Double Reinforcement Learning for Efficient and Robust Off-Policy Evaluation.

Nathan Kallus, Masatoshi Uehara

Browse the full ICML paper archive.