neural Invertible Variable-degree Optical Aberrations Correction
ABSTRACT: Optical aberrations of optical systems cause significant degradation of
imaging quality. Aberration correction by sophisticated lens designs and
special glass materials generally incurs high cost of manufacturing and the
increase in the weight of optical systems, thus recent work has shifted to
aberration correction with deep learning-based post-processing. Though
real-world optical aberrations vary in degree, existing methods cannot
eliminate variable-degree aberrations well, especially for the severe degrees
of degradation. Also, previous methods use a single feed-forward neural network
and suffer from information loss in the output. To address the issues, we
propose a novel aberration correction method with an invertible architecture by
leveraging its information-lossless property. Within the architecture, we
develop conditional invertible blocks to allow the processing of aberrations
with variable degrees. Our method is evaluated on both a synthetic dataset from
physics-based imaging simulation and a real captured dataset. Quantitative and
qualitative experimental results demonstrate that our method outperforms
compared methods in correcting variable-degree optical aberrations.
| no_new_dataset | 0.710597 |
2304.05565 | Anabella Doctor | Anabella C. Doctor | A Predictive Model using Machine Learning Algorithm in Identifying
Students Probability on Passing Semestral Course | null | International Journal of Computing Sciences Research (ISSN print:
2546-0552; ISSN online: 2546-115X) Vol. 7, pp. 1830-1856 | 10.25147/ijcsr.2017.001.1.135 | null | cs.LG cs.CV cs.CY | http://creativecommons.org/licenses/by/4.0/ | This study aims to determine a predictive model to learn students probability
to pass their courses taken at the earliest stage of the semester. To
successfully discover a good predictive model with high acceptability,
accurate, and precision rate which delivers a useful outcome for decision
making in education systems, in improving the processes of conveying knowledge
and uplifting students academic performance, the proponent applies and strictly
followed the CRISP-DM (Cross-Industry Standard Process for Data Mining)
methodology. This study employs classification for data mining techniques, and
decision tree for algorithm. With the utilization of the newly discovered
predictive model, the prediction of students probabilities to pass the current
courses they take gives 0.7619 accuracy, 0.8333 precision, 0.8823 recall, and
0.8571 f1 score, which shows that the model used in the prediction is reliable,
accurate, and recommendable. Considering the indicators and the results, it can
be noted that the prediction model used in this study is highly acceptable. The
data mining techniques provides effective and efficient innovative tools in
analyzing and predicting student performances. The model used in this study
will greatly affect the way educators understand and identify the weakness of
their students in the class, the way they improved the effectiveness of their
learning processes gearing to their students, bring down academic failure
rates, and help institution administrators modify their learning system
outcomes. Further study for the inclusion of some students demographic
information, vast amount of data within the dataset, automated and manual
process of predictive criteria indicators where the students can regulate to
which criteria, they must improve more for them to pass their courses taken at
the end of the semester as early as midterm period are highly needed.
| [
{
"version": "v1",
"created": "Wed, 12 Apr 2023 01:57:08 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Doctor",
"Anabella C.",
""
]
] | TITLE: A Predictive Model using Machine Learning Algorithm in Identifying
Students Probability on Passing Semestral Course
ABSTRACT: This study aims to determine a predictive model to learn students probability
to pass their courses taken at the earliest stage of the semester. To
successfully discover a good predictive model with high acceptability,
accurate, and precision rate which delivers a useful outcome for decision
making in education systems, in improving the processes of conveying knowledge
and uplifting students academic performance, the proponent applies and strictly
followed the CRISP-DM (Cross-Industry Standard Process for Data Mining)
methodology. This study employs classification for data mining techniques, and
decision tree for algorithm. With the utilization of the newly discovered
predictive model, the prediction of students probabilities to pass the current
courses they take gives 0.7619 accuracy, 0.8333 precision, 0.8823 recall, and
0.8571 f1 score, which shows that the model used in the prediction is reliable,
accurate, and recommendable. Considering the indicators and the results, it can
be noted that the prediction model used in this study is highly acceptable. The
data mining techniques provides effective and efficient innovative tools in
analyzing and predicting student performances. The model used in this study
will greatly affect the way educators understand and identify the weakness of
their students in the class, the way they improved the effectiveness of their
learning processes gearing to their students, bring down academic failure
rates, and help institution administrators modify their learning system
outcomes. Further study for the inclusion of some students demographic
information, vast amount of data within the dataset, automated and manual
process of predictive criteria indicators where the students can regulate to
which criteria, they must improve more for them to pass their courses taken at
the end of the semester as early as midterm period are highly needed.
