Fetching the paper…

Near-Optimal Offline Reinforcement Learning via Double Variance Reduction · Around