AI RESEARCH
Near-Optimal Primal-Dual Algorithm for Learning Linear Mixture CMDPs with Adversarial Rewards
arXiv CS.LG
•
ArXi:2603.27884v1 Announce Type: new We study safe reinforcement learning in finite-horizon linear mixture constrained Marko decision processes (CMDPs) with adversarial rewards under full-information feedback and an unknown transition kernel. We propose a primal-dual policy optimization algorithm that achieves regret and constraint violation bounds of $\widetilde{O}(\sqrt{d^2 H^3 K})$ under mild conditions, where $d$ is the feature dimension, $H$ is the horizon, and $K$ is the number of episodes.