Policy Resilience to Environment Poisoning Attacks on Reinforcement Learning
Xu, Hang, Qu, Xinghua, Rabinovich, Zinovi
–arXiv.org Artificial Intelligence
This paper investigates policy resilience to training-environment poisoning attacks on reinforcement learning (RL) policies, with the goal of recovering the deployment performance of a poisoned RL policy. Due to the fact that the policy resilience is an add-on concern to RL algorithms, it should be resource-efficient, time-conserving, and widely applicable without compromising the performance of RL algorithms. This paper proposes such a policy-resilience mechanism based on an idea of knowledge sharing. We summarize the policy resilience as three stages: preparation, diagnosis, recovery. Specifically, we design the mechanism as a federated architecture coupled with a meta-learning manner, pursuing an efficient extraction and sharing of the environment knowledge. With the shared knowledge, a poisoned agent can quickly identify the deployment condition and accordingly recover its policy performance. We empirically evaluate the resilience mechanism for both model-based and model-free RL algorithms, showing its effectiveness and efficiency in restoring the deployment performance of a poisoned policy.
arXiv.org Artificial Intelligence
Apr-24-2023
- Country:
- Asia
- Middle East > Jordan (0.04)
- Singapore > Central Region
- Singapore (0.04)
- Europe > Russia
- Central Federal District > Moscow Oblast > Moscow (0.04)
- North America > United States
- California (0.04)
- Hawaii (0.04)
- New York (0.04)
- Asia
- Genre:
- Research Report (1.00)
- Industry:
- Education > Educational Technology
- Energy (1.00)
- Information Technology > Security & Privacy (1.00)
- Technology: