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
2304.03200
Emilio Francesquini
Matheus Campos Fernandes, Fabr\'icio Olivetti de Fran\c{c}a, Emilio Francesquini
HOTGP -- Higher-Order Typed Genetic Programming
null
null
10.1145/3583131.3590464
null
cs.NE cs.AI
http://creativecommons.org/licenses/by-nc-sa/4.0/
Program synthesis is the process of generating a computer program following a set of specifications, which can be a high-level description of the problem and/or a set of input-output examples. The synthesis can be modeled as a search problem in which the search space is the set of all the programs valid under a grammar. As the search space is vast, brute force is usually not viable and search heuristics, such as genetic programming, also have difficulty navigating it without any guidance. In this paper we present HOTGP, a new genetic programming algorithm that synthesizes pure, typed, and functional programs. HOTGP leverages the knowledge provided by the rich data-types associated with the specification and the built-in grammar to constrain the search space and improve the performance of the synthesis. The grammar is based on Haskell's standard base library (the synthesized code can be directly compiled using any standard Haskell compiler) and includes support for higher-order functions, $\lambda$-functions, and parametric polymorphism. Experimental results show that, when compared to $6$ state-of-the-art algorithms using a standard set of benchmarks, HOTGP is competitive and capable of synthesizing the correct programs more frequently than any other of the evaluated algorithms.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 16:23:34 GMT" } ]
2023-04-07T00:00:00
[ [ "Fernandes", "Matheus Campos", "" ], [ "de França", "Fabrício Olivetti", "" ], [ "Francesquini", "Emilio", "" ] ]
TITLE: HOTGP -- Higher-Order Typed Genetic Programming ABSTRACT: Program synthesis is the process of generating a computer program following a set of specifications, which can be a high-level description of the problem and/or a set of input-output examples. The synthesis can be modeled as a search problem in which the search space is the set of all the programs valid under a grammar. As the search space is vast, brute force is usually not viable and search heuristics, such as genetic programming, also have difficulty navigating it without any guidance. In this paper we present HOTGP, a new genetic programming algorithm that synthesizes pure, typed, and functional programs. HOTGP leverages the knowledge provided by the rich data-types associated with the specification and the built-in grammar to constrain the search space and improve the performance of the synthesis. The grammar is based on Haskell's standard base library (the synthesized code can be directly compiled using any standard Haskell compiler) and includes support for higher-order functions, $\lambda$-functions, and parametric polymorphism. Experimental results show that, when compared to $6$ state-of-the-art algorithms using a standard set of benchmarks, HOTGP is competitive and capable of synthesizing the correct programs more frequently than any other of the evaluated algorithms.
no_new_dataset
0.706437
2304.03202
Yihe Dong
Yihe Dong, Sercan O. Arik
SLM: End-to-end Feature Selection via Sparse Learnable Masks
null
null
null
null
cs.LG
http://creativecommons.org/licenses/by/4.0/
Feature selection has been widely used to alleviate compute requirements during training, elucidate model interpretability, and improve model generalizability. We propose SLM -- Sparse Learnable Masks -- a canonical approach for end-to-end feature selection that scales well with respect to both the feature dimension and the number of samples. At the heart of SLM lies a simple but effective learnable sparse mask, which learns which features to select, and gives rise to a novel objective that provably maximizes the mutual information (MI) between the selected features and the labels, which can be derived from a quadratic relaxation of mutual information from first principles. In addition, we derive a scaling mechanism that allows SLM to precisely control the number of features selected, through a novel use of sparsemax. This allows for more effective learning as demonstrated in ablation studies. Empirically, SLM achieves state-of-the-art results against a variety of competitive baselines on eight benchmark datasets, often by a significant margin, especially on those with real-world challenges such as class imbalance.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 16:25:43 GMT" } ]
2023-04-07T00:00:00
[ [ "Dong", "Yihe", "" ], [ "Arik", "Sercan O.", "" ] ]
TITLE: SLM: End-to-end Feature Selection via Sparse Learnable Masks ABSTRACT: Feature selection has been widely used to alleviate compute requirements during training, elucidate model interpretability, and improve model generalizability. We propose SLM -- Sparse Learnable Masks -- a canonical approach for end-to-end feature selection that scales well with respect to both the feature dimension and the number of samples. At the heart of SLM lies a simple but effective learnable sparse mask, which learns which features to select, and gives rise to a novel objective that provably maximizes the mutual information (MI) between the selected features and the labels, which can be derived from a quadratic relaxation of mutual information from first principles. In addition, we derive a scaling mechanism that allows SLM to precisely control the number of features selected, through a novel use of sparsemax. This allows for more effective learning as demonstrated in ablation studies. Empirically, SLM achieves state-of-the-art results against a variety of competitive baselines on eight benchmark datasets, often by a significant margin, especially on those with real-world challenges such as class imbalance.
no_new_dataset
0.710007
2304.03208
Joel Hestness
Nolan Dey, Gurpreet Gosal, Zhiming (Charles) Chen, Hemant Khachane, William Marshall, Ribhu Pathria, Marvin Tom, Joel Hestness
Cerebras-GPT: Open Compute-Optimal Language Models Trained on the Cerebras Wafer-Scale Cluster
13 pages main text, 16 pages appendix, 13 figures
null
null
null
cs.LG cs.CL
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
We study recent research advances that improve large language models through efficient pre-training and scaling, and open datasets and tools. We combine these advances to introduce Cerebras-GPT, a family of open compute-optimal language models scaled from 111M to 13B parameters. We train Cerebras-GPT models on the Eleuther Pile dataset following DeepMind Chinchilla scaling rules for efficient pre-training (highest accuracy for a given compute budget). We characterize the predictable power-law scaling and compare Cerebras-GPT with other publicly-available models to show all Cerebras-GPT models have state-of-the-art training efficiency on both pre-training and downstream objectives. We describe our learnings including how Maximal Update Parameterization ($\mu$P) can further improve large model scaling, improving accuracy and hyperparameter predictability at scale. We release our pre-trained models and code, making this paper the first open and reproducible work comparing compute-optimal model scaling to models trained on fixed dataset sizes. Cerebras-GPT models are available on HuggingFace: https://huggingface.co/cerebras.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 16:43:16 GMT" } ]
2023-04-07T00:00:00
[ [ "Dey", "Nolan", "", "Charles" ], [ "Gosal", "Gurpreet", "", "Charles" ], [ "Zhiming", "", "", "Charles" ], [ "Chen", "", "" ], [ "Khachane", "Hemant", "" ], [ "Marshall", "William", "" ], [ "Pathria", "Ribhu", "" ], [ "Tom", "Marvin", "" ], [ "Hestness", "Joel", "" ] ]
TITLE: Cerebras-GPT: Open Compute-Optimal Language Models Trained on the Cerebras Wafer-Scale Cluster ABSTRACT: We study recent research advances that improve large language models through efficient pre-training and scaling, and open datasets and tools. We combine these advances to introduce Cerebras-GPT, a family of open compute-optimal language models scaled from 111M to 13B parameters. We train Cerebras-GPT models on the Eleuther Pile dataset following DeepMind Chinchilla scaling rules for efficient pre-training (highest accuracy for a given compute budget). We characterize the predictable power-law scaling and compare Cerebras-GPT with other publicly-available models to show all Cerebras-GPT models have state-of-the-art training efficiency on both pre-training and downstream objectives. We describe our learnings including how Maximal Update Parameterization ($\mu$P) can further improve large model scaling, improving accuracy and hyperparameter predictability at scale. We release our pre-trained models and code, making this paper the first open and reproducible work comparing compute-optimal model scaling to models trained on fixed dataset sizes. Cerebras-GPT models are available on HuggingFace: https://huggingface.co/cerebras.
no_new_dataset
0.710434
2304.03210
Ashka Shah
Ashka Shah, Arvind Ramanathan, Valerie Hayot-Sasson, Rick Stevens
Causal Discovery and Optimal Experimental Design for Genome-Scale Biological Network Recovery
To be published in Platform for Advanced Scientific Computing 2023 (PASC23) conference proceedings
null
null
null
q-bio.MN cs.DC
http://creativecommons.org/licenses/by/4.0/
Causal discovery of genome-scale networks is important for identifying pathways from genes to observable traits - e.g. differences in cell function, disease, drug resistance and others. Causal learners based on graphical models rely on interventional samples to orient edges in the network. However, these models have not been shown to scale up the size of the genome, which are on the order of 1e3-1e4 genes. We introduce a new learner, SP-GIES, that jointly learns from interventional and observational datasets and achieves almost 4x speedup against an existing learner for 1,000 node networks. SP-GIES achieves an AUC-PR score of 0.91 on 1,000 node networks, and scales up to 2,000 node networks - this is 4x larger than existing works. We also show how SP-GIES improves downstream optimal experimental design strategies for selecting interventional experiments to perform on the system. This is an important step forward in realizing causal discovery at scale via autonomous experimental design.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 16:44:19 GMT" } ]
2023-04-07T00:00:00
[ [ "Shah", "Ashka", "" ], [ "Ramanathan", "Arvind", "" ], [ "Hayot-Sasson", "Valerie", "" ], [ "Stevens", "Rick", "" ] ]
TITLE: Causal Discovery and Optimal Experimental Design for Genome-Scale Biological Network Recovery ABSTRACT: Causal discovery of genome-scale networks is important for identifying pathways from genes to observable traits - e.g. differences in cell function, disease, drug resistance and others. Causal learners based on graphical models rely on interventional samples to orient edges in the network. However, these models have not been shown to scale up the size of the genome, which are on the order of 1e3-1e4 genes. We introduce a new learner, SP-GIES, that jointly learns from interventional and observational datasets and achieves almost 4x speedup against an existing learner for 1,000 node networks. SP-GIES achieves an AUC-PR score of 0.91 on 1,000 node networks, and scales up to 2,000 node networks - this is 4x larger than existing works. We also show how SP-GIES improves downstream optimal experimental design strategies for selecting interventional experiments to perform on the system. This is an important step forward in realizing causal discovery at scale via autonomous experimental design.
no_new_dataset
0.70724
2304.03218
Mitchell Pavlak
Mitchell Pavlak, Nathan Drenkow, Nicholas Petrick, Mohammad Mehdi Farhangi, Mathias Unberath
Data AUDIT: Identifying Attribute Utility- and Detectability-Induced Bias in Task Models
11 pages, 5 figures
null
null
null
cs.LG cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
To safely deploy deep learning-based computer vision models for computer-aided detection and diagnosis, we must ensure that they are robust and reliable. Towards that goal, algorithmic auditing has received substantial attention. To guide their audit procedures, existing methods rely on heuristic approaches or high-level objectives (e.g., non-discrimination in regards to protected attributes, such as sex, gender, or race). However, algorithms may show bias with respect to various attributes beyond the more obvious ones, and integrity issues related to these more subtle attributes can have serious consequences. To enable the generation of actionable, data-driven hypotheses which identify specific dataset attributes likely to induce model bias, we contribute a first technique for the rigorous, quantitative screening of medical image datasets. Drawing from literature in the causal inference and information theory domains, our procedure decomposes the risks associated with dataset attributes in terms of their detectability and utility (defined as the amount of information knowing the attribute gives about a task label). To demonstrate the effectiveness and sensitivity of our method, we develop a variety of datasets with synthetically inserted artifacts with different degrees of association to the target label that allow evaluation of inherited model biases via comparison of performance against true counterfactual examples. Using these datasets and results from hundreds of trained models, we show our screening method reliably identifies nearly imperceptible bias-inducing artifacts. Lastly, we apply our method to the natural attributes of a popular skin-lesion dataset and demonstrate its success. Our approach provides a means to perform more systematic algorithmic audits and guide future data collection efforts in pursuit of safer and more reliable models.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 16:50:15 GMT" } ]
2023-04-07T00:00:00
[ [ "Pavlak", "Mitchell", "" ], [ "Drenkow", "Nathan", "" ], [ "Petrick", "Nicholas", "" ], [ "Farhangi", "Mohammad Mehdi", "" ], [ "Unberath", "Mathias", "" ] ]
TITLE: Data AUDIT: Identifying Attribute Utility- and Detectability-Induced Bias in Task Models ABSTRACT: To safely deploy deep learning-based computer vision models for computer-aided detection and diagnosis, we must ensure that they are robust and reliable. Towards that goal, algorithmic auditing has received substantial attention. To guide their audit procedures, existing methods rely on heuristic approaches or high-level objectives (e.g., non-discrimination in regards to protected attributes, such as sex, gender, or race). However, algorithms may show bias with respect to various attributes beyond the more obvious ones, and integrity issues related to these more subtle attributes can have serious consequences. To enable the generation of actionable, data-driven hypotheses which identify specific dataset attributes likely to induce model bias, we contribute a first technique for the rigorous, quantitative screening of medical image datasets. Drawing from literature in the causal inference and information theory domains, our procedure decomposes the risks associated with dataset attributes in terms of their detectability and utility (defined as the amount of information knowing the attribute gives about a task label). To demonstrate the effectiveness and sensitivity of our method, we develop a variety of datasets with synthetically inserted artifacts with different degrees of association to the target label that allow evaluation of inherited model biases via comparison of performance against true counterfactual examples. Using these datasets and results from hundreds of trained models, we show our screening method reliably identifies nearly imperceptible bias-inducing artifacts. Lastly, we apply our method to the natural attributes of a popular skin-lesion dataset and demonstrate its success. Our approach provides a means to perform more systematic algorithmic audits and guide future data collection efforts in pursuit of safer and more reliable models.
no_new_dataset
0.708011
2304.03220
Ahsan Mahmood
Ahsan Mahmood, Junier Oliva, Martin Styner
Anomaly Detection via Gumbel Noise Score Matching
null
null
null
null
cs.LG cs.AI cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
We propose Gumbel Noise Score Matching (GNSM), a novel unsupervised method to detect anomalies in categorical data. GNSM accomplishes this by estimating the scores, i.e. the gradients of log likelihoods w.r.t.~inputs, of continuously relaxed categorical distributions. We test our method on a suite of anomaly detection tabular datasets. GNSM achieves a consistently high performance across all experiments. We further demonstrate the flexibility of GNSM by applying it to image data where the model is tasked to detect poor segmentation predictions. Images ranked anomalous by GNSM show clear segmentation failures, with the outputs of GNSM strongly correlating with segmentation metrics computed on ground-truth. We outline the score matching training objective utilized by GNSM and provide an open-source implementation of our work.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 16:52:00 GMT" } ]
2023-04-07T00:00:00
[ [ "Mahmood", "Ahsan", "" ], [ "Oliva", "Junier", "" ], [ "Styner", "Martin", "" ] ]
TITLE: Anomaly Detection via Gumbel Noise Score Matching ABSTRACT: We propose Gumbel Noise Score Matching (GNSM), a novel unsupervised method to detect anomalies in categorical data. GNSM accomplishes this by estimating the scores, i.e. the gradients of log likelihoods w.r.t.~inputs, of continuously relaxed categorical distributions. We test our method on a suite of anomaly detection tabular datasets. GNSM achieves a consistently high performance across all experiments. We further demonstrate the flexibility of GNSM by applying it to image data where the model is tasked to detect poor segmentation predictions. Images ranked anomalous by GNSM show clear segmentation failures, with the outputs of GNSM strongly correlating with segmentation metrics computed on ground-truth. We outline the score matching training objective utilized by GNSM and provide an open-source implementation of our work.
no_new_dataset
0.711205
2304.03239
Hamid Usefi
Ali Amelia, Lourdes Pena-Castillo, Hamid Usefi
Assessing the Reproducibility of Machine-learning-based Biomarker Discovery in Parkinson's Disease
20 pages, 4 figures
null
null
null
q-bio.GN cs.LG
http://creativecommons.org/licenses/by/4.0/
Genome-Wide Association Studies (GWAS) help identify genetic variations in people with diseases such as Parkinson's disease (PD), which are less common in those without the disease. Thus, GWAS data can be used to identify genetic variations associated with the disease. Feature selection and machine learning approaches can be used to analyze GWAS data and identify potential disease biomarkers. However, GWAS studies have technical variations that affect the reproducibility of identified biomarkers, such as differences in genotyping platforms and selection criteria for individuals to be genotyped. To address this issue, we collected five GWAS datasets from the database of Genotypes and Phenotypes (dbGaP) and explored several data integration strategies. We evaluated the agreement among different strategies in terms of the Single Nucleotide Polymorphisms (SNPs) that were identified as potential PD biomarkers. Our results showed a low concordance of biomarkers discovered using different datasets or integration strategies. However, we identified fifty SNPs that were identified at least twice, which could potentially serve as novel PD biomarkers. These SNPs are indirectly linked to PD in the literature but have not been directly associated with PD before. These findings open up new potential avenues of investigation.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 17:21:10 GMT" } ]
2023-04-07T00:00:00
[ [ "Amelia", "Ali", "" ], [ "Pena-Castillo", "Lourdes", "" ], [ "Usefi", "Hamid", "" ] ]
TITLE: Assessing the Reproducibility of Machine-learning-based Biomarker Discovery in Parkinson's Disease ABSTRACT: Genome-Wide Association Studies (GWAS) help identify genetic variations in people with diseases such as Parkinson's disease (PD), which are less common in those without the disease. Thus, GWAS data can be used to identify genetic variations associated with the disease. Feature selection and machine learning approaches can be used to analyze GWAS data and identify potential disease biomarkers. However, GWAS studies have technical variations that affect the reproducibility of identified biomarkers, such as differences in genotyping platforms and selection criteria for individuals to be genotyped. To address this issue, we collected five GWAS datasets from the database of Genotypes and Phenotypes (dbGaP) and explored several data integration strategies. We evaluated the agreement among different strategies in terms of the Single Nucleotide Polymorphisms (SNPs) that were identified as potential PD biomarkers. Our results showed a low concordance of biomarkers discovered using different datasets or integration strategies. However, we identified fifty SNPs that were identified at least twice, which could potentially serve as novel PD biomarkers. These SNPs are indirectly linked to PD in the literature but have not been directly associated with PD before. These findings open up new potential avenues of investigation.
no_new_dataset
0.707998
2304.03243
Daniel McDuff
Daniel McDuff, Theodore Curran, Achuta Kadambi
Synthetic Data in Healthcare
null
null
null
null
cs.AI cs.LG stat.AP
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Synthetic data are becoming a critical tool for building artificially intelligent systems. Simulators provide a way of generating data systematically and at scale. These data can then be used either exclusively, or in conjunction with real data, for training and testing systems. Synthetic data are particularly attractive in cases where the availability of ``real'' training examples might be a bottleneck. While the volume of data in healthcare is growing exponentially, creating datasets for novel tasks and/or that reflect a diverse set of conditions and causal relationships is not trivial. Furthermore, these data are highly sensitive and often patient specific. Recent research has begun to illustrate the potential for synthetic data in many areas of medicine, but no systematic review of the literature exists. In this paper, we present the cases for physical and statistical simulations for creating data and the proposed applications in healthcare and medicine. We discuss that while synthetics can promote privacy, equity, safety and continual and causal learning, they also run the risk of introducing flaws, blind spots and propagating or exaggerating biases.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 17:23:39 GMT" } ]
2023-04-07T00:00:00
[ [ "McDuff", "Daniel", "" ], [ "Curran", "Theodore", "" ], [ "Kadambi", "Achuta", "" ] ]
TITLE: Synthetic Data in Healthcare ABSTRACT: Synthetic data are becoming a critical tool for building artificially intelligent systems. Simulators provide a way of generating data systematically and at scale. These data can then be used either exclusively, or in conjunction with real data, for training and testing systems. Synthetic data are particularly attractive in cases where the availability of ``real'' training examples might be a bottleneck. While the volume of data in healthcare is growing exponentially, creating datasets for novel tasks and/or that reflect a diverse set of conditions and causal relationships is not trivial. Furthermore, these data are highly sensitive and often patient specific. Recent research has begun to illustrate the potential for synthetic data in many areas of medicine, but no systematic review of the literature exists. In this paper, we present the cases for physical and statistical simulations for creating data and the proposed applications in healthcare and medicine. We discuss that while synthetics can promote privacy, equity, safety and continual and causal learning, they also run the risk of introducing flaws, blind spots and propagating or exaggerating biases.
no_new_dataset
0.712013
2304.03265
Francesco Montagna
Francesco Montagna, Nicoletta Noceti, Lorenzo Rosasco, Kun Zhang, Francesco Locatello
Causal Discovery with Score Matching on Additive Models with Arbitrary Noise
null
2nd Conference on Causal Learning and Reasoning (CLeaR 2023)
null
null
cs.LG stat.ME
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Causal discovery methods are intrinsically constrained by the set of assumptions needed to ensure structure identifiability. Moreover additional restrictions are often imposed in order to simplify the inference task: this is the case for the Gaussian noise assumption on additive non-linear models, which is common to many causal discovery approaches. In this paper we show the shortcomings of inference under this hypothesis, analyzing the risk of edge inversion under violation of Gaussianity of the noise terms. Then, we propose a novel method for inferring the topological ordering of the variables in the causal graph, from data generated according to an additive non-linear model with a generic noise distribution. This leads to NoGAM (Not only Gaussian Additive noise Models), a causal discovery algorithm with a minimal set of assumptions and state of the art performance, experimentally benchmarked on synthetic data.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 17:50:46 GMT" } ]
2023-04-07T00:00:00
[ [ "Montagna", "Francesco", "" ], [ "Noceti", "Nicoletta", "" ], [ "Rosasco", "Lorenzo", "" ], [ "Zhang", "Kun", "" ], [ "Locatello", "Francesco", "" ] ]
TITLE: Causal Discovery with Score Matching on Additive Models with Arbitrary Noise ABSTRACT: Causal discovery methods are intrinsically constrained by the set of assumptions needed to ensure structure identifiability. Moreover additional restrictions are often imposed in order to simplify the inference task: this is the case for the Gaussian noise assumption on additive non-linear models, which is common to many causal discovery approaches. In this paper we show the shortcomings of inference under this hypothesis, analyzing the risk of edge inversion under violation of Gaussianity of the noise terms. Then, we propose a novel method for inferring the topological ordering of the variables in the causal graph, from data generated according to an additive non-linear model with a generic noise distribution. This leads to NoGAM (Not only Gaussian Additive noise Models), a causal discovery algorithm with a minimal set of assumptions and state of the art performance, experimentally benchmarked on synthetic data.
no_new_dataset
0.711212
2304.03266
Zian Wang
Zian Wang, Tianchang Shen, Jun Gao, Shengyu Huang, Jacob Munkberg, Jon Hasselgren, Zan Gojcic, Wenzheng Chen, Sanja Fidler
Neural Fields meet Explicit Geometric Representation for Inverse Rendering of Urban Scenes
CVPR 2023. Project page: https://nv-tlabs.github.io/fegr/
null
null
null
cs.CV
http://creativecommons.org/licenses/by-nc-sa/4.0/
Reconstruction and intrinsic decomposition of scenes from captured imagery would enable many applications such as relighting and virtual object insertion. Recent NeRF based methods achieve impressive fidelity of 3D reconstruction, but bake the lighting and shadows into the radiance field, while mesh-based methods that facilitate intrinsic decomposition through differentiable rendering have not yet scaled to the complexity and scale of outdoor scenes. We present a novel inverse rendering framework for large urban scenes capable of jointly reconstructing the scene geometry, spatially-varying materials, and HDR lighting from a set of posed RGB images with optional depth. Specifically, we use a neural field to account for the primary rays, and use an explicit mesh (reconstructed from the underlying neural field) for modeling secondary rays that produce higher-order lighting effects such as cast shadows. By faithfully disentangling complex geometry and materials from lighting effects, our method enables photorealistic relighting with specular and shadow effects on several outdoor datasets. Moreover, it supports physics-based scene manipulations such as virtual object insertion with ray-traced shadow casting.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 17:51:54 GMT" } ]
2023-04-07T00:00:00
[ [ "Wang", "Zian", "" ], [ "Shen", "Tianchang", "" ], [ "Gao", "Jun", "" ], [ "Huang", "Shengyu", "" ], [ "Munkberg", "Jacob", "" ], [ "Hasselgren", "Jon", "" ], [ "Gojcic", "Zan", "" ], [ "Chen", "Wenzheng", "" ], [ "Fidler", "Sanja", "" ] ]
TITLE: Neural Fields meet Explicit Geometric Representation for Inverse Rendering of Urban Scenes ABSTRACT: Reconstruction and intrinsic decomposition of scenes from captured imagery would enable many applications such as relighting and virtual object insertion. Recent NeRF based methods achieve impressive fidelity of 3D reconstruction, but bake the lighting and shadows into the radiance field, while mesh-based methods that facilitate intrinsic decomposition through differentiable rendering have not yet scaled to the complexity and scale of outdoor scenes. We present a novel inverse rendering framework for large urban scenes capable of jointly reconstructing the scene geometry, spatially-varying materials, and HDR lighting from a set of posed RGB images with optional depth. Specifically, we use a neural field to account for the primary rays, and use an explicit mesh (reconstructed from the underlying neural field) for modeling secondary rays that produce higher-order lighting effects such as cast shadows. By faithfully disentangling complex geometry and materials from lighting effects, our method enables photorealistic relighting with specular and shadow effects on several outdoor datasets. Moreover, it supports physics-based scene manipulations such as virtual object insertion with ray-traced shadow casting.
no_new_dataset
0.709831
2304.03280
Akshay Krishnan
Akshay Krishnan, Amit Raj, Xianling Zhang, Alexandra Carlson, Nathan Tseng, Sandhya Sridhar, Nikita Jaipuria, James Hays
LANe: Lighting-Aware Neural Fields for Compositional Scene Synthesis
Project website: https://lane-composition.github.io
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Neural fields have recently enjoyed great success in representing and rendering 3D scenes. However, most state-of-the-art implicit representations model static or dynamic scenes as a whole, with minor variations. Existing work on learning disentangled world and object neural fields do not consider the problem of composing objects into different world neural fields in a lighting-aware manner. We present Lighting-Aware Neural Field (LANe) for the compositional synthesis of driving scenes in a physically consistent manner. Specifically, we learn a scene representation that disentangles the static background and transient elements into a world-NeRF and class-specific object-NeRFs to allow compositional synthesis of multiple objects in the scene. Furthermore, we explicitly designed both the world and object models to handle lighting variation, which allows us to compose objects into scenes with spatially varying lighting. This is achieved by constructing a light field of the scene and using it in conjunction with a learned shader to modulate the appearance of the object NeRFs. We demonstrate the performance of our model on a synthetic dataset of diverse lighting conditions rendered with the CARLA simulator, as well as a novel real-world dataset of cars collected at different times of the day. Our approach shows that it outperforms state-of-the-art compositional scene synthesis on the challenging dataset setup, via composing object-NeRFs learned from one scene into an entirely different scene whilst still respecting the lighting variations in the novel scene. For more results, please visit our project website https://lane-composition.github.io/.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 17:59:25 GMT" } ]
2023-04-07T00:00:00
[ [ "Krishnan", "Akshay", "" ], [ "Raj", "Amit", "" ], [ "Zhang", "Xianling", "" ], [ "Carlson", "Alexandra", "" ], [ "Tseng", "Nathan", "" ], [ "Sridhar", "Sandhya", "" ], [ "Jaipuria", "Nikita", "" ], [ "Hays", "James", "" ] ]
TITLE: LANe: Lighting-Aware Neural Fields for Compositional Scene Synthesis ABSTRACT: Neural fields have recently enjoyed great success in representing and rendering 3D scenes. However, most state-of-the-art implicit representations model static or dynamic scenes as a whole, with minor variations. Existing work on learning disentangled world and object neural fields do not consider the problem of composing objects into different world neural fields in a lighting-aware manner. We present Lighting-Aware Neural Field (LANe) for the compositional synthesis of driving scenes in a physically consistent manner. Specifically, we learn a scene representation that disentangles the static background and transient elements into a world-NeRF and class-specific object-NeRFs to allow compositional synthesis of multiple objects in the scene. Furthermore, we explicitly designed both the world and object models to handle lighting variation, which allows us to compose objects into scenes with spatially varying lighting. This is achieved by constructing a light field of the scene and using it in conjunction with a learned shader to modulate the appearance of the object NeRFs. We demonstrate the performance of our model on a synthetic dataset of diverse lighting conditions rendered with the CARLA simulator, as well as a novel real-world dataset of cars collected at different times of the day. Our approach shows that it outperforms state-of-the-art compositional scene synthesis on the challenging dataset setup, via composing object-NeRFs learned from one scene into an entirely different scene whilst still respecting the lighting variations in the novel scene. For more results, please visit our project website https://lane-composition.github.io/.
new_dataset
0.718792
2304.03282
Mingyu Ding
Mingyu Ding, Yikang Shen, Lijie Fan, Zhenfang Chen, Zitian Chen, Ping Luo, Joshua B. Tenenbaum, Chuang Gan
Visual Dependency Transformers: Dependency Tree Emerges from Reversed Attention
CVPR 2023
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Humans possess a versatile mechanism for extracting structured representations of our visual world. When looking at an image, we can decompose the scene into entities and their parts as well as obtain the dependencies between them. To mimic such capability, we propose Visual Dependency Transformers (DependencyViT) that can induce visual dependencies without any labels. We achieve that with a novel neural operator called \emph{reversed attention} that can naturally capture long-range visual dependencies between image patches. Specifically, we formulate it as a dependency graph where a child token in reversed attention is trained to attend to its parent tokens and send information following a normalized probability distribution rather than gathering information in conventional self-attention. With such a design, hierarchies naturally emerge from reversed attention layers, and a dependency tree is progressively induced from leaf nodes to the root node unsupervisedly. DependencyViT offers several appealing benefits. (i) Entities and their parts in an image are represented by different subtrees, enabling part partitioning from dependencies; (ii) Dynamic visual pooling is made possible. The leaf nodes which rarely send messages can be pruned without hindering the model performance, based on which we propose the lightweight DependencyViT-Lite to reduce the computational and memory footprints; (iii) DependencyViT works well on both self- and weakly-supervised pretraining paradigms on ImageNet, and demonstrates its effectiveness on 8 datasets and 5 tasks, such as unsupervised part and saliency segmentation, recognition, and detection.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 17:59:26 GMT" } ]
2023-04-07T00:00:00
[ [ "Ding", "Mingyu", "" ], [ "Shen", "Yikang", "" ], [ "Fan", "Lijie", "" ], [ "Chen", "Zhenfang", "" ], [ "Chen", "Zitian", "" ], [ "Luo", "Ping", "" ], [ "Tenenbaum", "Joshua B.", "" ], [ "Gan", "Chuang", "" ] ]
TITLE: Visual Dependency Transformers: Dependency Tree Emerges from Reversed Attention ABSTRACT: Humans possess a versatile mechanism for extracting structured representations of our visual world. When looking at an image, we can decompose the scene into entities and their parts as well as obtain the dependencies between them. To mimic such capability, we propose Visual Dependency Transformers (DependencyViT) that can induce visual dependencies without any labels. We achieve that with a novel neural operator called \emph{reversed attention} that can naturally capture long-range visual dependencies between image patches. Specifically, we formulate it as a dependency graph where a child token in reversed attention is trained to attend to its parent tokens and send information following a normalized probability distribution rather than gathering information in conventional self-attention. With such a design, hierarchies naturally emerge from reversed attention layers, and a dependency tree is progressively induced from leaf nodes to the root node unsupervisedly. DependencyViT offers several appealing benefits. (i) Entities and their parts in an image are represented by different subtrees, enabling part partitioning from dependencies; (ii) Dynamic visual pooling is made possible. The leaf nodes which rarely send messages can be pruned without hindering the model performance, based on which we propose the lightweight DependencyViT-Lite to reduce the computational and memory footprints; (iii) DependencyViT works well on both self- and weakly-supervised pretraining paradigms on ImageNet, and demonstrates its effectiveness on 8 datasets and 5 tasks, such as unsupervised part and saliency segmentation, recognition, and detection.
no_new_dataset
0.711656
2304.03285
Hadi AlZayer
Hadi Alzayer, Abdullah Abuolaim, Leung Chun Chan, Yang Yang, Ying Chen Lou, Jia-Bin Huang, Abhishek Kar
$\text{DC}^2$: Dual-Camera Defocus Control by Learning to Refocus
CVPR 2023. See the project page at https://defocus-control.github.io
null
null
null
cs.CV
http://creativecommons.org/licenses/by/4.0/
Smartphone cameras today are increasingly approaching the versatility and quality of professional cameras through a combination of hardware and software advancements. However, fixed aperture remains a key limitation, preventing users from controlling the depth of field (DoF) of captured images. At the same time, many smartphones now have multiple cameras with different fixed apertures -- specifically, an ultra-wide camera with wider field of view and deeper DoF and a higher resolution primary camera with shallower DoF. In this work, we propose $\text{DC}^2$, a system for defocus control for synthetically varying camera aperture, focus distance and arbitrary defocus effects by fusing information from such a dual-camera system. Our key insight is to leverage real-world smartphone camera dataset by using image refocus as a proxy task for learning to control defocus. Quantitative and qualitative evaluations on real-world data demonstrate our system's efficacy where we outperform state-of-the-art on defocus deblurring, bokeh rendering, and image refocus. Finally, we demonstrate creative post-capture defocus control enabled by our method, including tilt-shift and content-based defocus effects.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 17:59:58 GMT" } ]
2023-04-07T00:00:00
[ [ "Alzayer", "Hadi", "" ], [ "Abuolaim", "Abdullah", "" ], [ "Chan", "Leung Chun", "" ], [ "Yang", "Yang", "" ], [ "Lou", "Ying Chen", "" ], [ "Huang", "Jia-Bin", "" ], [ "Kar", "Abhishek", "" ] ]
TITLE: $\text{DC}^2$: Dual-Camera Defocus Control by Learning to Refocus ABSTRACT: Smartphone cameras today are increasingly approaching the versatility and quality of professional cameras through a combination of hardware and software advancements. However, fixed aperture remains a key limitation, preventing users from controlling the depth of field (DoF) of captured images. At the same time, many smartphones now have multiple cameras with different fixed apertures -- specifically, an ultra-wide camera with wider field of view and deeper DoF and a higher resolution primary camera with shallower DoF. In this work, we propose $\text{DC}^2$, a system for defocus control for synthetically varying camera aperture, focus distance and arbitrary defocus effects by fusing information from such a dual-camera system. Our key insight is to leverage real-world smartphone camera dataset by using image refocus as a proxy task for learning to control defocus. Quantitative and qualitative evaluations on real-world data demonstrate our system's efficacy where we outperform state-of-the-art on defocus deblurring, bokeh rendering, and image refocus. Finally, we demonstrate creative post-capture defocus control enabled by our method, including tilt-shift and content-based defocus effects.
no_new_dataset
0.706418
1902.08950
Shengfan Wang
Shengfan Wang, Xin Jiang, Jie Zhao, Xiaoman Wang, Weiguo Zhou and Yunhui Liu
Efficient Fully Convolution Neural Network for Generating Pixel Wise Robotic Grasps With High Resolution Images
Submitted to ROBIO 2019
null
null
null
cs.RO
http://creativecommons.org/licenses/by-nc-sa/4.0/
This paper presents an efficient neural network model to generate robotic grasps with high resolution images. The proposed model uses fully convolution neural network to generate robotic grasps for each pixel using 400 $\times$ 400 high resolution RGB-D images. It first down-sample the images to get features and then up-sample those features to the original size of the input as well as combines local and global features from different feature maps. Compared to other regression or classification methods for detecting robotic grasps, our method looks more like the segmentation methods which solves the problem through pixel-wise ways. We use Cornell Grasp Dataset to train and evaluate the model and get high accuracy about 94.42% for image-wise and 91.02% for object-wise and fast prediction time about 8ms. We also demonstrate that without training on the multiple objects dataset, our model can directly output robotic grasps candidates for different objects because of the pixel wise implementation.
