Near-Optimal Policy Identification in Robust Constrained Markov Decision Processes via Epigraph Form Jan 1, 0001 Go to Project Site Photo by rawpixel on Unsplash Reinforcement Learning Toshinori Kitamura Ph.D. student My research interests include reinforcement learning theory and deep learning. Related A Policy Gradient Primal-Dual Algorithm for Constrained MDPs with Uniform PAC Guarantees Japanese Translation of "CMPUT 653 Theoretical Foundations of Reinforcement Learning" PyTorch-RL-IL (rlil) ShinRL Variance Weighted MDVI