Simultaneously Learning Stochastic and Adversarial
    Episodic MDPs with Known Transition

          Tiancheng Jin                  Haipeng Luo
      University of Southern California        University of Southern California
       tiancheng.jin@usc.edu               haipengl@usc.edu

                        Abstract
     This work studies the problem of learning episodic Markov Decision Processes
     with known transition and bandit feedback. We develop the first algorithm with a
     “best-o ...