1 code implementation • 1 Mar 2025 • Zhuo Ouyang, Kaiwen Hu, Qi Zhang, Yifei Wang, Yisen Wang
In this paper, we develop an in-depth theoretical understanding of the projection head from the information-theoretic perspective.
no code implementations • 11 Feb 2025 • Yuyang Wu, Yifei Wang, Tianqi Du, Stefanie Jegelka, Yisen Wang
We theoretically prove the existence of an optimal CoT length and derive a scaling law for this optimal length based on model capability and task difficulty.
no code implementations • 4 Feb 2025 • Yisen Wang, HanQin Cai, Longxiu Huang
Multidimensional signal recovery has also been studied, but primarily in scenarios where the driving operator is a convolution operator.
no code implementations • 3 Jan 2025 • Mingjie Li, Wai Man Si, Michael Backes, Yang Zhang, Yisen Wang
As advancements in large language models (LLMs) continue and the demand for personalized models increases, parameter-efficient fine-tuning (PEFT) methods (e. g., LoRA) will become essential due to their efficiency in reducing computation costs.
no code implementations • 2 Jan 2025 • Yi-Ge Zhang, Jingyi Cui, Qiran Li, Yisen Wang
Unsupervised contrastive learning has shown significant performance improvements in recent years, often approaching or even rivaling supervised learning in various tasks.
no code implementations • 2 Jan 2025 • Jingyi Cui, Yi-Ge Zhang, Hengyu Liu, Yisen Wang
In particular, we for the first time derive a general robust condition for arbitrary contrastive losses, which serves as a criterion to verify the theoretical robustness of a supervised contrastive loss against label noise.
no code implementations • 26 Nov 2024 • Xiao Lin, Mingjie Li, Yisen Wang
Graph Neural Networks (GNNs) have garnered significant attention from researchers due to their outstanding performance in handling graph-related tasks, such as social network analysis, protein design, and so on.
1 code implementation • 10 Nov 2024 • Yifei Wang, Kaiwen Hu, Sharut Gupta, Ziyu Ye, Yisen Wang, Stefanie Jegelka
Given this limitation, there has been a surge of interest in equivariant self-supervised learning (E-SSL) that learns features to be augmentation-aware.
1 code implementation • 5 Nov 2024 • Qixun Wang, Yifei Wang, Yisen Wang, Xianghua Ying
Enhancing node-level Out-Of-Distribution (OOD) generalization on graphs remains a crucial area of research.
1 code implementation • 31 Oct 2024 • Lizhe Fang, Yifei Wang, Zhaoyang Liu, Chenheng Zhang, Stefanie Jegelka, Jinyang Gao, Bolin Ding, Yisen Wang
To address this, we propose \textbf{LongPPL}, a novel metric that focuses on key tokens by employing a long-short context contrastive method to identify them.
1 code implementation • 27 Oct 2024 • Qi Zhang, Yifei Wang, Jingyi Cui, Xiang Pan, Qi Lei, Stefanie Jegelka, Yisen Wang
In this work, we challenge the prevailing belief of the accuracy-interpretability tradeoff, showing that monosemantic features not only enhance interpretability but also bring concrete gains in model performance.
no code implementations • 13 Oct 2024 • Qixun Wang, Yifei Wang, Yisen Wang, Xianghua Ying
To achieve this, we conduct synthetic experiments where the objective is to learn OOD mathematical functions through ICL using a GPT-2 model.
1 code implementation • 11 Oct 2024 • Yisen Wang, Yichuan Mo, Dongxian Wu, Mingjie Li, Xingjun Ma, Zhouchen Lin
Specifically, in ResNet-like models (with skip connections), we find that using more gradients from the skip connections rather than the residual modules according to a decay factor during backpropagation allows one to craft adversarial examples with high transferability.
1 code implementation • 11 Oct 2024 • Zijun Wang, Haoqin Tu, Jieru Mei, Bingchen Zhao, Yisen Wang, Cihang Xie
This paper studies the vulnerabilities of transformer-based Large Language Models (LLMs) to jailbreaking attacks, focusing specifically on the optimization-based Greedy Coordinate Gradient (GCG) strategy.
no code implementations • 1 Oct 2024 • Lexiang Hu, Yisen Wang, Zhouchen Lin
Kolmogorov-Arnold Networks (KANs) have seen great success in scientific domains thanks to spline activation functions, becoming an alternative to Multi-Layer Perceptrons (MLPs).
