State Perturbation

State perturbation research explores how systems respond to changes in their input or internal parameters, aiming to improve robustness and efficiency across various applications. Current research focuses on developing algorithms and model architectures that mitigate the negative effects of perturbations, including techniques like sharpness-aware minimization in federated learning, input transformations in reinforcement learning, and Bayesian methods for graph neural networks. This work is significant because it enhances the reliability and performance of AI systems in real-world scenarios characterized by uncertainty and noise, impacting fields such as robotics, autonomous driving, and drug discovery.

Papers

August 26, 2024

Neighborhood and Global Perturbations Supported SAM in Federated Learning: From Local Tweaks To Global Awareness
Boyuan Li, Zihao Peng, Yafei Li, Mingliang Xu, Shengbo Chen, Baofeng Ji, Cong Shen
Global Attention Sharpness Aware Minimization Faster Convergence Neighborhood Level Local Optimum State Perturbation

July 31, 2024

On the Perturbed States for Transformed Input-robust Reinforcement Learning
Tung M. Luu, Haeyong Kang, Tri Ton, Thanh Nguyen, Chang D. Yoo
Reinforcement Learning Robust Reinforcement Learning Robust Agent State Perturbation

July 15, 2024

Flow Perturbation to Accelerate Unbiased Sampling of Boltzmann distribution
Xin Peng, Ang Gao
High Dimensional Boltzmann Distribution Random Perturbation Unbiased Sampling State Perturbation

June 5, 2024

Enhancing the Resilience of Graph Neural Networks to Topological Perturbations in Sparse Graphs
Shuqi He, Jun Zhuang, Ding Wang, Luyao Peng, Jun Song
Graph Neural Network Node Classification Tiny Refinement Elicit Resilience Sparse Graph Label Propagation State Perturbation Topology Based Label Label Transition

September 22, 2023

Walking-by-Logic: Signal Temporal Logic-Guided Model Predictive Control for Bipedal Locomotion Resilient to External Perturbations
Zhaoyuan Gu, Rongming Guo, William Yates, Yipu Chen, Ye Zhao
Predictive Control Trajectory Optimization Bipedal Locomotion Robust Locomotion Push Recovery State Perturbation

September 20, 2023

Safety Guaranteed Robust Multi-Agent Reinforcement Learning with Hierarchical Control for Connected and Automated Vehicles
Zhili Zhang, H M Sabbir Ahmad, Ehsan Sabouni, Yanchao Sun, Furong Huang, Wenchao Li, Fei Miao
Multi Agent Reinforcement Learning Control Barrier Function Safe Reinforcement Learning Robust Policy Connected Autonomous Vehicle Multi Agent Proximal Policy Optimization State Perturbation

June 9, 2023

Robustness Testing for Multi-Agent Reinforcement Learning: State Perturbations on Critical Agents
Ziyuan Zhou, Guanjun Liu
Multi Agent Reinforcement Learning Adversarial Perturbation State Perturbation Critical Agent Robustness Testing

May 31, 2023

Latent Exploration for Reinforcement Learning
Alberto Silvio Chiappa, Alessandro Marin Vargas, Ann Zixiang Huang, Alexander Mathis
Reinforcement Learning Pure Exploration State Perturbation Action Noise Latent Exploration

June 19, 2022

Robust Imitation Learning against Variations in Environment Dynamics
Jongseong Chae, Seungyul Han, Whiyoung Jung, Myungsik Cho, Sungho Choi, Youngchul Sung
Native Robustness Category Wise Variation Environment Dynamic Multi Expert Robust Imitation State Perturbation

State Perturbation

Papers

Neighborhood and Global Perturbations Supported SAM in Federated Learning: From Local Tweaks To Global Awareness

On the Perturbed States for Transformed Input-robust Reinforcement Learning

Flow Perturbation to Accelerate Unbiased Sampling of Boltzmann distribution

Enhancing the Resilience of Graph Neural Networks to Topological Perturbations in Sparse Graphs

Walking-by-Logic: Signal Temporal Logic-Guided Model Predictive Control for Bipedal Locomotion Resilient to External Perturbations

Safety Guaranteed Robust Multi-Agent Reinforcement Learning with Hierarchical Control for Connected and Automated Vehicles

Robustness Testing for Multi-Agent Reinforcement Learning: State Perturbations on Critical Agents

Latent Exploration for Reinforcement Learning

Robust Imitation Learning against Variations in Environment Dynamics