Optimal Transport Perturbations for Safe Reinforcement Learning with Robustness Guarantees

arXiv:2301.13375v2 Announce Type: replace
Abstract: 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 …

abstract algorithms applications arxiv assumptions collection cs.ai cs.lg data data collection decision decision making deployment environment general making performance process reinforcement reinforcement learning robust robustness safety stat.ml training transport trustworthy type world