1 code implementation • 9 Sep 2024 • Yichuan Mo, Hui Huang, Mingjie Li, Ang Li, Yisen Wang
Additionally, with the reversed trigger, we propose backdoor detection from the noise space, introducing the first backdoor input detection approach for diffusion models and a novel model detection algorithm that calculates the KL divergence between reversed and benign distributions.
no code implementations • 16 Jul 2024 • Yisen Wang, Yao Teng, LiMin Wang
Recognition and generation are two fundamental tasks in computer vision, which are often investigated separately in the exiting literature.
1 code implementation • 12 Jul 2024 • Tianqi Du, Yifei Wang, Yisen Wang
Furthermore, we propose a novel metric named TCAS, which is specifically designed to assess the effectiveness of discrete tokens within the MIM framework.
no code implementations • 1 Jul 2024 • Qi Zhang, Tianqi Du, Haotian Huang, Yifei Wang, Yisen Wang
In recent years, the rise of generative self-supervised learning (SSL) paradigms has exhibited impressive performance across visual, language, and multi-modal domains.
1 code implementation • 14 Jun 2024 • Ang Li, Yichuan Mo, Mingjie Li, Yisen Wang
Drawing on these insights, we propose a simple yet effective method called \textbf{Prompt-Independent Defense (PID)} to safeguard privacy against LDMs.
no code implementations • 28 May 2024 • Yifei Wang, Yuyang Wu, Zeming Wei, Stefanie Jegelka, Yisen Wang
Going beyond mimicking limited human experiences, recent studies show initial evidence that, like humans, large language models (LLMs) are capable of improving their abilities purely by self-correction, i. e., correcting previous responses through self-examination, in certain circumstances.
2 code implementations • 28 May 2024 • George Ma, Yifei Wang, Derek Lim, Stefanie Jegelka, Yisen Wang
In this work, we introduce a canonicalization perspective that provides an essential and complete view of the design of frames.
1 code implementation • 10 Apr 2024 • Yifei Wang, Wenhan Ma, Stefanie Jegelka, Yisen Wang
Relying only on unlabeled data, Self-supervised learning (SSL) can learn rich features in an economical and scalable way.
no code implementations • 20 Mar 2024 • Jinmin Li, Kuofeng Gao, Yang Bai, Jingyun Zhang, Shu-Tao Xia, Yisen Wang
Despite the remarkable performance of video-based large language models (LLMs), their adversarial threat remains unexplored.
1 code implementation • 19 Mar 2024 • Yifei Wang, Qi Zhang, Yaoyu Guo, Yisen Wang
In this paper, we propose Non-negative Contrastive Learning (NCL), a renaissance of Non-negative Matrix Factorization (NMF) aimed at deriving interpretable features.
1 code implementation • 19 Mar 2024 • Yifei Wang, Jizhe Zhang, Yisen Wang
Contrastive Learning (CL) has emerged as one of the most successful paradigms for unsupervised visual representation learning, yet it often depends on intensive manual data augmentations.
2 code implementations • 9 Feb 2024 • Yichuan Mo, Yuji Wang, Zeming Wei, Yisen Wang
To achieve our defense goal whilst maintaining natural performance, we optimize the control prompt with both adversarial and benign prompts.
1 code implementation • NeurIPS 2023 • Xiaojun Guo, Yifei Wang, Zeming Wei, Yisen Wang
With the prosperity of contrastive learning for visual representation learning (VCL), it is also adapted to the graph domain and yields promising performance.
1 code implementation • NeurIPS 2023 • Ang Li, Yifei Wang, Yiwen Guo, Yisen Wang
A well-known theory by \citet{ilyas2019adversarial} explains adversarial vulnerability from a data perspective by showing that one can extract non-robust features from adversarial examples and these features alone are useful for classification.
3 code implementations • NeurIPS 2023 • Jiangyan Ma, Yifei Wang, Yisen Wang
However, from a theoretical perspective, the universal expressive power of spectral embedding comes at the price of losing two important invariance properties of graphs, sign and basis invariance, which also limits its effectiveness on graph data.
no code implementations • 25 Oct 2023 • Chen Liu, Hongyu Zang, Xin Li, Yong Heng, Yifei Wang, Zhen Fang, Yisen Wang, Mingzhong Wang
Image-based Reinforcement Learning is a practical yet challenging task.
no code implementations • 10 Oct 2023 • Zeming Wei, Yifei Wang, Ang Li, Yichuan Mo, Yisen Wang
Large Language Models (LLMs) have shown remarkable success in various tasks, yet their safety and the risk of generating harmful content remain pressing concerns.
no code implementations • 29 Aug 2023 • Hong Zhu, Runpeng Yu, Xing Tang, Yifei Wang, Yuan Fang, Yisen Wang
Data in the real-world classification problems are always imbalanced or long-tailed, wherein the majority classes have the most of the samples that dominate the model training.
no code implementations • 7 Jun 2023 • Jingyi Cui, Weiran Huang, Yifei Wang, Yisen Wang
Therefore, to explore the mechanical differences between semi-supervised and noisy-labeled information in helping contrastive learning, we establish a unified theoretical framework of contrastive learning under weak supervision.
