An Optimal Algorithm for the Stochastic Bandits While Knowing the Near-Optimal Mean Reward | IEEE Journals & Magazine | IEEE Xplore