68 citations · 501 across the 111 of their papers we have counts for
18 papers · 1 filter
Safety Correction from Baseline: Towards the Risk-aware Policy in Robotics via Dual-agent Reinforcement Learning
Linrui Zhang, Zichen Yan, Li Shen +3
Learning a risk-aware policy is essential but rather challenging in unstructured robotic tasks. Safe reinforcement learning methods open up new possibilities to tackle this problem…
Evaluating Model-free Reinforcement Learning toward Safety-critical Tasks
Linrui Zhang, Qin Zhang, Li Shen +3
Safety comes first in many real-world applications involving autonomous agents. Despite a large number of reinforcement learning (RL) methods focusing on safety-critical tasks, the…
Adversarial Auto-Augment with Label Preservation: A Representation Learning Principle Guided Approach
Kaiwen Yang, Yanchao Sun, Jiahao Su +5
Data augmentation is a critical contributing factor to the success of deep learning but heavily relies on prior domain knowledge which is not always available. Recent works on auto…
Make Sharpness-Aware Minimization Stronger: A Sparsified Perturbation Approach
Peng Mi, Li Shen, Tianhe Ren +4
Deep neural networks often suffer from poor generalization caused by complex and non-convex loss landscapes. One of the popular solutions is Sharpness-Aware Minimization (SAM), whi…
Benefits of Permutation-Equivariance in Auction Mechanisms
Tian Qin, Fengxiang He, Dingfeng Shi +2
Designing an incentive-compatible auction mechanism that maximizes the auctioneer's revenue while minimizes the bidders' ex-post regret is an important yet intricate problem in eco…
Bridged Transformer for Vision and Point Cloud 3D Object Detection
Yikai Wang, TengQi Ye, Lele Cao +4
3D object detection is a crucial research topic in computer vision, which usually uses 3D point clouds as input in conventional setups. Recently, there is a trend of leveraging mul…
Super-model ecosystem: A domain-adaptation perspective
Fengxiang He, Dacheng Tao
This paper attempts to establish the theoretical foundation for the emerging super-model paradigm via domain adaptation, where one first trains a very large-scale model, {\it i.e.}…
Shortcut Learning of Large Language Models in Natural Language Understanding
Mengnan Du, Fengxiang He, Na Zou +2
Large language models (LLMs) have achieved state-of-the-art performance on a series of natural language understanding tasks. However, these LLMs might rely on dataset bias and arti…
Penalized Proximal Policy Optimization for Safe Reinforcement Learning
Linrui Zhang, Li Shen, Long Yang +4
Safe reinforcement learning aims to learn the optimal policy while satisfying safety constraints, which is essential in real-world applications. However, current algorithms still s…
A Survey on Gradient Inversion: Attacks, Defenses and Future Directions
Rui Zhang, Song Guo, Junxiao Wang +2
Recent studies have shown that the training samples can be recovered from gradients, which are called Gradient Inversion (GradInv) attacks. However, there remains a lack of extensi…
Topology-aware Generalization of Decentralized SGD
Tongtian Zhu, Fengxiang He, Lan Zhang +3
This paper studies the algorithmic stability and generalizability of decentralized stochastic gradient descent (D-SGD). We prove that the consensus model learned by D-SGD is $\math…
Understanding Deep Learning via Decision Boundary
Shiye Lei, Fengxiang He, Yancheng Yuan +1
This paper discovers that the neural network with lower decision boundary (DB) variability has better generalizability. Two new notions, algorithm DB variability and -data…
DisPFL: Towards Communication-Efficient Personalized Federated Learning via Decentralized Sparse Training
Rong Dai, Li Shen, Fengxiang He +2
Personalized federated learning is proposed to handle the data heterogeneity problem amongst clients by learning dedicated tailored local models for each user. However, existing wo…
Bridging Cross-Lingual Gaps During Leveraging the Multilingual Sequence-to-Sequence Pretraining for Text Generation and Understanding
Changtong Zan, Liang Ding, Li Shen +3
For multilingual sequence-to-sequence pretrained language models (multilingual Seq2Seq PLMs), e.g. mBART, the self-supervised pretraining task is trained on a wide range of monolin…
Robust Unlearnable Examples: Protecting Data Against Adversarial Learning
Shaopeng Fu, Fengxiang He, Yang Liu +2
The tremendous amount of accessible data in cyberspace face the risk of being unauthorized used for training deep learning models. To address this concern, methods are proposed to…
Knowledge Removal in Sampling-based Bayesian Inference
Shaopeng Fu, Fengxiang He, Dacheng Tao
The right to be forgotten has been legislated in many countries, but its enforcement in the AI industry would cause unbearable costs. When single data deletion requests come, compa…
Fine-tuning Global Model via Data-Free Knowledge Distillation for Non-IID Federated Learning
Lin Zhang, Li Shen, Liang Ding +2
Federated Learning (FL) is an emerging distributed learning paradigm under privacy constraint. Data heterogeneity is one of the main challenges in FL, which results in slow converg…
Achieving Personalized Federated Learning with Sparse Local Models
Tiansheng Huang, Shiwei Liu, Li Shen +3
Federated learning (FL) is vulnerable to heterogeneously distributed data, since a common global model in FL may not adapt to the heterogeneous data distribution of each user. To c…