1 code implementation • 7 Jun 2023 • Qi Zhang, Yifei Wang, Yisen Wang
Multi-modal contrastive learning (MMCL) has recently garnered considerable interest due to its superior performance in visual tasks, achieved by embedding multi-modal data, such as visual-language pairs.
1 code implementation • CVPR 2023 • Zeming Wei, Yifei Wang, Yiwen Guo, Yisen Wang
Adversarial training has been widely acknowledged as the most effective method to improve the adversarial robustness against adversarial examples for Deep Neural Networks (DNNs).
1 code implementation • CVPR 2023 • Hongjun Wang, Yisen Wang
The parameters of base learners are collected and combined to form a global learner at intervals during the training process.
2 code implementations • 12 Mar 2023 • Xiaojun Guo, Yifei Wang, Tianqi Du, Yisen Wang
Instead of characterizing oversmoothing from the view of complete collapse in which representations converge to a single point, we dive into a more general perspective of dimensional collapse in which representations lie in a narrow cone.
Ranked #8 on
Node Property Prediction
on ogbn-arxiv
1 code implementation • 8 Mar 2023 • Yifei Wang, Qi Zhang, Tianqi Du, Jiansheng Yang, Zhouchen Lin, Yisen Wang
In recent years, contrastive learning achieves impressive results on self-supervised visual representation learning, but there still lacks a rigorous understanding of its learning dynamics.
1 code implementation • 4 Mar 2023 • Zhijian Zhuo, Yifei Wang, Jinwen Ma, Yisen Wang
In this work, we propose a unified theoretical understanding for existing variants of non-contrastive learning.
1 code implementation • 2 Mar 2023 • Rundong Luo, Yifei Wang, Yisen Wang
Motivated by this observation, we revisit existing self-AT methods and discover an inherent dilemma that affects self-AT robustness: either strong or weak data augmentations are harmful to self-AT, and a medium strength is insufficient to bridge the gap.
no code implementations • 2 Mar 2023 • Xuyang Zhao, Tianqi Du, Yisen Wang, Jun Yao, Weiran Huang
Moreover, we show that contrastive learning fails to learn domain-invariant features, which limits its transferability.
1 code implementation • ICCV 2023 • Qingyan Meng, Mingqing Xiao, Shen Yan, Yisen Wang, Zhouchen Lin, Zhi-Quan Luo
In particular, our method achieves state-of-the-art accuracy on ImageNet, while the memory cost and training time are reduced by more than 70% and 50%, respectively, compared with BPTT.
1 code implementation • 1 Feb 2023 • Mingqing Xiao, Qingyan Meng, Zongpeng Zhang, Yisen Wang, Zhouchen Lin
In this paper, we study spike-based implicit differentiation on the equilibrium state (SPIDE) that extends the recently proposed training method, implicit differentiation on the equilibrium state (IDE), for supervised learning with purely spike-based computation, which demonstrates the potential for energy-efficient training of SNNs.
no code implementations • 18 Dec 2022 • Shiji Xin, Yifei Wang, Jingtong Su, Yisen Wang
Extensive experiments show that our proposed DAT can effectively remove domain-varying features and improve OOD generalization under both correlation shift and diversity shift.
2 code implementations • 15 Oct 2022 • Qi Zhang, Yifei Wang, Yisen Wang
Masked Autoencoders (MAE) based on a reconstruction task have risen to be a promising paradigm for self-supervised learning (SSL) and achieve state-of-the-art performance across different benchmark datasets.
1 code implementation • 14 Oct 2022 • Yichuan Mo, Dongxian Wu, Yifei Wang, Yiwen Guo, Yisen Wang
We find, when randomly masking gradients from some attention blocks or masking perturbations on some patches during adversarial training, the adversarial robustness of ViTs can be remarkably improved, which may potentially open up a line of work to explore the architectural information inside the newly designed models like ViTs.
