Pessimistic Nonlinear Least-Squares Value Iteration for Offline Reinforcement Learning.

Qiwei Di, Heyang Zhao, Jiafan He, Quanquan Gu

Browse the full ICLR paper archive.