Fetching the paper…

Towards Instance-Optimal Offline Reinforcement Learning with Pessimism · Around