1 code implementation • 13 Oct 2022 • Qixun Wang, Yifei Wang, Hong Zhu, Yisen Wang
In this paper, we empirically show that sample-wise AT has limited improvement on OOD performance.
no code implementations • 24 Jul 2022 • Quanshi Zhang, Xin Wang, Jie Ren, Xu Cheng, Shuyun Lin, Yisen Wang, Xiangming Zhu
This paper summarizes the common mechanism shared by twelve previous transferability-boosting methods in a unified view, i. e., these methods all reduce game-theoretic interactions between regional adversarial perturbations.
1 code implementation • 14 Jul 2022 • Chen Chen, Yisen Wang, Honghua Chen, Xuefeng Yan, Dayong Ren, Yanwen Guo, Haoran Xie, Fu Lee Wang, Mingqiang Wei
Semantic segmentation of point clouds, aiming to assign each point a semantic category, is critical to 3D scene understanding. Despite of significant advances in recent years, most of existing methods still suffer from either the object-level misclassification or the boundary-level ambiguity.
1 code implementation • 29 Jun 2022 • Qi Chen, Yifei Wang, Yisen Wang, Jiansheng Yang, Zhouchen Lin
Moreover, we show that the optimization-induced variants of our models can boost the performance and improve training stability and efficiency as well.
1 code implementation • CVPR 2022 • Qingyan Meng, Mingqing Xiao, Shen Yan, Yisen Wang, Zhouchen Lin, Zhi-Quan Luo
In this paper, we propose the Differentiation on Spike Representation (DSR) method, which could achieve high performance that is competitive to ANNs yet with low latency.
no code implementations • 26 Mar 2022 • Sha Yuan, Hanyu Zhao, Shuai Zhao, Jiahong Leng, Yangxiao Liang, Xiaozhi Wang, Jifan Yu, Xin Lv, Zhou Shao, Jiaao He, Yankai Lin, Xu Han, Zhenghao Liu, Ning Ding, Yongming Rao, Yizhao Gao, Liang Zhang, Ming Ding, Cong Fang, Yisen Wang, Mingsheng Long, Jing Zhang, Yinpeng Dong, Tianyu Pang, Peng Cui, Lingxiao Huang, Zheng Liang, HuaWei Shen, HUI ZHANG, Quanshi Zhang, Qingxiu Dong, Zhixing Tan, Mingxuan Wang, Shuo Wang, Long Zhou, Haoran Li, Junwei Bao, Yingwei Pan, Weinan Zhang, Zhou Yu, Rui Yan, Chence Shi, Minghao Xu, Zuobai Zhang, Guoqiang Wang, Xiang Pan, Mengjie Li, Xiaoyu Chu, Zijun Yao, Fangwei Zhu, Shulin Cao, Weicheng Xue, Zixuan Ma, Zhengyan Zhang, Shengding Hu, Yujia Qin, Chaojun Xiao, Zheni Zeng, Ganqu Cui, Weize Chen, Weilin Zhao, Yuan YAO, Peng Li, Wenzhao Zheng, Wenliang Zhao, Ziyi Wang, Borui Zhang, Nanyi Fei, Anwen Hu, Zenan Ling, Haoyang Li, Boxi Cao, Xianpei Han, Weidong Zhan, Baobao Chang, Hao Sun, Jiawen Deng, Chujie Zheng, Juanzi Li, Lei Hou, Xigang Cao, Jidong Zhai, Zhiyuan Liu, Maosong Sun, Jiwen Lu, Zhiwu Lu, Qin Jin, Ruihua Song, Ji-Rong Wen, Zhouchen Lin, LiWei Wang, Hang Su, Jun Zhu, Zhifang Sui, Jiajun Zhang, Yang Liu, Xiaodong He, Minlie Huang, Jian Tang, Jie Tang
With the rapid development of deep learning, training Big Models (BMs) for multiple downstream tasks becomes a popular paradigm.
1 code implementation • 25 Mar 2022 • Yifei Wang, Qi Zhang, Yisen Wang, Jiansheng Yang, Zhouchen Lin
Our theory suggests an alternative understanding of contrastive learning: the role of aligning positive samples is more like a surrogate task than an ultimate goal, and the overlapped augmented views (i. e., the chaos) create a ladder for contrastive learning to gradually learn class-separated representations.
no code implementations • ICLR 2022 • Yifei Wang, Yisen Wang, Jiansheng Yang, Zhouchen Lin
On the other hand, our unified framework can be extended to the unsupervised scenario, which interprets unsupervised contrastive learning as an important sampling of CEM.
