홈 › Papers

Rethinking Adversarial Attacks in Reinforcement Learning from Policy Distribution Perspective

2025-01-07 · Tianyang Duan, Zongyuan Zhang, Zheng Lin, Yue Gao, Ling Xiong, Yong Cui, Hongbin Liang, Xianhao Chen, Heming Cui, Dong Huang

Deep Reinforcement Learning (DRL) suffers from uncertainties and inaccuracies in the observation signal in realworld applications. Adversarial attack is an effective method for evaluating the robustness of DRL agents. However, existing attack methods targeting individual sampled actions have limited impacts on the overall policy distribution, particularly in continuous action spaces. To address these limitations, we propose the Distribution-Aware Projected Gradient Descent attack (DAPGD). DAPGD uses distribution similarity as the gradient perturbation input to attack the policy network, which leverages the entire policy distribution rather than relying on individual samples. We utilize the Bhattacharyya distance in DAPGD to measure policy similarity, enabling sensitive detection of subtle but critical differences between probability distributions. Our experiment results demonstrate that DAPGD achieves SOTA results compared to the baselines in three robot navigation tasks, achieving an average 22.03% higher reward drop compared to the best baseline.

📄 PDF Abstract BibTeX arXiv:2501.03562

Code (0)

등록된 구현이 없습니다.

Tasks

Adversarial AttackDeep Reinforcement Learningreinforcement-learning Reinforcement Learning Robot Navigation

Rethinking Robustness Assessment: Adversarial Attacks on Learning-based Quadrupedal Locomotion Controllers

2024-05-21 · Fan Shi, Chong Zhang, Takahiro Miki, Joonho Lee 외

Legged locomotion has recently achieved remarkable success with the progress of machine learning techniques, especially deep reinforcement learning (RL). Controllers employing neural networks have demonstrated empirical …

Deep Reinforcement Learning Reinforcement Learning (RL)

Rethinking Adversarial Policies: A Generalized Attack Formulation and Provable Defense in RL

2023-05-27 · Xiangyu Liu, Souradip Chakraborty, Yanchao Sun, Furong Huang

Most existing works focus on direct perturbations to the victim's state/action or the underlying transition dynamics to demonstrate the vulnerability of reinforcement learning agents to adversarial attacks. However, such…

Detecting Adversarial Attacks on Neural Network Policies with Visual Foresight

2017-10-02 · Yen-Chen Lin, Ming-Yu Liu, Min Sun, Jia-Bin Huang

Deep reinforcement learning has shown promising results in learning control policies for complex sequential decision-making tasks. However, these neural network-based policies are known to be vulnerable to adversarial ex…

Autonomous Vehicles Decision Making Deep Reinforcement Learning reinforcement-learning+3

Vulnerability of Deep Reinforcement Learning to Policy Induction Attacks

2017-01-16 · Vahid Behzadan, Arslan Munir

Deep learning classifiers are known to be inherently vulnerable to manipulation by intentionally perturbed inputs, named adversarial examples. In this work, we establish that reinforcement learning techniques based on De…

Deep Reinforcement Learning reinforcement-learning Reinforcement Learning Reinforcement Learning (RL)

Whatever Does Not Kill Deep Reinforcement Learning, Makes It Stronger

2017-12-23 · Vahid Behzadan, Arslan Munir

Recent developments have established the vulnerability of deep Reinforcement Learning (RL) to policy manipulation attacks via adversarial perturbations. In this paper, we investigate the robustness and resilience of deep…