More Benefits of Being Distributional: Second-Order Bounds for Reinforcement Learning.

Kaiwen Wang, Owen Oertell, Alekh Agarwal, Nathan Kallus, Wen Sun

Browse the full ICML paper archive.