[ { "version": "v1", "created": "Sun, 24 Feb 2019 14:12:03 GMT" }, { "version": "v2", "created": "Wed, 18 Sep 2019 08:38:04 GMT" } ]
2023-04-06T00:00:00
[ [ "Wang", "Shengfan", "" ], [ "Jiang", "Xin", "" ], [ "Zhao", "Jie", "" ], [ "Wang", "Xiaoman", "" ], [ "Zhou", "Weiguo", "" ], [ "Liu", "Yunhui", "" ] ]
TITLE: Efficient Fully Convolution Neural Network for Generating Pixel Wise Robotic Grasps With High Resolution Images ABSTRACT: This paper presents an efficient neural network model to generate robotic grasps with high resolution images. The proposed model uses fully convolution neural network to generate robotic grasps for each pixel using 400 $\times$ 400 high resolution RGB-D images. It first down-sample the images to get features and then up-sample those features to the original size of the input as well as combines local and global features from different feature maps. Compared to other regression or classification methods for detecting robotic grasps, our method looks more like the segmentation methods which solves the problem through pixel-wise ways. We use Cornell Grasp Dataset to train and evaluate the model and get high accuracy about 94.42% for image-wise and 91.02% for object-wise and fast prediction time about 8ms. We also demonstrate that without training on the multiple objects dataset, our model can directly output robotic grasps candidates for different objects because of the pixel wise implementation.
no_new_dataset
0.713394
1905.01583
Qi Wang
Yuan Yuan, Zhitong Xiong, and Qi Wang
VSSA-NET: Vertical Spatial Sequence Attention Network for Traffic Sign Detection
null
null
10.1109/TIP.2019.2896952
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Although traffic sign detection has been studied for years and great progress has been made with the rise of deep learning technique, there are still many problems remaining to be addressed. For complicated real-world traffic scenes, there are two main challenges. Firstly, traffic signs are usually small size objects, which makes it more difficult to detect than large ones; Secondly, it is hard to distinguish false targets which resemble real traffic signs in complex street scenes without context information. To handle these problems, we propose a novel end-to-end deep learning method for traffic sign detection in complex environments. Our contributions are as follows: 1) We propose a multi-resolution feature fusion network architecture which exploits densely connected deconvolution layers with skip connections, and can learn more effective features for the small size object; 2) We frame the traffic sign detection as a spatial sequence classification and regression task, and propose a vertical spatial sequence attention (VSSA) module to gain more context information for better detection performance. To comprehensively evaluate the proposed method, we do experiments on several traffic sign datasets as well as the general object detection dataset and the results have shown the effectiveness of our proposed method.
[ { "version": "v1", "created": "Sun, 5 May 2019 02:16:43 GMT" } ]
2023-04-06T00:00:00
[ [ "Yuan", "Yuan", "" ], [ "Xiong", "Zhitong", "" ], [ "Wang", "Qi", "" ] ]
TITLE: VSSA-NET: Vertical Spatial Sequence Attention Network for Traffic Sign Detection ABSTRACT: Although traffic sign detection has been studied for years and great progress has been made with the rise of deep learning technique, there are still many problems remaining to be addressed. For complicated real-world traffic scenes, there are two main challenges. Firstly, traffic signs are usually small size objects, which makes it more difficult to detect than large ones; Secondly, it is hard to distinguish false targets which resemble real traffic signs in complex street scenes without context information. To handle these problems, we propose a novel end-to-end deep learning method for traffic sign detection in complex environments. Our contributions are as follows: 1) We propose a multi-resolution feature fusion network architecture which exploits densely connected deconvolution layers with skip connections, and can learn more effective features for the small size object; 2) We frame the traffic sign detection as a spatial sequence classification and regression task, and propose a vertical spatial sequence attention (VSSA) module to gain more context information for better detection performance. To comprehensively evaluate the proposed method, we do experiments on several traffic sign datasets as well as the general object detection dataset and the results have shown the effectiveness of our proposed method.
no_new_dataset
0.711262
1912.11368
Badong Chen
Yunfei Zheng, Badong Chen, Shiyuan Wang, and Weiqun Wang
Broad Learning System Based on Maximum Correntropy Criterion
null
null
null
null
cs.LG stat.ML
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
As an effective and efficient discriminative learning method, Broad Learning System (BLS) has received increasing attention due to its outstanding performance in various regression and classification problems. However, the standard BLS is derived under the minimum mean square error (MMSE) criterion, which is, of course, not always a good choice due to its sensitivity to outliers. To enhance the robustness of BLS, we propose in this work to adopt the maximum correntropy criterion (MCC) to train the output weights, obtaining a correntropy based broad learning system (C-BLS). Thanks to the inherent superiorities of MCC, the proposed C-BLS is expected to achieve excellent robustness to outliers while maintaining the original performance of the standard BLS in Gaussian or noise-free environment. In addition, three alternative incremental learning algorithms, derived from a weighted regularized least-squares solution rather than pseudoinverse formula, for C-BLS are developed.With the incremental learning algorithms, the system can be updated quickly without the entire retraining process from the beginning, when some new samples arrive or the network deems to be expanded. Experiments on various regression and classification datasets are reported to demonstrate the desirable performance of the new methods.
[ { "version": "v1", "created": "Tue, 24 Dec 2019 13:56:55 GMT" } ]
2023-04-06T00:00:00
[ [ "Zheng", "Yunfei", "" ], [ "Chen", "Badong", "" ], [ "Wang", "Shiyuan", "" ], [ "Wang", "Weiqun", "" ] ]
TITLE: Broad Learning System Based on Maximum Correntropy Criterion ABSTRACT: As an effective and efficient discriminative learning method, Broad Learning System (BLS) has received increasing attention due to its outstanding performance in various regression and classification problems. However, the standard BLS is derived under the minimum mean square error (MMSE) criterion, which is, of course, not always a good choice due to its sensitivity to outliers. To enhance the robustness of BLS, we propose in this work to adopt the maximum correntropy criterion (MCC) to train the output weights, obtaining a correntropy based broad learning system (C-BLS). Thanks to the inherent superiorities of MCC, the proposed C-BLS is expected to achieve excellent robustness to outliers while maintaining the original performance of the standard BLS in Gaussian or noise-free environment. In addition, three alternative incremental learning algorithms, derived from a weighted regularized least-squares solution rather than pseudoinverse formula, for C-BLS are developed.With the incremental learning algorithms, the system can be updated quickly without the entire retraining process from the beginning, when some new samples arrive or the network deems to be expanded. Experiments on various regression and classification datasets are reported to demonstrate the desirable performance of the new methods.
no_new_dataset
0.710597
2007.10631
Xueping Wang
Xueping Wang, Sujoy Paul, Dripta S. Raychaudhuri, Min Liu, Yaonan Wang and Amit K. Roy-Chowdhury
Learning Person Re-identification Models from Videos with Weak Supervision
null
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Most person re-identification methods, being supervised techniques, suffer from the burden of massive annotation requirement. Unsupervised methods overcome this need for labeled data, but perform poorly compared to the supervised alternatives. In order to cope with this issue, we introduce the problem of learning person re-identification models from videos with weak supervision. The weak nature of the supervision arises from the requirement of video-level labels, i.e. person identities who appear in the video, in contrast to the more precise framelevel annotations. Towards this goal, we propose a multiple instance attention learning framework for person re-identification using such video-level labels. Specifically, we first cast the video person re-identification task into a multiple instance learning setting, in which person images in a video are collected into a bag. The relations between videos with similar labels can be utilized to identify persons, on top of that, we introduce a co-person attention mechanism which mines the similarity correlations between videos with person identities in common. The attention weights are obtained based on all person images instead of person tracklets in a video, making our learned model less affected by noisy annotations. Extensive experiments demonstrate the superiority of the proposed method over the related methods on two weakly labeled person re-identification datasets.
[ { "version": "v1", "created": "Tue, 21 Jul 2020 07:23:32 GMT" } ]
2023-04-06T00:00:00
[ [ "Wang", "Xueping", "" ], [ "Paul", "Sujoy", "" ], [ "Raychaudhuri", "Dripta S.", "" ], [ "Liu", "Min", "" ], [ "Wang", "Yaonan", "" ], [ "Roy-Chowdhury", "Amit K.", "" ] ]
TITLE: Learning Person Re-identification Models from Videos with Weak Supervision ABSTRACT: Most person re-identification methods, being supervised techniques, suffer from the burden of massive annotation requirement. Unsupervised methods overcome this need for labeled data, but perform poorly compared to the supervised alternatives. In order to cope with this issue, we introduce the problem of learning person re-identification models from videos with weak supervision. The weak nature of the supervision arises from the requirement of video-level labels, i.e. person identities who appear in the video, in contrast to the more precise framelevel annotations. Towards this goal, we propose a multiple instance attention learning framework for person re-identification using such video-level labels. Specifically, we first cast the video person re-identification task into a multiple instance learning setting, in which person images in a video are collected into a bag. The relations between videos with similar labels can be utilized to identify persons, on top of that, we introduce a co-person attention mechanism which mines the similarity correlations between videos with person identities in common. The attention weights are obtained based on all person images instead of person tracklets in a video, making our learned model less affected by noisy annotations. Extensive experiments demonstrate the superiority of the proposed method over the related methods on two weakly labeled person re-identification datasets.
no_new_dataset
0.710791
2009.02571
Rahul Dubey Dr
Param Khakhar and, Rahul Kumar Dubey
The Integrity of Machine Learning Algorithms against Software Defect Prediction
7 pages, 4 figures
null
null
null
cs.SE cs.LG stat.ML
http://creativecommons.org/licenses/by/4.0/
The increased computerization in recent years has resulted in the production of a variety of different software, however measures need to be taken to ensure that the produced software isn't defective. Many researchers have worked in this area and have developed different Machine Learning-based approaches that predict whether the software is defective or not. This issue can't be resolved simply by using different conventional classifiers because the dataset is highly imbalanced i.e the number of defective samples detected is extremely less as compared to the number of non-defective samples. Therefore, to address this issue, certain sophisticated methods are required. The different methods developed by the researchers can be broadly classified into Resampling based methods, Cost-sensitive learning-based methods, and Ensemble Learning. Among these methods. This report analyses the performance of the Online Sequential Extreme Learning Machine (OS-ELM) proposed by Liang et.al. against several classifiers such as Logistic Regression, Support Vector Machine, Random Forest, and Na\"ive Bayes after oversampling the data. OS-ELM trains faster than conventional deep neural networks and it always converges to the globally optimal solution. A comparison is performed on the original dataset as well as the over-sampled data set. The oversampling technique used is Cluster-based Over-Sampling with Noise Filtering. This technique is better than several state-of-the-art techniques for oversampling. The analysis is carried out on 3 projects KC1, PC4 and PC3 carried out by the NASA group. The metrics used for measurement are recall and balanced accuracy. The results are higher for OS-ELM as compared to other classifiers in both scenarios.
[ { "version": "v1", "created": "Sat, 5 Sep 2020 17:26:56 GMT" } ]
2023-04-06T00:00:00
[ [ "and", "Param Khakhar", "" ], [ "Dubey", "Rahul Kumar", "" ] ]
TITLE: The Integrity of Machine Learning Algorithms against Software Defect Prediction ABSTRACT: The increased computerization in recent years has resulted in the production of a variety of different software, however measures need to be taken to ensure that the produced software isn't defective. Many researchers have worked in this area and have developed different Machine Learning-based approaches that predict whether the software is defective or not. This issue can't be resolved simply by using different conventional classifiers because the dataset is highly imbalanced i.e the number of defective samples detected is extremely less as compared to the number of non-defective samples. Therefore, to address this issue, certain sophisticated methods are required. The different methods developed by the researchers can be broadly classified into Resampling based methods, Cost-sensitive learning-based methods, and Ensemble Learning. Among these methods. This report analyses the performance of the Online Sequential Extreme Learning Machine (OS-ELM) proposed by Liang et.al. against several classifiers such as Logistic Regression, Support Vector Machine, Random Forest, and Na\"ive Bayes after oversampling the data. OS-ELM trains faster than conventional deep neural networks and it always converges to the globally optimal solution. A comparison is performed on the original dataset as well as the over-sampled data set. The oversampling technique used is Cluster-based Over-Sampling with Noise Filtering. This technique is better than several state-of-the-art techniques for oversampling. The analysis is carried out on 3 projects KC1, PC4 and PC3 carried out by the NASA group. The metrics used for measurement are recall and balanced accuracy. The results are higher for OS-ELM as compared to other classifiers in both scenarios.
no_new_dataset
0.710641
2009.10434
Haoyu Tang
Haoyu Tang, Jihua Zhu, Meng Liu, Zan Gao, and Zhiyong Cheng
Frame-wise Cross-modal Matching for Video Moment Retrieval
12 pages; accepted by IEEE TMM
IEEE Transactions on Multimedia 2021
10.1109/TMM.2021.3063631
null
cs.CV cs.MM
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Video moment retrieval targets at retrieving a moment in a video for a given language query. The challenges of this task include 1) the requirement of localizing the relevant moment in an untrimmed video, and 2) bridging the semantic gap between textual query and video contents. To tackle those problems, early approaches adopt the sliding window or uniform sampling to collect video clips first and then match each clip with the query. Obviously, these strategies are time-consuming and often lead to unsatisfied accuracy in localization due to the unpredictable length of the golden moment. To avoid the limitations, researchers recently attempt to directly predict the relevant moment boundaries without the requirement to generate video clips first. One mainstream approach is to generate a multimodal feature vector for the target query and video frames (e.g., concatenation) and then use a regression approach upon the multimodal feature vector for boundary detection. Although some progress has been achieved by this approach, we argue that those methods have not well captured the cross-modal interactions between the query and video frames. In this paper, we propose an Attentive Cross-modal Relevance Matching (ACRM) model which predicts the temporal boundaries based on an interaction modeling. In addition, an attention module is introduced to assign higher weights to query words with richer semantic cues, which are considered to be more important for finding relevant video contents. Another contribution is that we propose an additional predictor to utilize the internal frames in the model training to improve the localization accuracy. Extensive experiments on two datasets TACoS and Charades-STA demonstrate the superiority of our method over several state-of-the-art methods. Ablation studies have been also conducted to examine the effectiveness of different modules in our ACRM model.
[ { "version": "v1", "created": "Tue, 22 Sep 2020 10:25:41 GMT" }, { "version": "v2", "created": "Thu, 22 Jul 2021 07:32:20 GMT" } ]
2023-04-06T00:00:00
[ [ "Tang", "Haoyu", "" ], [ "Zhu", "Jihua", "" ], [ "Liu", "Meng", "" ], [ "Gao", "Zan", "" ], [ "Cheng", "Zhiyong", "" ] ]
TITLE: Frame-wise Cross-modal Matching for Video Moment Retrieval ABSTRACT: Video moment retrieval targets at retrieving a moment in a video for a given language query. The challenges of this task include 1) the requirement of localizing the relevant moment in an untrimmed video, and 2) bridging the semantic gap between textual query and video contents. To tackle those problems, early approaches adopt the sliding window or uniform sampling to collect video clips first and then match each clip with the query. Obviously, these strategies are time-consuming and often lead to unsatisfied accuracy in localization due to the unpredictable length of the golden moment. To avoid the limitations, researchers recently attempt to directly predict the relevant moment boundaries without the requirement to generate video clips first. One mainstream approach is to generate a multimodal feature vector for the target query and video frames (e.g., concatenation) and then use a regression approach upon the multimodal feature vector for boundary detection. Although some progress has been achieved by this approach, we argue that those methods have not well captured the cross-modal interactions between the query and video frames. In this paper, we propose an Attentive Cross-modal Relevance Matching (ACRM) model which predicts the temporal boundaries based on an interaction modeling. In addition, an attention module is introduced to assign higher weights to query words with richer semantic cues, which are considered to be more important for finding relevant video contents. Another contribution is that we propose an additional predictor to utilize the internal frames in the model training to improve the localization accuracy. Extensive experiments on two datasets TACoS and Charades-STA demonstrate the superiority of our method over several state-of-the-art methods. Ablation studies have been also conducted to examine the effectiveness of different modules in our ACRM model.
no_new_dataset
0.710622
2009.13964
Yusheng Su
Yusheng Su, Xu Han, Zhengyan Zhang, Peng Li, Zhiyuan Liu, Yankai Lin, Jie Zhou and Maosong Sun
CokeBERT: Contextual Knowledge Selection and Embedding towards Enhanced Pre-Trained Language Models
null
AI Open 2021
10.1016/j.aiopen.2021.06.004
null
cs.CL cs.AI
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Several recent efforts have been devoted to enhancing pre-trained language models (PLMs) by utilizing extra heterogeneous knowledge in knowledge graphs (KGs) and achieved consistent improvements on various knowledge-driven NLP tasks. However, most of these knowledge-enhanced PLMs embed static sub-graphs of KGs ("knowledge context"), regardless of that the knowledge required by PLMs may change dynamically according to specific text ("textual context"). In this paper, we propose a novel framework named Coke to dynamically select contextual knowledge and embed knowledge context according to textual context for PLMs, which can avoid the effect of redundant and ambiguous knowledge in KGs that cannot match the input text. Our experimental results show that Coke outperforms various baselines on typical knowledge-driven NLP tasks, indicating the effectiveness of utilizing dynamic knowledge context for language understanding. Besides the performance improvements, the dynamically selected knowledge in Coke can describe the semantics of text-related knowledge in a more interpretable form than the conventional PLMs. Our source code and datasets will be available to provide more details for Coke.
[ { "version": "v1", "created": "Tue, 29 Sep 2020 12:29:04 GMT" }, { "version": "v2", "created": "Wed, 30 Sep 2020 09:31:29 GMT" }, { "version": "v3", "created": "Thu, 1 Oct 2020 09:04:48 GMT" }, { "version": "v4", "created": "Sat, 5 Dec 2020 15:25:11 GMT" }, { "version": "v5", "created": "Wed, 5 Apr 2023 07:55:56 GMT" } ]
2023-04-06T00:00:00
[ [ "Su", "Yusheng", "" ], [ "Han", "Xu", "" ], [ "Zhang", "Zhengyan", "" ], [ "Li", "Peng", "" ], [ "Liu", "Zhiyuan", "" ], [ "Lin", "Yankai", "" ], [ "Zhou", "Jie", "" ], [ "Sun", "Maosong", "" ] ]
TITLE: CokeBERT: Contextual Knowledge Selection and Embedding towards Enhanced Pre-Trained Language Models ABSTRACT: Several recent efforts have been devoted to enhancing pre-trained language models (PLMs) by utilizing extra heterogeneous knowledge in knowledge graphs (KGs) and achieved consistent improvements on various knowledge-driven NLP tasks. However, most of these knowledge-enhanced PLMs embed static sub-graphs of KGs ("knowledge context"), regardless of that the knowledge required by PLMs may change dynamically according to specific text ("textual context"). In this paper, we propose a novel framework named Coke to dynamically select contextual knowledge and embed knowledge context according to textual context for PLMs, which can avoid the effect of redundant and ambiguous knowledge in KGs that cannot match the input text. Our experimental results show that Coke outperforms various baselines on typical knowledge-driven NLP tasks, indicating the effectiveness of utilizing dynamic knowledge context for language understanding. Besides the performance improvements, the dynamically selected knowledge in Coke can describe the semantics of text-related knowledge in a more interpretable form than the conventional PLMs. Our source code and datasets will be available to provide more details for Coke.
no_new_dataset
0.706589
2010.04971
Issa Annamoradnejad
Navid Khezrian, Jafar Habibi, Issa Annamoradnejad
Tag Recommendation for Online Q&A Communities based on BERT Pre-Training Technique
5 pages, initial results
The CSI Journal on Computer Science and Engineering, 2021, vol. 18, no. 2
null
null
cs.CL cs.IR
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Online Q&A and open source communities use tags and keywords to index, categorize, and search for specific content. The most obvious advantage of tag recommendation is the correct classification of information. In this study, we used the BERT pre-training technique in tag recommendation task for online Q&A and open-source communities for the first time. Our evaluation on freecode datasets show that the proposed method, called TagBERT, is more accurate compared to deep learning and other baseline methods. Moreover, our model achieved a high stability by solving the problem of previous researches, where increasing the number of tag recommendations significantly reduced model performance.
[ { "version": "v1", "created": "Sat, 10 Oct 2020 10:52:22 GMT" } ]
2023-04-06T00:00:00
[ [ "Khezrian", "Navid", "" ], [ "Habibi", "Jafar", "" ], [ "Annamoradnejad", "Issa", "" ] ]
TITLE: Tag Recommendation for Online Q&A Communities based on BERT Pre-Training Technique ABSTRACT: Online Q&A and open source communities use tags and keywords to index, categorize, and search for specific content. The most obvious advantage of tag recommendation is the correct classification of information. In this study, we used the BERT pre-training technique in tag recommendation task for online Q&A and open-source communities for the first time. Our evaluation on freecode datasets show that the proposed method, called TagBERT, is more accurate compared to deep learning and other baseline methods. Moreover, our model achieved a high stability by solving the problem of previous researches, where increasing the number of tag recommendations significantly reduced model performance.
no_new_dataset
0.711481
2109.04013
Yuwen Li
Yuwen Li and Ludmil Zikatanov
New stabilized $P_1\times P_0$ finite element methods for nearly inviscid and incompressible flows
null
Computer Methods in Applied Mechanics and Engineering 393 (2022)
10.1016/j.cma.2022.114815
null
math.NA cs.NA
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
This work proposes a new stabilized $P_1\times P_0$ finite element method for solving the incompressible Navier--Stokes equations. The numerical scheme is based on a reduced Bernardi--Raugel element with statically condensed face bubbles and is pressure-robust in the small viscosity regime. For the Stokes problem, an error estimate uniform with respect to the kinematic viscosity is shown. For the Navier--Stokes equation, the nonlinear convection term is discretized using an edge-averaged finite element method. In comparison with classical schemes, the proposed method does not require tuning of parameters and is validated for competitiveness on several benchmark problems in 2 and 3 dimensional space.
[ { "version": "v1", "created": "Thu, 9 Sep 2021 03:20:21 GMT" }, { "version": "v2", "created": "Mon, 13 Sep 2021 14:19:39 GMT" } ]
2023-04-06T00:00:00
[ [ "Li", "Yuwen", "" ], [ "Zikatanov", "Ludmil", "" ] ]
TITLE: New stabilized $P_1\times P_0$ finite element methods for nearly inviscid and incompressible flows ABSTRACT: This work proposes a new stabilized $P_1\times P_0$ finite element method for solving the incompressible Navier--Stokes equations. The numerical scheme is based on a reduced Bernardi--Raugel element with statically condensed face bubbles and is pressure-robust in the small viscosity regime. For the Stokes problem, an error estimate uniform with respect to the kinematic viscosity is shown. For the Navier--Stokes equation, the nonlinear convection term is discretized using an edge-averaged finite element method. In comparison with classical schemes, the proposed method does not require tuning of parameters and is validated for competitiveness on several benchmark problems in 2 and 3 dimensional space.
no_new_dataset
0.709453
2109.07704
Yucheng Ding
Yucheng Ding, Chaoyue Niu, Fan Wu, Shaojie Tang, Chengfei Lv, Yanghe Feng, Guihai Chen
Federated Submodel Optimization for Hot and Cold Data Features
null
null
null
null
cs.LG cs.AI
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
We study practical data characteristics underlying federated learning, where non-i.i.d. data from clients have sparse features, and a certain client's local data normally involves only a small part of the full model, called a submodel. Due to data sparsity, the classical federated averaging (FedAvg) algorithm or its variants will be severely slowed down, because when updating the global model, each client's zero update of the full model excluding its submodel is inaccurately aggregated. Therefore, we propose federated submodel averaging (FedSubAvg), ensuring that the expectation of the global update of each model parameter is equal to the average of the local updates of the clients who involve it. We theoretically proved the convergence rate of FedSubAvg by deriving an upper bound under a new metric called the element-wise gradient norm. In particular, this new metric can characterize the convergence of federated optimization over sparse data, while the conventional metric of squared gradient norm used in FedAvg and its variants cannot. We extensively evaluated FedSubAvg over both public and industrial datasets. The evaluation results demonstrate that FedSubAvg significantly outperforms FedAvg and its variants.
[ { "version": "v1", "created": "Thu, 16 Sep 2021 03:54:54 GMT" }, { "version": "v2", "created": "Tue, 28 Sep 2021 02:45:16 GMT" }, { "version": "v3", "created": "Sat, 5 Feb 2022 13:43:47 GMT" }, { "version": "v4", "created": "Wed, 5 Apr 2023 12:38:32 GMT" } ]
2023-04-06T00:00:00
[ [ "Ding", "Yucheng", "" ], [ "Niu", "Chaoyue", "" ], [ "Wu", "Fan", "" ], [ "Tang", "Shaojie", "" ], [ "Lv", "Chengfei", "" ], [ "Feng", "Yanghe", "" ], [ "Chen", "Guihai", "" ] ]
TITLE: Federated Submodel Optimization for Hot and Cold Data Features ABSTRACT: We study practical data characteristics underlying federated learning, where non-i.i.d. data from clients have sparse features, and a certain client's local data normally involves only a small part of the full model, called a submodel. Due to data sparsity, the classical federated averaging (FedAvg) algorithm or its variants will be severely slowed down, because when updating the global model, each client's zero update of the full model excluding its submodel is inaccurately aggregated. Therefore, we propose federated submodel averaging (FedSubAvg), ensuring that the expectation of the global update of each model parameter is equal to the average of the local updates of the clients who involve it. We theoretically proved the convergence rate of FedSubAvg by deriving an upper bound under a new metric called the element-wise gradient norm. In particular, this new metric can characterize the convergence of federated optimization over sparse data, while the conventional metric of squared gradient norm used in FedAvg and its variants cannot. We extensively evaluated FedSubAvg over both public and industrial datasets. The evaluation results demonstrate that FedSubAvg significantly outperforms FedAvg and its variants.
no_new_dataset
0.708042
2110.15610
Yiheng Liu
Yiheng Liu, Wengang Zhou, Qiaokang Xie, Houqiang Li
Unsupervised Person Re-Identification with Wireless Positioning under Weak Scene Labeling
Accepted by TPAMI 2022
null
null
null
cs.CV
http://creativecommons.org/licenses/by/4.0/
Existing unsupervised person re-identification methods only rely on visual clues to match pedestrians under different cameras. Since visual data is essentially susceptible to occlusion, blur, clothing changes, etc., a promising solution is to introduce heterogeneous data to make up for the defect of visual data. Some works based on full-scene labeling introduce wireless positioning to assist cross-domain person re-identification, but their GPS labeling of entire monitoring scenes is laborious. To this end, we propose to explore unsupervised person re-identification with both visual data and wireless positioning trajectories under weak scene labeling, in which we only need to know the locations of the cameras. Specifically, we propose a novel unsupervised multimodal training framework (UMTF), which models the complementarity of visual data and wireless information. Our UMTF contains a multimodal data association strategy (MMDA) and a multimodal graph neural network (MMGN). MMDA explores potential data associations in unlabeled multimodal data, while MMGN propagates multimodal messages in the video graph based on the adjacency matrix learned from histogram statistics of wireless data. Thanks to the robustness of the wireless data to visual noise and the collaboration of various modules, UMTF is capable of learning a model free of the human label on data. Extensive experimental results conducted on two challenging datasets, i.e., WP-ReID and DukeMTMC-VideoReID demonstrate the effectiveness of the proposed method.
[ { "version": "v1", "created": "Fri, 29 Oct 2021 08:25:44 GMT" }, { "version": "v2", "created": "Wed, 5 Apr 2023 11:07:23 GMT" } ]
2023-04-06T00:00:00
[ [ "Liu", "Yiheng", "" ], [ "Zhou", "Wengang", "" ], [ "Xie", "Qiaokang", "" ], [ "Li", "Houqiang", "" ] ]
TITLE: Unsupervised Person Re-Identification with Wireless Positioning under Weak Scene Labeling ABSTRACT: Existing unsupervised person re-identification methods only rely on visual clues to match pedestrians under different cameras. Since visual data is essentially susceptible to occlusion, blur, clothing changes, etc., a promising solution is to introduce heterogeneous data to make up for the defect of visual data. Some works based on full-scene labeling introduce wireless positioning to assist cross-domain person re-identification, but their GPS labeling of entire monitoring scenes is laborious. To this end, we propose to explore unsupervised person re-identification with both visual data and wireless positioning trajectories under weak scene labeling, in which we only need to know the locations of the cameras. Specifically, we propose a novel unsupervised multimodal training framework (UMTF), which models the complementarity of visual data and wireless information. Our UMTF contains a multimodal data association strategy (MMDA) and a multimodal graph neural network (MMGN). MMDA explores potential data associations in unlabeled multimodal data, while MMGN propagates multimodal messages in the video graph based on the adjacency matrix learned from histogram statistics of wireless data. Thanks to the robustness of the wireless data to visual noise and the collaboration of various modules, UMTF is capable of learning a model free of the human label on data. Extensive experimental results conducted on two challenging datasets, i.e., WP-ReID and DukeMTMC-VideoReID demonstrate the effectiveness of the proposed method.
no_new_dataset
0.710038
2111.10221
Luojun Lin
Luojun Lin, Han Xie, Zhishu Sun, Weijie Chen, Wenxi Liu, Yuanlong Yu, Lei Zhang
Semi-Supervised Domain Generalization with Evolving Intermediate Domain
13 pages, 13 figures
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Domain Generalization (DG) aims to generalize a model trained on multiple source domains to an unseen target domain. The source domains always require precise annotations, which can be cumbersome or even infeasible to obtain in practice due to the vast amount of data involved. Web data, however, offers an opportunity to access large amounts of unlabeled data with rich style information, which can be leveraged to improve DG. From this perspective, we introduce a novel paradigm of DG, termed as Semi-Supervised Domain Generalization (SSDG), to explore how the labeled and unlabeled source domains can interact, and establish two settings, including the close-set and open-set SSDG. The close-set SSDG is based on existing public DG datasets, while the open-set SSDG, built on the newly-collected web-crawled datasets, presents a novel yet realistic challenge that pushes the limits of current technologies. A natural approach of SSDG is to transfer knowledge from labeled data to unlabeled data via pseudo labeling, and train the model on both labeled and pseudo-labeled data for generalization. Since there are conflicting goals between domain-oriented pseudo labeling and out-of-domain generalization, we develop a pseudo labeling phase and a generalization phase independently for SSDG. Unfortunately, due to the large domain gap, the pseudo labels provided in the pseudo labeling phase inevitably contain noise, which has negative affect on the subsequent generalization phase. Therefore, to improve the quality of pseudo labels and further enhance generalizability, we propose a cyclic learning framework to encourage a positive feedback between these two phases, utilizing an evolving intermediate domain that bridges the labeled and unlabeled domains in a curriculum learning manner...
[ { "version": "v1", "created": "Fri, 19 Nov 2021 13:55:57 GMT" }, { "version": "v2", "created": "Fri, 27 May 2022 08:45:26 GMT" }, { "version": "v3", "created": "Wed, 5 Apr 2023 07:28:18 GMT" } ]
2023-04-06T00:00:00
[ [ "Lin", "Luojun", "" ], [ "Xie", "Han", "" ], [ "Sun", "Zhishu", "" ], [ "Chen", "Weijie", "" ], [ "Liu", "Wenxi", "" ], [ "Yu", "Yuanlong", "" ], [ "Zhang", "Lei", "" ] ]
TITLE: Semi-Supervised Domain Generalization with Evolving Intermediate Domain ABSTRACT: Domain Generalization (DG) aims to generalize a model trained on multiple source domains to an unseen target domain. The source domains always require precise annotations, which can be cumbersome or even infeasible to obtain in practice due to the vast amount of data involved. Web data, however, offers an opportunity to access large amounts of unlabeled data with rich style information, which can be leveraged to improve DG. From this perspective, we introduce a novel paradigm of DG, termed as Semi-Supervised Domain Generalization (SSDG), to explore how the labeled and unlabeled source domains can interact, and establish two settings, including the close-set and open-set SSDG. The close-set SSDG is based on existing public DG datasets, while the open-set SSDG, built on the newly-collected web-crawled datasets, presents a novel yet realistic challenge that pushes the limits of current technologies. A natural approach of SSDG is to transfer knowledge from labeled data to unlabeled data via pseudo labeling, and train the model on both labeled and pseudo-labeled data for generalization. Since there are conflicting goals between domain-oriented pseudo labeling and out-of-domain generalization, we develop a pseudo labeling phase and a generalization phase independently for SSDG. Unfortunately, due to the large domain gap, the pseudo labels provided in the pseudo labeling phase inevitably contain noise, which has negative affect on the subsequent generalization phase. Therefore, to improve the quality of pseudo labels and further enhance generalizability, we propose a cyclic learning framework to encourage a positive feedback between these two phases, utilizing an evolving intermediate domain that bridges the labeled and unlabeled domains in a curriculum learning manner...
no_new_dataset
0.710038
2112.00250
Zhonghao Chen
Hongmin Gao, Zhonghao Chen, and Chenming Li
Shallow Network Based on Depthwise Over-Parameterized Convolution for Hyperspectral Image Classification
null
null
10.1109/LGRS.2021.3133598
null
cs.CV eess.IV
http://creativecommons.org/licenses/by/4.0/
Recently, convolutional neural network (CNN) techniques have gained popularity as a tool for hyperspectral image classification (HSIC). To improve the feature extraction efficiency of HSIC under the condition of limited samples, the current methods generally use deep models with plenty of layers. However, deep network models are prone to overfitting and gradient vanishing problems when samples are limited. In addition, the spatial resolution decreases severely with deeper depth, which is very detrimental to spatial edge feature extraction. Therefore, this letter proposes a shallow model for HSIC, which is called depthwise over-parameterized convolutional neural network (DOCNN). To ensure the effective extraction of the shallow model, the depthwise over-parameterized convolution (DO-Conv) kernel is introduced to extract the discriminative features. The depthwise over-parameterized Convolution kernel is composed of a standard convolution kernel and a depthwise convolution kernel, which can extract the spatial feature of the different channels individually and fuse the spatial features of the whole channels simultaneously. Moreover, to further reduce the loss of spatial edge features due to the convolution operation, a dense residual connection (DRC) structure is proposed to apply to the feature extraction part of the whole network. Experimental results obtained from three benchmark data sets show that the proposed method outperforms other state-of-the-art methods in terms of classification accuracy and computational efficiency.
