Optimal Transport Perturbations for Safe Reinforcement Learning with Robustness Guarantees
This work addresses the critical need for safe and robust reinforcement learning algorithms in decision-making applications, though it appears incremental as it builds on existing safe reinforcement learning techniques with a novel robustness component.
The paper tackles the problem of ensuring robustness and safety in deep reinforcement learning for real-world deployment by introducing a framework that uses optimal transport cost uncertainty sets to handle environment disturbances. The result is an approach that demonstrates robust performance and significantly improves safety in continuous control tasks compared to standard methods.
Robustness and safety are critical for the trustworthy deployment of deep reinforcement learning. Real-world decision making applications require algorithms that can guarantee robust performance and safety in the presence of general environment disturbances, while making limited assumptions on the data collection process during training. In order to accomplish this goal, we introduce a safe reinforcement learning framework that incorporates robustness through the use of an optimal transport cost uncertainty set. We provide an efficient implementation based on applying Optimal Transport Perturbations to construct worst-case virtual state transitions, which does not impact data collection during training and does not require detailed simulator access. In experiments on continuous control tasks with safety constraints, our approach demonstrates robust performance while significantly improving safety at deployment time compared to standard safe reinforcement learning.