| no_new_dataset | 0.709202 |
2304.05568 | Jiabao Ji | Jiabao Ji, Guanhua Zhang, Zhaowen Wang, Bairu Hou, Zhifei Zhang, Brian
Price, Shiyu Chang | Improving Diffusion Models for Scene Text Editing with Dual Encoders | 22 pages, 19 figures | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by/4.0/ | Scene text editing is a challenging task that involves modifying or inserting
specified texts in an image while maintaining its natural and realistic
appearance. Most previous approaches to this task rely on style-transfer models
that crop out text regions and feed them into image transfer models, such as
GANs. However, these methods are limited in their ability to change text style
and are unable to insert texts into images. Recent advances in diffusion models
have shown promise in overcoming these limitations with text-conditional image
editing. However, our empirical analysis reveals that state-of-the-art
diffusion models struggle with rendering correct text and controlling text
style. To address these problems, we propose DIFFSTE to improve pre-trained
diffusion models with a dual encoder design, which includes a character encoder
for better text legibility and an instruction encoder for better style control.
An instruction tuning framework is introduced to train our model to learn the
mapping from the text instruction to the corresponding image with either the
specified style or the style of the surrounding texts in the background. Such a
training method further brings our method the zero-shot generalization ability
to the following three scenarios: generating text with unseen font variation,
e.g., italic and bold, mixing different fonts to construct a new font, and
using more relaxed forms of natural language as the instructions to guide the
generation task. We evaluate our approach on five datasets and demonstrate its
superior performance in terms of text correctness, image naturalness, and style
controllability. Our code is publicly available.
https://github.com/UCSB-NLP-Chang/DiffSTE
| [
{
"version": "v1",
"created": "Wed, 12 Apr 2023 02:08:34 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Ji",
"Jiabao",
""
],
[
"Zhang",
"Guanhua",
""
],
[
"Wang",
"Zhaowen",
""
],
[
"Hou",
"Bairu",
""
],
[
"Zhang",
"Zhifei",
""
],
[
"Price",
"Brian",
""
],
[
"Chang",
"Shiyu",
""
]
] | TITLE: Improving Diffusion Models for Scene Text Editing with Dual Encoders
ABSTRACT: Scene text editing is a challenging task that involves modifying or inserting
specified texts in an image while maintaining its natural and realistic
appearance. Most previous approaches to this task rely on style-transfer models
that crop out text regions and feed them into image transfer models, such as
GANs. However, these methods are limited in their ability to change text style
and are unable to insert texts into images. Recent advances in diffusion models
have shown promise in overcoming these limitations with text-conditional image
editing. However, our empirical analysis reveals that state-of-the-art
diffusion models struggle with rendering correct text and controlling text
style. To address these problems, we propose DIFFSTE to improve pre-trained
diffusion models with a dual encoder design, which includes a character encoder
for better text legibility and an instruction encoder for better style control.
An instruction tuning framework is introduced to train our model to learn the
mapping from the text instruction to the corresponding image with either the
specified style or the style of the surrounding texts in the background. Such a
training method further brings our method the zero-shot generalization ability
to the following three scenarios: generating text with unseen font variation,
e.g., italic and bold, mixing different fonts to construct a new font, and
using more relaxed forms of natural language as the instructions to guide the
generation task. We evaluate our approach on five datasets and demonstrate its
superior performance in terms of text correctness, image naturalness, and style
controllability. Our code is publicly available.