[ { "version": "v1", "created": "Wed, 1 Dec 2021 03:10:02 GMT" } ]
2023-04-06T00:00:00
[ [ "Gao", "Hongmin", "" ], [ "Chen", "Zhonghao", "" ], [ "Li", "Chenming", "" ] ]
TITLE: Shallow Network Based on Depthwise Over-Parameterized Convolution for Hyperspectral Image Classification ABSTRACT: Recently, convolutional neural network (CNN) techniques have gained popularity as a tool for hyperspectral image classification (HSIC). To improve the feature extraction efficiency of HSIC under the condition of limited samples, the current methods generally use deep models with plenty of layers. However, deep network models are prone to overfitting and gradient vanishing problems when samples are limited. In addition, the spatial resolution decreases severely with deeper depth, which is very detrimental to spatial edge feature extraction. Therefore, this letter proposes a shallow model for HSIC, which is called depthwise over-parameterized convolutional neural network (DOCNN). To ensure the effective extraction of the shallow model, the depthwise over-parameterized convolution (DO-Conv) kernel is introduced to extract the discriminative features. The depthwise over-parameterized Convolution kernel is composed of a standard convolution kernel and a depthwise convolution kernel, which can extract the spatial feature of the different channels individually and fuse the spatial features of the whole channels simultaneously. Moreover, to further reduce the loss of spatial edge features due to the convolution operation, a dense residual connection (DRC) structure is proposed to apply to the feature extraction part of the whole network. Experimental results obtained from three benchmark data sets show that the proposed method outperforms other state-of-the-art methods in terms of classification accuracy and computational efficiency.
no_new_dataset
0.71086
2202.03086
Mathieu De Coster
Mathieu De Coster, Dimitar Shterionov, Mieke Van Herreweghe, Joni Dambre
Machine Translation from Signed to Spoken Languages: State of the Art and Challenges
This is the version of the article submitted to peer review to Universal Access in the Information Society. Please refer to "De Coster, M., Shterionov, D., Van Herreweghe, M. et al. Machine translation from signed to spoken languages: state of the art and challenges. Univ Access Inf Soc (2023)." for the published and updated version
null
10.1007/s10209-023-00992-1
null
cs.CL
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Automatic translation from signed to spoken languages is an interdisciplinary research domain, lying on the intersection of computer vision, machine translation and linguistics. Nevertheless, research in this domain is performed mostly by computer scientists in isolation. As the domain is becoming increasingly popular - the majority of scientific papers on the topic of sign language translation have been published in the past three years - we provide an overview of the state of the art as well as some required background in the different related disciplines. We give a high-level introduction to sign language linguistics and machine translation to illustrate the requirements of automatic sign language translation. We present a systematic literature review to illustrate the state of the art in the domain and then, harking back to the requirements, lay out several challenges for future research. We find that significant advances have been made on the shoulders of spoken language machine translation research. However, current approaches are often not linguistically motivated or are not adapted to the different input modality of sign languages. We explore challenges related to the representation of sign language data, the collection of datasets, the need for interdisciplinary research and requirements for moving beyond research, towards applications. Based on our findings, we advocate for interdisciplinary research and to base future research on linguistic analysis of sign languages. Furthermore, the inclusion of deaf and hearing end users of sign language translation applications in use case identification, data collection and evaluation is of the utmost importance in the creation of useful sign language translation models. We recommend iterative, human-in-the-loop, design and development of sign language translation models.
[ { "version": "v1", "created": "Mon, 7 Feb 2022 11:54:07 GMT" }, { "version": "v2", "created": "Tue, 15 Mar 2022 11:19:27 GMT" }, { "version": "v3", "created": "Tue, 22 Mar 2022 14:41:52 GMT" }, { "version": "v4", "created": "Wed, 5 Apr 2023 06:30:14 GMT" } ]
2023-04-06T00:00:00
[ [ "De Coster", "Mathieu", "" ], [ "Shterionov", "Dimitar", "" ], [ "Van Herreweghe", "Mieke", "" ], [ "Dambre", "Joni", "" ] ]
TITLE: Machine Translation from Signed to Spoken Languages: State of the Art and Challenges ABSTRACT: Automatic translation from signed to spoken languages is an interdisciplinary research domain, lying on the intersection of computer vision, machine translation and linguistics. Nevertheless, research in this domain is performed mostly by computer scientists in isolation. As the domain is becoming increasingly popular - the majority of scientific papers on the topic of sign language translation have been published in the past three years - we provide an overview of the state of the art as well as some required background in the different related disciplines. We give a high-level introduction to sign language linguistics and machine translation to illustrate the requirements of automatic sign language translation. We present a systematic literature review to illustrate the state of the art in the domain and then, harking back to the requirements, lay out several challenges for future research. We find that significant advances have been made on the shoulders of spoken language machine translation research. However, current approaches are often not linguistically motivated or are not adapted to the different input modality of sign languages. We explore challenges related to the representation of sign language data, the collection of datasets, the need for interdisciplinary research and requirements for moving beyond research, towards applications. Based on our findings, we advocate for interdisciplinary research and to base future research on linguistic analysis of sign languages. Furthermore, the inclusion of deaf and hearing end users of sign language translation applications in use case identification, data collection and evaluation is of the utmost importance in the creation of useful sign language translation models. We recommend iterative, human-in-the-loop, design and development of sign language translation models.
no_new_dataset
0.709849
2203.00131
Yunhe Gao
Yunhe Gao, Mu Zhou, Di Liu, Zhennan Yan, Shaoting Zhang, Dimitris N. Metaxas
A Data-scalable Transformer for Medical Image Segmentation: Architecture, Model Efficiency, and Benchmark
null
null
null
null
eess.IV cs.CV
http://creativecommons.org/licenses/by-nc-nd/4.0/
Transformers have demonstrated remarkable performance in natural language processing and computer vision. However, existing vision Transformers struggle to learn from limited medical data and are unable to generalize on diverse medical image tasks. To tackle these challenges, we present MedFormer, a data-scalable Transformer designed for generalizable 3D medical image segmentation. Our approach incorporates three key elements: a desirable inductive bias, hierarchical modeling with linear-complexity attention, and multi-scale feature fusion that integrates spatial and semantic information globally. MedFormer can learn across tiny- to large-scale data without pre-training. Comprehensive experiments demonstrate MedFormer's potential as a versatile segmentation backbone, outperforming CNNs and vision Transformers on seven public datasets covering multiple modalities (e.g., CT and MRI) and various medical targets (e.g., healthy organs, diseased tissues, and tumors). We provide public access to our models and evaluation pipeline, offering solid baselines and unbiased comparisons to advance a wide range of downstream clinical applications.
[ { "version": "v1", "created": "Mon, 28 Feb 2022 22:59:42 GMT" }, { "version": "v2", "created": "Thu, 3 Mar 2022 03:22:10 GMT" }, { "version": "v3", "created": "Fri, 20 May 2022 05:02:32 GMT" }, { "version": "v4", "created": "Sun, 26 Jun 2022 17:28:50 GMT" }, { "version": "v5", "created": "Wed, 5 Apr 2023 01:56:21 GMT" } ]
2023-04-06T00:00:00
[ [ "Gao", "Yunhe", "" ], [ "Zhou", "Mu", "" ], [ "Liu", "Di", "" ], [ "Yan", "Zhennan", "" ], [ "Zhang", "Shaoting", "" ], [ "Metaxas", "Dimitris N.", "" ] ]
TITLE: A Data-scalable Transformer for Medical Image Segmentation: Architecture, Model Efficiency, and Benchmark ABSTRACT: Transformers have demonstrated remarkable performance in natural language processing and computer vision. However, existing vision Transformers struggle to learn from limited medical data and are unable to generalize on diverse medical image tasks. To tackle these challenges, we present MedFormer, a data-scalable Transformer designed for generalizable 3D medical image segmentation. Our approach incorporates three key elements: a desirable inductive bias, hierarchical modeling with linear-complexity attention, and multi-scale feature fusion that integrates spatial and semantic information globally. MedFormer can learn across tiny- to large-scale data without pre-training. Comprehensive experiments demonstrate MedFormer's potential as a versatile segmentation backbone, outperforming CNNs and vision Transformers on seven public datasets covering multiple modalities (e.g., CT and MRI) and various medical targets (e.g., healthy organs, diseased tissues, and tumors). We provide public access to our models and evaluation pipeline, offering solid baselines and unbiased comparisons to advance a wide range of downstream clinical applications.
no_new_dataset
0.707436
2203.05711
Yidan Sun Miss
Yidan Sun, Qin Chao, Yangfeng Ji and Boyang Li
Synopses of Movie Narratives: a Video-Language Dataset for Story Understanding
25 pages, 17 figures
null
null
null
cs.CV cs.CL cs.LG
http://creativecommons.org/licenses/by/4.0/
Despite recent advances of AI, story understanding remains an open and under-investigated problem. We collect, preprocess, and publicly release a video-language story dataset, Synopses of Movie Narratives (SyMoN), containing 5,193 video summaries of popular movies and TV series with a total length of 869 hours. SyMoN captures naturalistic storytelling videos made by human creators and intended for a human audience. As a prototypical and naturalistic story dataset, SyMoN features high coverage of multimodal story events and abundant mental-state descriptions. Its use of storytelling techniques cause cross-domain semantic gaps that provide appropriate challenges to existing models. We establish benchmarks on video-text retrieval and zero-shot alignment on movie summary videos, which showcase the importance of in-domain data and long-term memory in story understanding. With SyMoN, we hope to lay the groundwork for progress in multimodal story understanding.
[ { "version": "v1", "created": "Fri, 11 Mar 2022 01:45:33 GMT" }, { "version": "v2", "created": "Mon, 3 Apr 2023 03:52:14 GMT" }, { "version": "v3", "created": "Tue, 4 Apr 2023 16:27:38 GMT" }, { "version": "v4", "created": "Wed, 5 Apr 2023 02:09:02 GMT" } ]
2023-04-06T00:00:00
[ [ "Sun", "Yidan", "" ], [ "Chao", "Qin", "" ], [ "Ji", "Yangfeng", "" ], [ "Li", "Boyang", "" ] ]
TITLE: Synopses of Movie Narratives: a Video-Language Dataset for Story Understanding ABSTRACT: Despite recent advances of AI, story understanding remains an open and under-investigated problem. We collect, preprocess, and publicly release a video-language story dataset, Synopses of Movie Narratives (SyMoN), containing 5,193 video summaries of popular movies and TV series with a total length of 869 hours. SyMoN captures naturalistic storytelling videos made by human creators and intended for a human audience. As a prototypical and naturalistic story dataset, SyMoN features high coverage of multimodal story events and abundant mental-state descriptions. Its use of storytelling techniques cause cross-domain semantic gaps that provide appropriate challenges to existing models. We establish benchmarks on video-text retrieval and zero-shot alignment on movie summary videos, which showcase the importance of in-domain data and long-term memory in story understanding. With SyMoN, we hope to lay the groundwork for progress in multimodal story understanding.
new_dataset
0.714404
2204.06766
Siyi Tang
Siyi Tang, Amara Tariq, Jared Dunnmon, Umesh Sharma, Praneetha Elugunti, Daniel Rubin, Bhavik N. Patel, Imon Banerjee
Multimodal spatiotemporal graph neural networks for improved prediction of 30-day all-cause hospital readmission
null
IEEE Journal of Biomedical and Health Informatics, vol. 27, no. 4, pp. 2071-2082, April 2023
10.1109/JBHI.2023.3236888
null
cs.LG cs.AI
http://creativecommons.org/licenses/by/4.0/
Measures to predict 30-day readmission are considered an important quality factor for hospitals as accurate predictions can reduce the overall cost of care by identifying high risk patients before they are discharged. While recent deep learning-based studies have shown promising empirical results on readmission prediction, several limitations exist that may hinder widespread clinical utility, such as (a) only patients with certain conditions are considered, (b) existing approaches do not leverage data temporality, (c) individual admissions are assumed independent of each other, which is unrealistic, (d) prior studies are usually limited to single source of data and single center data. To address these limitations, we propose a multimodal, modality-agnostic spatiotemporal graph neural network (MM-STGNN) for prediction of 30-day all-cause hospital readmission that fuses multimodal in-patient longitudinal data. By training and evaluating our methods using longitudinal chest radiographs and electronic health records from two independent centers, we demonstrate that MM-STGNN achieves AUROC of 0.79 on both primary and external datasets. Furthermore, MM-STGNN significantly outperforms the current clinical reference standard, LACE+ score (AUROC=0.61), on the primary dataset. For subset populations of patients with heart and vascular disease, our model also outperforms baselines on predicting 30-day readmission (e.g., 3.7 point improvement in AUROC in patients with heart disease). Lastly, qualitative model interpretability analysis indicates that while patients' primary diagnoses were not explicitly used to train the model, node features crucial for model prediction directly reflect patients' primary diagnoses. Importantly, our MM-STGNN is agnostic to node feature modalities and could be utilized to integrate multimodal data for triaging patients in various downstream resource allocation tasks.
[ { "version": "v1", "created": "Thu, 14 Apr 2022 05:50:07 GMT" } ]
2023-04-06T00:00:00
[ [ "Tang", "Siyi", "" ], [ "Tariq", "Amara", "" ], [ "Dunnmon", "Jared", "" ], [ "Sharma", "Umesh", "" ], [ "Elugunti", "Praneetha", "" ], [ "Rubin", "Daniel", "" ], [ "Patel", "Bhavik N.", "" ], [ "Banerjee", "Imon", "" ] ]
TITLE: Multimodal spatiotemporal graph neural networks for improved prediction of 30-day all-cause hospital readmission ABSTRACT: Measures to predict 30-day readmission are considered an important quality factor for hospitals as accurate predictions can reduce the overall cost of care by identifying high risk patients before they are discharged. While recent deep learning-based studies have shown promising empirical results on readmission prediction, several limitations exist that may hinder widespread clinical utility, such as (a) only patients with certain conditions are considered, (b) existing approaches do not leverage data temporality, (c) individual admissions are assumed independent of each other, which is unrealistic, (d) prior studies are usually limited to single source of data and single center data. To address these limitations, we propose a multimodal, modality-agnostic spatiotemporal graph neural network (MM-STGNN) for prediction of 30-day all-cause hospital readmission that fuses multimodal in-patient longitudinal data. By training and evaluating our methods using longitudinal chest radiographs and electronic health records from two independent centers, we demonstrate that MM-STGNN achieves AUROC of 0.79 on both primary and external datasets. Furthermore, MM-STGNN significantly outperforms the current clinical reference standard, LACE+ score (AUROC=0.61), on the primary dataset. For subset populations of patients with heart and vascular disease, our model also outperforms baselines on predicting 30-day readmission (e.g., 3.7 point improvement in AUROC in patients with heart disease). Lastly, qualitative model interpretability analysis indicates that while patients' primary diagnoses were not explicitly used to train the model, node features crucial for model prediction directly reflect patients' primary diagnoses. Importantly, our MM-STGNN is agnostic to node feature modalities and could be utilized to integrate multimodal data for triaging patients in various downstream resource allocation tasks.
no_new_dataset
0.709372
2206.11600
Jorge Fernandez-De-Cossio-Diaz
Jorge Fernandez-de-Cossio-Diaz, Simona Cocco, Remi Monasson
Disentangling representations in Restricted Boltzmann Machines without adversaries
Minor corrections. Accepted for publication in Physical Review X
null
10.1103/PhysRevX.13.021003
null
cs.LG cond-mat.dis-nn
http://creativecommons.org/licenses/by/4.0/
A goal of unsupervised machine learning is to build representations of complex high-dimensional data, with simple relations to their properties. Such disentangled representations make easier to interpret the significant latent factors of variation in the data, as well as to generate new data with desirable features. Methods for disentangling representations often rely on an adversarial scheme, in which representations are tuned to avoid discriminators from being able to reconstruct information about the data properties (labels). Unfortunately adversarial training is generally difficult to implement in practice. Here we propose a simple, effective way of disentangling representations without any need to train adversarial discriminators, and apply our approach to Restricted Boltzmann Machines (RBM), one of the simplest representation-based generative models. Our approach relies on the introduction of adequate constraints on the weights during training, which allows us to concentrate information about labels on a small subset of latent variables. The effectiveness of the approach is illustrated with four examples: the CelebA dataset of facial images, the two-dimensional Ising model, the MNIST dataset of handwritten digits, and the taxonomy of protein families. In addition, we show how our framework allows for analytically computing the cost, in terms of log-likelihood of the data, associated to the disentanglement of their representations.
[ { "version": "v1", "created": "Thu, 23 Jun 2022 10:24:20 GMT" }, { "version": "v2", "created": "Fri, 1 Jul 2022 07:58:59 GMT" }, { "version": "v3", "created": "Mon, 16 Jan 2023 09:44:11 GMT" }, { "version": "v4", "created": "Wed, 8 Mar 2023 23:13:41 GMT" } ]
2023-04-06T00:00:00
[ [ "Fernandez-de-Cossio-Diaz", "Jorge", "" ], [ "Cocco", "Simona", "" ], [ "Monasson", "Remi", "" ] ]
TITLE: Disentangling representations in Restricted Boltzmann Machines without adversaries ABSTRACT: A goal of unsupervised machine learning is to build representations of complex high-dimensional data, with simple relations to their properties. Such disentangled representations make easier to interpret the significant latent factors of variation in the data, as well as to generate new data with desirable features. Methods for disentangling representations often rely on an adversarial scheme, in which representations are tuned to avoid discriminators from being able to reconstruct information about the data properties (labels). Unfortunately adversarial training is generally difficult to implement in practice. Here we propose a simple, effective way of disentangling representations without any need to train adversarial discriminators, and apply our approach to Restricted Boltzmann Machines (RBM), one of the simplest representation-based generative models. Our approach relies on the introduction of adequate constraints on the weights during training, which allows us to concentrate information about labels on a small subset of latent variables. The effectiveness of the approach is illustrated with four examples: the CelebA dataset of facial images, the two-dimensional Ising model, the MNIST dataset of handwritten digits, and the taxonomy of protein families. In addition, we show how our framework allows for analytically computing the cost, in terms of log-likelihood of the data, associated to the disentanglement of their representations.
no_new_dataset
0.709202
2209.07976
Jiri Sedlar
Jiri Sedlar, Karla Stepanova, Radoslav Skoviera, Jan K. Behrens, Matus Tuna, Gabriela Sejnova, Josef Sivic, Robert Babuska
Imitrob: Imitation Learning Dataset for Training and Evaluating 6D Object Pose Estimators
The dataset and code are publicly available at http://imitrob.ciirc.cvut.cz/imitrobdataset.php
IEEE Robotics and Automation Letters, vol. 8, no. 5, pp. 2788-2795, 2023
10.1109/LRA.2023.3259735
null
cs.RO cs.CV
http://creativecommons.org/licenses/by-nc-sa/4.0/
This paper introduces a dataset for training and evaluating methods for 6D pose estimation of hand-held tools in task demonstrations captured by a standard RGB camera. Despite the significant progress of 6D pose estimation methods, their performance is usually limited for heavily occluded objects, which is a common case in imitation learning, where the object is typically partially occluded by the manipulating hand. Currently, there is a lack of datasets that would enable the development of robust 6D pose estimation methods for these conditions. To overcome this problem, we collect a new dataset (Imitrob) aimed at 6D pose estimation in imitation learning and other applications where a human holds a tool and performs a task. The dataset contains image sequences of nine different tools and twelve manipulation tasks with two camera viewpoints, four human subjects, and left/right hand. Each image is accompanied by an accurate ground truth measurement of the 6D object pose obtained by the HTC Vive motion tracking device. The use of the dataset is demonstrated by training and evaluating a recent 6D object pose estimation method (DOPE) in various setups.
[ { "version": "v1", "created": "Fri, 16 Sep 2022 14:43:46 GMT" }, { "version": "v2", "created": "Mon, 19 Sep 2022 15:34:06 GMT" }, { "version": "v3", "created": "Wed, 5 Apr 2023 17:30:35 GMT" } ]
2023-04-06T00:00:00
[ [ "Sedlar", "Jiri", "" ], [ "Stepanova", "Karla", "" ], [ "Skoviera", "Radoslav", "" ], [ "Behrens", "Jan K.", "" ], [ "Tuna", "Matus", "" ], [ "Sejnova", "Gabriela", "" ], [ "Sivic", "Josef", "" ], [ "Babuska", "Robert", "" ] ]
TITLE: Imitrob: Imitation Learning Dataset for Training and Evaluating 6D Object Pose Estimators ABSTRACT: This paper introduces a dataset for training and evaluating methods for 6D pose estimation of hand-held tools in task demonstrations captured by a standard RGB camera. Despite the significant progress of 6D pose estimation methods, their performance is usually limited for heavily occluded objects, which is a common case in imitation learning, where the object is typically partially occluded by the manipulating hand. Currently, there is a lack of datasets that would enable the development of robust 6D pose estimation methods for these conditions. To overcome this problem, we collect a new dataset (Imitrob) aimed at 6D pose estimation in imitation learning and other applications where a human holds a tool and performs a task. The dataset contains image sequences of nine different tools and twelve manipulation tasks with two camera viewpoints, four human subjects, and left/right hand. Each image is accompanied by an accurate ground truth measurement of the 6D object pose obtained by the HTC Vive motion tracking device. The use of the dataset is demonstrated by training and evaluating a recent 6D object pose estimation method (DOPE) in various setups.
new_dataset
0.715989
2209.09532
Shihe Wang
Shihe Wang, Jianfeng Ren, Xiaoyu Lian, Ruibin Bai, Xudong Jiang
Boosting the Discriminant Power of Naive Bayes
Accepted by 2022 International Conference on Pattern Recognition
null
10.1109/ICPR56361.2022.9956358
null
cs.LG
http://creativecommons.org/licenses/by/4.0/
Naive Bayes has been widely used in many applications because of its simplicity and ability in handling both numerical data and categorical data. However, lack of modeling of correlations between features limits its performance. In addition, noise and outliers in the real-world dataset also greatly degrade the classification performance. In this paper, we propose a feature augmentation method employing a stack auto-encoder to reduce the noise in the data and boost the discriminant power of naive Bayes. The proposed stack auto-encoder consists of two auto-encoders for different purposes. The first encoder shrinks the initial features to derive a compact feature representation in order to remove the noise and redundant information. The second encoder boosts the discriminant power of the features by expanding them into a higher-dimensional space so that different classes of samples could be better separated in the higher-dimensional space. By integrating the proposed feature augmentation method with the regularized naive Bayes, the discrimination power of the model is greatly enhanced. The proposed method is evaluated on a set of machine-learning benchmark datasets. The experimental results show that the proposed method significantly and consistently outperforms the state-of-the-art naive Bayes classifiers.
[ { "version": "v1", "created": "Tue, 20 Sep 2022 08:02:54 GMT" } ]
2023-04-06T00:00:00
[ [ "Wang", "Shihe", "" ], [ "Ren", "Jianfeng", "" ], [ "Lian", "Xiaoyu", "" ], [ "Bai", "Ruibin", "" ], [ "Jiang", "Xudong", "" ] ]
TITLE: Boosting the Discriminant Power of Naive Bayes ABSTRACT: Naive Bayes has been widely used in many applications because of its simplicity and ability in handling both numerical data and categorical data. However, lack of modeling of correlations between features limits its performance. In addition, noise and outliers in the real-world dataset also greatly degrade the classification performance. In this paper, we propose a feature augmentation method employing a stack auto-encoder to reduce the noise in the data and boost the discriminant power of naive Bayes. The proposed stack auto-encoder consists of two auto-encoders for different purposes. The first encoder shrinks the initial features to derive a compact feature representation in order to remove the noise and redundant information. The second encoder boosts the discriminant power of the features by expanding them into a higher-dimensional space so that different classes of samples could be better separated in the higher-dimensional space. By integrating the proposed feature augmentation method with the regularized naive Bayes, the discrimination power of the model is greatly enhanced. The proposed method is evaluated on a set of machine-learning benchmark datasets. The experimental results show that the proposed method significantly and consistently outperforms the state-of-the-art naive Bayes classifiers.
no_new_dataset
0.710635
2209.10095
Shihe Wang
Shihe Wang, Jianfeng Ren, Ruibin Bai, Yuan Yao, Xudong Jiang
A Max-relevance-min-divergence Criterion for Data Discretization with Applications on Naive Bayes
Under major revision of Pattern Recognition
null
null
null
cs.LG cs.DM cs.IT math.IT
http://creativecommons.org/licenses/by/4.0/
In many classification models, data is discretized to better estimate its distribution. Existing discretization methods often target at maximizing the discriminant power of discretized data, while overlooking the fact that the primary target of data discretization in classification is to improve the generalization performance. As a result, the data tend to be over-split into many small bins since the data without discretization retain the maximal discriminant information. Thus, we propose a Max-Dependency-Min-Divergence (MDmD) criterion that maximizes both the discriminant information and generalization ability of the discretized data. More specifically, the Max-Dependency criterion maximizes the statistical dependency between the discretized data and the classification variable while the Min-Divergence criterion explicitly minimizes the JS-divergence between the training data and the validation data for a given discretization scheme. The proposed MDmD criterion is technically appealing, but it is difficult to reliably estimate the high-order joint distributions of attributes and the classification variable. We hence further propose a more practical solution, Max-Relevance-Min-Divergence (MRmD) discretization scheme, where each attribute is discretized separately, by simultaneously maximizing the discriminant information and the generalization ability of the discretized data. The proposed MRmD is compared with the state-of-the-art discretization algorithms under the naive Bayes classification framework on 45 machine-learning benchmark datasets. It significantly outperforms all the compared methods on most of the datasets.
[ { "version": "v1", "created": "Tue, 20 Sep 2022 07:45:00 GMT" }, { "version": "v2", "created": "Wed, 5 Apr 2023 02:30:11 GMT" } ]
2023-04-06T00:00:00
[ [ "Wang", "Shihe", "" ], [ "Ren", "Jianfeng", "" ], [ "Bai", "Ruibin", "" ], [ "Yao", "Yuan", "" ], [ "Jiang", "Xudong", "" ] ]
TITLE: A Max-relevance-min-divergence Criterion for Data Discretization with Applications on Naive Bayes ABSTRACT: In many classification models, data is discretized to better estimate its distribution. Existing discretization methods often target at maximizing the discriminant power of discretized data, while overlooking the fact that the primary target of data discretization in classification is to improve the generalization performance. As a result, the data tend to be over-split into many small bins since the data without discretization retain the maximal discriminant information. Thus, we propose a Max-Dependency-Min-Divergence (MDmD) criterion that maximizes both the discriminant information and generalization ability of the discretized data. More specifically, the Max-Dependency criterion maximizes the statistical dependency between the discretized data and the classification variable while the Min-Divergence criterion explicitly minimizes the JS-divergence between the training data and the validation data for a given discretization scheme. The proposed MDmD criterion is technically appealing, but it is difficult to reliably estimate the high-order joint distributions of attributes and the classification variable. We hence further propose a more practical solution, Max-Relevance-Min-Divergence (MRmD) discretization scheme, where each attribute is discretized separately, by simultaneously maximizing the discriminant information and the generalization ability of the discretized data. The proposed MRmD is compared with the state-of-the-art discretization algorithms under the naive Bayes classification framework on 45 machine-learning benchmark datasets. It significantly outperforms all the compared methods on most of the datasets.
no_new_dataset
0.709837
2209.13136
Pranav Shetty
Pranav Shetty, Arunkumar Chitteth Rajan, Christopher Kuenneth, Sonkakshi Gupta, Lakshmi Prerana Panchumarti, Lauren Holm, Chao Zhang, and Rampi Ramprasad
A general-purpose material property data extraction pipeline from large polymer corpora using Natural Language Processing
null
null
10.1038/s41524-023-01003-w
null
cs.CL cond-mat.mtrl-sci cond-mat.soft
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
The ever-increasing number of materials science articles makes it hard to infer chemistry-structure-property relations from published literature. We used natural language processing (NLP) methods to automatically extract material property data from the abstracts of polymer literature. As a component of our pipeline, we trained MaterialsBERT, a language model, using 2.4 million materials science abstracts, which outperforms other baseline models in three out of five named entity recognition datasets when used as the encoder for text. Using this pipeline, we obtained ~300,000 material property records from ~130,000 abstracts in 60 hours. The extracted data was analyzed for a diverse range of applications such as fuel cells, supercapacitors, and polymer solar cells to recover non-trivial insights. The data extracted through our pipeline is made available through a web platform at https://polymerscholar.org which can be used to locate material property data recorded in abstracts conveniently. This work demonstrates the feasibility of an automatic pipeline that starts from published literature and ends with a complete set of extracted material property information.
[ { "version": "v1", "created": "Tue, 27 Sep 2022 03:47:03 GMT" } ]
2023-04-06T00:00:00
[ [ "Shetty", "Pranav", "" ], [ "Rajan", "Arunkumar Chitteth", "" ], [ "Kuenneth", "Christopher", "" ], [ "Gupta", "Sonkakshi", "" ], [ "Panchumarti", "Lakshmi Prerana", "" ], [ "Holm", "Lauren", "" ], [ "Zhang", "Chao", "" ], [ "Ramprasad", "Rampi", "" ] ]
TITLE: A general-purpose material property data extraction pipeline from large polymer corpora using Natural Language Processing ABSTRACT: The ever-increasing number of materials science articles makes it hard to infer chemistry-structure-property relations from published literature. We used natural language processing (NLP) methods to automatically extract material property data from the abstracts of polymer literature. As a component of our pipeline, we trained MaterialsBERT, a language model, using 2.4 million materials science abstracts, which outperforms other baseline models in three out of five named entity recognition datasets when used as the encoder for text. Using this pipeline, we obtained ~300,000 material property records from ~130,000 abstracts in 60 hours. The extracted data was analyzed for a diverse range of applications such as fuel cells, supercapacitors, and polymer solar cells to recover non-trivial insights. The data extracted through our pipeline is made available through a web platform at https://polymerscholar.org which can be used to locate material property data recorded in abstracts conveniently. This work demonstrates the feasibility of an automatic pipeline that starts from published literature and ends with a complete set of extracted material property information.
no_new_dataset
0.711663
2210.00192
Han Ruihua
Ruihua Han, Shuai Wang, Shuaijun Wang, Zeqing Zhang, Qianru Zhang, Yonina C. Eldar, Qi Hao, Jia Pan
RDA: An Accelerated Collision Free Motion Planner for Autonomous Navigation in Cluttered Environments
Published in: IEEE Robotics and Automation Letters ( Volume: 8, Issue: 3, March 2023) (https://ieeexplore.ieee.org/document/10036019)
null
10.1109/LRA.2023.3242138
null
cs.RO
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Autonomous motion planning is challenging in multi-obstacle environments due to nonconvex collision avoidance constraints. Directly applying numerical solvers to these nonconvex formulations fails to exploit the constraint structures, resulting in excessive computation time. In this paper, we present an accelerated collision-free motion planner, namely regularized dual alternating direction method of multipliers (RDADMM or RDA for short), for the model predictive control (MPC) based motion planning problem. The proposed RDA addresses nonconvex motion planning via solving a smooth biconvex reformulation via duality and allows the collision avoidance constraints to be computed in parallel for each obstacle to reduce computation time significantly. We validate the performance of the RDA planner through path-tracking experiments with car-like robots in both simulation and real-world settings. Experimental results show that the proposed method generates smooth collision-free trajectories with less computation time compared with other benchmarks and performs robustly in cluttered environments. The source code is available at https://github.com/hanruihua/RDA_planner.
[ { "version": "v1", "created": "Sat, 1 Oct 2022 04:47:02 GMT" }, { "version": "v2", "created": "Thu, 15 Dec 2022 07:53:19 GMT" }, { "version": "v3", "created": "Wed, 1 Feb 2023 09:49:42 GMT" }, { "version": "v4", "created": "Wed, 5 Apr 2023 02:25:53 GMT" } ]
2023-04-06T00:00:00
[ [ "Han", "Ruihua", "" ], [ "Wang", "Shuai", "" ], [ "Wang", "Shuaijun", "" ], [ "Zhang", "Zeqing", "" ], [ "Zhang", "Qianru", "" ], [ "Eldar", "Yonina C.", "" ], [ "Hao", "Qi", "" ], [ "Pan", "Jia", "" ] ]
TITLE: RDA: An Accelerated Collision Free Motion Planner for Autonomous Navigation in Cluttered Environments ABSTRACT: Autonomous motion planning is challenging in multi-obstacle environments due to nonconvex collision avoidance constraints. Directly applying numerical solvers to these nonconvex formulations fails to exploit the constraint structures, resulting in excessive computation time. In this paper, we present an accelerated collision-free motion planner, namely regularized dual alternating direction method of multipliers (RDADMM or RDA for short), for the model predictive control (MPC) based motion planning problem. The proposed RDA addresses nonconvex motion planning via solving a smooth biconvex reformulation via duality and allows the collision avoidance constraints to be computed in parallel for each obstacle to reduce computation time significantly. We validate the performance of the RDA planner through path-tracking experiments with car-like robots in both simulation and real-world settings. Experimental results show that the proposed method generates smooth collision-free trajectories with less computation time compared with other benchmarks and performs robustly in cluttered environments. The source code is available at https://github.com/hanruihua/RDA_planner.
no_new_dataset
0.709447
2210.00875
Yiming Li
Yiming Li, Yang Bai, Yong Jiang, Yong Yang, Shu-Tao Xia, Bo Li
Untargeted Backdoor Watermark: Towards Harmless and Stealthy Dataset Copyright Protection
This work is accepted by the NeurIPS 2022 (Oral, TOP 2%). The first two authors contributed equally to this work. 25 pages. We have fixed some typos in the previous version
null
null
null
cs.CR cs.AI cs.CV cs.LG
http://creativecommons.org/licenses/by/4.0/
Deep neural networks (DNNs) have demonstrated their superiority in practice. Arguably, the rapid development of DNNs is largely benefited from high-quality (open-sourced) datasets, based on which researchers and developers can easily evaluate and improve their learning methods. Since the data collection is usually time-consuming or even expensive, how to protect their copyrights is of great significance and worth further exploration. In this paper, we revisit dataset ownership verification. We find that existing verification methods introduced new security risks in DNNs trained on the protected dataset, due to the targeted nature of poison-only backdoor watermarks. To alleviate this problem, in this work, we explore the untargeted backdoor watermarking scheme, where the abnormal model behaviors are not deterministic. Specifically, we introduce two dispersibilities and prove their correlation, based on which we design the untargeted backdoor watermark under both poisoned-label and clean-label settings. We also discuss how to use the proposed untargeted backdoor watermark for dataset ownership verification. Experiments on benchmark datasets verify the effectiveness of our methods and their resistance to existing backdoor defenses. Our codes are available at \url{https://github.com/THUYimingLi/Untargeted_Backdoor_Watermark}.