1 code implementation • ICLR 2022 • Hongjun Wang, Yisen Wang
In this work, we are dedicated to the weight states of models through the training process and devise a simple but powerful \emph{Self-Ensemble Adversarial Training} (SEAT) method for yielding a robust classifier by averaging weights of history models.
no code implementations • 15 Dec 2021 • Yisen Wang, Xingjun Ma, James Bailey, JinFeng Yi, BoWen Zhou, Quanquan Gu
In this paper, we propose such a criterion, namely First-Order Stationary Condition for constrained optimization (FOSC), to quantitatively evaluate the convergence quality of adversarial examples found in the inner maximization.
1 code implementation • NeurIPS 2021 • Jie Ren, Die Zhang, Yisen Wang, Lu Chen, Zhanpeng Zhou, Yiting Chen, Xu Cheng, Xin Wang, Meng Zhou, Jie Shi, Quanshi Zhang
This paper provides a unified view to explain different adversarial attacks and defense methods, i. e. the view of multi-order interactions between input variables of DNNs.
1 code implementation • NeurIPS 2021 • Lingshen He, Yuxuan Chen, Zhengyang Shen, Yiming Dong, Yisen Wang, Zhouchen Lin
Group equivariant CNNs (G-CNNs) that incorporate more equivariance can significantly improve the performance of conventional CNNs.
1 code implementation • NeurIPS 2021 • Dantong Niu, Ruohao Guo, Yisen Wang
Images, captured by a camera, play a critical role in training Deep Neural Networks (DNNs).
no code implementations • NeurIPS 2021 • Lingshen He, Yiming Dong, Yisen Wang, DaCheng Tao, Zhouchen Lin
Attention mechanism has shown great performance and efficiency in a lot of deep learning models, in which relative position encoding plays a crucial role.
no code implementations • NeurIPS 2021 • Yang Bai, Xin Yan, Yong Jiang, Shu-Tao Xia, Yisen Wang
Adversarial robustness has received increasing attention along with the study of adversarial examples.
1 code implementation • 25 Nov 2021 • Yang Bai, Xin Yan, Yong Jiang, Shu-Tao Xia, Yisen Wang
Adversarial robustness has received increasing attention along with the study of adversarial examples.
no code implementations • 19 Nov 2021 • Zhirui Wang, Yifei Wang, Yisen Wang
Adversarial training is widely believed to be a reliable approach to improve model robustness against adversarial attack.
1 code implementation • NeurIPS 2021 • Chen Ma, Xiangyu Guo, Li Chen, Jun-Hai Yong, Yisen Wang
In this paper, we propose a novel geometric-based approach called Tangent Attack (TA), which identifies an optimal tangent point of a virtual hemisphere located on the decision boundary to reduce the distortion of the attack.
1 code implementation • NeurIPS 2021 • Zhengyang Geng, Xin-Yu Zhang, Shaojie Bai, Yisen Wang, Zhouchen Lin
This paper focuses on training implicit models of infinite layers.
1 code implementation • 5 Nov 2021 • Jie Ren, Die Zhang, Yisen Wang, Lu Chen, Zhanpeng Zhou, Yiting Chen, Xu Cheng, Xin Wang, Meng Zhou, Jie Shi, Quanshi Zhang
This paper provides a unified view to explain different adversarial attacks and defense methods, \emph{i. e.} the view of multi-order interactions between input variables of DNNs.
no code implementations • NeurIPS 2021 • Yifei Wang, Zhengyang Geng, Feng Jiang, Chuming Li, Yisen Wang, Jiansheng Yang, Zhouchen Lin
Multi-view methods learn representations by aligning multiple views of the same image and their performance largely depends on the choice of data augmentation.
2 code implementations • NeurIPS 2021 • Dongxian Wu, Yisen Wang
As deep neural networks (DNNs) are growing larger, their requirements for computational resources become huge, which makes outsourcing training more popular.
1 code implementation • 20 Oct 2021 • Dantong Niu, Ruohao Guo, Yisen Wang
Images, captured by a camera, play a critical role in training Deep Neural Networks (DNNs).
no code implementations • 7 Oct 2021 • Wenbin Ouyang, Yisen Wang, Paul Weng, Shaochen Han
Since training on large instances is impractical, we design a novel deep RL approach with a focus on generalizability.