https://github.com/UCSB-NLP-Chang/DiffSTE
| no_new_dataset | 0.710653 |
2304.05578 | Wei Tan | Wei Tan, Jionghao Lin, David Lang, Guanliang Chen, Dragan Gasevic, Lan
Du, Wray Buntine | Does Informativeness Matter? Active Learning for Educational Dialogue
Act Classification | 12 pages full paper, The 24th International Conference on Artificial
Intelligence in Education, AIED 2023 | null | null | null | cs.CL cs.AI cs.LG | http://creativecommons.org/licenses/by/4.0/ | Dialogue Acts (DAs) can be used to explain what expert tutors do and what
students know during the tutoring process. Most empirical studies adopt the
random sampling method to obtain sentence samples for manual annotation of DAs,
which are then used to train DA classifiers. However, these studies have paid
little attention to sample informativeness, which can reflect the information
quantity of the selected samples and inform the extent to which a classifier
can learn patterns. Notably, the informativeness level may vary among the
samples and the classifier might only need a small amount of low informative
samples to learn the patterns. Random sampling may overlook sample
informativeness, which consumes human labelling costs and contributes less to
training the classifiers. As an alternative, researchers suggest employing
statistical sampling methods of Active Learning (AL) to identify the
informative samples for training the classifiers. However, the use of AL
methods in educational DA classification tasks is under-explored. In this
paper, we examine the informativeness of annotated sentence samples. Then, the
study investigates how the AL methods can select informative samples to support
DA classifiers in the AL sampling process. The results reveal that most
annotated sentences present low informativeness in the training dataset and the
patterns of these sentences can be easily captured by the DA classifier. We
also demonstrate how AL methods can reduce the cost of manual annotation in the
AL sampling process.
| [
{
"version": "v1",
"created": "Wed, 12 Apr 2023 02:42:20 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Tan",
"Wei",
""
],
[
"Lin",
"Jionghao",
""
],
[
"Lang",
"David",
""
],
[
"Chen",
"Guanliang",
""
],
[
"Gasevic",
"Dragan",
""
],
[
"Du",
"Lan",
""
],
[
"Buntine",
"Wray",
""
]
] | TITLE: Does Informativeness Matter? Active Learning for Educational Dialogue
Act Classification
ABSTRACT: Dialogue Acts (DAs) can be used to explain what expert tutors do and what
students know during the tutoring process. Most empirical studies adopt the
random sampling method to obtain sentence samples for manual annotation of DAs,
which are then used to train DA classifiers. However, these studies have paid
little attention to sample informativeness, which can reflect the information
quantity of the selected samples and inform the extent to which a classifier
can learn patterns. Notably, the informativeness level may vary among the
samples and the classifier might only need a small amount of low informative
samples to learn the patterns. Random sampling may overlook sample
informativeness, which consumes human labelling costs and contributes less to
training the classifiers. As an alternative, researchers suggest employing
statistical sampling methods of Active Learning (AL) to identify the
informative samples for training the classifiers. However, the use of AL
methods in educational DA classification tasks is under-explored. In this
paper, we examine the informativeness of annotated sentence samples. Then, the
study investigates how the AL methods can select informative samples to support
DA classifiers in the AL sampling process. The results reveal that most
annotated sentences present low informativeness in the training dataset and the
patterns of these sentences can be easily captured by the DA classifier. We
also demonstrate how AL methods can reduce the cost of manual annotation in the
AL sampling process.
| no_new_dataset | 0.711638 |
2304.05591 | Siddharth Suresh | Siddharth Suresh, Kushin Mukherjee, Timothy T. Rogers | Semantic Feature Verification in FLAN-T5 | To appear as a Tiny Paper at ICLR 2023 | null | null | null | cs.CL cs.AI cs.LG | http://creativecommons.org/licenses/by/4.0/ | This study evaluates the potential of a large language model for aiding in
generation of semantic feature norms - a critical tool for evaluating
conceptual structure in cognitive science. Building from an existing
human-generated dataset, we show that machine-verified norms capture aspects of
conceptual structure beyond what is expressed in human norms alone, and better
explain human judgments of semantic similarity amongst items that are distally
related. The results suggest that LLMs can greatly enhance traditional methods
of semantic feature norm verification, with implications for our understanding
of conceptual representation in humans and machines.