[ { "version": "v1", "created": "Tue, 27 Sep 2022 12:56:56 GMT" }, { "version": "v2", "created": "Thu, 30 Mar 2023 10:52:55 GMT" }, { "version": "v3", "created": "Wed, 5 Apr 2023 13:32:57 GMT" } ]
2023-04-06T00:00:00
[ [ "Li", "Yiming", "" ], [ "Bai", "Yang", "" ], [ "Jiang", "Yong", "" ], [ "Yang", "Yong", "" ], [ "Xia", "Shu-Tao", "" ], [ "Li", "Bo", "" ] ]
TITLE: Untargeted Backdoor Watermark: Towards Harmless and Stealthy Dataset Copyright Protection ABSTRACT: Deep neural networks (DNNs) have demonstrated their superiority in practice. Arguably, the rapid development of DNNs is largely benefited from high-quality (open-sourced) datasets, based on which researchers and developers can easily evaluate and improve their learning methods. Since the data collection is usually time-consuming or even expensive, how to protect their copyrights is of great significance and worth further exploration. In this paper, we revisit dataset ownership verification. We find that existing verification methods introduced new security risks in DNNs trained on the protected dataset, due to the targeted nature of poison-only backdoor watermarks. To alleviate this problem, in this work, we explore the untargeted backdoor watermarking scheme, where the abnormal model behaviors are not deterministic. Specifically, we introduce two dispersibilities and prove their correlation, based on which we design the untargeted backdoor watermark under both poisoned-label and clean-label settings. We also discuss how to use the proposed untargeted backdoor watermark for dataset ownership verification. Experiments on benchmark datasets verify the effectiveness of our methods and their resistance to existing backdoor defenses. Our codes are available at \url{https://github.com/THUYimingLi/Untargeted_Backdoor_Watermark}.
no_new_dataset
0.708458
2210.08323
Haoran Xu
Haoran Xu, Li Jiang, Jianxiong Li, Xianyuan Zhan
A Policy-Guided Imitation Approach for Offline Reinforcement Learning
Oral @ NeurIPS 2022; An extended version with more experiments & correct some experiment details in previous version
null
null
null
cs.LG cs.AI
http://creativecommons.org/licenses/by-nc-sa/4.0/
Offline reinforcement learning (RL) methods can generally be categorized into two types: RL-based and Imitation-based. RL-based methods could in principle enjoy out-of-distribution generalization but suffer from erroneous off-policy evaluation. Imitation-based methods avoid off-policy evaluation but are too conservative to surpass the dataset. In this study, we propose an alternative approach, inheriting the training stability of imitation-style methods while still allowing logical out-of-distribution generalization. We decompose the conventional reward-maximizing policy in offline RL into a guide-policy and an execute-policy. During training, the guide-poicy and execute-policy are learned using only data from the dataset, in a supervised and decoupled manner. During evaluation, the guide-policy guides the execute-policy by telling where it should go so that the reward can be maximized, serving as the \textit{Prophet}. By doing so, our algorithm allows \textit{state-compositionality} from the dataset, rather than \textit{action-compositionality} conducted in prior imitation-style methods. We dumb this new approach Policy-guided Offline RL (\texttt{POR}). \texttt{POR} demonstrates the state-of-the-art performance on D4RL, a standard benchmark for offline RL. We also highlight the benefits of \texttt{POR} in terms of improving with supplementary suboptimal data and easily adapting to new tasks by only changing the guide-poicy.
[ { "version": "v1", "created": "Sat, 15 Oct 2022 15:54:28 GMT" }, { "version": "v2", "created": "Wed, 26 Oct 2022 18:06:58 GMT" }, { "version": "v3", "created": "Wed, 5 Apr 2023 04:58:45 GMT" } ]
2023-04-06T00:00:00
[ [ "Xu", "Haoran", "" ], [ "Jiang", "Li", "" ], [ "Li", "Jianxiong", "" ], [ "Zhan", "Xianyuan", "" ] ]
TITLE: A Policy-Guided Imitation Approach for Offline Reinforcement Learning ABSTRACT: Offline reinforcement learning (RL) methods can generally be categorized into two types: RL-based and Imitation-based. RL-based methods could in principle enjoy out-of-distribution generalization but suffer from erroneous off-policy evaluation. Imitation-based methods avoid off-policy evaluation but are too conservative to surpass the dataset. In this study, we propose an alternative approach, inheriting the training stability of imitation-style methods while still allowing logical out-of-distribution generalization. We decompose the conventional reward-maximizing policy in offline RL into a guide-policy and an execute-policy. During training, the guide-poicy and execute-policy are learned using only data from the dataset, in a supervised and decoupled manner. During evaluation, the guide-policy guides the execute-policy by telling where it should go so that the reward can be maximized, serving as the \textit{Prophet}. By doing so, our algorithm allows \textit{state-compositionality} from the dataset, rather than \textit{action-compositionality} conducted in prior imitation-style methods. We dumb this new approach Policy-guided Offline RL (\texttt{POR}). \texttt{POR} demonstrates the state-of-the-art performance on D4RL, a standard benchmark for offline RL. We also highlight the benefits of \texttt{POR} in terms of improving with supplementary suboptimal data and easily adapting to new tasks by only changing the guide-poicy.
no_new_dataset
0.707019
2211.03263
Bonaventure F. P. Dossou
Bonaventure F. P. Dossou, Atnafu Lambebo Tonja, Oreen Yousuf, Salomey Osei, Abigail Oppong, Iyanuoluwa Shode, Oluwabusayo Olufunke Awoyomi, Chris Chinenye Emezue
AfroLM: A Self-Active Learning-based Multilingual Pretrained Language Model for 23 African Languages
Third Workshop on Simple and Efficient Natural Language Processing, EMNLP 2022
null
null
null
cs.CL cs.AI cs.LG
http://creativecommons.org/licenses/by/4.0/
In recent years, multilingual pre-trained language models have gained prominence due to their remarkable performance on numerous downstream Natural Language Processing tasks (NLP). However, pre-training these large multilingual language models requires a lot of training data, which is not available for African Languages. Active learning is a semi-supervised learning algorithm, in which a model consistently and dynamically learns to identify the most beneficial samples to train itself on, in order to achieve better optimization and performance on downstream tasks. Furthermore, active learning effectively and practically addresses real-world data scarcity. Despite all its benefits, active learning, in the context of NLP and especially multilingual language models pretraining, has received little consideration. In this paper, we present AfroLM, a multilingual language model pretrained from scratch on 23 African languages (the largest effort to date) using our novel self-active learning framework. Pretrained on a dataset significantly (14x) smaller than existing baselines, AfroLM outperforms many multilingual pretrained language models (AfriBERTa, XLMR-base, mBERT) on various NLP downstream tasks (NER, text classification, and sentiment analysis). Additional out-of-domain sentiment analysis experiments show that \textbf{AfroLM} is able to generalize well across various domains. We release the code source, and our datasets used in our framework at https://github.com/bonaventuredossou/MLM_AL.
[ { "version": "v1", "created": "Mon, 7 Nov 2022 02:15:25 GMT" }, { "version": "v2", "created": "Wed, 23 Nov 2022 19:10:18 GMT" } ]
2023-04-06T00:00:00
[ [ "Dossou", "Bonaventure F. P.", "" ], [ "Tonja", "Atnafu Lambebo", "" ], [ "Yousuf", "Oreen", "" ], [ "Osei", "Salomey", "" ], [ "Oppong", "Abigail", "" ], [ "Shode", "Iyanuoluwa", "" ], [ "Awoyomi", "Oluwabusayo Olufunke", "" ], [ "Emezue", "Chris Chinenye", "" ] ]
TITLE: AfroLM: A Self-Active Learning-based Multilingual Pretrained Language Model for 23 African Languages ABSTRACT: In recent years, multilingual pre-trained language models have gained prominence due to their remarkable performance on numerous downstream Natural Language Processing tasks (NLP). However, pre-training these large multilingual language models requires a lot of training data, which is not available for African Languages. Active learning is a semi-supervised learning algorithm, in which a model consistently and dynamically learns to identify the most beneficial samples to train itself on, in order to achieve better optimization and performance on downstream tasks. Furthermore, active learning effectively and practically addresses real-world data scarcity. Despite all its benefits, active learning, in the context of NLP and especially multilingual language models pretraining, has received little consideration. In this paper, we present AfroLM, a multilingual language model pretrained from scratch on 23 African languages (the largest effort to date) using our novel self-active learning framework. Pretrained on a dataset significantly (14x) smaller than existing baselines, AfroLM outperforms many multilingual pretrained language models (AfriBERTa, XLMR-base, mBERT) on various NLP downstream tasks (NER, text classification, and sentiment analysis). Additional out-of-domain sentiment analysis experiments show that \textbf{AfroLM} is able to generalize well across various domains. We release the code source, and our datasets used in our framework at https://github.com/bonaventuredossou/MLM_AL.
no_new_dataset
0.709604
2211.15259
Till Bungert
Paul F. Jaeger, Carsten T. L\"uth, Lukas Klein and Till J. Bungert
A Call to Reflect on Evaluation Practices for Failure Detection in Image Classification
null
ICLR 2023 (oral)
null
null
cs.CV cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Reliable application of machine learning-based decision systems in the wild is one of the major challenges currently investigated by the field. A large portion of established approaches aims to detect erroneous predictions by means of assigning confidence scores. This confidence may be obtained by either quantifying the model's predictive uncertainty, learning explicit scoring functions, or assessing whether the input is in line with the training distribution. Curiously, while these approaches all state to address the same eventual goal of detecting failures of a classifier upon real-life application, they currently constitute largely separated research fields with individual evaluation protocols, which either exclude a substantial part of relevant methods or ignore large parts of relevant failure sources. In this work, we systematically reveal current pitfalls caused by these inconsistencies and derive requirements for a holistic and realistic evaluation of failure detection. To demonstrate the relevance of this unified perspective, we present a large-scale empirical study for the first time enabling benchmarking confidence scoring functions w.r.t all relevant methods and failure sources. The revelation of a simple softmax response baseline as the overall best performing method underlines the drastic shortcomings of current evaluation in the abundance of publicized research on confidence scoring. Code and trained models are at https://github.com/IML-DKFZ/fd-shifts.
[ { "version": "v1", "created": "Mon, 28 Nov 2022 12:25:27 GMT" }, { "version": "v2", "created": "Wed, 5 Apr 2023 08:39:40 GMT" } ]
2023-04-06T00:00:00
[ [ "Jaeger", "Paul F.", "" ], [ "Lüth", "Carsten T.", "" ], [ "Klein", "Lukas", "" ], [ "Bungert", "Till J.", "" ] ]
TITLE: A Call to Reflect on Evaluation Practices for Failure Detection in Image Classification ABSTRACT: Reliable application of machine learning-based decision systems in the wild is one of the major challenges currently investigated by the field. A large portion of established approaches aims to detect erroneous predictions by means of assigning confidence scores. This confidence may be obtained by either quantifying the model's predictive uncertainty, learning explicit scoring functions, or assessing whether the input is in line with the training distribution. Curiously, while these approaches all state to address the same eventual goal of detecting failures of a classifier upon real-life application, they currently constitute largely separated research fields with individual evaluation protocols, which either exclude a substantial part of relevant methods or ignore large parts of relevant failure sources. In this work, we systematically reveal current pitfalls caused by these inconsistencies and derive requirements for a holistic and realistic evaluation of failure detection. To demonstrate the relevance of this unified perspective, we present a large-scale empirical study for the first time enabling benchmarking confidence scoring functions w.r.t all relevant methods and failure sources. The revelation of a simple softmax response baseline as the overall best performing method underlines the drastic shortcomings of current evaluation in the abundance of publicized research on confidence scoring. Code and trained models are at https://github.com/IML-DKFZ/fd-shifts.
no_new_dataset
0.703791
2211.16197
Luke Rowe
Luke Rowe, Martin Ethier, Eli-Henry Dykhne, Krzysztof Czarnecki
FJMP: Factorized Joint Multi-Agent Motion Prediction over Learned Directed Acyclic Interaction Graphs
CVPR 2023
null
null
null
cs.CV cs.RO
http://creativecommons.org/licenses/by/4.0/
Predicting the future motion of road agents is a critical task in an autonomous driving pipeline. In this work, we address the problem of generating a set of scene-level, or joint, future trajectory predictions in multi-agent driving scenarios. To this end, we propose FJMP, a Factorized Joint Motion Prediction framework for multi-agent interactive driving scenarios. FJMP models the future scene interaction dynamics as a sparse directed interaction graph, where edges denote explicit interactions between agents. We then prune the graph into a directed acyclic graph (DAG) and decompose the joint prediction task into a sequence of marginal and conditional predictions according to the partial ordering of the DAG, where joint future trajectories are decoded using a directed acyclic graph neural network (DAGNN). We conduct experiments on the INTERACTION and Argoverse 2 datasets and demonstrate that FJMP produces more accurate and scene-consistent joint trajectory predictions than non-factorized approaches, especially on the most interactive and kinematically interesting agents. FJMP ranks 1st on the multi-agent test leaderboard of the INTERACTION dataset.
[ { "version": "v1", "created": "Sun, 27 Nov 2022 18:59:17 GMT" }, { "version": "v2", "created": "Wed, 5 Apr 2023 00:23:12 GMT" } ]
2023-04-06T00:00:00
[ [ "Rowe", "Luke", "" ], [ "Ethier", "Martin", "" ], [ "Dykhne", "Eli-Henry", "" ], [ "Czarnecki", "Krzysztof", "" ] ]
TITLE: FJMP: Factorized Joint Multi-Agent Motion Prediction over Learned Directed Acyclic Interaction Graphs ABSTRACT: Predicting the future motion of road agents is a critical task in an autonomous driving pipeline. In this work, we address the problem of generating a set of scene-level, or joint, future trajectory predictions in multi-agent driving scenarios. To this end, we propose FJMP, a Factorized Joint Motion Prediction framework for multi-agent interactive driving scenarios. FJMP models the future scene interaction dynamics as a sparse directed interaction graph, where edges denote explicit interactions between agents. We then prune the graph into a directed acyclic graph (DAG) and decompose the joint prediction task into a sequence of marginal and conditional predictions according to the partial ordering of the DAG, where joint future trajectories are decoded using a directed acyclic graph neural network (DAGNN). We conduct experiments on the INTERACTION and Argoverse 2 datasets and demonstrate that FJMP produces more accurate and scene-consistent joint trajectory predictions than non-factorized approaches, especially on the most interactive and kinematically interesting agents. FJMP ranks 1st on the multi-agent test leaderboard of the INTERACTION dataset.
no_new_dataset
0.708219
2212.05051
Feng Cheng
Feng Cheng, Xizi Wang, Jie Lei, David Crandall, Mohit Bansal, Gedas Bertasius
VindLU: A Recipe for Effective Video-and-Language Pretraining
CVPR 2023. Project page: https://klauscc.github.io/vindlu.html
null
null
null
cs.CV
http://creativecommons.org/licenses/by/4.0/
The last several years have witnessed remarkable progress in video-and-language (VidL) understanding. However, most modern VidL approaches use complex and specialized model architectures and sophisticated pretraining protocols, making the reproducibility, analysis and comparisons of these frameworks difficult. Hence, instead of proposing yet another new VidL model, this paper conducts a thorough empirical study demystifying the most important factors in the VidL model design. Among the factors that we investigate are (i) the spatiotemporal architecture design, (ii) the multimodal fusion schemes, (iii) the pretraining objectives, (iv) the choice of pretraining data, (v) pretraining and finetuning protocols, and (vi) dataset and model scaling. Our empirical study reveals that the most important design factors include: temporal modeling, video-to-text multimodal fusion, masked modeling objectives, and joint training on images and videos. Using these empirical insights, we then develop a step-by-step recipe, dubbed VindLU, for effective VidL pretraining. Our final model trained using our recipe achieves comparable or better than state-of-the-art results on several VidL tasks without relying on external CLIP pretraining. In particular, on the text-to-video retrieval task, our approach obtains 61.2% on DiDeMo, and 55.0% on ActivityNet, outperforming current SOTA by 7.8% and 6.1% respectively. Furthermore, our model also obtains state-of-the-art video question-answering results on ActivityNet-QA, MSRVTT-QA, MSRVTT-MC and TVQA. Our code and pretrained models are publicly available at: https://github.com/klauscc/VindLU.
[ { "version": "v1", "created": "Fri, 9 Dec 2022 18:54:05 GMT" }, { "version": "v2", "created": "Wed, 5 Apr 2023 17:56:15 GMT" } ]
2023-04-06T00:00:00
[ [ "Cheng", "Feng", "" ], [ "Wang", "Xizi", "" ], [ "Lei", "Jie", "" ], [ "Crandall", "David", "" ], [ "Bansal", "Mohit", "" ], [ "Bertasius", "Gedas", "" ] ]
TITLE: VindLU: A Recipe for Effective Video-and-Language Pretraining ABSTRACT: The last several years have witnessed remarkable progress in video-and-language (VidL) understanding. However, most modern VidL approaches use complex and specialized model architectures and sophisticated pretraining protocols, making the reproducibility, analysis and comparisons of these frameworks difficult. Hence, instead of proposing yet another new VidL model, this paper conducts a thorough empirical study demystifying the most important factors in the VidL model design. Among the factors that we investigate are (i) the spatiotemporal architecture design, (ii) the multimodal fusion schemes, (iii) the pretraining objectives, (iv) the choice of pretraining data, (v) pretraining and finetuning protocols, and (vi) dataset and model scaling. Our empirical study reveals that the most important design factors include: temporal modeling, video-to-text multimodal fusion, masked modeling objectives, and joint training on images and videos. Using these empirical insights, we then develop a step-by-step recipe, dubbed VindLU, for effective VidL pretraining. Our final model trained using our recipe achieves comparable or better than state-of-the-art results on several VidL tasks without relying on external CLIP pretraining. In particular, on the text-to-video retrieval task, our approach obtains 61.2% on DiDeMo, and 55.0% on ActivityNet, outperforming current SOTA by 7.8% and 6.1% respectively. Furthermore, our model also obtains state-of-the-art video question-answering results on ActivityNet-QA, MSRVTT-QA, MSRVTT-MC and TVQA. Our code and pretrained models are publicly available at: https://github.com/klauscc/VindLU.
no_new_dataset
0.710427
2212.05190
Alexandre Larouche
Alexandre Larouche, Audrey Durand, Richard Khoury and Caroline Sirois
Neural Bandits for Data Mining: Searching for Dangerous Polypharmacy
This article is presented at the W3PHIAI-23 workshop at AAAI-2023
null
null
null
cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Polypharmacy, most often defined as the simultaneous consumption of five or more drugs at once, is a prevalent phenomenon in the older population. Some of these polypharmacies, deemed inappropriate, may be associated with adverse health outcomes such as death or hospitalization. Considering the combinatorial nature of the problem as well as the size of claims database and the cost to compute an exact association measure for a given drug combination, it is impossible to investigate every possible combination of drugs. Therefore, we propose to optimize the search for potentially inappropriate polypharmacies (PIPs). To this end, we propose the OptimNeuralTS strategy, based on Neural Thompson Sampling and differential evolution, to efficiently mine claims datasets and build a predictive model of the association between drug combinations and health outcomes. We benchmark our method using two datasets generated by an internally developed simulator of polypharmacy data containing 500 drugs and 100 000 distinct combinations. Empirically, our method can detect up to 72% of PIPs while maintaining an average precision score of 99% using 30 000 time steps.
[ { "version": "v1", "created": "Sat, 10 Dec 2022 03:43:23 GMT" }, { "version": "v2", "created": "Tue, 10 Jan 2023 19:06:53 GMT" }, { "version": "v3", "created": "Wed, 5 Apr 2023 15:25:54 GMT" } ]
2023-04-06T00:00:00
[ [ "Larouche", "Alexandre", "" ], [ "Durand", "Audrey", "" ], [ "Khoury", "Richard", "" ], [ "Sirois", "Caroline", "" ] ]
TITLE: Neural Bandits for Data Mining: Searching for Dangerous Polypharmacy ABSTRACT: Polypharmacy, most often defined as the simultaneous consumption of five or more drugs at once, is a prevalent phenomenon in the older population. Some of these polypharmacies, deemed inappropriate, may be associated with adverse health outcomes such as death or hospitalization. Considering the combinatorial nature of the problem as well as the size of claims database and the cost to compute an exact association measure for a given drug combination, it is impossible to investigate every possible combination of drugs. Therefore, we propose to optimize the search for potentially inappropriate polypharmacies (PIPs). To this end, we propose the OptimNeuralTS strategy, based on Neural Thompson Sampling and differential evolution, to efficiently mine claims datasets and build a predictive model of the association between drug combinations and health outcomes. We benchmark our method using two datasets generated by an internally developed simulator of polypharmacy data containing 500 drugs and 100 000 distinct combinations. Empirically, our method can detect up to 72% of PIPs while maintaining an average precision score of 99% using 30 000 time steps.
no_new_dataset
0.708395
2212.05199
Lijun Yu
Lijun Yu, Yong Cheng, Kihyuk Sohn, Jos\'e Lezama, Han Zhang, Huiwen Chang, Alexander G. Hauptmann, Ming-Hsuan Yang, Yuan Hao, Irfan Essa, Lu Jiang
MAGVIT: Masked Generative Video Transformer
CVPR 2023 highlight
null
null
null
cs.CV
http://creativecommons.org/licenses/by/4.0/
We introduce the MAsked Generative VIdeo Transformer, MAGVIT, to tackle various video synthesis tasks with a single model. We introduce a 3D tokenizer to quantize a video into spatial-temporal visual tokens and propose an embedding method for masked video token modeling to facilitate multi-task learning. We conduct extensive experiments to demonstrate the quality, efficiency, and flexibility of MAGVIT. Our experiments show that (i) MAGVIT performs favorably against state-of-the-art approaches and establishes the best-published FVD on three video generation benchmarks, including the challenging Kinetics-600. (ii) MAGVIT outperforms existing methods in inference time by two orders of magnitude against diffusion models and by 60x against autoregressive models. (iii) A single MAGVIT model supports ten diverse generation tasks and generalizes across videos from different visual domains. The source code and trained models will be released to the public at https://magvit.cs.cmu.edu.
[ { "version": "v1", "created": "Sat, 10 Dec 2022 04:26:32 GMT" }, { "version": "v2", "created": "Wed, 5 Apr 2023 02:32:59 GMT" } ]
2023-04-06T00:00:00
[ [ "Yu", "Lijun", "" ], [ "Cheng", "Yong", "" ], [ "Sohn", "Kihyuk", "" ], [ "Lezama", "José", "" ], [ "Zhang", "Han", "" ], [ "Chang", "Huiwen", "" ], [ "Hauptmann", "Alexander G.", "" ], [ "Yang", "Ming-Hsuan", "" ], [ "Hao", "Yuan", "" ], [ "Essa", "Irfan", "" ], [ "Jiang", "Lu", "" ] ]
TITLE: MAGVIT: Masked Generative Video Transformer ABSTRACT: We introduce the MAsked Generative VIdeo Transformer, MAGVIT, to tackle various video synthesis tasks with a single model. We introduce a 3D tokenizer to quantize a video into spatial-temporal visual tokens and propose an embedding method for masked video token modeling to facilitate multi-task learning. We conduct extensive experiments to demonstrate the quality, efficiency, and flexibility of MAGVIT. Our experiments show that (i) MAGVIT performs favorably against state-of-the-art approaches and establishes the best-published FVD on three video generation benchmarks, including the challenging Kinetics-600. (ii) MAGVIT outperforms existing methods in inference time by two orders of magnitude against diffusion models and by 60x against autoregressive models. (iii) A single MAGVIT model supports ten diverse generation tasks and generalizes across videos from different visual domains. The source code and trained models will be released to the public at https://magvit.cs.cmu.edu.
no_new_dataset
0.708855
2301.00366
Kunal Chaturvedi
Kunal Chaturvedi, Ali Braytee, Jun Li, Mukesh Prasad
SS-CPGAN: Self-Supervised Cut-and-Pasting Generative Adversarial Network for Object Segmentation
null
null
10.3390/s23073649
null
cs.CV cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
This paper proposes a novel self-supervised based Cut-and-Paste GAN to perform foreground object segmentation and generate realistic composite images without manual annotations. We accomplish this goal by a simple yet effective self-supervised approach coupled with the U-Net based discriminator. The proposed method extends the ability of the standard discriminators to learn not only the global data representations via classification (real/fake) but also learn semantic and structural information through pseudo labels created using the self-supervised task. The proposed method empowers the generator to create meaningful masks by forcing it to learn informative per-pixel as well as global image feedback from the discriminator. Our experiments demonstrate that our proposed method significantly outperforms the state-of-the-art methods on the standard benchmark datasets.
[ { "version": "v1", "created": "Sun, 1 Jan 2023 07:42:50 GMT" }, { "version": "v2", "created": "Wed, 22 Mar 2023 18:11:14 GMT" }, { "version": "v3", "created": "Tue, 4 Apr 2023 18:44:03 GMT" } ]
2023-04-06T00:00:00
[ [ "Chaturvedi", "Kunal", "" ], [ "Braytee", "Ali", "" ], [ "Li", "Jun", "" ], [ "Prasad", "Mukesh", "" ] ]
TITLE: SS-CPGAN: Self-Supervised Cut-and-Pasting Generative Adversarial Network for Object Segmentation ABSTRACT: This paper proposes a novel self-supervised based Cut-and-Paste GAN to perform foreground object segmentation and generate realistic composite images without manual annotations. We accomplish this goal by a simple yet effective self-supervised approach coupled with the U-Net based discriminator. The proposed method extends the ability of the standard discriminators to learn not only the global data representations via classification (real/fake) but also learn semantic and structural information through pseudo labels created using the self-supervised task. The proposed method empowers the generator to create meaningful masks by forcing it to learn informative per-pixel as well as global image feedback from the discriminator. Our experiments demonstrate that our proposed method significantly outperforms the state-of-the-art methods on the standard benchmark datasets.
no_new_dataset
0.710584
2301.02938
Leimin Tian
Leimin Tian, Kerry He, Shiyu Xu, Akansel Cosgun, Dana Kuli\'c
Crafting with a Robot Assistant: Use Social Cues to Inform Adaptive Handovers in Human-Robot Collaboration
accepted at HRI 2023
null
null
null
cs.RO cs.HC
http://creativecommons.org/licenses/by/4.0/
We study human-robot handovers in a naturalistic collaboration scenario, where a mobile manipulator robot assists a person during a crafting session by providing and retrieving objects used for wooden piece assembly (functional activities) and painting (creative activities). We collect quantitative and qualitative data from 20 participants in a Wizard-of-Oz study, generating the Functional And Creative Tasks Human-Robot Collaboration dataset (the FACT HRC dataset), available to the research community. This work illustrates how social cues and task context inform the temporal-spatial coordination in human-robot handovers, and how human-robot collaboration is shaped by and in turn influences people's functional and creative activities.
[ { "version": "v1", "created": "Sat, 7 Jan 2023 21:19:31 GMT" }, { "version": "v2", "created": "Wed, 5 Apr 2023 02:56:46 GMT" } ]
2023-04-06T00:00:00
[ [ "Tian", "Leimin", "" ], [ "He", "Kerry", "" ], [ "Xu", "Shiyu", "" ], [ "Cosgun", "Akansel", "" ], [ "Kulić", "Dana", "" ] ]
TITLE: Crafting with a Robot Assistant: Use Social Cues to Inform Adaptive Handovers in Human-Robot Collaboration ABSTRACT: We study human-robot handovers in a naturalistic collaboration scenario, where a mobile manipulator robot assists a person during a crafting session by providing and retrieving objects used for wooden piece assembly (functional activities) and painting (creative activities). We collect quantitative and qualitative data from 20 participants in a Wizard-of-Oz study, generating the Functional And Creative Tasks Human-Robot Collaboration dataset (the FACT HRC dataset), available to the research community. This work illustrates how social cues and task context inform the temporal-spatial coordination in human-robot handovers, and how human-robot collaboration is shaped by and in turn influences people's functional and creative activities.
new_dataset
0.710616
2301.08140
Alistair Weld
Alistair Weld, Joao Cartucho, Chi Xu, Joseph Davids and Stamatia Giannarou
Regularising disparity estimation via multi task learning with structured light reconstruction
null
Computer Methods in Biomechanics and Biomedical Engineering: Imaging & Visualization 2022
10.1080/21681163.2022.2156391
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
3D reconstruction is a useful tool for surgical planning and guidance. However, the lack of available medical data stunts research and development in this field, as supervised deep learning methods for accurate disparity estimation rely heavily on large datasets containing ground truth information. Alternative approaches to supervision have been explored, such as self-supervision, which can reduce or remove entirely the need for ground truth. However, no proposed alternatives have demonstrated performance capabilities close to what would be expected from a supervised setup. This work aims to alleviate this issue. In this paper, we investigate the learning of structured light projections to enhance the development of direct disparity estimation networks. We show for the first time that it is possible to accurately learn the projection of structured light on a scene, implicitly learning disparity. Secondly, we \textcolor{black}{explore the use of a multi task learning (MTL) framework for the joint training of structured light and disparity. We present results which show that MTL with structured light improves disparity training; without increasing the number of model parameters. Our MTL setup outperformed the single task learning (STL) network in every validation test. Notably, in the medical generalisation test, the STL error was 1.4 times worse than that of the best MTL performance. The benefit of using MTL is emphasised when the training data is limited.} A dataset containing stereoscopic images, disparity maps and structured light projections on medical phantoms and ex vivo tissue was created for evaluation together with virtual scenes. This dataset will be made publicly available in the future.
[ { "version": "v1", "created": "Thu, 19 Jan 2023 15:54:52 GMT" } ]
2023-04-06T00:00:00
[ [ "Weld", "Alistair", "" ], [ "Cartucho", "Joao", "" ], [ "Xu", "Chi", "" ], [ "Davids", "Joseph", "" ], [ "Giannarou", "Stamatia", "" ] ]
TITLE: Regularising disparity estimation via multi task learning with structured light reconstruction ABSTRACT: 3D reconstruction is a useful tool for surgical planning and guidance. However, the lack of available medical data stunts research and development in this field, as supervised deep learning methods for accurate disparity estimation rely heavily on large datasets containing ground truth information. Alternative approaches to supervision have been explored, such as self-supervision, which can reduce or remove entirely the need for ground truth. However, no proposed alternatives have demonstrated performance capabilities close to what would be expected from a supervised setup. This work aims to alleviate this issue. In this paper, we investigate the learning of structured light projections to enhance the development of direct disparity estimation networks. We show for the first time that it is possible to accurately learn the projection of structured light on a scene, implicitly learning disparity. Secondly, we \textcolor{black}{explore the use of a multi task learning (MTL) framework for the joint training of structured light and disparity. We present results which show that MTL with structured light improves disparity training; without increasing the number of model parameters. Our MTL setup outperformed the single task learning (STL) network in every validation test. Notably, in the medical generalisation test, the STL error was 1.4 times worse than that of the best MTL performance. The benefit of using MTL is emphasised when the training data is limited.} A dataset containing stereoscopic images, disparity maps and structured light projections on medical phantoms and ex vivo tissue was created for evaluation together with virtual scenes. This dataset will be made publicly available in the future.
new_dataset
0.719732
2301.13622
Jakub Tomczak
Kamil Deja, Tomasz Trzcinski, Jakub M. Tomczak
Learning Data Representations with Joint Diffusion Models
Code: https://github.com/KamilDeja/joint_diffusion
null
null
null
cs.LG cs.CV stat.ML
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Joint machine learning models that allow synthesizing and classifying data often offer uneven performance between those tasks or are unstable to train. In this work, we depart from a set of empirical observations that indicate the usefulness of internal representations built by contemporary deep diffusion-based generative models not only for generating but also predicting. We then propose to extend the vanilla diffusion model with a classifier that allows for stable joint end-to-end training with shared parameterization between those objectives. The resulting joint diffusion model outperforms recent state-of-the-art hybrid methods in terms of both classification and generation quality on all evaluated benchmarks. On top of our joint training approach, we present how we can directly benefit from shared generative and discriminative representations by introducing a method for visual counterfactual explanations.
[ { "version": "v1", "created": "Tue, 31 Jan 2023 13:29:19 GMT" }, { "version": "v2", "created": "Wed, 5 Apr 2023 13:09:54 GMT" } ]
2023-04-06T00:00:00
[ [ "Deja", "Kamil", "" ], [ "Trzcinski", "Tomasz", "" ], [ "Tomczak", "Jakub M.", "" ] ]
TITLE: Learning Data Representations with Joint Diffusion Models ABSTRACT: Joint machine learning models that allow synthesizing and classifying data often offer uneven performance between those tasks or are unstable to train. In this work, we depart from a set of empirical observations that indicate the usefulness of internal representations built by contemporary deep diffusion-based generative models not only for generating but also predicting. We then propose to extend the vanilla diffusion model with a classifier that allows for stable joint end-to-end training with shared parameterization between those objectives. The resulting joint diffusion model outperforms recent state-of-the-art hybrid methods in terms of both classification and generation quality on all evaluated benchmarks. On top of our joint training approach, we present how we can directly benefit from shared generative and discriminative representations by introducing a method for visual counterfactual explanations.
no_new_dataset
0.70978
2302.05894
Loukas Ilias
Michail Chatzianastasis, Loukas Ilias, Dimitris Askounis, Michalis Vazirgiannis
Neural Architecture Search with Multimodal Fusion Methods for Diagnosing Dementia
Accepted at ICASSP 2023
null
null
null
cs.LG
http://creativecommons.org/licenses/by-nc-nd/4.0/
Alzheimer's dementia (AD) affects memory, thinking, and language, deteriorating person's life. An early diagnosis is very important as it enables the person to receive medical help and ensure quality of life. Therefore, leveraging spontaneous speech in conjunction with machine learning methods for recognizing AD patients has emerged into a hot topic. Most of the previous works employ Convolutional Neural Networks (CNNs), to process the input signal. However, finding a CNN architecture is a time-consuming process and requires domain expertise. Moreover, the researchers introduce early and late fusion approaches for fusing different modalities or concatenate the representations of the different modalities during training, thus the inter-modal interactions are not captured. To tackle these limitations, first we exploit a Neural Architecture Search (NAS) method to automatically find a high performing CNN architecture. Next, we exploit several fusion methods, including Multimodal Factorized Bilinear Pooling and Tucker Decomposition, to combine both speech and text modalities. To the best of our knowledge, there is no prior work exploiting a NAS approach and these fusion methods in the task of dementia detection from spontaneous speech. We perform extensive experiments on the ADReSS Challenge dataset and show the effectiveness of our approach over state-of-the-art methods.
[ { "version": "v1", "created": "Sun, 12 Feb 2023 11:25:29 GMT" }, { "version": "v2", "created": "Wed, 5 Apr 2023 16:50:50 GMT" } ]
2023-04-06T00:00:00
[ [ "Chatzianastasis", "Michail", "" ], [ "Ilias", "Loukas", "" ], [ "Askounis", "Dimitris", "" ], [ "Vazirgiannis", "Michalis", "" ] ]
TITLE: Neural Architecture Search with Multimodal Fusion Methods for Diagnosing Dementia ABSTRACT: Alzheimer's dementia (AD) affects memory, thinking, and language, deteriorating person's life. An early diagnosis is very important as it enables the person to receive medical help and ensure quality of life. Therefore, leveraging spontaneous speech in conjunction with machine learning methods for recognizing AD patients has emerged into a hot topic. Most of the previous works employ Convolutional Neural Networks (CNNs), to process the input signal. However, finding a CNN architecture is a time-consuming process and requires domain expertise. Moreover, the researchers introduce early and late fusion approaches for fusing different modalities or concatenate the representations of the different modalities during training, thus the inter-modal interactions are not captured. To tackle these limitations, first we exploit a Neural Architecture Search (NAS) method to automatically find a high performing CNN architecture. Next, we exploit several fusion methods, including Multimodal Factorized Bilinear Pooling and Tucker Decomposition, to combine both speech and text modalities. To the best of our knowledge, there is no prior work exploiting a NAS approach and these fusion methods in the task of dementia detection from spontaneous speech. We perform extensive experiments on the ADReSS Challenge dataset and show the effectiveness of our approach over state-of-the-art methods.
no_new_dataset
0.708231
2302.12126
Yunyong Ko
Yunyong Ko, Seongeun Ryu, Soeun Han, Youngseung Jeon, Jaehoon Kim, Sohyun Park, Kyungsik Han, Hanghang Tong, Sang-Wook Kim
KHAN: Knowledge-Aware Hierarchical Attention Networks for Accurate Political Stance Prediction
12 pages, 5 figures, 10 tables, the Web Conference 2023 (WWW)
null
10.1145/3543507.3583300
null
cs.CL cs.AI cs.LG
http://creativecommons.org/licenses/by/4.0/
The political stance prediction for news articles has been widely studied to mitigate the echo chamber effect -- people fall into their thoughts and reinforce their pre-existing beliefs. The previous works for the political stance problem focus on (1) identifying political factors that could reflect the political stance of a news article and (2) capturing those factors effectively. Despite their empirical successes, they are not sufficiently justified in terms of how effective their identified factors are in the political stance prediction. Motivated by this, in this work, we conduct a user study to investigate important factors in political stance prediction, and observe that the context and tone of a news article (implicit) and external knowledge for real-world entities appearing in the article (explicit) are important in determining its political stance. Based on this observation, we propose a novel knowledge-aware approach to political stance prediction (KHAN), employing (1) hierarchical attention networks (HAN) to learn the relationships among words and sentences in three different levels and (2) knowledge encoding (KE) to incorporate external knowledge for real-world entities into the process of political stance prediction. Also, to take into account the subtle and important difference between opposite political stances, we build two independent political knowledge graphs (KG) (i.e., KG-lib and KG-con) by ourselves and learn to fuse the different political knowledge. Through extensive evaluations on three real-world datasets, we demonstrate the superiority of DASH in terms of (1) accuracy, (2) efficiency, and (3) effectiveness.
