Fetching the paper…

DOPE: Doubly Optimistic and Pessimistic Exploration for Safe Reinforcement Learning · Around