1 code implementation • NeurIPS 2021 • Hanxun Huang, Yisen Wang, Sarah Monazam Erfani, Quanquan Gu, James Bailey, Xingjun Ma
Specifically, we make the following key observations: 1) more parameters (higher model capacity) does not necessarily help adversarial robustness; 2) reducing capacity at the last stage (the last group of blocks) of the network can actually improve adversarial robustness; and 3) under the same parameter budget, there exists an optimal architectural configuration for adversarial robustness.
no code implementations • 6 Oct 2021 • Wenbin Ouyang, Yisen Wang, Shaochen Han, Zhejian Jin, Paul Weng
In this work, we propose a novel approach named MAGIC that includes a deep learning architecture and a DRL training method.
no code implementations • 29 Sep 2021 • Lu Chen, Renjie Chen, Hang Guo, Yuan Luo, Quanshi Zhang, Yisen Wang
Adversarial examples have attracted significant attention over the years, yet a sufficient understanding is in lack, especially when analyzing their performances in combination with adversarial training.
no code implementations • ICLR 2022 • Mingjie Li, Yisen Wang, Xingyu Xie, Zhouchen Lin
Works have shown the strong connections between some implicit models and optimization problems.
1 code implementation • NeurIPS 2021 • Mingqing Xiao, Qingyan Meng, Zongpeng Zhang, Yisen Wang, Zhouchen Lin
In this work, we consider feedback spiking neural networks, which are more brain-like, and propose a novel training method that does not rely on the exact reverse of the forward computation.
no code implementations • 29 Sep 2021 • Zhirui Wang, Yifei Wang, Yisen Wang
Adversarial training is widely believed to be a reliable approach to improve model robustness against adversarial attack.
no code implementations • 29 Sep 2021 • Renjie Chen, Yuan Luo, Yisen Wang
After adversarial training was proposed, a series of works focus on improving the compunational efficiency of adversarial training for deep neural networks (DNNs).
no code implementations • 29 Sep 2021 • Yiwen Kou, Qinyuan Zheng, Yisen Wang
In this paper, we introduce a framework that is able to deal with robustness properties of arbitrary smoothing measures including those with bounded support set by using Wasserstein distance as well as total variation distance.
no code implementations • 29 Sep 2021 • Xiaojun Guo, Xingjun Ma, Yisen Wang
Many real-world problems can be formulated as graphs and solved by graph learning techniques.
no code implementations • ICLR 2022 • Yifei Wang, Qi Zhang, Yisen Wang, Jiansheng Yang, Zhouchen Lin
Our work suggests an alternative understanding of contrastive learning: the role of aligning positive samples is more like a surrogate task than an ultimate goal, and it is the overlapping augmented views (i. e., the chaos) that create a ladder for contrastive learning to gradually learn class-separated representations.
no code implementations • 29 Sep 2021 • Shiji Xin, Yifei Wang, Jingtong Su, Yisen Wang
Extensive experiments show that our proposed DAT can effectively remove the domain-varying features and improve OOD generalization on both correlation shift and diversity shift tasks.
1 code implementation • 1 Jul 2021 • Yifei Wang, Yisen Wang, Jiansheng Yang, Zhouchen Lin
Recently, sampling methods have been successfully applied to enhance the sample quality of Generative Adversarial Networks (GANs).
no code implementations • ICML Workshop AML 2021 • Yifei Wang, Yisen Wang, Jiansheng Yang, Zhouchen Lin
Based on these, we propose principled adversarial sampling algorithms in both supervised and unsupervised scenarios.
no code implementations • ICML Workshop AML 2021 • Nodens Koren, Xingjun Ma, Qiuhong Ke, Yisen Wang, James Bailey
Understanding the actions of both humans and artificial intelligence (AI) agents is important before modern AI systems can be fully integrated into our daily life.
1 code implementation • 10 Jun 2021 • Hongwei Wen, Jingyi Cui, Hanyuan Hang, Jiabin Liu, Yisen Wang, Zhouchen Lin
As an important branch of weakly supervised learning, partial label learning deals with data where each instance is assigned with a set of candidate labels, whereas only one of them is true.
no code implementations • 10 Jun 2021 • Jingyi Cui, Hanyuan Hang, Yisen Wang, Zhouchen Lin
In this paper, we propose a density estimation algorithm called \textit{Gradient Boosting Histogram Transform} (GBHT), where we adopt the \textit{Negative Log Likelihood} as the loss function to make the boosting procedure available for the unsupervised tasks.
