Adaptive Sampling for Best Policy Identification
                  in Markov Decision Processes

                    Aymen Al Marjani 1 Alexandre Proutiere 2

              Abstract                certainty. This paper, as most related work in this field, fo-
                                  cuses on systems and control objectives that are modelled as
   We investigate the problem of best-policy identifi-
                                  a standard discounted Markov Decision Processes (MDPs)
  ...