Towards Robust Deep Reinforcement Learning against Environmental State Perturbation
This addresses robustness issues for DRL agents in real-world applications, but it is incremental as it builds on existing adversarial training methods.
The paper tackles the problem of environmental state perturbations in Deep Reinforcement Learning (DRL), which are natural in embodied scenarios, and proposes a defense framework called Boosted Adversarial Training (BAT) that significantly enhances agent robustness across various situations.
Adversarial attacks and robustness in Deep Reinforcement Learning (DRL) have been widely studied in various threat models; however, few consider environmental state perturbations, which are natural in embodied scenarios. To improve the robustness of DRL agents, we formulate the problem of environmental state perturbation, introducing a preliminary non-targeted attack method as a calibration adversary, and then propose a defense framework, named Boosted Adversarial Training (BAT), which first tunes the agents via supervised learning to avoid catastrophic failure and subsequently adversarially trains the agent with reinforcement learning. Extensive experimental results substantiate the vulnerability of mainstream agents under environmental state perturbations and the effectiveness of our proposed attack. The defense results demonstrate that while existing robust reinforcement learning algorithms may not be suitable, our BAT framework can significantly enhance the robustness of agents against environmental state perturbations across various situations.