Regret Minimization for Reinforcement Learning by
    Evaluating the Optimal Bias Function

             Zihan Zhang                  Xiangyang Ji
           Tsinghua University              Tsinghua University
       zihan-zh17@mails.tsinghua.edu.cn           xyji@tsinghua.edu.cn

                         Abstract
       We present an algorithm based on the Optimism in the Face of Uncertainty (OFU)
       principle which is able to learn Reinforcement Learning (RL) modeled by Markov
       deci ...