[ { "version": "v1", "created": "Thu, 23 Feb 2023 16:09:42 GMT" }, { "version": "v2", "created": "Wed, 1 Mar 2023 20:32:01 GMT" }, { "version": "v3", "created": "Tue, 4 Apr 2023 20:03:36 GMT" } ]
2023-04-06T00:00:00
[ [ "Ko", "Yunyong", "" ], [ "Ryu", "Seongeun", "" ], [ "Han", "Soeun", "" ], [ "Jeon", "Youngseung", "" ], [ "Kim", "Jaehoon", "" ], [ "Park", "Sohyun", "" ], [ "Han", "Kyungsik", "" ], [ "Tong", "Hanghang", "" ], [ "Kim", "Sang-Wook", "" ] ]
TITLE: KHAN: Knowledge-Aware Hierarchical Attention Networks for Accurate Political Stance Prediction ABSTRACT: The political stance prediction for news articles has been widely studied to mitigate the echo chamber effect -- people fall into their thoughts and reinforce their pre-existing beliefs. The previous works for the political stance problem focus on (1) identifying political factors that could reflect the political stance of a news article and (2) capturing those factors effectively. Despite their empirical successes, they are not sufficiently justified in terms of how effective their identified factors are in the political stance prediction. Motivated by this, in this work, we conduct a user study to investigate important factors in political stance prediction, and observe that the context and tone of a news article (implicit) and external knowledge for real-world entities appearing in the article (explicit) are important in determining its political stance. Based on this observation, we propose a novel knowledge-aware approach to political stance prediction (KHAN), employing (1) hierarchical attention networks (HAN) to learn the relationships among words and sentences in three different levels and (2) knowledge encoding (KE) to incorporate external knowledge for real-world entities into the process of political stance prediction. Also, to take into account the subtle and important difference between opposite political stances, we build two independent political knowledge graphs (KG) (i.e., KG-lib and KG-con) by ourselves and learn to fuse the different political knowledge. Through extensive evaluations on three real-world datasets, we demonstrate the superiority of DASH in terms of (1) accuracy, (2) efficiency, and (3) effectiveness.
no_new_dataset
0.709177
2302.12252
Muzammal Naseer
Muzammal Naseer, Ahmad Mahmood, Salman Khan, and Fahad Khan
Boosting Adversarial Transferability using Dynamic Cues
International Conference on Learning Representations (ICLR'23), Code:https://bit.ly/3Xd9gRQ
null
null
null
cs.CV cs.LG
http://creativecommons.org/licenses/by/4.0/
The transferability of adversarial perturbations between image models has been extensively studied. In this case, an attack is generated from a known surrogate \eg, the ImageNet trained model, and transferred to change the decision of an unknown (black-box) model trained on an image dataset. However, attacks generated from image models do not capture the dynamic nature of a moving object or a changing scene due to a lack of temporal cues within image models. This leads to reduced transferability of adversarial attacks from representation-enriched \emph{image} models such as Supervised Vision Transformers (ViTs), Self-supervised ViTs (\eg, DINO), and Vision-language models (\eg, CLIP) to black-box \emph{video} models. In this work, we induce dynamic cues within the image models without sacrificing their original performance on images. To this end, we optimize \emph{temporal prompts} through frozen image models to capture motion dynamics. Our temporal prompts are the result of a learnable transformation that allows optimizing for temporal gradients during an adversarial attack to fool the motion dynamics. Specifically, we introduce spatial (image) and temporal (video) cues within the same source model through task-specific prompts. Attacking such prompts maximizes the adversarial transferability from image-to-video and image-to-image models using the attacks designed for image models. Our attack results indicate that the attacker does not need specialized architectures, \eg, divided space-time attention, 3D convolutions, or multi-view convolution networks for different data modalities. Image models are effective surrogates to optimize an adversarial attack to fool black-box models in a changing environment over time. Code is available at https://bit.ly/3Xd9gRQ
[ { "version": "v1", "created": "Thu, 23 Feb 2023 18:59:56 GMT" }, { "version": "v2", "created": "Tue, 4 Apr 2023 19:46:08 GMT" } ]
2023-04-06T00:00:00
[ [ "Naseer", "Muzammal", "" ], [ "Mahmood", "Ahmad", "" ], [ "Khan", "Salman", "" ], [ "Khan", "Fahad", "" ] ]
TITLE: Boosting Adversarial Transferability using Dynamic Cues ABSTRACT: The transferability of adversarial perturbations between image models has been extensively studied. In this case, an attack is generated from a known surrogate \eg, the ImageNet trained model, and transferred to change the decision of an unknown (black-box) model trained on an image dataset. However, attacks generated from image models do not capture the dynamic nature of a moving object or a changing scene due to a lack of temporal cues within image models. This leads to reduced transferability of adversarial attacks from representation-enriched \emph{image} models such as Supervised Vision Transformers (ViTs), Self-supervised ViTs (\eg, DINO), and Vision-language models (\eg, CLIP) to black-box \emph{video} models. In this work, we induce dynamic cues within the image models without sacrificing their original performance on images. To this end, we optimize \emph{temporal prompts} through frozen image models to capture motion dynamics. Our temporal prompts are the result of a learnable transformation that allows optimizing for temporal gradients during an adversarial attack to fool the motion dynamics. Specifically, we introduce spatial (image) and temporal (video) cues within the same source model through task-specific prompts. Attacking such prompts maximizes the adversarial transferability from image-to-video and image-to-image models using the attacks designed for image models. Our attack results indicate that the attacker does not need specialized architectures, \eg, divided space-time attention, 3D convolutions, or multi-view convolution networks for different data modalities. Image models are effective surrogates to optimize an adversarial attack to fool black-box models in a changing environment over time. Code is available at https://bit.ly/3Xd9gRQ
no_new_dataset
0.711982
2303.02760
Xuan Ju
Xuan Ju, Ailing Zeng, Jianan Wang, Qiang Xu, Lei Zhang
Human-Art: A Versatile Human-Centric Dataset Bridging Natural and Artificial Scenes
CVPR2023
null
null
null
cs.CV
http://creativecommons.org/licenses/by-nc-nd/4.0/
Humans have long been recorded in a variety of forms since antiquity. For example, sculptures and paintings were the primary media for depicting human beings before the invention of cameras. However, most current human-centric computer vision tasks like human pose estimation and human image generation focus exclusively on natural images in the real world. Artificial humans, such as those in sculptures, paintings, and cartoons, are commonly neglected, making existing models fail in these scenarios. As an abstraction of life, art incorporates humans in both natural and artificial scenes. We take advantage of it and introduce the Human-Art dataset to bridge related tasks in natural and artificial scenarios. Specifically, Human-Art contains 50k high-quality images with over 123k person instances from 5 natural and 15 artificial scenarios, which are annotated with bounding boxes, keypoints, self-contact points, and text information for humans represented in both 2D and 3D. It is, therefore, comprehensive and versatile for various downstream tasks. We also provide a rich set of baseline results and detailed analyses for related tasks, including human detection, 2D and 3D human pose estimation, image generation, and motion transfer. As a challenging dataset, we hope Human-Art can provide insights for relevant research and open up new research questions.
[ { "version": "v1", "created": "Sun, 5 Mar 2023 20:05:21 GMT" }, { "version": "v2", "created": "Wed, 5 Apr 2023 07:36:48 GMT" } ]
2023-04-06T00:00:00
[ [ "Ju", "Xuan", "" ], [ "Zeng", "Ailing", "" ], [ "Wang", "Jianan", "" ], [ "Xu", "Qiang", "" ], [ "Zhang", "Lei", "" ] ]
TITLE: Human-Art: A Versatile Human-Centric Dataset Bridging Natural and Artificial Scenes ABSTRACT: Humans have long been recorded in a variety of forms since antiquity. For example, sculptures and paintings were the primary media for depicting human beings before the invention of cameras. However, most current human-centric computer vision tasks like human pose estimation and human image generation focus exclusively on natural images in the real world. Artificial humans, such as those in sculptures, paintings, and cartoons, are commonly neglected, making existing models fail in these scenarios. As an abstraction of life, art incorporates humans in both natural and artificial scenes. We take advantage of it and introduce the Human-Art dataset to bridge related tasks in natural and artificial scenarios. Specifically, Human-Art contains 50k high-quality images with over 123k person instances from 5 natural and 15 artificial scenarios, which are annotated with bounding boxes, keypoints, self-contact points, and text information for humans represented in both 2D and 3D. It is, therefore, comprehensive and versatile for various downstream tasks. We also provide a rich set of baseline results and detailed analyses for related tasks, including human detection, 2D and 3D human pose estimation, image generation, and motion transfer. As a challenging dataset, we hope Human-Art can provide insights for relevant research and open up new research questions.
new_dataset
0.715206
2303.03750
Vukosi Marivate
Richard Lastrucci, Isheanesu Dzingirai, Jenalea Rajab, Andani Madodonga, Matimba Shingange, Daniel Njini, Vukosi Marivate
Preparing the Vuk'uzenzele and ZA-gov-multilingual South African multilingual corpora
Accepted and to appear at Fourth workshop on Resources for African Indigenous Languages (RAIL) at EACL 2023
null
null
null
cs.CL
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
This paper introduces two multilingual government themed corpora in various South African languages. The corpora were collected by gathering the South African Government newspaper (Vuk'uzenzele), as well as South African government speeches (ZA-gov-multilingual), that are translated into all 11 South African official languages. The corpora can be used for a myriad of downstream NLP tasks. The corpora were created to allow researchers to study the language used in South African government publications, with a focus on understanding how South African government officials communicate with their constituents. In this paper we highlight the process of gathering, cleaning and making available the corpora. We create parallel sentence corpora for Neural Machine Translation (NMT) tasks using Language-Agnostic Sentence Representations (LASER) embeddings. With these aligned sentences we then provide NMT benchmarks for 9 indigenous languages by fine-tuning a massively multilingual pre-trained language model.
[ { "version": "v1", "created": "Tue, 7 Mar 2023 09:20:09 GMT" }, { "version": "v2", "created": "Wed, 5 Apr 2023 09:39:32 GMT" } ]
2023-04-06T00:00:00
[ [ "Lastrucci", "Richard", "" ], [ "Dzingirai", "Isheanesu", "" ], [ "Rajab", "Jenalea", "" ], [ "Madodonga", "Andani", "" ], [ "Shingange", "Matimba", "" ], [ "Njini", "Daniel", "" ], [ "Marivate", "Vukosi", "" ] ]
TITLE: Preparing the Vuk'uzenzele and ZA-gov-multilingual South African multilingual corpora ABSTRACT: This paper introduces two multilingual government themed corpora in various South African languages. The corpora were collected by gathering the South African Government newspaper (Vuk'uzenzele), as well as South African government speeches (ZA-gov-multilingual), that are translated into all 11 South African official languages. The corpora can be used for a myriad of downstream NLP tasks. The corpora were created to allow researchers to study the language used in South African government publications, with a focus on understanding how South African government officials communicate with their constituents. In this paper we highlight the process of gathering, cleaning and making available the corpora. We create parallel sentence corpora for Neural Machine Translation (NMT) tasks using Language-Agnostic Sentence Representations (LASER) embeddings. With these aligned sentences we then provide NMT benchmarks for 9 indigenous languages by fine-tuning a massively multilingual pre-trained language model.
no_new_dataset
0.689554
2303.04803
Jiarui Xu
Jiarui Xu, Sifei Liu, Arash Vahdat, Wonmin Byeon, Xiaolong Wang, Shalini De Mello
Open-Vocabulary Panoptic Segmentation with Text-to-Image Diffusion Models
CVPR 2023 Highlight. Project page and code: https://jerryxu.net/ODISE
null
null
null
cs.CV
http://creativecommons.org/licenses/by/4.0/
We present ODISE: Open-vocabulary DIffusion-based panoptic SEgmentation, which unifies pre-trained text-image diffusion and discriminative models to perform open-vocabulary panoptic segmentation. Text-to-image diffusion models have the remarkable ability to generate high-quality images with diverse open-vocabulary language descriptions. This demonstrates that their internal representation space is highly correlated with open concepts in the real world. Text-image discriminative models like CLIP, on the other hand, are good at classifying images into open-vocabulary labels. We leverage the frozen internal representations of both these models to perform panoptic segmentation of any category in the wild. Our approach outperforms the previous state of the art by significant margins on both open-vocabulary panoptic and semantic segmentation tasks. In particular, with COCO training only, our method achieves 23.4 PQ and 30.0 mIoU on the ADE20K dataset, with 8.3 PQ and 7.9 mIoU absolute improvement over the previous state of the art. We open-source our code and models at https://github.com/NVlabs/ODISE .
[ { "version": "v1", "created": "Wed, 8 Mar 2023 18:58:26 GMT" }, { "version": "v2", "created": "Thu, 9 Mar 2023 01:37:25 GMT" }, { "version": "v3", "created": "Sat, 11 Mar 2023 09:39:35 GMT" }, { "version": "v4", "created": "Wed, 5 Apr 2023 17:40:38 GMT" } ]
2023-04-06T00:00:00
[ [ "Xu", "Jiarui", "" ], [ "Liu", "Sifei", "" ], [ "Vahdat", "Arash", "" ], [ "Byeon", "Wonmin", "" ], [ "Wang", "Xiaolong", "" ], [ "De Mello", "Shalini", "" ] ]
TITLE: Open-Vocabulary Panoptic Segmentation with Text-to-Image Diffusion Models ABSTRACT: We present ODISE: Open-vocabulary DIffusion-based panoptic SEgmentation, which unifies pre-trained text-image diffusion and discriminative models to perform open-vocabulary panoptic segmentation. Text-to-image diffusion models have the remarkable ability to generate high-quality images with diverse open-vocabulary language descriptions. This demonstrates that their internal representation space is highly correlated with open concepts in the real world. Text-image discriminative models like CLIP, on the other hand, are good at classifying images into open-vocabulary labels. We leverage the frozen internal representations of both these models to perform panoptic segmentation of any category in the wild. Our approach outperforms the previous state of the art by significant margins on both open-vocabulary panoptic and semantic segmentation tasks. In particular, with COCO training only, our method achieves 23.4 PQ and 30.0 mIoU on the ADE20K dataset, with 8.3 PQ and 7.9 mIoU absolute improvement over the previous state of the art. We open-source our code and models at https://github.com/NVlabs/ODISE .
no_new_dataset
0.712188
2303.06241
Vipul Gupta
Vipul Gupta, Apurva Narayan
Do we need entire training data for adversarial training?
6 pages, 4 figures
null
null
null
cs.CV cs.AI cs.LG
http://creativecommons.org/licenses/by/4.0/
Deep Neural Networks (DNNs) are being used to solve a wide range of problems in many domains including safety-critical domains like self-driving cars and medical imagery. DNNs suffer from vulnerability against adversarial attacks. In the past few years, numerous approaches have been proposed to tackle this problem by training networks using adversarial training. Almost all the approaches generate adversarial examples for the entire training dataset, thus increasing the training time drastically. We show that we can decrease the training time for any adversarial training algorithm by using only a subset of training data for adversarial training. To select the subset, we filter the adversarially-prone samples from the training data. We perform a simple adversarial attack on all training examples to filter this subset. In this attack, we add a small perturbation to each pixel and a few grid lines to the input image. We perform adversarial training on the adversarially-prone subset and mix it with vanilla training performed on the entire dataset. Our results show that when our method-agnostic approach is plugged into FGSM, we achieve a speedup of 3.52x on MNIST and 1.98x on the CIFAR-10 dataset with comparable robust accuracy. We also test our approach on state-of-the-art Free adversarial training and achieve a speedup of 1.2x in training time with a marginal drop in robust accuracy on the ImageNet dataset.
[ { "version": "v1", "created": "Fri, 10 Mar 2023 23:21:05 GMT" }, { "version": "v2", "created": "Wed, 5 Apr 2023 00:07:46 GMT" } ]
2023-04-06T00:00:00
[ [ "Gupta", "Vipul", "" ], [ "Narayan", "Apurva", "" ] ]
TITLE: Do we need entire training data for adversarial training? ABSTRACT: Deep Neural Networks (DNNs) are being used to solve a wide range of problems in many domains including safety-critical domains like self-driving cars and medical imagery. DNNs suffer from vulnerability against adversarial attacks. In the past few years, numerous approaches have been proposed to tackle this problem by training networks using adversarial training. Almost all the approaches generate adversarial examples for the entire training dataset, thus increasing the training time drastically. We show that we can decrease the training time for any adversarial training algorithm by using only a subset of training data for adversarial training. To select the subset, we filter the adversarially-prone samples from the training data. We perform a simple adversarial attack on all training examples to filter this subset. In this attack, we add a small perturbation to each pixel and a few grid lines to the input image. We perform adversarial training on the adversarially-prone subset and mix it with vanilla training performed on the entire dataset. Our results show that when our method-agnostic approach is plugged into FGSM, we achieve a speedup of 3.52x on MNIST and 1.98x on the CIFAR-10 dataset with comparable robust accuracy. We also test our approach on state-of-the-art Free adversarial training and achieve a speedup of 1.2x in training time with a marginal drop in robust accuracy on the ImageNet dataset.
no_new_dataset
0.709233
2303.11668
Yongjin Jeon
Yongjin Jeon, Youngtack Oh, Doyoung Jeong, Hyunguk Choi, Junsik Kim
Focus or Not: A Baseline for Anomaly Event Detection On the Open Public Places with Satellite Images
I am withdrawing my submission due to issues with content modification
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
In recent years, monitoring the world wide area with satellite images has been emerged as an important issue. Site monitoring task can be divided into two independent tasks; 1) Change Detection and 2) Anomaly Event Detection. Unlike to change detection research is actively conducted based on the numerous datasets(\eg LEVIR-CD, WHU-CD, S2Looking, xView2 and etc...) to meet up the expectations of industries or governments, research on AI models for detecting anomaly events is passively and rarely conducted. In this paper, we introduce a novel satellite imagery dataset(AED-RS) for detecting anomaly events on the open public places. AED-RS Dataset contains satellite images of normal and abnormal situations of 8 open public places from all over the world. Each places are labeled with different criteria based on the difference of characteristics of each places. With this dataset, we introduce a baseline model for our dataset TB-FLOW, which can be trained in weakly-supervised manner and shows reasonable performance on the AED-RS Dataset compared with the other NF(Normalizing-Flow) based anomaly detection models. Our dataset and code will be publicly open in \url{https://github.com/SIAnalytics/RS_AnomalyDetection.git}.
[ { "version": "v1", "created": "Tue, 21 Mar 2023 08:23:05 GMT" }, { "version": "v2", "created": "Wed, 5 Apr 2023 02:39:39 GMT" } ]
2023-04-06T00:00:00
[ [ "Jeon", "Yongjin", "" ], [ "Oh", "Youngtack", "" ], [ "Jeong", "Doyoung", "" ], [ "Choi", "Hyunguk", "" ], [ "Kim", "Junsik", "" ] ]
TITLE: Focus or Not: A Baseline for Anomaly Event Detection On the Open Public Places with Satellite Images ABSTRACT: In recent years, monitoring the world wide area with satellite images has been emerged as an important issue. Site monitoring task can be divided into two independent tasks; 1) Change Detection and 2) Anomaly Event Detection. Unlike to change detection research is actively conducted based on the numerous datasets(\eg LEVIR-CD, WHU-CD, S2Looking, xView2 and etc...) to meet up the expectations of industries or governments, research on AI models for detecting anomaly events is passively and rarely conducted. In this paper, we introduce a novel satellite imagery dataset(AED-RS) for detecting anomaly events on the open public places. AED-RS Dataset contains satellite images of normal and abnormal situations of 8 open public places from all over the world. Each places are labeled with different criteria based on the difference of characteristics of each places. With this dataset, we introduce a baseline model for our dataset TB-FLOW, which can be trained in weakly-supervised manner and shows reasonable performance on the AED-RS Dataset compared with the other NF(Normalizing-Flow) based anomaly detection models. Our dataset and code will be publicly open in \url{https://github.com/SIAnalytics/RS_AnomalyDetection.git}.
new_dataset
0.716039
2303.16976
Vishal Asnani
Vishal Asnani, Xi Yin, Tal Hassner, Xiaoming Liu
MaLP: Manipulation Localization Using a Proactive Scheme
Published at Conference on Computer Vision and Pattern Recognition 2023
null
null
null
cs.CV
http://creativecommons.org/licenses/by-nc-sa/4.0/
Advancements in the generation quality of various Generative Models (GMs) has made it necessary to not only perform binary manipulation detection but also localize the modified pixels in an image. However, prior works termed as passive for manipulation localization exhibit poor generalization performance over unseen GMs and attribute modifications. To combat this issue, we propose a proactive scheme for manipulation localization, termed MaLP. We encrypt the real images by adding a learned template. If the image is manipulated by any GM, this added protection from the template not only aids binary detection but also helps in identifying the pixels modified by the GM. The template is learned by leveraging local and global-level features estimated by a two-branch architecture. We show that MaLP performs better than prior passive works. We also show the generalizability of MaLP by testing on 22 different GMs, providing a benchmark for future research on manipulation localization. Finally, we show that MaLP can be used as a discriminator for improving the generation quality of GMs. Our models/codes are available at www.github.com/vishal3477/pro_loc.
[ { "version": "v1", "created": "Wed, 29 Mar 2023 19:17:06 GMT" }, { "version": "v2", "created": "Tue, 4 Apr 2023 18:30:45 GMT" } ]
2023-04-06T00:00:00
[ [ "Asnani", "Vishal", "" ], [ "Yin", "Xi", "" ], [ "Hassner", "Tal", "" ], [ "Liu", "Xiaoming", "" ] ]
TITLE: MaLP: Manipulation Localization Using a Proactive Scheme ABSTRACT: Advancements in the generation quality of various Generative Models (GMs) has made it necessary to not only perform binary manipulation detection but also localize the modified pixels in an image. However, prior works termed as passive for manipulation localization exhibit poor generalization performance over unseen GMs and attribute modifications. To combat this issue, we propose a proactive scheme for manipulation localization, termed MaLP. We encrypt the real images by adding a learned template. If the image is manipulated by any GM, this added protection from the template not only aids binary detection but also helps in identifying the pixels modified by the GM. The template is learned by leveraging local and global-level features estimated by a two-branch architecture. We show that MaLP performs better than prior passive works. We also show the generalizability of MaLP by testing on 22 different GMs, providing a benchmark for future research on manipulation localization. Finally, we show that MaLP can be used as a discriminator for improving the generation quality of GMs. Our models/codes are available at www.github.com/vishal3477/pro_loc.
no_new_dataset
0.711425
2303.16999
Dominic Masters
Zhiyi Li, Douglas Orr, Valeriu Ohan, Godfrey Da costa, Tom Murray, Adam Sanders, Deniz Beker, Dominic Masters
PopSparse: Accelerated block sparse matrix multiplication on IPU
null
null
null
null
cs.LG
http://creativecommons.org/licenses/by-sa/4.0/
Reducing the computational cost of running large scale neural networks using sparsity has attracted great attention in the deep learning community. While much success has been achieved in reducing FLOP and parameter counts while maintaining acceptable task performance, achieving actual speed improvements has typically been much more difficult, particularly on general purpose accelerators (GPAs) such as NVIDIA GPUs using low precision number formats. In this work we introduce PopSparse, a library that enables fast sparse operations on Graphcore IPUs by leveraging both the unique hardware characteristics of IPUs as well as any block structure defined in the data. We target two different types of sparsity: static, where the sparsity pattern is fixed at compile-time; and dynamic, where it can change each time the model is run. We present benchmark results for matrix multiplication for both of these modes on IPU with a range of block sizes, matrix sizes and densities. Results indicate that the PopSparse implementations are faster than dense matrix multiplications on IPU at a range of sparsity levels with large matrix size and block size. Furthermore, static sparsity in general outperforms dynamic sparsity. While previous work on GPAs has shown speedups only for very high sparsity (typically 99\% and above), the present work demonstrates that our static sparse implementation outperforms equivalent dense calculations in FP16 at lower sparsity (around 90%). IPU code is available to view and run at ipu.dev/sparsity-benchmarks, GPU code will be made available shortly.
[ { "version": "v1", "created": "Wed, 29 Mar 2023 20:00:19 GMT" }, { "version": "v2", "created": "Wed, 5 Apr 2023 13:43:15 GMT" } ]
2023-04-06T00:00:00
[ [ "Li", "Zhiyi", "" ], [ "Orr", "Douglas", "" ], [ "Ohan", "Valeriu", "" ], [ "Da costa", "Godfrey", "" ], [ "Murray", "Tom", "" ], [ "Sanders", "Adam", "" ], [ "Beker", "Deniz", "" ], [ "Masters", "Dominic", "" ] ]
TITLE: PopSparse: Accelerated block sparse matrix multiplication on IPU ABSTRACT: Reducing the computational cost of running large scale neural networks using sparsity has attracted great attention in the deep learning community. While much success has been achieved in reducing FLOP and parameter counts while maintaining acceptable task performance, achieving actual speed improvements has typically been much more difficult, particularly on general purpose accelerators (GPAs) such as NVIDIA GPUs using low precision number formats. In this work we introduce PopSparse, a library that enables fast sparse operations on Graphcore IPUs by leveraging both the unique hardware characteristics of IPUs as well as any block structure defined in the data. We target two different types of sparsity: static, where the sparsity pattern is fixed at compile-time; and dynamic, where it can change each time the model is run. We present benchmark results for matrix multiplication for both of these modes on IPU with a range of block sizes, matrix sizes and densities. Results indicate that the PopSparse implementations are faster than dense matrix multiplications on IPU at a range of sparsity levels with large matrix size and block size. Furthermore, static sparsity in general outperforms dynamic sparsity. While previous work on GPAs has shown speedups only for very high sparsity (typically 99\% and above), the present work demonstrates that our static sparse implementation outperforms equivalent dense calculations in FP16 at lower sparsity (around 90%). IPU code is available to view and run at ipu.dev/sparsity-benchmarks, GPU code will be made available shortly.
no_new_dataset
0.702734
2303.17802
Kazuhiro Fukui
Takumi Kanai, Naoya Sogi, Atsuto Maki, Kazuhiro Fukui
Time-series Anomaly Detection based on Difference Subspace between Signal Subspaces
8pages, an acknowledgement was added to v1
null
null
null
cs.LG cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
This paper proposes a new method for anomaly detection in time-series data by incorporating the concept of difference subspace into the singular spectrum analysis (SSA). The key idea is to monitor slight temporal variations of the difference subspace between two signal subspaces corresponding to the past and present time-series data, as anomaly score. It is a natural generalization of the conventional SSA-based method which measures the minimum angle between the two signal subspaces as the degree of changes. By replacing the minimum angle with the difference subspace, our method boosts the performance while using the SSA-based framework as it can capture the whole structural difference between the two subspaces in its magnitude and direction. We demonstrate our method's effectiveness through performance evaluations on public time-series datasets.
[ { "version": "v1", "created": "Fri, 31 Mar 2023 05:22:56 GMT" }, { "version": "v2", "created": "Wed, 5 Apr 2023 02:13:48 GMT" } ]
2023-04-06T00:00:00
[ [ "Kanai", "Takumi", "" ], [ "Sogi", "Naoya", "" ], [ "Maki", "Atsuto", "" ], [ "Fukui", "Kazuhiro", "" ] ]
TITLE: Time-series Anomaly Detection based on Difference Subspace between Signal Subspaces ABSTRACT: This paper proposes a new method for anomaly detection in time-series data by incorporating the concept of difference subspace into the singular spectrum analysis (SSA). The key idea is to monitor slight temporal variations of the difference subspace between two signal subspaces corresponding to the past and present time-series data, as anomaly score. It is a natural generalization of the conventional SSA-based method which measures the minimum angle between the two signal subspaces as the degree of changes. By replacing the minimum angle with the difference subspace, our method boosts the performance while using the SSA-based framework as it can capture the whole structural difference between the two subspaces in its magnitude and direction. We demonstrate our method's effectiveness through performance evaluations on public time-series datasets.
no_new_dataset
0.711481
2303.18027
Jungo Kasai
Jungo Kasai, Yuhei Kasai, Keisuke Sakaguchi, Yutaro Yamada, Dragomir Radev
Evaluating GPT-4 and ChatGPT on Japanese Medical Licensing Examinations
Added results from the March 2023 exam
null
null
null
cs.CL
http://creativecommons.org/licenses/by/4.0/
As large language models (LLMs) gain popularity among speakers of diverse languages, we believe that it is crucial to benchmark them to better understand model behaviors, failures, and limitations in languages beyond English. In this work, we evaluate LLM APIs (ChatGPT, GPT-3, and GPT-4) on the Japanese national medical licensing examinations from the past five years, including the current year. Our team comprises native Japanese-speaking NLP researchers and a practicing cardiologist based in Japan. Our experiments show that GPT-4 outperforms ChatGPT and GPT-3 and passes all six years of the exams, highlighting LLMs' potential in a language that is typologically distant from English. However, our evaluation also exposes critical limitations of the current LLM APIs. First, LLMs sometimes select prohibited choices that should be strictly avoided in medical practice in Japan, such as suggesting euthanasia. Further, our analysis shows that the API costs are generally higher and the maximum context size is smaller for Japanese because of the way non-Latin scripts are currently tokenized in the pipeline. We release our benchmark as Igaku QA as well as all model outputs and exam metadata. We hope that our results and benchmark will spur progress on more diverse applications of LLMs. Our benchmark is available at https://github.com/jungokasai/IgakuQA.
[ { "version": "v1", "created": "Fri, 31 Mar 2023 13:04:47 GMT" }, { "version": "v2", "created": "Wed, 5 Apr 2023 07:53:37 GMT" } ]
2023-04-06T00:00:00
[ [ "Kasai", "Jungo", "" ], [ "Kasai", "Yuhei", "" ], [ "Sakaguchi", "Keisuke", "" ], [ "Yamada", "Yutaro", "" ], [ "Radev", "Dragomir", "" ] ]
TITLE: Evaluating GPT-4 and ChatGPT on Japanese Medical Licensing Examinations ABSTRACT: As large language models (LLMs) gain popularity among speakers of diverse languages, we believe that it is crucial to benchmark them to better understand model behaviors, failures, and limitations in languages beyond English. In this work, we evaluate LLM APIs (ChatGPT, GPT-3, and GPT-4) on the Japanese national medical licensing examinations from the past five years, including the current year. Our team comprises native Japanese-speaking NLP researchers and a practicing cardiologist based in Japan. Our experiments show that GPT-4 outperforms ChatGPT and GPT-3 and passes all six years of the exams, highlighting LLMs' potential in a language that is typologically distant from English. However, our evaluation also exposes critical limitations of the current LLM APIs. First, LLMs sometimes select prohibited choices that should be strictly avoided in medical practice in Japan, such as suggesting euthanasia. Further, our analysis shows that the API costs are generally higher and the maximum context size is smaller for Japanese because of the way non-Latin scripts are currently tokenized in the pipeline. We release our benchmark as Igaku QA as well as all model outputs and exam metadata. We hope that our results and benchmark will spur progress on more diverse applications of LLMs. Our benchmark is available at https://github.com/jungokasai/IgakuQA.
no_new_dataset
0.564339
2303.18139
Thomas Tanay
Thomas Tanay and Ale\v{s} Leonardis and Matteo Maggioni
Efficient View Synthesis and 3D-based Multi-Frame Denoising with Multiplane Feature Representations
Accepted at CVPR 2023
null
null
null
cs.CV
http://creativecommons.org/licenses/by/4.0/
While current multi-frame restoration methods combine information from multiple input images using 2D alignment techniques, recent advances in novel view synthesis are paving the way for a new paradigm relying on volumetric scene representations. In this work, we introduce the first 3D-based multi-frame denoising method that significantly outperforms its 2D-based counterparts with lower computational requirements. Our method extends the multiplane image (MPI) framework for novel view synthesis by introducing a learnable encoder-renderer pair manipulating multiplane representations in feature space. The encoder fuses information across views and operates in a depth-wise manner while the renderer fuses information across depths and operates in a view-wise manner. The two modules are trained end-to-end and learn to separate depths in an unsupervised way, giving rise to Multiplane Feature (MPF) representations. Experiments on the Spaces and Real Forward-Facing datasets as well as on raw burst data validate our approach for view synthesis, multi-frame denoising, and view synthesis under noisy conditions.
[ { "version": "v1", "created": "Fri, 31 Mar 2023 15:23:35 GMT" }, { "version": "v2", "created": "Wed, 5 Apr 2023 11:08:37 GMT" } ]
2023-04-06T00:00:00
[ [ "Tanay", "Thomas", "" ], [ "Leonardis", "Aleš", "" ], [ "Maggioni", "Matteo", "" ] ]
TITLE: Efficient View Synthesis and 3D-based Multi-Frame Denoising with Multiplane Feature Representations ABSTRACT: While current multi-frame restoration methods combine information from multiple input images using 2D alignment techniques, recent advances in novel view synthesis are paving the way for a new paradigm relying on volumetric scene representations. In this work, we introduce the first 3D-based multi-frame denoising method that significantly outperforms its 2D-based counterparts with lower computational requirements. Our method extends the multiplane image (MPI) framework for novel view synthesis by introducing a learnable encoder-renderer pair manipulating multiplane representations in feature space. The encoder fuses information across views and operates in a depth-wise manner while the renderer fuses information across depths and operates in a view-wise manner. The two modules are trained end-to-end and learn to separate depths in an unsupervised way, giving rise to Multiplane Feature (MPF) representations. Experiments on the Spaces and Real Forward-Facing datasets as well as on raw burst data validate our approach for view synthesis, multi-frame denoising, and view synthesis under noisy conditions.
no_new_dataset
0.709831
2304.01352
Karen Avetisyan
Karen Avetisyan, Arthur Malajyan, Tsolak Ghukasyan, Arutyun Avetisyan
A Simple and Effective Method of Cross-Lingual Plagiarism Detection
null
null
null
null
cs.CL cs.IR
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
We present a simple cross-lingual plagiarism detection method applicable to a large number of languages. The presented approach leverages open multilingual thesauri for candidate retrieval task and pre-trained multilingual BERT-based language models for detailed analysis. The method does not rely on machine translation and word sense disambiguation when in use, and therefore is suitable for a large number of languages, including under-resourced languages. The effectiveness of the proposed approach is demonstrated for several existing and new benchmarks, achieving state-of-the-art results for French, Russian, and Armenian languages.
