Provably Efficient Reinforcement Learning via Surprise Bound.

Hanlin Zhu, Ruosong Wang, Jason D. Lee

Browse the full AISTATS paper archive.