Provably Efficient Q-Learning
            with Low Switching Cost

     Yu Bai        Tengyang Xie  Nan Jiang           Yu-Xiang Wang
  Stanford University          UIUC                UC Santa Barbara
  yub@stanford.edu     {tx10, nanjiang}@illinois.edu        yuxiangw@cs.ucsb.edu

                       Abstract
     We take initial steps in studying PAC-MDP algorithms with limited adaptivity,
     that is, algorithms that change its exploration policy as infrequently as possible
     during r ...