[ { "version": "v1", "created": "Mon, 3 Apr 2023 20:27:10 GMT" }, { "version": "v2", "created": "Wed, 5 Apr 2023 09:23:17 GMT" } ]
2023-04-06T00:00:00
[ [ "Avetisyan", "Karen", "" ], [ "Malajyan", "Arthur", "" ], [ "Ghukasyan", "Tsolak", "" ], [ "Avetisyan", "Arutyun", "" ] ]
TITLE: A Simple and Effective Method of Cross-Lingual Plagiarism Detection ABSTRACT: We present a simple cross-lingual plagiarism detection method applicable to a large number of languages. The presented approach leverages open multilingual thesauri for candidate retrieval task and pre-trained multilingual BERT-based language models for detailed analysis. The method does not rely on machine translation and word sense disambiguation when in use, and therefore is suitable for a large number of languages, including under-resourced languages. The effectiveness of the proposed approach is demonstrated for several existing and new benchmarks, achieving state-of-the-art results for French, Russian, and Armenian languages.
no_new_dataset
0.705011
2304.01487
Kavita Kumari
Alessandro Pegoraro, Kavita Kumari, Hossein Fereidooni, Ahmad-Reza Sadeghi
To ChatGPT, or not to ChatGPT: That is the question!
null
null
null
null
cs.LG cs.AI cs.CL
http://creativecommons.org/licenses/by-sa/4.0/
ChatGPT has become a global sensation. As ChatGPT and other Large Language Models (LLMs) emerge, concerns of misusing them in various ways increase, such as disseminating fake news, plagiarism, manipulating public opinion, cheating, and fraud. Hence, distinguishing AI-generated from human-generated becomes increasingly essential. Researchers have proposed various detection methodologies, ranging from basic binary classifiers to more complex deep-learning models. Some detection techniques rely on statistical characteristics or syntactic patterns, while others incorporate semantic or contextual information to improve accuracy. The primary objective of this study is to provide a comprehensive and contemporary assessment of the most recent techniques in ChatGPT detection. Additionally, we evaluated other AI-generated text detection tools that do not specifically claim to detect ChatGPT-generated content to assess their performance in detecting ChatGPT-generated content. For our evaluation, we have curated a benchmark dataset consisting of prompts from ChatGPT and humans, including diverse questions from medical, open Q&A, and finance domains and user-generated responses from popular social networking platforms. The dataset serves as a reference to assess the performance of various techniques in detecting ChatGPT-generated content. Our evaluation results demonstrate that none of the existing methods can effectively detect ChatGPT-generated content.
[ { "version": "v1", "created": "Tue, 4 Apr 2023 03:04:28 GMT" }, { "version": "v2", "created": "Wed, 5 Apr 2023 04:28:41 GMT" } ]
2023-04-06T00:00:00
[ [ "Pegoraro", "Alessandro", "" ], [ "Kumari", "Kavita", "" ], [ "Fereidooni", "Hossein", "" ], [ "Sadeghi", "Ahmad-Reza", "" ] ]
TITLE: To ChatGPT, or not to ChatGPT: That is the question! ABSTRACT: ChatGPT has become a global sensation. As ChatGPT and other Large Language Models (LLMs) emerge, concerns of misusing them in various ways increase, such as disseminating fake news, plagiarism, manipulating public opinion, cheating, and fraud. Hence, distinguishing AI-generated from human-generated becomes increasingly essential. Researchers have proposed various detection methodologies, ranging from basic binary classifiers to more complex deep-learning models. Some detection techniques rely on statistical characteristics or syntactic patterns, while others incorporate semantic or contextual information to improve accuracy. The primary objective of this study is to provide a comprehensive and contemporary assessment of the most recent techniques in ChatGPT detection. Additionally, we evaluated other AI-generated text detection tools that do not specifically claim to detect ChatGPT-generated content to assess their performance in detecting ChatGPT-generated content. For our evaluation, we have curated a benchmark dataset consisting of prompts from ChatGPT and humans, including diverse questions from medical, open Q&A, and finance domains and user-generated responses from popular social networking platforms. The dataset serves as a reference to assess the performance of various techniques in detecting ChatGPT-generated content. Our evaluation results demonstrate that none of the existing methods can effectively detect ChatGPT-generated content.
new_dataset
0.714454
2304.01686
Dang Pham Bang
Bang-Dang Pham, Phong Tran, Anh Tran, Cuong Pham, Rang Nguyen, Minh Hoai
HyperCUT: Video Sequence from a Single Blurry Image using Unsupervised Ordering
Accepted to CVPR 2023
null
null
null
cs.CV cs.AI
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
We consider the challenging task of training models for image-to-video deblurring, which aims to recover a sequence of sharp images corresponding to a given blurry image input. A critical issue disturbing the training of an image-to-video model is the ambiguity of the frame ordering since both the forward and backward sequences are plausible solutions. This paper proposes an effective self-supervised ordering scheme that allows training high-quality image-to-video deblurring models. Unlike previous methods that rely on order-invariant losses, we assign an explicit order for each video sequence, thus avoiding the order-ambiguity issue. Specifically, we map each video sequence to a vector in a latent high-dimensional space so that there exists a hyperplane such that for every video sequence, the vectors extracted from it and its reversed sequence are on different sides of the hyperplane. The side of the vectors will be used to define the order of the corresponding sequence. Last but not least, we propose a real-image dataset for the image-to-video deblurring problem that covers a variety of popular domains, including face, hand, and street. Extensive experimental results confirm the effectiveness of our method. Code and data are available at https://github.com/VinAIResearch/HyperCUT.git
[ { "version": "v1", "created": "Tue, 4 Apr 2023 10:29:42 GMT" }, { "version": "v2", "created": "Wed, 5 Apr 2023 12:27:50 GMT" } ]
2023-04-06T00:00:00
[ [ "Pham", "Bang-Dang", "" ], [ "Tran", "Phong", "" ], [ "Tran", "Anh", "" ], [ "Pham", "Cuong", "" ], [ "Nguyen", "Rang", "" ], [ "Hoai", "Minh", "" ] ]
TITLE: HyperCUT: Video Sequence from a Single Blurry Image using Unsupervised Ordering ABSTRACT: We consider the challenging task of training models for image-to-video deblurring, which aims to recover a sequence of sharp images corresponding to a given blurry image input. A critical issue disturbing the training of an image-to-video model is the ambiguity of the frame ordering since both the forward and backward sequences are plausible solutions. This paper proposes an effective self-supervised ordering scheme that allows training high-quality image-to-video deblurring models. Unlike previous methods that rely on order-invariant losses, we assign an explicit order for each video sequence, thus avoiding the order-ambiguity issue. Specifically, we map each video sequence to a vector in a latent high-dimensional space so that there exists a hyperplane such that for every video sequence, the vectors extracted from it and its reversed sequence are on different sides of the hyperplane. The side of the vectors will be used to define the order of the corresponding sequence. Last but not least, we propose a real-image dataset for the image-to-video deblurring problem that covers a variety of popular domains, including face, hand, and street. Extensive experimental results confirm the effectiveness of our method. Code and data are available at https://github.com/VinAIResearch/HyperCUT.git
new_dataset
0.651355
2304.01905
Jing Liu
Saumya Y. Sahai, Jing Liu, Thejaswi Muniyappa, Kanthashree M. Sathyendra, Anastasios Alexandridis, Grant P. Strimel, Ross McGowan, Ariya Rastrow, Feng-Ju Chang, Athanasios Mouchtaris and Siegfried Kunzmann
Dual-Attention Neural Transducers for Efficient Wake Word Spotting in Speech Recognition
Accepted to Proc. IEEE ICASSP 2023
null
null
null
cs.SD cs.CL cs.LG eess.AS
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
We present dual-attention neural biasing, an architecture designed to boost Wake Words (WW) recognition and improve inference time latency on speech recognition tasks. This architecture enables a dynamic switch for its runtime compute paths by exploiting WW spotting to select which branch of its attention networks to execute for an input audio frame. With this approach, we effectively improve WW spotting accuracy while saving runtime compute cost as defined by floating point operations (FLOPs). Using an in-house de-identified dataset, we demonstrate that the proposed dual-attention network can reduce the compute cost by $90\%$ for WW audio frames, with only $1\%$ increase in the number of parameters. This architecture improves WW F1 score by $16\%$ relative and improves generic rare word error rate by $3\%$ relative compared to the baselines.
[ { "version": "v1", "created": "Mon, 3 Apr 2023 01:19:39 GMT" }, { "version": "v2", "created": "Wed, 5 Apr 2023 01:22:38 GMT" } ]
2023-04-06T00:00:00
[ [ "Sahai", "Saumya Y.", "" ], [ "Liu", "Jing", "" ], [ "Muniyappa", "Thejaswi", "" ], [ "Sathyendra", "Kanthashree M.", "" ], [ "Alexandridis", "Anastasios", "" ], [ "Strimel", "Grant P.", "" ], [ "McGowan", "Ross", "" ], [ "Rastrow", "Ariya", "" ], [ "Chang", "Feng-Ju", "" ], [ "Mouchtaris", "Athanasios", "" ], [ "Kunzmann", "Siegfried", "" ] ]
TITLE: Dual-Attention Neural Transducers for Efficient Wake Word Spotting in Speech Recognition ABSTRACT: We present dual-attention neural biasing, an architecture designed to boost Wake Words (WW) recognition and improve inference time latency on speech recognition tasks. This architecture enables a dynamic switch for its runtime compute paths by exploiting WW spotting to select which branch of its attention networks to execute for an input audio frame. With this approach, we effectively improve WW spotting accuracy while saving runtime compute cost as defined by floating point operations (FLOPs). Using an in-house de-identified dataset, we demonstrate that the proposed dual-attention network can reduce the compute cost by $90\%$ for WW audio frames, with only $1\%$ increase in the number of parameters. This architecture improves WW F1 score by $16\%$ relative and improves generic rare word error rate by $3\%$ relative compared to the baselines.
no_new_dataset
0.666985
2304.02015
Zheng Yuan
Zheng Yuan, Hongyi Yuan, Chuanqi Tan, Wei Wang, Songfang Huang
How well do Large Language Models perform in Arithmetic tasks?
null
null
null
null
cs.CL cs.AI
http://creativecommons.org/licenses/by/4.0/
Large language models have emerged abilities including chain-of-thought to answer math word problems step by step. Solving math word problems not only requires abilities to disassemble problems via chain-of-thought but also needs to calculate arithmetic expressions correctly for each step. To the best of our knowledge, there is no work to focus on evaluating the arithmetic ability of large language models. In this work, we propose an arithmetic dataset MATH 401 to test the latest large language models including GPT-4, ChatGPT, InstrctGPT, Galactica, and LLaMA with various arithmetic expressions and provide a detailed analysis of the ability of large language models. MATH 401 and evaluation codes are released at \url{https://github.com/GanjinZero/math401-llm}.
[ { "version": "v1", "created": "Thu, 16 Mar 2023 09:28:15 GMT" } ]
2023-04-06T00:00:00
[ [ "Yuan", "Zheng", "" ], [ "Yuan", "Hongyi", "" ], [ "Tan", "Chuanqi", "" ], [ "Wang", "Wei", "" ], [ "Huang", "Songfang", "" ] ]
TITLE: How well do Large Language Models perform in Arithmetic tasks? ABSTRACT: Large language models have emerged abilities including chain-of-thought to answer math word problems step by step. Solving math word problems not only requires abilities to disassemble problems via chain-of-thought but also needs to calculate arithmetic expressions correctly for each step. To the best of our knowledge, there is no work to focus on evaluating the arithmetic ability of large language models. In this work, we propose an arithmetic dataset MATH 401 to test the latest large language models including GPT-4, ChatGPT, InstrctGPT, Galactica, and LLaMA with various arithmetic expressions and provide a detailed analysis of the ability of large language models. MATH 401 and evaluation codes are released at \url{https://github.com/GanjinZero/math401-llm}.
new_dataset
0.714653
2304.02023
Atalanti A. Mastakouri
Numair Sani, Atalanti A. Mastakouri, Dominik Janzing
Bounding probabilities of causation through the causal marginal problem
null
null
null
null
stat.ML cs.IT math.IT stat.ME
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Probabilities of Causation play a fundamental role in decision making in law, health care and public policy. Nevertheless, their point identification is challenging, requiring strong assumptions such as monotonicity. In the absence of such assumptions, existing work requires multiple observations of datasets that contain the same treatment and outcome variables, in order to establish bounds on these probabilities. However, in many clinical trials and public policy evaluation cases, there exist independent datasets that examine the effect of a different treatment each on the same outcome variable. Here, we outline how to significantly tighten existing bounds on the probabilities of causation, by imposing counterfactual consistency between SCMs constructed from such independent datasets ('causal marginal problem'). Next, we describe a new information theoretic approach on falsification of counterfactual probabilities, using conditional mutual information to quantify counterfactual influence. The latter generalises to arbitrary discrete variables and number of treatments, and renders the causal marginal problem more interpretable. Since the question of 'tight enough' is left to the user, we provide an additional method of inference when the bounds are unsatisfactory: A maximum entropy based method that defines a metric for the space of plausible SCMs and proposes the entropy maximising SCM for inferring counterfactuals in the absence of more information.
[ { "version": "v1", "created": "Tue, 4 Apr 2023 12:16:38 GMT" } ]
2023-04-06T00:00:00
[ [ "Sani", "Numair", "" ], [ "Mastakouri", "Atalanti A.", "" ], [ "Janzing", "Dominik", "" ] ]
TITLE: Bounding probabilities of causation through the causal marginal problem ABSTRACT: Probabilities of Causation play a fundamental role in decision making in law, health care and public policy. Nevertheless, their point identification is challenging, requiring strong assumptions such as monotonicity. In the absence of such assumptions, existing work requires multiple observations of datasets that contain the same treatment and outcome variables, in order to establish bounds on these probabilities. However, in many clinical trials and public policy evaluation cases, there exist independent datasets that examine the effect of a different treatment each on the same outcome variable. Here, we outline how to significantly tighten existing bounds on the probabilities of causation, by imposing counterfactual consistency between SCMs constructed from such independent datasets ('causal marginal problem'). Next, we describe a new information theoretic approach on falsification of counterfactual probabilities, using conditional mutual information to quantify counterfactual influence. The latter generalises to arbitrary discrete variables and number of treatments, and renders the causal marginal problem more interpretable. Since the question of 'tight enough' is left to the user, we provide an additional method of inference when the bounds are unsatisfactory: A maximum entropy based method that defines a metric for the space of plausible SCMs and proposes the entropy maximising SCM for inferring counterfactuals in the absence of more information.
no_new_dataset
0.709831
2304.02064
Qi Chen
Qi Chen, Mario Marchand
Algorithm-Dependent Bounds for Representation Learning of Multi-Source Domain Adaptation
null
null
null
null
cs.LG stat.ML
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
We use information-theoretic tools to derive a novel analysis of Multi-source Domain Adaptation (MDA) from the representation learning perspective. Concretely, we study joint distribution alignment for supervised MDA with few target labels and unsupervised MDA with pseudo labels, where the latter is relatively hard and less commonly studied. We further provide algorithm-dependent generalization bounds for these two settings, where the generalization is characterized by the mutual information between the parameters and the data. Then we propose a novel deep MDA algorithm, implicitly addressing the target shift through joint alignment. Finally, the mutual information bounds are extended to this algorithm providing a non-vacuous gradient-norm estimation. The proposed algorithm has comparable performance to the state-of-the-art on target-shifted MDA benchmark with improved memory efficiency.
[ { "version": "v1", "created": "Tue, 4 Apr 2023 18:32:20 GMT" } ]
2023-04-06T00:00:00
[ [ "Chen", "Qi", "" ], [ "Marchand", "Mario", "" ] ]
TITLE: Algorithm-Dependent Bounds for Representation Learning of Multi-Source Domain Adaptation ABSTRACT: We use information-theoretic tools to derive a novel analysis of Multi-source Domain Adaptation (MDA) from the representation learning perspective. Concretely, we study joint distribution alignment for supervised MDA with few target labels and unsupervised MDA with pseudo labels, where the latter is relatively hard and less commonly studied. We further provide algorithm-dependent generalization bounds for these two settings, where the generalization is characterized by the mutual information between the parameters and the data. Then we propose a novel deep MDA algorithm, implicitly addressing the target shift through joint alignment. Finally, the mutual information bounds are extended to this algorithm providing a non-vacuous gradient-norm estimation. The proposed algorithm has comparable performance to the state-of-the-art on target-shifted MDA benchmark with improved memory efficiency.
no_new_dataset
0.709466
2304.02080
Vladislav Lialin
Vladislav Lialin, Stephen Rawls, David Chan, Shalini Ghosh, Anna Rumshisky, Wael Hamza
Scalable and Accurate Self-supervised Multimodal Representation Learning without Aligned Video and Text Data
null
2023 IEEE/CVF Winter Conference on Applications of Computer Vision Workshops (WACVW)
10.1109/WACVW58289.2023.00043
null
cs.CV cs.CL
http://creativecommons.org/licenses/by/4.0/
Scaling up weakly-supervised datasets has shown to be highly effective in the image-text domain and has contributed to most of the recent state-of-the-art computer vision and multimodal neural networks. However, existing large-scale video-text datasets and mining techniques suffer from several limitations, such as the scarcity of aligned data, the lack of diversity in the data, and the difficulty of collecting aligned data. Currently popular video-text data mining approach via automatic speech recognition (ASR) used in HowTo100M provides low-quality captions that often do not refer to the video content. Other mining approaches do not provide proper language descriptions (video tags) and are biased toward short clips (alt text). In this work, we show how recent advances in image captioning allow us to pre-train high-quality video models without any parallel video-text data. We pre-train several video captioning models that are based on an OPT language model and a TimeSformer visual backbone. We fine-tune these networks on several video captioning datasets. First, we demonstrate that image captioning pseudolabels work better for pre-training than the existing HowTo100M ASR captions. Second, we show that pre-training on both images and videos produces a significantly better network (+4 CIDER on MSR-VTT) than pre-training on a single modality. Our methods are complementary to the existing pre-training or data mining approaches and can be used in a variety of settings. Given the efficacy of the pseudolabeling method, we are planning to publicly release the generated captions.
[ { "version": "v1", "created": "Tue, 4 Apr 2023 19:11:05 GMT" } ]
2023-04-06T00:00:00
[ [ "Lialin", "Vladislav", "" ], [ "Rawls", "Stephen", "" ], [ "Chan", "David", "" ], [ "Ghosh", "Shalini", "" ], [ "Rumshisky", "Anna", "" ], [ "Hamza", "Wael", "" ] ]
TITLE: Scalable and Accurate Self-supervised Multimodal Representation Learning without Aligned Video and Text Data ABSTRACT: Scaling up weakly-supervised datasets has shown to be highly effective in the image-text domain and has contributed to most of the recent state-of-the-art computer vision and multimodal neural networks. However, existing large-scale video-text datasets and mining techniques suffer from several limitations, such as the scarcity of aligned data, the lack of diversity in the data, and the difficulty of collecting aligned data. Currently popular video-text data mining approach via automatic speech recognition (ASR) used in HowTo100M provides low-quality captions that often do not refer to the video content. Other mining approaches do not provide proper language descriptions (video tags) and are biased toward short clips (alt text). In this work, we show how recent advances in image captioning allow us to pre-train high-quality video models without any parallel video-text data. We pre-train several video captioning models that are based on an OPT language model and a TimeSformer visual backbone. We fine-tune these networks on several video captioning datasets. First, we demonstrate that image captioning pseudolabels work better for pre-training than the existing HowTo100M ASR captions. Second, we show that pre-training on both images and videos produces a significantly better network (+4 CIDER on MSR-VTT) than pre-training on a single modality. Our methods are complementary to the existing pre-training or data mining approaches and can be used in a variety of settings. Given the efficacy of the pseudolabeling method, we are planning to publicly release the generated captions.
no_new_dataset
0.710409
2304.02135
Thanh-Dat Truong
Thanh-Dat Truong, Ngan Le, Bhiksha Raj, Jackson Cothren, Khoa Luu
FREDOM: Fairness Domain Adaptation Approach to Semantic Scene Understanding
Accepted to CVPR'23
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Although Domain Adaptation in Semantic Scene Segmentation has shown impressive improvement in recent years, the fairness concerns in the domain adaptation have yet to be well defined and addressed. In addition, fairness is one of the most critical aspects when deploying the segmentation models into human-related real-world applications, e.g., autonomous driving, as any unfair predictions could influence human safety. In this paper, we propose a novel Fairness Domain Adaptation (FREDOM) approach to semantic scene segmentation. In particular, from the proposed formulated fairness objective, a new adaptation framework will be introduced based on the fair treatment of class distributions. Moreover, to generally model the context of structural dependency, a new conditional structural constraint is introduced to impose the consistency of predicted segmentation. Thanks to the proposed Conditional Structure Network, the self-attention mechanism has sufficiently modeled the structural information of segmentation. Through the ablation studies, the proposed method has shown the performance improvement of the segmentation models and promoted fairness in the model predictions. The experimental results on the two standard benchmarks, i.e., SYNTHIA $\to$ Cityscapes and GTA5 $\to$ Cityscapes, have shown that our method achieved State-of-the-Art (SOTA) performance.
[ { "version": "v1", "created": "Tue, 4 Apr 2023 21:35:10 GMT" } ]
2023-04-06T00:00:00
[ [ "Truong", "Thanh-Dat", "" ], [ "Le", "Ngan", "" ], [ "Raj", "Bhiksha", "" ], [ "Cothren", "Jackson", "" ], [ "Luu", "Khoa", "" ] ]
TITLE: FREDOM: Fairness Domain Adaptation Approach to Semantic Scene Understanding ABSTRACT: Although Domain Adaptation in Semantic Scene Segmentation has shown impressive improvement in recent years, the fairness concerns in the domain adaptation have yet to be well defined and addressed. In addition, fairness is one of the most critical aspects when deploying the segmentation models into human-related real-world applications, e.g., autonomous driving, as any unfair predictions could influence human safety. In this paper, we propose a novel Fairness Domain Adaptation (FREDOM) approach to semantic scene segmentation. In particular, from the proposed formulated fairness objective, a new adaptation framework will be introduced based on the fair treatment of class distributions. Moreover, to generally model the context of structural dependency, a new conditional structural constraint is introduced to impose the consistency of predicted segmentation. Thanks to the proposed Conditional Structure Network, the self-attention mechanism has sufficiently modeled the structural information of segmentation. Through the ablation studies, the proposed method has shown the performance improvement of the segmentation models and promoted fairness in the model predictions. The experimental results on the two standard benchmarks, i.e., SYNTHIA $\to$ Cityscapes and GTA5 $\to$ Cityscapes, have shown that our method achieved State-of-the-Art (SOTA) performance.
no_new_dataset
0.709585
2304.02144
Siyi Guo
Siyi Guo, Negar Mokhberian, Kristina Lerman
A Data Fusion Framework for Multi-Domain Morality Learning
null
null
null
null
cs.CL
http://creativecommons.org/licenses/by/4.0/
Language models can be trained to recognize the moral sentiment of text, creating new opportunities to study the role of morality in human life. As interest in language and morality has grown, several ground truth datasets with moral annotations have been released. However, these datasets vary in the method of data collection, domain, topics, instructions for annotators, etc. Simply aggregating such heterogeneous datasets during training can yield models that fail to generalize well. We describe a data fusion framework for training on multiple heterogeneous datasets that improve performance and generalizability. The model uses domain adversarial training to align the datasets in feature space and a weighted loss function to deal with label shift. We show that the proposed framework achieves state-of-the-art performance in different datasets compared to prior works in morality inference.
[ { "version": "v1", "created": "Tue, 4 Apr 2023 22:05:02 GMT" } ]
2023-04-06T00:00:00
[ [ "Guo", "Siyi", "" ], [ "Mokhberian", "Negar", "" ], [ "Lerman", "Kristina", "" ] ]
TITLE: A Data Fusion Framework for Multi-Domain Morality Learning ABSTRACT: Language models can be trained to recognize the moral sentiment of text, creating new opportunities to study the role of morality in human life. As interest in language and morality has grown, several ground truth datasets with moral annotations have been released. However, these datasets vary in the method of data collection, domain, topics, instructions for annotators, etc. Simply aggregating such heterogeneous datasets during training can yield models that fail to generalize well. We describe a data fusion framework for training on multiple heterogeneous datasets that improve performance and generalizability. The model uses domain adversarial training to align the datasets in feature space and a weighted loss function to deal with label shift. We show that the proposed framework achieves state-of-the-art performance in different datasets compared to prior works in morality inference.
no_new_dataset
0.587056
2304.02147
Alec Diaz-Arias
Alec Diaz-Arias and Dmitriy Shin
ConvFormer: Parameter Reduction in Transformer Models for 3D Human Pose Estimation by Leveraging Dynamic Multi-Headed Convolutional Attention
null
null
null
null
cs.CV cs.LG
http://creativecommons.org/licenses/by-nc-sa/4.0/
Recently, fully-transformer architectures have replaced the defacto convolutional architecture for the 3D human pose estimation task. In this paper we propose \textbf{\textit{ConvFormer}}, a novel convolutional transformer that leverages a new \textbf{\textit{dynamic multi-headed convolutional self-attention}} mechanism for monocular 3D human pose estimation. We designed a spatial and temporal convolutional transformer to comprehensively model human joint relations within individual frames and globally across the motion sequence. Moreover, we introduce a novel notion of \textbf{\textit{temporal joints profile}} for our temporal ConvFormer that fuses complete temporal information immediately for a local neighborhood of joint features. We have quantitatively and qualitatively validated our method on three common benchmark datasets: Human3.6M, MPI-INF-3DHP, and HumanEva. Extensive experiments have been conducted to identify the optimal hyper-parameter set. These experiments demonstrated that we achieved a \textbf{significant parameter reduction relative to prior transformer models} while attaining State-of-the-Art (SOTA) or near SOTA on all three datasets. Additionally, we achieved SOTA for Protocol III on H36M for both GT and CPN detection inputs. Finally, we obtained SOTA on all three metrics for the MPI-INF-3DHP dataset and for all three subjects on HumanEva under Protocol II.
[ { "version": "v1", "created": "Tue, 4 Apr 2023 22:23:50 GMT" } ]
2023-04-06T00:00:00
[ [ "Diaz-Arias", "Alec", "" ], [ "Shin", "Dmitriy", "" ] ]
TITLE: ConvFormer: Parameter Reduction in Transformer Models for 3D Human Pose Estimation by Leveraging Dynamic Multi-Headed Convolutional Attention ABSTRACT: Recently, fully-transformer architectures have replaced the defacto convolutional architecture for the 3D human pose estimation task. In this paper we propose \textbf{\textit{ConvFormer}}, a novel convolutional transformer that leverages a new \textbf{\textit{dynamic multi-headed convolutional self-attention}} mechanism for monocular 3D human pose estimation. We designed a spatial and temporal convolutional transformer to comprehensively model human joint relations within individual frames and globally across the motion sequence. Moreover, we introduce a novel notion of \textbf{\textit{temporal joints profile}} for our temporal ConvFormer that fuses complete temporal information immediately for a local neighborhood of joint features. We have quantitatively and qualitatively validated our method on three common benchmark datasets: Human3.6M, MPI-INF-3DHP, and HumanEva. Extensive experiments have been conducted to identify the optimal hyper-parameter set. These experiments demonstrated that we achieved a \textbf{significant parameter reduction relative to prior transformer models} while attaining State-of-the-Art (SOTA) or near SOTA on all three datasets. Additionally, we achieved SOTA for Protocol III on H36M for both GT and CPN detection inputs. Finally, we obtained SOTA on all three metrics for the MPI-INF-3DHP dataset and for all three subjects on HumanEva under Protocol II.
no_new_dataset
0.711663
2304.02160
Ke Chen
Ke Chen, Gordon Wichern, Fran\c{c}ois G. Germain, Jonathan Le Roux
Pac-HuBERT: Self-Supervised Music Source Separation via Primitive Auditory Clustering and Hidden-Unit BERT
5 pages, 2 figures, 3 tables
null
null
null
cs.SD cs.LG eess.AS
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
In spite of the progress in music source separation research, the small amount of publicly-available clean source data remains a constant limiting factor for performance. Thus, recent advances in self-supervised learning present a largely-unexplored opportunity for improving separation models by leveraging unlabelled music data. In this paper, we propose a self-supervised learning framework for music source separation inspired by the HuBERT speech representation model. We first investigate the potential impact of the original HuBERT model by inserting an adapted version of it into the well-known Demucs V2 time-domain separation model architecture. We then propose a time-frequency-domain self-supervised model, Pac-HuBERT (for primitive auditory clustering HuBERT), that we later use in combination with a Res-U-Net decoder for source separation. Pac-HuBERT uses primitive auditory features of music as unsupervised clustering labels to initialize the self-supervised pretraining process using the Free Music Archive (FMA) dataset. The resulting framework achieves better source-to-distortion ratio (SDR) performance on the MusDB18 test set than the original Demucs V2 and Res-U-Net models. We further demonstrate that it can boost performance with small amounts of supervised data. Ultimately, our proposed framework is an effective solution to the challenge of limited clean source data for music source separation.
[ { "version": "v1", "created": "Tue, 4 Apr 2023 23:19:53 GMT" } ]
2023-04-06T00:00:00
[ [ "Chen", "Ke", "" ], [ "Wichern", "Gordon", "" ], [ "Germain", "François G.", "" ], [ "Roux", "Jonathan Le", "" ] ]
TITLE: Pac-HuBERT: Self-Supervised Music Source Separation via Primitive Auditory Clustering and Hidden-Unit BERT ABSTRACT: In spite of the progress in music source separation research, the small amount of publicly-available clean source data remains a constant limiting factor for performance. Thus, recent advances in self-supervised learning present a largely-unexplored opportunity for improving separation models by leveraging unlabelled music data. In this paper, we propose a self-supervised learning framework for music source separation inspired by the HuBERT speech representation model. We first investigate the potential impact of the original HuBERT model by inserting an adapted version of it into the well-known Demucs V2 time-domain separation model architecture. We then propose a time-frequency-domain self-supervised model, Pac-HuBERT (for primitive auditory clustering HuBERT), that we later use in combination with a Res-U-Net decoder for source separation. Pac-HuBERT uses primitive auditory features of music as unsupervised clustering labels to initialize the self-supervised pretraining process using the Free Music Archive (FMA) dataset. The resulting framework achieves better source-to-distortion ratio (SDR) performance on the MusDB18 test set than the original Demucs V2 and Res-U-Net models. We further demonstrate that it can boost performance with small amounts of supervised data. Ultimately, our proposed framework is an effective solution to the challenge of limited clean source data for music source separation.
no_new_dataset
0.711205
2304.02173
Zhi-Qi Cheng
Zhi-Qi Cheng, Qi Dai, Siyao Li, Jingdong Sun, Teruko Mitamura, Alexander G. Hauptmann
ChartReader: A Unified Framework for Chart Derendering and Comprehension without Heuristic Rules
null
null
null
null
cs.CV cs.AI cs.MM
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Charts are a powerful tool for visually conveying complex data, but their comprehension poses a challenge due to the diverse chart types and intricate components. Existing chart comprehension methods suffer from either heuristic rules or an over-reliance on OCR systems, resulting in suboptimal performance. To address these issues, we present ChartReader, a unified framework that seamlessly integrates chart derendering and comprehension tasks. Our approach includes a transformer-based chart component detection module and an extended pre-trained vision-language model for chart-to-X tasks. By learning the rules of charts automatically from annotated datasets, our approach eliminates the need for manual rule-making, reducing effort and enhancing accuracy.~We also introduce a data variable replacement technique and extend the input and position embeddings of the pre-trained model for cross-task training. We evaluate ChartReader on Chart-to-Table, ChartQA, and Chart-to-Text tasks, demonstrating its superiority over existing methods. Our proposed framework can significantly reduce the manual effort involved in chart analysis, providing a step towards a universal chart understanding model. Moreover, our approach offers opportunities for plug-and-play integration with mainstream LLMs such as T5 and TaPas, extending their capability to chart comprehension tasks. The code is available at https://github.com/zhiqic/ChartReader.
[ { "version": "v1", "created": "Wed, 5 Apr 2023 00:25:27 GMT" } ]
2023-04-06T00:00:00
[ [ "Cheng", "Zhi-Qi", "" ], [ "Dai", "Qi", "" ], [ "Li", "Siyao", "" ], [ "Sun", "Jingdong", "" ], [ "Mitamura", "Teruko", "" ], [ "Hauptmann", "Alexander G.", "" ] ]
TITLE: ChartReader: A Unified Framework for Chart Derendering and Comprehension without Heuristic Rules ABSTRACT: Charts are a powerful tool for visually conveying complex data, but their comprehension poses a challenge due to the diverse chart types and intricate components. Existing chart comprehension methods suffer from either heuristic rules or an over-reliance on OCR systems, resulting in suboptimal performance. To address these issues, we present ChartReader, a unified framework that seamlessly integrates chart derendering and comprehension tasks. Our approach includes a transformer-based chart component detection module and an extended pre-trained vision-language model for chart-to-X tasks. By learning the rules of charts automatically from annotated datasets, our approach eliminates the need for manual rule-making, reducing effort and enhancing accuracy.~We also introduce a data variable replacement technique and extend the input and position embeddings of the pre-trained model for cross-task training. We evaluate ChartReader on Chart-to-Table, ChartQA, and Chart-to-Text tasks, demonstrating its superiority over existing methods. Our proposed framework can significantly reduce the manual effort involved in chart analysis, providing a step towards a universal chart understanding model. Moreover, our approach offers opportunities for plug-and-play integration with mainstream LLMs such as T5 and TaPas, extending their capability to chart comprehension tasks. The code is available at https://github.com/zhiqic/ChartReader.
no_new_dataset
0.70939
2304.02187
Chenyan Wu
Chenyan Wu, Dolzodmaa Davaasuren, Tal Shafir, Rachelle Tsachor, James Z. Wang
Bodily expressed emotion understanding through integrating Laban movement analysis
null
null
null
null
cs.AI
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Body movements carry important information about a person's emotions or mental state and are essential in daily communication. Enhancing the ability of machines to understand emotions expressed through body language can improve the communication of assistive robots with children and elderly users, provide psychiatric professionals with quantitative diagnostic and prognostic assistance, and aid law enforcement in identifying deception. This study develops a high-quality human motor element dataset based on the Laban Movement Analysis movement coding system and utilizes that to jointly learn about motor elements and emotions. Our long-term ambition is to integrate knowledge from computing, psychology, and performing arts to enable automated understanding and analysis of emotion and mental state through body language. This work serves as a launchpad for further research into recognizing emotions through analysis of human movement.
[ { "version": "v1", "created": "Wed, 5 Apr 2023 02:07:15 GMT" } ]
2023-04-06T00:00:00
[ [ "Wu", "Chenyan", "" ], [ "Davaasuren", "Dolzodmaa", "" ], [ "Shafir", "Tal", "" ], [ "Tsachor", "Rachelle", "" ], [ "Wang", "James Z.", "" ] ]
TITLE: Bodily expressed emotion understanding through integrating Laban movement analysis ABSTRACT: Body movements carry important information about a person's emotions or mental state and are essential in daily communication. Enhancing the ability of machines to understand emotions expressed through body language can improve the communication of assistive robots with children and elderly users, provide psychiatric professionals with quantitative diagnostic and prognostic assistance, and aid law enforcement in identifying deception. This study develops a high-quality human motor element dataset based on the Laban Movement Analysis movement coding system and utilizes that to jointly learn about motor elements and emotions. Our long-term ambition is to integrate knowledge from computing, psychology, and performing arts to enable automated understanding and analysis of emotion and mental state through body language. This work serves as a launchpad for further research into recognizing emotions through analysis of human movement.
new_dataset
0.714479
2304.02189
Soumyabrata Dey
A. Ravishankar Rao, Daniel Clarke, Subrata Garai, Soumyabrata Dey
A system for exploring big data: an iterative k-means searchlight for outlier detection on open health data
2018 International Joint Conference on Neural Networks (IJCNN)
null
10.1109/IJCNN.2018.8489448
null
cs.LG
http://creativecommons.org/licenses/by-nc-sa/4.0/
The interactive exploration of large and evolving datasets is challenging as relationships between underlying variables may not be fully understood. There may be hidden trends and patterns in the data that are worthy of further exploration and analysis. We present a system that methodically explores multiple combinations of variables using a searchlight technique and identifies outliers. An iterative k-means clustering algorithm is applied to features derived through a split-apply-combine paradigm used in the database literature. Outliers are identified as singleton or small clusters. This algorithm is swept across the dataset in a searchlight manner. The dimensions that contain outliers are combined in pairs with other dimensions using a susbset scan technique to gain further insight into the outliers. We illustrate this system by anaylzing open health care data released by New York State. We apply our iterative k-means searchlight followed by subset scanning. Several anomalous trends in the data are identified, including cost overruns at specific hospitals, and increases in diagnoses such as suicides. These constitute novel findings in the literature, and are of potential use to regulatory agencies, policy makers and concerned citizens.
