Fetching the paper…

Model-Based Reinforcement Learning with a Generative Model is Minimax Optimal · Around