Lee, K.; Kim, K.-E. Tighter Value Function Bounds for Bayesian Reinforcement Learning. AAAI 2015, 29.