| [
{
"version": "v1",
"created": "Wed, 12 Apr 2023 03:37:57 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Suresh",
"Siddharth",
""
],
[
"Mukherjee",
"Kushin",
""
],
[
"Rogers",
"Timothy T.",
""
]
] | TITLE: Semantic Feature Verification in FLAN-T5
ABSTRACT: This study evaluates the potential of a large language model for aiding in
generation of semantic feature norms - a critical tool for evaluating
conceptual structure in cognitive science. Building from an existing
human-generated dataset, we show that machine-verified norms capture aspects of
conceptual structure beyond what is expressed in human norms alone, and better
explain human judgments of semantic similarity amongst items that are distally
related. The results suggest that LLMs can greatly enhance traditional methods
of semantic feature norm verification, with implications for our understanding
of conceptual representation in humans and machines.
| new_dataset | 0.713675 |
2304.05610 | Wei Xiao | Dejian Meng, Wei Xiao, Lijun Zhang, Zhuang Zhang, and Zihao Liu | Vehicle Trajectory Prediction based Predictive Collision Risk Assessment
for Autonomous Driving in Highway Scenarios | manuscript submitted to IEEE Transactions on Intelligent
Transportation Systems | null | null | null | cs.RO | http://creativecommons.org/licenses/by-nc-nd/4.0/ | For driving safely and efficiently in highway scenarios, autonomous vehicles
(AVs) must be able to predict future behaviors of surrounding object vehicles
(OVs), and assess collision risk accurately for reasonable decision-making.
Aiming at autonomous driving in highway scenarios, a predictive collision risk
assessment method based on trajectory prediction of OVs is proposed in this
paper. Firstly, the vehicle trajectory prediction is formulated as a sequence
generation task with long short-term memory (LSTM) encoder-decoder framework.
Convolutional social pooling (CSP) and graph attention network (GAN) are
adopted for extracting local spatial vehicle interactions and distant spatial
vehicle interactions, respectively. Then, two basic risk metrics,
time-to-collision (TTC) and minimal distance margin (MDM), are calculated
between the predicted trajectory of OV and the candidate trajectory of AV.
Consequently, a time-continuous risk function is constructed with temporal and
spatial risk metrics. Finally, the vehicle trajectory prediction model
CSP-GAN-LSTM is evaluated on two public highway datasets. The quantitative
results indicate that the proposed CSP-GAN-LSTM model outperforms the existing
state-of-the-art (SOTA) methods in terms of position prediction accuracy.
Besides, simulation results in typical highway scenarios further validate the
feasibility and effectiveness of the proposed predictive collision risk
assessment method.
| [
{
"version": "v1",
"created": "Wed, 12 Apr 2023 05:02:48 GMT"
}
] | 2023-04-13T00:00:00 | [
[
"Meng",
"Dejian",
""
],
[
"Xiao",
"Wei",
""
],
[
"Zhang",
"Lijun",
""
],
[
"Zhang",
"Zhuang",
""
],
[
"Liu",
"Zihao",
""
]
] | TITLE: Vehicle Trajectory Prediction based Predictive Collision Risk Assessment
for Autonomous Driving in Highway Scenarios
ABSTRACT: For driving safely and efficiently in highway scenarios, autonomous vehicles
(AVs) must be able to predict future behaviors of surrounding object vehicles
(OVs), and assess collision risk accurately for reasonable decision-making.
Aiming at autonomous driving in highway scenarios, a predictive collision risk
assessment method based on trajectory prediction of OVs is proposed in this
paper. Firstly, the vehicle trajectory prediction is formulated as a sequence
generation task with long short-term memory (LSTM) encoder-decoder framework.
Convolutional social pooling (CSP) and graph attention network (GAN) are
adopted for extracting local spatial vehicle interactions and distant spatial
vehicle interactions, respectively. Then, two basic risk metrics,
time-to-collision (TTC) and minimal distance margin (MDM), are calculated
between the predicted trajectory of OV and the candidate trajectory of AV.
Consequently, a time-continuous risk function is constructed with temporal and
spatial risk metrics. Finally, the vehicle trajectory prediction model
CSP-GAN-LSTM is evaluated on two public highway datasets. The quantitative
results indicate that the proposed CSP-GAN-LSTM model outperforms the existing
state-of-the-art (SOTA) methods in terms of position prediction accuracy.
Besides, simulation results in typical highway scenarios further validate the
feasibility and effectiveness of the proposed predictive collision risk
assessment method.
| no_new_dataset | 0.710258 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.