The Reward Biased Method: An Optimism based Approach for Reinforcement Learning | IEEE Conference Publication | IEEE Xplore