no code implementations • 5 Jun 2021 • Dinghuai Zhang, Kartik Ahuja, Yilun Xu, Yisen Wang, Aaron Courville
Can models with particular structure avoid being biased towards spurious correlation in out-of-distribution (OOD) generalization?
no code implementations • 29 May 2021 • Qi Tian, Kun Kuang, Kelu Jiang, Fei Wu, Yisen Wang
Adversarial training is one of the most effective approaches to improve model robustness against adversarial examples.
no code implementations • 27 May 2021 • Xingyu Xie, Qiuhao Wang, Zenan Ling, Xia Li, Yisen Wang, Guangcan Liu, Zhouchen Lin
In this paper, we investigate an emerging question: can an implicit equilibrium model's equilibrium point be regarded as the solution of an optimization problem?
1 code implementation • 12 Mar 2021 • Jie Ren, Die Zhang, Yisen Wang, Lu Chen, Zhanpeng Zhou, Yiting Chen, Xu Cheng, Xin Wang, Meng Zhou, Jie Shi, Quanshi Zhang
This paper provides a unified view to explain different adversarial attacks and defense methods, i. e. the view of multi-order interactions between input variables of DNNs.
1 code implementation • ICLR 2021 • Yang Bai, Yuyuan Zeng, Yong Jiang, Shu-Tao Xia, Xingjun Ma, Yisen Wang
The study of adversarial examples and their activation has attracted significant attention for secure and robust learning with deep neural networks (DNNs).
1 code implementation • NeurIPS 2021 • Yifei Wang, Yisen Wang, Jiansheng Yang, Zhouchen Lin
Graph Convolutional Networks (GCNs) have attracted more and more attentions in recent years.
no code implementations • 18 Jan 2021 • Shihao Zhao, Xingjun Ma, Yisen Wang, James Bailey, Bo Li, Yu-Gang Jiang
In this paper, we focus on image classification and propose a method to visualize and understand the class-wise knowledge (patterns) learned by DNNs under three different settings including natural, backdoor and adversarial.
no code implementations • 17 Jan 2021 • Nodens Koren, Qiuhong Ke, Yisen Wang, James Bailey, Xingjun Ma
Understanding the actions of both humans and artificial intelligence (AI) agents is important before modern AI systems can be fully integrated into our daily life.
1 code implementation • ICLR 2021 • Hanxun Huang, Xingjun Ma, Sarah Monazam Erfani, James Bailey, Yisen Wang
This paper raises the question: \emph{can data be made unlearnable for deep learning models?}
no code implementations • ICLR 2021 • Xin Wang, Jie Ren, Shuyun Lin, Xiangming Zhu, Yisen Wang, Quanshi Zhang
We discover and prove the negative correlation between the adversarial transferability and the interaction inside adversarial perturbations.
no code implementations • 1 Jan 2021 • Yifei Wang, Yisen Wang, Jiansheng Yang, Zhouchen Lin
Recently, sampling methods have been successfully applied to enhance the sample quality of Generative Adversarial Networks (GANs).
no code implementations • 1 Jan 2021 • Qi Tian, Kun Kuang, Fei Wu, Yisen Wang
Adversarial training is one of the most effective approaches to improve model robustness against adversarial examples.
1 code implementation • 8 Oct 2020 • Xin Wang, Jie Ren, Shuyun Lin, Xiangming Zhu, Yisen Wang, Quanshi Zhang
We discover and prove the negative correlation between the adversarial transferability and the interaction inside adversarial perturbations.
1 code implementation • ECCV 2020 • Yang Bai, Yuyuan Zeng, Yong Jiang, Yisen Wang, Shu-Tao Xia, Weiwei Guo
Deep neural networks (DNNs) have demonstrated excellent performance on various tasks, however they are under the risk of adversarial examples that can be easily generated when the target model is accessible to an attacker (white-box setting).
no code implementations • 1 Jul 2020 • Dongxian Wu, Yisen Wang, Zhuobin Zheng, Shu-Tao Xia
Deep neural networks (DNNs) exhibit great success on many tasks with the help of large-scale well annotated datasets.
4 code implementations • ICML 2020 • Xingjun Ma, Hanxun Huang, Yisen Wang, Simone Romano, Sarah Erfani, James Bailey
However, in practice, simply being robust is not sufficient for a loss function to train accurate DNNs.
Ranked #32 on
Image Classification
on mini WebVision 1.0
(ImageNet Top-1 Accuracy metric)
2 code implementations • ICLR 2020 • Yisen Wang, Difan Zou, Jin-Feng Yi, James Bailey, Xingjun Ma, Quanquan Gu
In this paper, we investigate the distinctive influence of misclassified and correctly classified examples on the final robustness of adversarial training.
