Improved Regret for Efficient Online Reinforcement Learning with Linear Function Approximation.

Uri Sherman, Tomer Koren, Yishay Mansour

Browse the full ICML paper archive.