[ { "version": "v1", "created": "Wed, 5 Apr 2023 02:09:15 GMT" } ]
2023-04-06T00:00:00
[ [ "Rao", "A. Ravishankar", "" ], [ "Clarke", "Daniel", "" ], [ "Garai", "Subrata", "" ], [ "Dey", "Soumyabrata", "" ] ]
TITLE: A system for exploring big data: an iterative k-means searchlight for outlier detection on open health data ABSTRACT: The interactive exploration of large and evolving datasets is challenging as relationships between underlying variables may not be fully understood. There may be hidden trends and patterns in the data that are worthy of further exploration and analysis. We present a system that methodically explores multiple combinations of variables using a searchlight technique and identifies outliers. An iterative k-means clustering algorithm is applied to features derived through a split-apply-combine paradigm used in the database literature. Outliers are identified as singleton or small clusters. This algorithm is swept across the dataset in a searchlight manner. The dimensions that contain outliers are combined in pairs with other dimensions using a susbset scan technique to gain further insight into the outliers. We illustrate this system by anaylzing open health care data released by New York State. We apply our iterative k-means searchlight followed by subset scanning. Several anomalous trends in the data are identified, including cost overruns at specific hospitals, and increases in diagnoses such as suicides. These constitute novel findings in the literature, and are of potential use to regulatory agencies, policy makers and concerned citizens.
no_new_dataset
0.710635
2304.02195
Sungmin Kang
Sungmin Kang, Bei Chen, Shin Yoo, Jian-Guang Lou
Explainable Automated Debugging via Large Language Model-driven Scientific Debugging
null
null
null
null
cs.SE
http://creativecommons.org/licenses/by/4.0/
Automated debugging techniques have the potential to reduce developer effort in debugging, and have matured enough to be adopted by industry. However, one critical issue with existing techniques is that, while developers want rationales for the provided automatic debugging results, existing techniques are ill-suited to provide them, as their deduction process differs significantly from that of human developers. Inspired by the way developers interact with code when debugging, we propose Automated Scientific Debugging (AutoSD), a technique that given buggy code and a bug-revealing test, prompts large language models to automatically generate hypotheses, uses debuggers to actively interact with buggy code, and thus automatically reach conclusions prior to patch generation. By aligning the reasoning of automated debugging more closely with that of human developers, we aim to produce intelligible explanations of how a specific patch has been generated, with the hope that the explanation will lead to more efficient and accurate developer decisions. Our empirical analysis on three program repair benchmarks shows that AutoSD performs competitively with other program repair baselines, and that it can indicate when it is confident in its results. Furthermore, we perform a human study with 20 participants, including six professional developers, to evaluate the utility of explanations from AutoSD. Participants with access to explanations could judge patch correctness in roughly the same time as those without, but their accuracy improved for five out of six real-world bugs studied: 70% of participants answered that they wanted explanations when using repair tools, while 55% answered that they were satisfied with the Scientific Debugging presentation.
[ { "version": "v1", "created": "Wed, 5 Apr 2023 02:34:10 GMT" } ]
2023-04-06T00:00:00
[ [ "Kang", "Sungmin", "" ], [ "Chen", "Bei", "" ], [ "Yoo", "Shin", "" ], [ "Lou", "Jian-Guang", "" ] ]
TITLE: Explainable Automated Debugging via Large Language Model-driven Scientific Debugging ABSTRACT: Automated debugging techniques have the potential to reduce developer effort in debugging, and have matured enough to be adopted by industry. However, one critical issue with existing techniques is that, while developers want rationales for the provided automatic debugging results, existing techniques are ill-suited to provide them, as their deduction process differs significantly from that of human developers. Inspired by the way developers interact with code when debugging, we propose Automated Scientific Debugging (AutoSD), a technique that given buggy code and a bug-revealing test, prompts large language models to automatically generate hypotheses, uses debuggers to actively interact with buggy code, and thus automatically reach conclusions prior to patch generation. By aligning the reasoning of automated debugging more closely with that of human developers, we aim to produce intelligible explanations of how a specific patch has been generated, with the hope that the explanation will lead to more efficient and accurate developer decisions. Our empirical analysis on three program repair benchmarks shows that AutoSD performs competitively with other program repair baselines, and that it can indicate when it is confident in its results. Furthermore, we perform a human study with 20 participants, including six professional developers, to evaluate the utility of explanations from AutoSD. Participants with access to explanations could judge patch correctness in roughly the same time as those without, but their accuracy improved for five out of six real-world bugs studied: 70% of participants answered that they wanted explanations when using repair tools, while 55% answered that they were satisfied with the Scientific Debugging presentation.
no_new_dataset
0.705608
2304.02205
Jifan Yu
Jifan Yu, Mengying Lu, Qingyang Zhong, Zijun Yao, Shangqing Tu, Zhengshan Liao, Xiaoya Li, Manli Li, Lei Hou, Hai-Tao Zheng, Juanzi Li, Jie Tang
MoocRadar: A Fine-grained and Multi-aspect Knowledge Repository for Improving Cognitive Student Modeling in MOOCs
Accepted by SIGIR 2023
null
null
null
cs.AI cs.IR
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Student modeling, the task of inferring a student's learning characteristics through their interactions with coursework, is a fundamental issue in intelligent education. Although the recent attempts from knowledge tracing and cognitive diagnosis propose several promising directions for improving the usability and effectiveness of current models, the existing public datasets are still insufficient to meet the need for these potential solutions due to their ignorance of complete exercising contexts, fine-grained concepts, and cognitive labels. In this paper, we present MoocRadar, a fine-grained, multi-aspect knowledge repository consisting of 2,513 exercise questions, 5,600 knowledge concepts, and over 12 million behavioral records. Specifically, we propose a framework to guarantee a high-quality and comprehensive annotation of fine-grained concepts and cognitive labels. The statistical and experimental results indicate that our dataset provides the basis for the future improvements of existing methods. Moreover, to support the convenient usage for researchers, we release a set of tools for data querying, model adaption, and even the extension of our repository, which are now available at https://github.com/THU-KEG/MOOC-Radar.
[ { "version": "v1", "created": "Wed, 5 Apr 2023 03:36:40 GMT" } ]
2023-04-06T00:00:00
[ [ "Yu", "Jifan", "" ], [ "Lu", "Mengying", "" ], [ "Zhong", "Qingyang", "" ], [ "Yao", "Zijun", "" ], [ "Tu", "Shangqing", "" ], [ "Liao", "Zhengshan", "" ], [ "Li", "Xiaoya", "" ], [ "Li", "Manli", "" ], [ "Hou", "Lei", "" ], [ "Zheng", "Hai-Tao", "" ], [ "Li", "Juanzi", "" ], [ "Tang", "Jie", "" ] ]
TITLE: MoocRadar: A Fine-grained and Multi-aspect Knowledge Repository for Improving Cognitive Student Modeling in MOOCs ABSTRACT: Student modeling, the task of inferring a student's learning characteristics through their interactions with coursework, is a fundamental issue in intelligent education. Although the recent attempts from knowledge tracing and cognitive diagnosis propose several promising directions for improving the usability and effectiveness of current models, the existing public datasets are still insufficient to meet the need for these potential solutions due to their ignorance of complete exercising contexts, fine-grained concepts, and cognitive labels. In this paper, we present MoocRadar, a fine-grained, multi-aspect knowledge repository consisting of 2,513 exercise questions, 5,600 knowledge concepts, and over 12 million behavioral records. Specifically, we propose a framework to guarantee a high-quality and comprehensive annotation of fine-grained concepts and cognitive labels. The statistical and experimental results indicate that our dataset provides the basis for the future improvements of existing methods. Moreover, to support the convenient usage for researchers, we release a set of tools for data querying, model adaption, and even the extension of our repository, which are now available at https://github.com/THU-KEG/MOOC-Radar.
new_dataset
0.718218
2304.02208
Soumyabrata Dey
A. Ravishankar Rao, Subrata Garai, Soumyabrata Dey, Hang Peng
PIKS: A Technique to Identify Actionable Trends for Policy-Makers Through Open Healthcare Data
null
SN COMPUT. SCI. 2, 477 (2021)
10.1007/s42979-021-00871-7
null
cs.LG cs.CY
http://creativecommons.org/licenses/by-nc-sa/4.0/
With calls for increasing transparency, governments are releasing greater amounts of data in multiple domains including finance, education and healthcare. The efficient exploratory analysis of healthcare data constitutes a significant challenge. Key concerns in public health include the quick identification and analysis of trends, and the detection of outliers. This allows policies to be rapidly adapted to changing circumstances. We present an efficient outlier detection technique, termed PIKS (Pruned iterative-k means searchlight), which combines an iterative k-means algorithm with a pruned searchlight based scan. We apply this technique to identify outliers in two publicly available healthcare datasets from the New York Statewide Planning and Research Cooperative System, and California's Office of Statewide Health Planning and Development. We provide a comparison of our technique with three other existing outlier detection techniques, consisting of auto-encoders, isolation forests and feature bagging. We identified outliers in conditions including suicide rates, immunity disorders, social admissions, cardiomyopathies, and pregnancy in the third trimester. We demonstrate that the PIKS technique produces results consistent with other techniques such as the auto-encoder. However, the auto-encoder needs to be trained, which requires several parameters to be tuned. In comparison, the PIKS technique has far fewer parameters to tune. This makes it advantageous for fast, "out-of-the-box" data exploration. The PIKS technique is scalable and can readily ingest new datasets. Hence, it can provide valuable, up-to-date insights to citizens, patients and policy-makers. We have made our code open source, and with the availability of open data, other researchers can easily reproduce and extend our work. This will help promote a deeper understanding of healthcare policies and public health issues.
[ { "version": "v1", "created": "Wed, 5 Apr 2023 03:45:39 GMT" } ]
2023-04-06T00:00:00
[ [ "Rao", "A. Ravishankar", "" ], [ "Garai", "Subrata", "" ], [ "Dey", "Soumyabrata", "" ], [ "Peng", "Hang", "" ] ]
TITLE: PIKS: A Technique to Identify Actionable Trends for Policy-Makers Through Open Healthcare Data ABSTRACT: With calls for increasing transparency, governments are releasing greater amounts of data in multiple domains including finance, education and healthcare. The efficient exploratory analysis of healthcare data constitutes a significant challenge. Key concerns in public health include the quick identification and analysis of trends, and the detection of outliers. This allows policies to be rapidly adapted to changing circumstances. We present an efficient outlier detection technique, termed PIKS (Pruned iterative-k means searchlight), which combines an iterative k-means algorithm with a pruned searchlight based scan. We apply this technique to identify outliers in two publicly available healthcare datasets from the New York Statewide Planning and Research Cooperative System, and California's Office of Statewide Health Planning and Development. We provide a comparison of our technique with three other existing outlier detection techniques, consisting of auto-encoders, isolation forests and feature bagging. We identified outliers in conditions including suicide rates, immunity disorders, social admissions, cardiomyopathies, and pregnancy in the third trimester. We demonstrate that the PIKS technique produces results consistent with other techniques such as the auto-encoder. However, the auto-encoder needs to be trained, which requires several parameters to be tuned. In comparison, the PIKS technique has far fewer parameters to tune. This makes it advantageous for fast, "out-of-the-box" data exploration. The PIKS technique is scalable and can readily ingest new datasets. Hence, it can provide valuable, up-to-date insights to citizens, patients and policy-makers. We have made our code open source, and with the availability of open data, other researchers can easily reproduce and extend our work. This will help promote a deeper understanding of healthcare policies and public health issues.
no_new_dataset
0.708231
2304.02211
Zhanyu Wang
Zhanyu Wang, Lingqiao Liu, Lei Wang, Luping Zhou
METransformer: Radiology Report Generation by Transformer with Multiple Learnable Expert Tokens
Accepted by CVPR2023
null
null
null
cs.CV
http://creativecommons.org/licenses/by-nc-nd/4.0/
In clinical scenarios, multi-specialist consultation could significantly benefit the diagnosis, especially for intricate cases. This inspires us to explore a "multi-expert joint diagnosis" mechanism to upgrade the existing "single expert" framework commonly seen in the current literature. To this end, we propose METransformer, a method to realize this idea with a transformer-based backbone. The key design of our method is the introduction of multiple learnable "expert" tokens into both the transformer encoder and decoder. In the encoder, each expert token interacts with both vision tokens and other expert tokens to learn to attend different image regions for image representation. These expert tokens are encouraged to capture complementary information by an orthogonal loss that minimizes their overlap. In the decoder, each attended expert token guides the cross-attention between input words and visual tokens, thus influencing the generated report. A metrics-based expert voting strategy is further developed to generate the final report. By the multi-experts concept, our model enjoys the merits of an ensemble-based approach but through a manner that is computationally more efficient and supports more sophisticated interactions among experts. Experimental results demonstrate the promising performance of our proposed model on two widely used benchmarks. Last but not least, the framework-level innovation makes our work ready to incorporate advances on existing "single-expert" models to further improve its performance.
[ { "version": "v1", "created": "Wed, 5 Apr 2023 03:54:28 GMT" } ]
2023-04-06T00:00:00
[ [ "Wang", "Zhanyu", "" ], [ "Liu", "Lingqiao", "" ], [ "Wang", "Lei", "" ], [ "Zhou", "Luping", "" ] ]
TITLE: METransformer: Radiology Report Generation by Transformer with Multiple Learnable Expert Tokens ABSTRACT: In clinical scenarios, multi-specialist consultation could significantly benefit the diagnosis, especially for intricate cases. This inspires us to explore a "multi-expert joint diagnosis" mechanism to upgrade the existing "single expert" framework commonly seen in the current literature. To this end, we propose METransformer, a method to realize this idea with a transformer-based backbone. The key design of our method is the introduction of multiple learnable "expert" tokens into both the transformer encoder and decoder. In the encoder, each expert token interacts with both vision tokens and other expert tokens to learn to attend different image regions for image representation. These expert tokens are encouraged to capture complementary information by an orthogonal loss that minimizes their overlap. In the decoder, each attended expert token guides the cross-attention between input words and visual tokens, thus influencing the generated report. A metrics-based expert voting strategy is further developed to generate the final report. By the multi-experts concept, our model enjoys the merits of an ensemble-based approach but through a manner that is computationally more efficient and supports more sophisticated interactions among experts. Experimental results demonstrate the promising performance of our proposed model on two widely used benchmarks. Last but not least, the framework-level innovation makes our work ready to incorporate advances on existing "single-expert" models to further improve its performance.
no_new_dataset
0.71086
2304.02214
Binbin Feng
Binbin Feng, Jun Li, Jianhua Xu
LogoNet: a fine-grained network for instance-level logo sketch retrieval
null
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Sketch-based image retrieval, which aims to use sketches as queries to retrieve images containing the same query instance, receives increasing attention in recent years. Although dramatic progress has been made in sketch retrieval, few efforts are devoted to logo sketch retrieval which is still hindered by the following challenges: Firstly, logo sketch retrieval is more difficult than typical sketch retrieval problem, since a logo sketch usually contains much less visual contents with only irregular strokes and lines. Secondly, instance-specific sketches demonstrate dramatic appearance variances, making them less identifiable when querying the same logo instance. Thirdly, there exist several sketch retrieval benchmarking datasets nowadays, whereas an instance-level logo sketch dataset is still publicly unavailable. To address the above-mentioned limitations, we make twofold contributions in this study for instance-level logo sketch retrieval. To begin with, we construct an instance-level logo sketch dataset containing 2k logo instances and exceeding 9k sketches. To our knowledge, this is the first publicly available instance-level logo sketch dataset. Next, we develop a fine-grained triple-branch CNN architecture based on hybrid attention mechanism termed LogoNet for accurate logo sketch retrieval. More specifically, we embed the hybrid attention mechanism into the triple-branch architecture for capturing the key query-specific information from the limited visual cues in the logo sketches. Experimental evaluations both on our assembled dataset and public benchmark datasets demonstrate the effectiveness of our proposed network.
[ { "version": "v1", "created": "Wed, 5 Apr 2023 04:03:02 GMT" } ]
2023-04-06T00:00:00
[ [ "Feng", "Binbin", "" ], [ "Li", "Jun", "" ], [ "Xu", "Jianhua", "" ] ]
TITLE: LogoNet: a fine-grained network for instance-level logo sketch retrieval ABSTRACT: Sketch-based image retrieval, which aims to use sketches as queries to retrieve images containing the same query instance, receives increasing attention in recent years. Although dramatic progress has been made in sketch retrieval, few efforts are devoted to logo sketch retrieval which is still hindered by the following challenges: Firstly, logo sketch retrieval is more difficult than typical sketch retrieval problem, since a logo sketch usually contains much less visual contents with only irregular strokes and lines. Secondly, instance-specific sketches demonstrate dramatic appearance variances, making them less identifiable when querying the same logo instance. Thirdly, there exist several sketch retrieval benchmarking datasets nowadays, whereas an instance-level logo sketch dataset is still publicly unavailable. To address the above-mentioned limitations, we make twofold contributions in this study for instance-level logo sketch retrieval. To begin with, we construct an instance-level logo sketch dataset containing 2k logo instances and exceeding 9k sketches. To our knowledge, this is the first publicly available instance-level logo sketch dataset. Next, we develop a fine-grained triple-branch CNN architecture based on hybrid attention mechanism termed LogoNet for accurate logo sketch retrieval. More specifically, we embed the hybrid attention mechanism into the triple-branch architecture for capturing the key query-specific information from the limited visual cues in the logo sketches. Experimental evaluations both on our assembled dataset and public benchmark datasets demonstrate the effectiveness of our proposed network.
new_dataset
0.715809
2304.02222
Fengyi Shen
Fengyi Shen, Akhil Gurram, Ziyuan Liu, He Wang, Alois Knoll
DiGA: Distil to Generalize and then Adapt for Domain Adaptive Semantic Segmentation
CVPR2023
2023 Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition
null
null
cs.CV
http://creativecommons.org/licenses/by-nc-sa/4.0/
Domain adaptive semantic segmentation methods commonly utilize stage-wise training, consisting of a warm-up and a self-training stage. However, this popular approach still faces several challenges in each stage: for warm-up, the widely adopted adversarial training often results in limited performance gain, due to blind feature alignment; for self-training, finding proper categorical thresholds is very tricky. To alleviate these issues, we first propose to replace the adversarial training in the warm-up stage by a novel symmetric knowledge distillation module that only accesses the source domain data and makes the model domain generalizable. Surprisingly, this domain generalizable warm-up model brings substantial performance improvement, which can be further amplified via our proposed cross-domain mixture data augmentation technique. Then, for the self-training stage, we propose a threshold-free dynamic pseudo-label selection mechanism to ease the aforementioned threshold problem and make the model better adapted to the target domain. Extensive experiments demonstrate that our framework achieves remarkable and consistent improvements compared to the prior arts on popular benchmarks. Codes and models are available at https://github.com/fy-vision/DiGA
[ { "version": "v1", "created": "Wed, 5 Apr 2023 04:32:02 GMT" } ]
2023-04-06T00:00:00
[ [ "Shen", "Fengyi", "" ], [ "Gurram", "Akhil", "" ], [ "Liu", "Ziyuan", "" ], [ "Wang", "He", "" ], [ "Knoll", "Alois", "" ] ]
TITLE: DiGA: Distil to Generalize and then Adapt for Domain Adaptive Semantic Segmentation ABSTRACT: Domain adaptive semantic segmentation methods commonly utilize stage-wise training, consisting of a warm-up and a self-training stage. However, this popular approach still faces several challenges in each stage: for warm-up, the widely adopted adversarial training often results in limited performance gain, due to blind feature alignment; for self-training, finding proper categorical thresholds is very tricky. To alleviate these issues, we first propose to replace the adversarial training in the warm-up stage by a novel symmetric knowledge distillation module that only accesses the source domain data and makes the model domain generalizable. Surprisingly, this domain generalizable warm-up model brings substantial performance improvement, which can be further amplified via our proposed cross-domain mixture data augmentation technique. Then, for the self-training stage, we propose a threshold-free dynamic pseudo-label selection mechanism to ease the aforementioned threshold problem and make the model better adapted to the target domain. Extensive experiments demonstrate that our framework achieves remarkable and consistent improvements compared to the prior arts on popular benchmarks. Codes and models are available at https://github.com/fy-vision/DiGA
no_new_dataset
0.710189
2304.02225
Junhuem Park
Junheum Park, Jintae Kim, Chang-Su Kim
BiFormer: Learning Bilateral Motion Estimation via Bilateral Transformer for 4K Video Frame Interpolation
Accepted to CVPR2023
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
A novel 4K video frame interpolator based on bilateral transformer (BiFormer) is proposed in this paper, which performs three steps: global motion estimation, local motion refinement, and frame synthesis. First, in global motion estimation, we predict symmetric bilateral motion fields at a coarse scale. To this end, we propose BiFormer, the first transformer-based bilateral motion estimator. Second, we refine the global motion fields efficiently using blockwise bilateral cost volumes (BBCVs). Third, we warp the input frames using the refined motion fields and blend them to synthesize an intermediate frame. Extensive experiments demonstrate that the proposed BiFormer algorithm achieves excellent interpolation performance on 4K datasets. The source codes are available at https://github.com/JunHeum/BiFormer.
[ { "version": "v1", "created": "Wed, 5 Apr 2023 04:52:23 GMT" } ]
2023-04-06T00:00:00
[ [ "Park", "Junheum", "" ], [ "Kim", "Jintae", "" ], [ "Kim", "Chang-Su", "" ] ]
TITLE: BiFormer: Learning Bilateral Motion Estimation via Bilateral Transformer for 4K Video Frame Interpolation ABSTRACT: A novel 4K video frame interpolator based on bilateral transformer (BiFormer) is proposed in this paper, which performs three steps: global motion estimation, local motion refinement, and frame synthesis. First, in global motion estimation, we predict symmetric bilateral motion fields at a coarse scale. To this end, we propose BiFormer, the first transformer-based bilateral motion estimator. Second, we refine the global motion fields efficiently using blockwise bilateral cost volumes (BBCVs). Third, we warp the input frames using the refined motion fields and blend them to synthesize an intermediate frame. Extensive experiments demonstrate that the proposed BiFormer algorithm achieves excellent interpolation performance on 4K datasets. The source codes are available at https://github.com/JunHeum/BiFormer.
no_new_dataset
0.712282
2304.02239
Hao Wang
Jinhao Li, Changlong Wang, Hao Wang
Optimal Energy Storage Scheduling for Wind Curtailment Reduction and Energy Arbitrage: A Deep Reinforcement Learning Approach
2023 IEEE Power & Energy Society General Meeting (PESGM). arXiv admin note: text overlap with arXiv:2212.13368
null
null
null
cs.LG cs.SY eess.SY
http://creativecommons.org/licenses/by/4.0/
Wind energy has been rapidly gaining popularity as a means for combating climate change. However, the variable nature of wind generation can undermine system reliability and lead to wind curtailment, causing substantial economic losses to wind power producers. Battery energy storage systems (BESS) that serve as onsite backup sources are among the solutions to mitigate wind curtailment. However, such an auxiliary role of the BESS might severely weaken its economic viability. This paper addresses the issue by proposing joint wind curtailment reduction and energy arbitrage for the BESS. We decouple the market participation of the co-located wind-battery system and develop a joint-bidding framework for the wind farm and BESS. It is challenging to optimize the joint-bidding because of the stochasticity of energy prices and wind generation. Therefore, we leverage deep reinforcement learning to maximize the overall revenue from the spot market while unlocking the BESS's potential in concurrently reducing wind curtailment and conducting energy arbitrage. We validate the proposed strategy using realistic wind farm data and demonstrate that our joint-bidding strategy responds better to wind curtailment and generates higher revenues than the optimization-based benchmark. Our simulations also reveal that the extra wind generation used to be curtailed can be an effective power source to charge the BESS, resulting in additional financial returns.
[ { "version": "v1", "created": "Wed, 5 Apr 2023 06:02:58 GMT" } ]
2023-04-06T00:00:00
[ [ "Li", "Jinhao", "" ], [ "Wang", "Changlong", "" ], [ "Wang", "Hao", "" ] ]
TITLE: Optimal Energy Storage Scheduling for Wind Curtailment Reduction and Energy Arbitrage: A Deep Reinforcement Learning Approach ABSTRACT: Wind energy has been rapidly gaining popularity as a means for combating climate change. However, the variable nature of wind generation can undermine system reliability and lead to wind curtailment, causing substantial economic losses to wind power producers. Battery energy storage systems (BESS) that serve as onsite backup sources are among the solutions to mitigate wind curtailment. However, such an auxiliary role of the BESS might severely weaken its economic viability. This paper addresses the issue by proposing joint wind curtailment reduction and energy arbitrage for the BESS. We decouple the market participation of the co-located wind-battery system and develop a joint-bidding framework for the wind farm and BESS. It is challenging to optimize the joint-bidding because of the stochasticity of energy prices and wind generation. Therefore, we leverage deep reinforcement learning to maximize the overall revenue from the spot market while unlocking the BESS's potential in concurrently reducing wind curtailment and conducting energy arbitrage. We validate the proposed strategy using realistic wind farm data and demonstrate that our joint-bidding strategy responds better to wind curtailment and generates higher revenues than the optimization-based benchmark. Our simulations also reveal that the extra wind generation used to be curtailed can be an effective power source to charge the BESS, resulting in additional financial returns.
no_new_dataset
0.710829
2304.02264
Nele Albers
Nele Albers, Mark A. Neerincx, Willem-Paul Brinkman
Persuading to Prepare for Quitting Smoking with a Virtual Coach: Using States and User Characteristics to Predict Behavior
10 pages, 6 figures, Camera-ready version for the 22nd International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2023)
null
null
null
cs.AI cs.HC
http://creativecommons.org/licenses/by/4.0/
Despite their prevalence in eHealth applications for behavior change, persuasive messages tend to have small effects on behavior. Conditions or states (e.g., confidence, knowledge, motivation) and characteristics (e.g., gender, age, personality) of persuadees are two promising components for more effective algorithms for choosing persuasive messages. However, it is not yet sufficiently clear how well considering these components allows one to predict behavior after persuasive attempts, especially in the long run. Since collecting data for many algorithm components is costly and places a burden on users, a better understanding of the impact of individual components in practice is welcome. This can help to make an informed decision on which components to use. We thus conducted a longitudinal study in which a virtual coach persuaded 671 daily smokers to do preparatory activities for quitting smoking and becoming more physically active, such as envisioning one's desired future self. Based on the collected data, we designed a Reinforcement Learning (RL)-approach that considers current and future states to maximize the effort people spend on their activities. Using this RL-approach, we found, based on leave-one-out cross-validation, that considering states helps to predict both behavior and future states. User characteristics and especially involvement in the activities, on the other hand, only help to predict behavior if used in combination with states rather than alone. We see these results as supporting the use of states and involvement in persuasion algorithms. Our dataset is available online.
[ { "version": "v1", "created": "Wed, 5 Apr 2023 07:31:29 GMT" } ]
2023-04-06T00:00:00
[ [ "Albers", "Nele", "" ], [ "Neerincx", "Mark A.", "" ], [ "Brinkman", "Willem-Paul", "" ] ]
TITLE: Persuading to Prepare for Quitting Smoking with a Virtual Coach: Using States and User Characteristics to Predict Behavior ABSTRACT: Despite their prevalence in eHealth applications for behavior change, persuasive messages tend to have small effects on behavior. Conditions or states (e.g., confidence, knowledge, motivation) and characteristics (e.g., gender, age, personality) of persuadees are two promising components for more effective algorithms for choosing persuasive messages. However, it is not yet sufficiently clear how well considering these components allows one to predict behavior after persuasive attempts, especially in the long run. Since collecting data for many algorithm components is costly and places a burden on users, a better understanding of the impact of individual components in practice is welcome. This can help to make an informed decision on which components to use. We thus conducted a longitudinal study in which a virtual coach persuaded 671 daily smokers to do preparatory activities for quitting smoking and becoming more physically active, such as envisioning one's desired future self. Based on the collected data, we designed a Reinforcement Learning (RL)-approach that considers current and future states to maximize the effort people spend on their activities. Using this RL-approach, we found, based on leave-one-out cross-validation, that considering states helps to predict both behavior and future states. User characteristics and especially involvement in the activities, on the other hand, only help to predict behavior if used in combination with states rather than alone. We see these results as supporting the use of states and involvement in persuasion algorithms. Our dataset is available online.
new_dataset
0.612136
2304.02273
Bowen Liu
Bowen Liu, Yu Chen, Rakesh Chowdary Machineni, Shiyu Liu, Hun-Seok Kim
MMVC: Learned Multi-Mode Video Compression with Block-based Prediction Mode Selection and Density-Adaptive Entropy Coding
null
null
null
null
eess.IV cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Learning-based video compression has been extensively studied over the past years, but it still has limitations in adapting to various motion patterns and entropy models. In this paper, we propose multi-mode video compression (MMVC), a block wise mode ensemble deep video compression framework that selects the optimal mode for feature domain prediction adapting to different motion patterns. Proposed multi-modes include ConvLSTM-based feature domain prediction, optical flow conditioned feature domain prediction, and feature propagation to address a wide range of cases from static scenes without apparent motions to dynamic scenes with a moving camera. We partition the feature space into blocks for temporal prediction in spatial block-based representations. For entropy coding, we consider both dense and sparse post-quantization residual blocks, and apply optional run-length coding to sparse residuals to improve the compression rate. In this sense, our method uses a dual-mode entropy coding scheme guided by a binary density map, which offers significant rate reduction surpassing the extra cost of transmitting the binary selection map. We validate our scheme with some of the most popular benchmarking datasets. Compared with state-of-the-art video compression schemes and standard codecs, our method yields better or competitive results measured with PSNR and MS-SSIM.
[ { "version": "v1", "created": "Wed, 5 Apr 2023 07:37:48 GMT" } ]
2023-04-06T00:00:00
[ [ "Liu", "Bowen", "" ], [ "Chen", "Yu", "" ], [ "Machineni", "Rakesh Chowdary", "" ], [ "Liu", "Shiyu", "" ], [ "Kim", "Hun-Seok", "" ] ]
TITLE: MMVC: Learned Multi-Mode Video Compression with Block-based Prediction Mode Selection and Density-Adaptive Entropy Coding ABSTRACT: Learning-based video compression has been extensively studied over the past years, but it still has limitations in adapting to various motion patterns and entropy models. In this paper, we propose multi-mode video compression (MMVC), a block wise mode ensemble deep video compression framework that selects the optimal mode for feature domain prediction adapting to different motion patterns. Proposed multi-modes include ConvLSTM-based feature domain prediction, optical flow conditioned feature domain prediction, and feature propagation to address a wide range of cases from static scenes without apparent motions to dynamic scenes with a moving camera. We partition the feature space into blocks for temporal prediction in spatial block-based representations. For entropy coding, we consider both dense and sparse post-quantization residual blocks, and apply optional run-length coding to sparse residuals to improve the compression rate. In this sense, our method uses a dual-mode entropy coding scheme guided by a binary density map, which offers significant rate reduction surpassing the extra cost of transmitting the binary selection map. We validate our scheme with some of the most popular benchmarking datasets. Compared with state-of-the-art video compression schemes and standard codecs, our method yields better or competitive results measured with PSNR and MS-SSIM.
no_new_dataset
0.711036
2304.02286
Borja Guerrero
Borja Guerrero Santillan
A step towards the applicability of algorithms based on invariant causal learning on observational data
null
null
null
null
cs.LG stat.ME
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Machine learning can benefit from causal discovery for interpretation and from causal inference for generalization. In this line of research, a few invariant learning algorithms for out-of-distribution (OOD) generalization have been proposed by using multiple training environments to find invariant relationships. Some of them are focused on causal discovery as Invariant Causal Prediction (ICP), which finds causal parents of a variable of interest, and some directly provide a causal optimal predictor that generalizes well in OOD environments as Invariant Risk Minimization (IRM). This group of algorithms works under the assumption of multiple environments that represent different interventions in the causal inference context. Those environments are not normally available when working with observational data and real-world applications. Here we propose a method to generate them in an efficient way. We assess the performance of this unsupervised learning problem by implementing ICP on simulated data. We also show how to apply ICP efficiently integrated with our method for causal discovery. Finally, we proposed an improved version of our method in combination with ICP for datasets with multiple covariates where ICP and other causal discovery methods normally degrade in performance.
[ { "version": "v1", "created": "Wed, 5 Apr 2023 08:15:57 GMT" } ]
2023-04-06T00:00:00
[ [ "Santillan", "Borja Guerrero", "" ] ]
TITLE: A step towards the applicability of algorithms based on invariant causal learning on observational data ABSTRACT: Machine learning can benefit from causal discovery for interpretation and from causal inference for generalization. In this line of research, a few invariant learning algorithms for out-of-distribution (OOD) generalization have been proposed by using multiple training environments to find invariant relationships. Some of them are focused on causal discovery as Invariant Causal Prediction (ICP), which finds causal parents of a variable of interest, and some directly provide a causal optimal predictor that generalizes well in OOD environments as Invariant Risk Minimization (IRM). This group of algorithms works under the assumption of multiple environments that represent different interventions in the causal inference context. Those environments are not normally available when working with observational data and real-world applications. Here we propose a method to generate them in an efficient way. We assess the performance of this unsupervised learning problem by implementing ICP on simulated data. We also show how to apply ICP efficiently integrated with our method for causal discovery. Finally, we proposed an improved version of our method in combination with ICP for datasets with multiple covariates where ICP and other causal discovery methods normally degrade in performance.
no_new_dataset
0.709202
2304.02296
Yeshwanth Kumar Adimoolam
Yeshwanth Kumar Adimoolam, Bodhiswatta Chatterjee, Charalambos Poullis, Melinos Averkiou
Efficient Deduplication and Leakage Detection in Large Scale Image Datasets with a focus on the CrowdAI Mapping Challenge Dataset
9 pages, 2 figures
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Recent advancements in deep learning and computer vision have led to widespread use of deep neural networks to extract building footprints from remote-sensing imagery. The success of such methods relies on the availability of large databases of high-resolution remote sensing images with high-quality annotations. The CrowdAI Mapping Challenge Dataset is one of these datasets that has been used extensively in recent years to train deep neural networks. This dataset consists of $ \sim\ $280k training images and $ \sim\ $60k testing images, with polygonal building annotations for all images. However, issues such as low-quality and incorrect annotations, extensive duplication of image samples, and data leakage significantly reduce the utility of deep neural networks trained on the dataset. Therefore, it is an imperative pre-condition to adopt a data validation pipeline that evaluates the quality of the dataset prior to its use. To this end, we propose a drop-in pipeline that employs perceptual hashing techniques for efficient de-duplication of the dataset and identification of instances of data leakage between training and testing splits. In our experiments, we demonstrate that nearly 250k($ \sim\ $90%) images in the training split were identical. Moreover, our analysis on the validation split demonstrates that roughly 56k of the 60k images also appear in the training split, resulting in a data leakage of 93%. The source code used for the analysis and de-duplication of the CrowdAI Mapping Challenge dataset is publicly available at https://github.com/yeshwanth95/CrowdAI_Hash_and_search .