3 code implementations • NeurIPS 2020 • Dongxian Wu, Shu-Tao Xia, Yisen Wang
The study on improving the robustness of deep neural networks against adversarial examples grows rapidly in recent years.
1 code implementation • CVPR 2020 • Ranjie Duan, Xingjun Ma, Yisen Wang, James Bailey, A. K. Qin, Yun Yang
Deep neural networks (DNNs) are known to be vulnerable to adversarial examples.
4 code implementations • ICLR 2020 • Dongxian Wu, Yisen Wang, Shu-Tao Xia, James Bailey, Xingjun Ma
We find that using more gradients from the skip connections rather than the residual modules according to a decay factor, allows one to craft adversarial examples with high transferability.
no code implementations • IJCNLP 2019 • Min Zeng, Yisen Wang, Yuan Luo
Based on which, we further find that there is redundancy among the dimensions of latent variable, and the lengths and sentence patterns of the responses can be strongly correlated to each dimension of the latent variable.
4 code implementations • ICCV 2019 • Yisen Wang, Xingjun Ma, Zaiyi Chen, Yuan Luo, Jin-Feng Yi, James Bailey
In this paper, we show that DNN learning with Cross Entropy (CE) exhibits overfitting to noisy labels on some classes ("easy" classes), but more surprisingly, it also suffers from significant under learning on some other classes ("hard" classes).
Ranked #43 on
Image Classification
on Clothing1M
no code implementations • 24 Jul 2019 • Xingjun Ma, Yuhao Niu, Lin Gu, Yisen Wang, Yitian Zhao, James Bailey, Feng Lu
This raises safety concerns about the deployment of these systems in clinical settings.
no code implementations • 10 Jun 2019 • Dong-Dong Chen, Yisen Wang, Jin-Feng Yi, Zaiyi Chen, Zhi-Hua Zhou
Unsupervised domain adaptation aims to transfer the classifier learned from the source domain to the target domain in an unsupervised manner.
no code implementations • 22 Jul 2018 • Yisen Wang, Bo Dai, Lingkai Kong, Sarah Monazam Erfani, James Bailey, Hongyuan Zha
Learning nonlinear dynamics from diffusion data is a challenging problem since the individuals observed may be different at different time points, generally following an aggregate behaviour.
2 code implementations • ICML 2018 • Xingjun Ma, Yisen Wang, Michael E. Houle, Shuo Zhou, Sarah M. Erfani, Shu-Tao Xia, Sudanthi Wijewickrema, James Bailey
Datasets with significant proportions of noisy (incorrect) class labels present challenges for training accurate Deep Neural Networks (DNNs).
Ranked #41 on
Image Classification
on mini WebVision 1.0
1 code implementation • CVPR 2018 • Weiyang Liu, Zhen Liu, Zhiding Yu, Bo Dai, Rongmei Lin, Yisen Wang, James M. Rehg, Le Song
Inner product-based convolution has been a central component of convolutional neural networks (CNNs) and the key to learning visual representations.
1 code implementation • CVPR 2018 • Yisen Wang, Weiyang Liu, Xingjun Ma, James Bailey, Hongyuan Zha, Le Song, Shu-Tao Xia
We refer to this more complex scenario as the \textbf{open-set noisy label} problem and show that it is nontrivial in order to make accurate predictions.
1 code implementation • ICLR 2018 • Xingjun Ma, Bo Li, Yisen Wang, Sarah M. Erfani, Sudanthi Wijewickrema, Grant Schoenebeck, Dawn Song, Michael E. Houle, James Bailey
Deep Neural Networks (DNNs) have recently been shown to be vulnerable against adversarial examples, which are carefully crafted instances that can mislead DNNs to make errors during prediction.
no code implementations • 24 Feb 2017 • Yisen Wang, Xuejiao Deng, Songbai Pu, Zhiheng Huang
Furthermore, we introduce a CTC-based system combination, which is different from the conventional frame-wise senone-based one.
Automatic Speech Recognition
Automatic Speech Recognition (ASR)
+1
no code implementations • 25 Nov 2015 • Yisen Wang, Chaobing Song, Shu-Tao Xia
In this paper, a Tsallis Entropy Criterion (TEC) algorithm is proposed to unify Shannon entropy, Gain Ratio and Gini index, which generalizes the split criteria of decision trees.