[ { "version": "v1", "created": "Wed, 5 Apr 2023 08:36:17 GMT" } ]
2023-04-06T00:00:00
[ [ "Adimoolam", "Yeshwanth Kumar", "" ], [ "Chatterjee", "Bodhiswatta", "" ], [ "Poullis", "Charalambos", "" ], [ "Averkiou", "Melinos", "" ] ]
TITLE: Efficient Deduplication and Leakage Detection in Large Scale Image Datasets with a focus on the CrowdAI Mapping Challenge Dataset ABSTRACT: Recent advancements in deep learning and computer vision have led to widespread use of deep neural networks to extract building footprints from remote-sensing imagery. The success of such methods relies on the availability of large databases of high-resolution remote sensing images with high-quality annotations. The CrowdAI Mapping Challenge Dataset is one of these datasets that has been used extensively in recent years to train deep neural networks. This dataset consists of $ \sim\ $280k training images and $ \sim\ $60k testing images, with polygonal building annotations for all images. However, issues such as low-quality and incorrect annotations, extensive duplication of image samples, and data leakage significantly reduce the utility of deep neural networks trained on the dataset. Therefore, it is an imperative pre-condition to adopt a data validation pipeline that evaluates the quality of the dataset prior to its use. To this end, we propose a drop-in pipeline that employs perceptual hashing techniques for efficient de-duplication of the dataset and identification of instances of data leakage between training and testing splits. In our experiments, we demonstrate that nearly 250k($ \sim\ $90%) images in the training split were identical. Moreover, our analysis on the validation split demonstrates that roughly 56k of the 60k images also appear in the training split, resulting in a data leakage of 93%. The source code used for the analysis and de-duplication of the CrowdAI Mapping Challenge dataset is publicly available at https://github.com/yeshwanth95/CrowdAI_Hash_and_search .
new_dataset
0.669907
2304.02309
Michael Stettler
Michael Stettler, Alexander Lappe, Nick Taubert, Martin Giese
Multi-Domain Norm-referenced Encoding Enables Data Efficient Transfer Learning of Facial Expression Recognition
null
null
null
null
cs.CV
http://creativecommons.org/licenses/by/4.0/
People can innately recognize human facial expressions in unnatural forms, such as when depicted on the unusual faces drawn in cartoons or when applied to an animal's features. However, current machine learning algorithms struggle with out-of-domain transfer in facial expression recognition (FER). We propose a biologically-inspired mechanism for such transfer learning, which is based on norm-referenced encoding, where patterns are encoded in terms of difference vectors relative to a domain-specific reference vector. By incorporating domain-specific reference frames, we demonstrate high data efficiency in transfer learning across multiple domains. Our proposed architecture provides an explanation for how the human brain might innately recognize facial expressions on varying head shapes (humans, monkeys, and cartoon avatars) without extensive training. Norm-referenced encoding also allows the intensity of the expression to be read out directly from neural unit activity, similar to face-selective neurons in the brain. Our model achieves a classification accuracy of 92.15\% on the FERG dataset with extreme data efficiency. We train our proposed mechanism with only 12 images, including a single image of each class (facial expression) and one image per domain (avatar). In comparison, the authors of the FERG dataset achieved a classification accuracy of 89.02\% with their FaceExpr model, which was trained on 43,000 images.
[ { "version": "v1", "created": "Wed, 5 Apr 2023 09:06:30 GMT" } ]
2023-04-06T00:00:00
[ [ "Stettler", "Michael", "" ], [ "Lappe", "Alexander", "" ], [ "Taubert", "Nick", "" ], [ "Giese", "Martin", "" ] ]
TITLE: Multi-Domain Norm-referenced Encoding Enables Data Efficient Transfer Learning of Facial Expression Recognition ABSTRACT: People can innately recognize human facial expressions in unnatural forms, such as when depicted on the unusual faces drawn in cartoons or when applied to an animal's features. However, current machine learning algorithms struggle with out-of-domain transfer in facial expression recognition (FER). We propose a biologically-inspired mechanism for such transfer learning, which is based on norm-referenced encoding, where patterns are encoded in terms of difference vectors relative to a domain-specific reference vector. By incorporating domain-specific reference frames, we demonstrate high data efficiency in transfer learning across multiple domains. Our proposed architecture provides an explanation for how the human brain might innately recognize facial expressions on varying head shapes (humans, monkeys, and cartoon avatars) without extensive training. Norm-referenced encoding also allows the intensity of the expression to be read out directly from neural unit activity, similar to face-selective neurons in the brain. Our model achieves a classification accuracy of 92.15\% on the FERG dataset with extreme data efficiency. We train our proposed mechanism with only 12 images, including a single image of each class (facial expression) and one image per domain (avatar). In comparison, the authors of the FERG dataset achieved a classification accuracy of 89.02\% with their FaceExpr model, which was trained on 43,000 images.
no_new_dataset
0.712651
2304.02317
Zhonghao Lyu
Zhonghao Lyu, Guangxu Zhu, Jie Xu, Bo Ai, Shuguang Cui
Semantic Communications for Image Recovery and Classification via Deep Joint Source and Channel Coding
null
null
null
null
cs.NI
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
With the recent advancements in edge artificial intelligence (AI), future sixth-generation (6G) networks need to support new AI tasks such as classification and clustering apart from data recovery. Motivated by the success of deep learning, the semantic-aware and task-oriented communications with deep joint source and channel coding (JSCC) have emerged as new paradigm shifts in 6G from the conventional data-oriented communications with separate source and channel coding (SSCC). However, most existing works focused on the deep JSCC designs for one task of data recovery or AI task execution independently, which cannot be transferred to other unintended tasks. Differently, this paper investigates the JSCC semantic communications to support multi-task services, by performing the image data recovery and classification task execution simultaneously. First, we propose a new end-to-end deep JSCC framework by unifying the coding rate reduction maximization and the mean square error (MSE) minimization in the loss function. Here, the coding rate reduction maximization facilitates the learning of discriminative features for enabling to perform classification tasks directly in the feature space, and the MSE minimization helps the learning of informative features for high-quality image data recovery. Next, to further improve the robustness against variational wireless channels, we propose a new gated deep JSCC design, in which a gated net is incorporated for adaptively pruning the output features to adjust their dimensions based on channel conditions. Finally, we present extensive numerical experiments to validate the performance of our proposed deep JSCC designs as compared to various benchmark schemes.
[ { "version": "v1", "created": "Wed, 5 Apr 2023 09:15:08 GMT" } ]
2023-04-06T00:00:00
[ [ "Lyu", "Zhonghao", "" ], [ "Zhu", "Guangxu", "" ], [ "Xu", "Jie", "" ], [ "Ai", "Bo", "" ], [ "Cui", "Shuguang", "" ] ]
TITLE: Semantic Communications for Image Recovery and Classification via Deep Joint Source and Channel Coding ABSTRACT: With the recent advancements in edge artificial intelligence (AI), future sixth-generation (6G) networks need to support new AI tasks such as classification and clustering apart from data recovery. Motivated by the success of deep learning, the semantic-aware and task-oriented communications with deep joint source and channel coding (JSCC) have emerged as new paradigm shifts in 6G from the conventional data-oriented communications with separate source and channel coding (SSCC). However, most existing works focused on the deep JSCC designs for one task of data recovery or AI task execution independently, which cannot be transferred to other unintended tasks. Differently, this paper investigates the JSCC semantic communications to support multi-task services, by performing the image data recovery and classification task execution simultaneously. First, we propose a new end-to-end deep JSCC framework by unifying the coding rate reduction maximization and the mean square error (MSE) minimization in the loss function. Here, the coding rate reduction maximization facilitates the learning of discriminative features for enabling to perform classification tasks directly in the feature space, and the MSE minimization helps the learning of informative features for high-quality image data recovery. Next, to further improve the robustness against variational wireless channels, we propose a new gated deep JSCC design, in which a gated net is incorporated for adaptively pruning the output features to adjust their dimensions based on channel conditions. Finally, we present extensive numerical experiments to validate the performance of our proposed deep JSCC designs as compared to various benchmark schemes.
no_new_dataset
0.707196
2304.02319
Arshdeep Singh
Arshdeep Singh and Mark D. Plumbley
Efficient CNNs via Passive Filter Pruning
null
null
null
null
cs.LG cs.AI cs.CV eess.SP
http://creativecommons.org/licenses/by/4.0/
Convolutional neural networks (CNNs) have shown state-of-the-art performance in various applications. However, CNNs are resource-hungry due to their requirement of high computational complexity and memory storage. Recent efforts toward achieving computational efficiency in CNNs involve filter pruning methods that eliminate some of the filters in CNNs based on the \enquote{importance} of the filters. The majority of existing filter pruning methods are either "active", which use a dataset and generate feature maps to quantify filter importance, or "passive", which compute filter importance using entry-wise norm of the filters without involving data. Under a high pruning ratio where large number of filters are to be pruned from the network, the entry-wise norm methods eliminate relatively smaller norm filters without considering the significance of the filters in producing the node output, resulting in degradation in the performance. To address this, we present a passive filter pruning method where the filters are pruned based on their contribution in producing output by considering the operator norm of the filters. The proposed pruning method generalizes better across various CNNs compared to that of the entry-wise norm-based pruning methods. In comparison to the existing active filter pruning methods, the proposed pruning method is at least 4.5 times faster in computing filter importance and is able to achieve similar performance compared to that of the active filter pruning methods. The efficacy of the proposed pruning method is evaluated on audio scene classification and image classification using various CNNs architecture such as VGGish, DCASE21_Net, VGG-16 and ResNet-50.
[ { "version": "v1", "created": "Wed, 5 Apr 2023 09:19:19 GMT" } ]
2023-04-06T00:00:00
[ [ "Singh", "Arshdeep", "" ], [ "Plumbley", "Mark D.", "" ] ]
TITLE: Efficient CNNs via Passive Filter Pruning ABSTRACT: Convolutional neural networks (CNNs) have shown state-of-the-art performance in various applications. However, CNNs are resource-hungry due to their requirement of high computational complexity and memory storage. Recent efforts toward achieving computational efficiency in CNNs involve filter pruning methods that eliminate some of the filters in CNNs based on the \enquote{importance} of the filters. The majority of existing filter pruning methods are either "active", which use a dataset and generate feature maps to quantify filter importance, or "passive", which compute filter importance using entry-wise norm of the filters without involving data. Under a high pruning ratio where large number of filters are to be pruned from the network, the entry-wise norm methods eliminate relatively smaller norm filters without considering the significance of the filters in producing the node output, resulting in degradation in the performance. To address this, we present a passive filter pruning method where the filters are pruned based on their contribution in producing output by considering the operator norm of the filters. The proposed pruning method generalizes better across various CNNs compared to that of the entry-wise norm-based pruning methods. In comparison to the existing active filter pruning methods, the proposed pruning method is at least 4.5 times faster in computing filter importance and is able to achieve similar performance compared to that of the active filter pruning methods. The efficacy of the proposed pruning method is evaluated on audio scene classification and image classification using various CNNs architecture such as VGGish, DCASE21_Net, VGG-16 and ResNet-50.
no_new_dataset
0.71282
2304.02321
Marlene Careil
Marl\`ene Careil, Jakob Verbeek, St\'ephane Lathuili\`ere
Few-shot Semantic Image Synthesis with Class Affinity Transfer
Accepted to CVPR 2023
null
null
null
cs.CV
http://creativecommons.org/licenses/by-nc-sa/4.0/
Semantic image synthesis aims to generate photo realistic images given a semantic segmentation map. Despite much recent progress, training them still requires large datasets of images annotated with per-pixel label maps that are extremely tedious to obtain. To alleviate the high annotation cost, we propose a transfer method that leverages a model trained on a large source dataset to improve the learning ability on small target datasets via estimated pairwise relations between source and target classes. The class affinity matrix is introduced as a first layer to the source model to make it compatible with the target label maps, and the source model is then further finetuned for the target domain. To estimate the class affinities we consider different approaches to leverage prior knowledge: semantic segmentation on the source domain, textual label embeddings, and self-supervised vision features. We apply our approach to GAN-based and diffusion-based architectures for semantic synthesis. Our experiments show that the different ways to estimate class affinity can be effectively combined, and that our approach significantly improves over existing state-of-the-art transfer approaches for generative image models.
[ { "version": "v1", "created": "Wed, 5 Apr 2023 09:24:45 GMT" } ]
2023-04-06T00:00:00
[ [ "Careil", "Marlène", "" ], [ "Verbeek", "Jakob", "" ], [ "Lathuilière", "Stéphane", "" ] ]
TITLE: Few-shot Semantic Image Synthesis with Class Affinity Transfer ABSTRACT: Semantic image synthesis aims to generate photo realistic images given a semantic segmentation map. Despite much recent progress, training them still requires large datasets of images annotated with per-pixel label maps that are extremely tedious to obtain. To alleviate the high annotation cost, we propose a transfer method that leverages a model trained on a large source dataset to improve the learning ability on small target datasets via estimated pairwise relations between source and target classes. The class affinity matrix is introduced as a first layer to the source model to make it compatible with the target label maps, and the source model is then further finetuned for the target domain. To estimate the class affinities we consider different approaches to leverage prior knowledge: semantic segmentation on the source domain, textual label embeddings, and self-supervised vision features. We apply our approach to GAN-based and diffusion-based architectures for semantic synthesis. Our experiments show that the different ways to estimate class affinity can be effectively combined, and that our approach significantly improves over existing state-of-the-art transfer approaches for generative image models.
no_new_dataset
0.711619
2304.02349
Jose Angel Sosa Martinez
Jose Sosa and David Hogg
Self-supervised 3D Human Pose Estimation from a Single Image
null
null
null
null
cs.CV
http://creativecommons.org/licenses/by/4.0/
We propose a new self-supervised method for predicting 3D human body pose from a single image. The prediction network is trained from a dataset of unlabelled images depicting people in typical poses and a set of unpaired 2D poses. By minimising the need for annotated data, the method has the potential for rapid application to pose estimation of other articulated structures (e.g. animals). The self-supervision comes from an earlier idea exploiting consistency between predicted pose under 3D rotation. Our method is a substantial advance on state-of-the-art self-supervised methods in training a mapping directly from images, without limb articulation constraints or any 3D empirical pose prior. We compare performance with state-of-the-art self-supervised methods using benchmark datasets that provide images and ground-truth 3D pose (Human3.6M, MPI-INF-3DHP). Despite the reduced requirement for annotated data, we show that the method outperforms on Human3.6M and matches performance on MPI-INF-3DHP. Qualitative results on a dataset of human hands show the potential for rapidly learning to predict 3D pose for articulated structures other than the human body.
[ { "version": "v1", "created": "Wed, 5 Apr 2023 10:26:21 GMT" } ]
2023-04-06T00:00:00
[ [ "Sosa", "Jose", "" ], [ "Hogg", "David", "" ] ]
TITLE: Self-supervised 3D Human Pose Estimation from a Single Image ABSTRACT: We propose a new self-supervised method for predicting 3D human body pose from a single image. The prediction network is trained from a dataset of unlabelled images depicting people in typical poses and a set of unpaired 2D poses. By minimising the need for annotated data, the method has the potential for rapid application to pose estimation of other articulated structures (e.g. animals). The self-supervision comes from an earlier idea exploiting consistency between predicted pose under 3D rotation. Our method is a substantial advance on state-of-the-art self-supervised methods in training a mapping directly from images, without limb articulation constraints or any 3D empirical pose prior. We compare performance with state-of-the-art self-supervised methods using benchmark datasets that provide images and ground-truth 3D pose (Human3.6M, MPI-INF-3DHP). Despite the reduced requirement for annotated data, we show that the method outperforms on Human3.6M and matches performance on MPI-INF-3DHP. Qualitative results on a dataset of human hands show the potential for rapidly learning to predict 3D pose for articulated structures other than the human body.
no_new_dataset
0.639061
2304.02383
Antoine Passemiers
Antoine Passemiers, Pietro Folco, Daniele Raimondi, Giovanni Birolo, Yves Moreau, Piero Fariselli
How good Neural Networks interpretation methods really are? A quantitative benchmark
null
null
null
null
cs.LG
http://creativecommons.org/licenses/by/4.0/
Saliency Maps (SMs) have been extensively used to interpret deep learning models decision by highlighting the features deemed relevant by the model. They are used on highly nonlinear problems, where linear feature selection (FS) methods fail at highlighting relevant explanatory variables. However, the reliability of gradient-based feature attribution methods such as SM has mostly been only qualitatively (visually) assessed, and quantitative benchmarks are currently missing, partially due to the lack of a definite ground truth on image data. Concerned about the apophenic biases introduced by visual assessment of these methods, in this paper we propose a synthetic quantitative benchmark for Neural Networks (NNs) interpretation methods. For this purpose, we built synthetic datasets with nonlinearly separable classes and increasing number of decoy (random) features, illustrating the challenge of FS in high-dimensional settings. We also compare these methods to conventional approaches such as mRMR or Random Forests. Our results show that our simple synthetic datasets are sufficient to challenge most of the benchmarked methods. TreeShap, mRMR and LassoNet are the best performing FS methods. We also show that, when quantifying the relevance of a few non linearly-entangled predictive features diluted in a large number of irrelevant noisy variables, neural network-based FS and interpretation methods are still far from being reliable.
[ { "version": "v1", "created": "Wed, 5 Apr 2023 11:47:27 GMT" } ]
2023-04-06T00:00:00
[ [ "Passemiers", "Antoine", "" ], [ "Folco", "Pietro", "" ], [ "Raimondi", "Daniele", "" ], [ "Birolo", "Giovanni", "" ], [ "Moreau", "Yves", "" ], [ "Fariselli", "Piero", "" ] ]
TITLE: How good Neural Networks interpretation methods really are? A quantitative benchmark ABSTRACT: Saliency Maps (SMs) have been extensively used to interpret deep learning models decision by highlighting the features deemed relevant by the model. They are used on highly nonlinear problems, where linear feature selection (FS) methods fail at highlighting relevant explanatory variables. However, the reliability of gradient-based feature attribution methods such as SM has mostly been only qualitatively (visually) assessed, and quantitative benchmarks are currently missing, partially due to the lack of a definite ground truth on image data. Concerned about the apophenic biases introduced by visual assessment of these methods, in this paper we propose a synthetic quantitative benchmark for Neural Networks (NNs) interpretation methods. For this purpose, we built synthetic datasets with nonlinearly separable classes and increasing number of decoy (random) features, illustrating the challenge of FS in high-dimensional settings. We also compare these methods to conventional approaches such as mRMR or Random Forests. Our results show that our simple synthetic datasets are sufficient to challenge most of the benchmarked methods. TreeShap, mRMR and LassoNet are the best performing FS methods. We also show that, when quantifying the relevance of a few non linearly-entangled predictive features diluted in a large number of irrelevant noisy variables, neural network-based FS and interpretation methods are still far from being reliable.
new_dataset
0.702071
2304.02389
Bo Qian
Bo Qian, Hao Chen, Xiangning Wang, Haoxuan Che, Gitaek Kwon, Jaeyoung Kim, Sungjin Choi, Seoyoung Shin, Felix Krause, Markus Unterdechler, Junlin Hou, Rui Feng, Yihao Li, Mostafa El Habib Daho, Qiang Wu, Ping Zhang, Xiaokang Yang, Yiyu Cai, Weiping Jia, Huating Li, Bin Sheng
DRAC: Diabetic Retinopathy Analysis Challenge with Ultra-Wide Optical Coherence Tomography Angiography Images
null
null
null
null
eess.IV cs.CV cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Computer-assisted automatic analysis of diabetic retinopathy (DR) is of great importance in reducing the risks of vision loss and even blindness. Ultra-wide optical coherence tomography angiography (UW-OCTA) is a non-invasive and safe imaging modality in DR diagnosis system, but there is a lack of publicly available benchmarks for model development and evaluation. To promote further research and scientific benchmarking for diabetic retinopathy analysis using UW-OCTA images, we organized a challenge named "DRAC - Diabetic Retinopathy Analysis Challenge" in conjunction with the 25th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI 2022). The challenge consists of three tasks: segmentation of DR lesions, image quality assessment and DR grading. The scientific community responded positively to the challenge, with 11, 12, and 13 teams from geographically diverse institutes submitting different solutions in these three tasks, respectively. This paper presents a summary and analysis of the top-performing solutions and results for each task of the challenge. The obtained results from top algorithms indicate the importance of data augmentation, model architecture and ensemble of networks in improving the performance of deep learning models. These findings have the potential to enable new developments in diabetic retinopathy analysis. The challenge remains open for post-challenge registrations and submissions for benchmarking future methodology developments.
[ { "version": "v1", "created": "Wed, 5 Apr 2023 12:04:55 GMT" } ]
2023-04-06T00:00:00
[ [ "Qian", "Bo", "" ], [ "Chen", "Hao", "" ], [ "Wang", "Xiangning", "" ], [ "Che", "Haoxuan", "" ], [ "Kwon", "Gitaek", "" ], [ "Kim", "Jaeyoung", "" ], [ "Choi", "Sungjin", "" ], [ "Shin", "Seoyoung", "" ], [ "Krause", "Felix", "" ], [ "Unterdechler", "Markus", "" ], [ "Hou", "Junlin", "" ], [ "Feng", "Rui", "" ], [ "Li", "Yihao", "" ], [ "Daho", "Mostafa El Habib", "" ], [ "Wu", "Qiang", "" ], [ "Zhang", "Ping", "" ], [ "Yang", "Xiaokang", "" ], [ "Cai", "Yiyu", "" ], [ "Jia", "Weiping", "" ], [ "Li", "Huating", "" ], [ "Sheng", "Bin", "" ] ]
TITLE: DRAC: Diabetic Retinopathy Analysis Challenge with Ultra-Wide Optical Coherence Tomography Angiography Images ABSTRACT: Computer-assisted automatic analysis of diabetic retinopathy (DR) is of great importance in reducing the risks of vision loss and even blindness. Ultra-wide optical coherence tomography angiography (UW-OCTA) is a non-invasive and safe imaging modality in DR diagnosis system, but there is a lack of publicly available benchmarks for model development and evaluation. To promote further research and scientific benchmarking for diabetic retinopathy analysis using UW-OCTA images, we organized a challenge named "DRAC - Diabetic Retinopathy Analysis Challenge" in conjunction with the 25th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI 2022). The challenge consists of three tasks: segmentation of DR lesions, image quality assessment and DR grading. The scientific community responded positively to the challenge, with 11, 12, and 13 teams from geographically diverse institutes submitting different solutions in these three tasks, respectively. This paper presents a summary and analysis of the top-performing solutions and results for each task of the challenge. The obtained results from top algorithms indicate the importance of data augmentation, model architecture and ensemble of networks in improving the performance of deep learning models. These findings have the potential to enable new developments in diabetic retinopathy analysis. The challenge remains open for post-challenge registrations and submissions for benchmarking future methodology developments.
no_new_dataset
0.698394
2304.02554
Mingqi Gao
Mingqi Gao, Jie Ruan, Renliang Sun, Xunjian Yin, Shiping Yang, Xiaojun Wan
Human-like Summarization Evaluation with ChatGPT
9 pages, 5 figures, in process
null
null
null
cs.CL
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Evaluating text summarization is a challenging problem, and existing evaluation metrics are far from satisfactory. In this study, we explored ChatGPT's ability to perform human-like summarization evaluation using four human evaluation methods on five datasets. We found that ChatGPT was able to complete annotations relatively smoothly using Likert scale scoring, pairwise comparison, Pyramid, and binary factuality evaluation. Additionally, it outperformed commonly used automatic evaluation metrics on some datasets. Furthermore, we discussed the impact of different prompts, compared its performance with that of human evaluation, and analyzed the generated explanations and invalid responses.
[ { "version": "v1", "created": "Wed, 5 Apr 2023 16:17:32 GMT" } ]
2023-04-06T00:00:00
[ [ "Gao", "Mingqi", "" ], [ "Ruan", "Jie", "" ], [ "Sun", "Renliang", "" ], [ "Yin", "Xunjian", "" ], [ "Yang", "Shiping", "" ], [ "Wan", "Xiaojun", "" ] ]
TITLE: Human-like Summarization Evaluation with ChatGPT ABSTRACT: Evaluating text summarization is a challenging problem, and existing evaluation metrics are far from satisfactory. In this study, we explored ChatGPT's ability to perform human-like summarization evaluation using four human evaluation methods on five datasets. We found that ChatGPT was able to complete annotations relatively smoothly using Likert scale scoring, pairwise comparison, Pyramid, and binary factuality evaluation. Additionally, it outperformed commonly used automatic evaluation metrics on some datasets. Furthermore, we discussed the impact of different prompts, compared its performance with that of human evaluation, and analyzed the generated explanations and invalid responses.
no_new_dataset
0.709466
2304.02556
Rui Shao
Rui Shao, Tianxing Wu, Ziwei Liu
Detecting and Grounding Multi-Modal Media Manipulation
CVPR 2023. Project page: https://rshaojimmy.github.io/Projects/MultiModal-DeepFake Code: https://github.com/rshaojimmy/MultiModal-DeepFake
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Misinformation has become a pressing issue. Fake media, in both visual and textual forms, is widespread on the web. While various deepfake detection and text fake news detection methods have been proposed, they are only designed for single-modality forgery based on binary classification, let alone analyzing and reasoning subtle forgery traces across different modalities. In this paper, we highlight a new research problem for multi-modal fake media, namely Detecting and Grounding Multi-Modal Media Manipulation (DGM^4). DGM^4 aims to not only detect the authenticity of multi-modal media, but also ground the manipulated content (i.e., image bounding boxes and text tokens), which requires deeper reasoning of multi-modal media manipulation. To support a large-scale investigation, we construct the first DGM^4 dataset, where image-text pairs are manipulated by various approaches, with rich annotation of diverse manipulations. Moreover, we propose a novel HierArchical Multi-modal Manipulation rEasoning tRansformer (HAMMER) to fully capture the fine-grained interaction between different modalities. HAMMER performs 1) manipulation-aware contrastive learning between two uni-modal encoders as shallow manipulation reasoning, and 2) modality-aware cross-attention by multi-modal aggregator as deep manipulation reasoning. Dedicated manipulation detection and grounding heads are integrated from shallow to deep levels based on the interacted multi-modal information. Finally, we build an extensive benchmark and set up rigorous evaluation metrics for this new research problem. Comprehensive experiments demonstrate the superiority of our model; several valuable observations are also revealed to facilitate future research in multi-modal media manipulation.
[ { "version": "v1", "created": "Wed, 5 Apr 2023 16:20:40 GMT" } ]
2023-04-06T00:00:00
[ [ "Shao", "Rui", "" ], [ "Wu", "Tianxing", "" ], [ "Liu", "Ziwei", "" ] ]
TITLE: Detecting and Grounding Multi-Modal Media Manipulation ABSTRACT: Misinformation has become a pressing issue. Fake media, in both visual and textual forms, is widespread on the web. While various deepfake detection and text fake news detection methods have been proposed, they are only designed for single-modality forgery based on binary classification, let alone analyzing and reasoning subtle forgery traces across different modalities. In this paper, we highlight a new research problem for multi-modal fake media, namely Detecting and Grounding Multi-Modal Media Manipulation (DGM^4). DGM^4 aims to not only detect the authenticity of multi-modal media, but also ground the manipulated content (i.e., image bounding boxes and text tokens), which requires deeper reasoning of multi-modal media manipulation. To support a large-scale investigation, we construct the first DGM^4 dataset, where image-text pairs are manipulated by various approaches, with rich annotation of diverse manipulations. Moreover, we propose a novel HierArchical Multi-modal Manipulation rEasoning tRansformer (HAMMER) to fully capture the fine-grained interaction between different modalities. HAMMER performs 1) manipulation-aware contrastive learning between two uni-modal encoders as shallow manipulation reasoning, and 2) modality-aware cross-attention by multi-modal aggregator as deep manipulation reasoning. Dedicated manipulation detection and grounding heads are integrated from shallow to deep levels based on the interacted multi-modal information. Finally, we build an extensive benchmark and set up rigorous evaluation metrics for this new research problem. Comprehensive experiments demonstrate the superiority of our model; several valuable observations are also revealed to facilitate future research in multi-modal media manipulation.
new_dataset
0.714466
2304.02569
Shengyu Huang
Liyuan Zhu, Yuru Jia, Shengyu Huang, Nicholas Meyer, Andreas Wieser, Konrad Schindler, Jordan Aaron
DEFLOW: Self-supervised 3D Motion Estimation of Debris Flow
Photogrammetric Computer Vision Workshop, CVPRW 2023, camera ready
null
null
null
cs.CV
http://creativecommons.org/licenses/by/4.0/
Existing work on scene flow estimation focuses on autonomous driving and mobile robotics, while automated solutions are lacking for motion in nature, such as that exhibited by debris flows. We propose DEFLOW, a model for 3D motion estimation of debris flows, together with a newly captured dataset. We adopt a novel multi-level sensor fusion architecture and self-supervision to incorporate the inductive biases of the scene. We further adopt a multi-frame temporal processing module to enable flow speed estimation over time. Our model achieves state-of-the-art optical flow and depth estimation on our dataset, and fully automates the motion estimation for debris flows. The source code and dataset are available at project page.
[ { "version": "v1", "created": "Wed, 5 Apr 2023 16:40:14 GMT" } ]
2023-04-06T00:00:00
[ [ "Zhu", "Liyuan", "" ], [ "Jia", "Yuru", "" ], [ "Huang", "Shengyu", "" ], [ "Meyer", "Nicholas", "" ], [ "Wieser", "Andreas", "" ], [ "Schindler", "Konrad", "" ], [ "Aaron", "Jordan", "" ] ]
TITLE: DEFLOW: Self-supervised 3D Motion Estimation of Debris Flow ABSTRACT: Existing work on scene flow estimation focuses on autonomous driving and mobile robotics, while automated solutions are lacking for motion in nature, such as that exhibited by debris flows. We propose DEFLOW, a model for 3D motion estimation of debris flows, together with a newly captured dataset. We adopt a novel multi-level sensor fusion architecture and self-supervision to incorporate the inductive biases of the scene. We further adopt a multi-frame temporal processing module to enable flow speed estimation over time. Our model achieves state-of-the-art optical flow and depth estimation on our dataset, and fully automates the motion estimation for debris flows. The source code and dataset are available at project page.
new_dataset
0.714584
2304.02577
Philip Aston
Temesgen Mehari, Ashish Sundar, Alen Bosnjakovic, Peter Harris, Steven E. Williams, Axel Loewe, Olaf Doessel, Claudia Nagel, Nils Strodthoff, Philip J. Aston
ECG Feature Importance Rankings: Cardiologists vs. Algorithms
null
null
null
null
physics.med-ph cs.LG eess.SP
http://creativecommons.org/licenses/by-nc-sa/4.0/
Feature importance methods promise to provide a ranking of features according to importance for a given classification task. A wide range of methods exist but their rankings often disagree and they are inherently difficult to evaluate due to a lack of ground truth beyond synthetic datasets. In this work, we put feature importance methods to the test on real-world data in the domain of cardiology, where we try to distinguish three specific pathologies from healthy subjects based on ECG features comparing to features used in cardiologists' decision rules as ground truth. Some methods generally performed well and others performed poorly, while some methods did well on some but not all of the problems considered.
[ { "version": "v1", "created": "Wed, 5 Apr 2023 16:48:24 GMT" } ]
2023-04-06T00:00:00
[ [ "Mehari", "Temesgen", "" ], [ "Sundar", "Ashish", "" ], [ "Bosnjakovic", "Alen", "" ], [ "Harris", "Peter", "" ], [ "Williams", "Steven E.", "" ], [ "Loewe", "Axel", "" ], [ "Doessel", "Olaf", "" ], [ "Nagel", "Claudia", "" ], [ "Strodthoff", "Nils", "" ], [ "Aston", "Philip J.", "" ] ]
TITLE: ECG Feature Importance Rankings: Cardiologists vs. Algorithms ABSTRACT: Feature importance methods promise to provide a ranking of features according to importance for a given classification task. A wide range of methods exist but their rankings often disagree and they are inherently difficult to evaluate due to a lack of ground truth beyond synthetic datasets. In this work, we put feature importance methods to the test on real-world data in the domain of cardiology, where we try to distinguish three specific pathologies from healthy subjects based on ECG features comparing to features used in cardiologists' decision rules as ground truth. Some methods generally performed well and others performed poorly, while some methods did well on some but not all of the problems considered.
no_new_dataset
0.713238
2304.02643
Alexander Kirillov
Alexander Kirillov, Eric Mintun, Nikhila Ravi, Hanzi Mao, Chloe Rolland, Laura Gustafson, Tete Xiao, Spencer Whitehead, Alexander C. Berg, Wan-Yen Lo, Piotr Doll\'ar, Ross Girshick
Segment Anything
Project web-page: https://segment-anything.com
null
null
null
cs.CV cs.AI cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
We introduce the Segment Anything (SA) project: a new task, model, and dataset for image segmentation. Using our efficient model in a data collection loop, we built the largest segmentation dataset to date (by far), with over 1 billion masks on 11M licensed and privacy respecting images. The model is designed and trained to be promptable, so it can transfer zero-shot to new image distributions and tasks. We evaluate its capabilities on numerous tasks and find that its zero-shot performance is impressive -- often competitive with or even superior to prior fully supervised results. We are releasing the Segment Anything Model (SAM) and corresponding dataset (SA-1B) of 1B masks and 11M images at https://segment-anything.com to foster research into foundation models for computer vision.
[ { "version": "v1", "created": "Wed, 5 Apr 2023 17:59:46 GMT" } ]
2023-04-06T00:00:00
[ [ "Kirillov", "Alexander", "" ], [ "Mintun", "Eric", "" ], [ "Ravi", "Nikhila", "" ], [ "Mao", "Hanzi", "" ], [ "Rolland", "Chloe", "" ], [ "Gustafson", "Laura", "" ], [ "Xiao", "Tete", "" ], [ "Whitehead", "Spencer", "" ], [ "Berg", "Alexander C.", "" ], [ "Lo", "Wan-Yen", "" ], [ "Dollár", "Piotr", "" ], [ "Girshick", "Ross", "" ] ]
TITLE: Segment Anything ABSTRACT: We introduce the Segment Anything (SA) project: a new task, model, and dataset for image segmentation. Using our efficient model in a data collection loop, we built the largest segmentation dataset to date (by far), with over 1 billion masks on 11M licensed and privacy respecting images. The model is designed and trained to be promptable, so it can transfer zero-shot to new image distributions and tasks. We evaluate its capabilities on numerous tasks and find that its zero-shot performance is impressive -- often competitive with or even superior to prior fully supervised results. We are releasing the Segment Anything Model (SAM) and corresponding dataset (SA-1B) of 1B masks and 11M images at https://segment-anything.com to foster research into foundation models for computer vision.
new_dataset
0.71227