Intrinsically Motivated Reinforcement Learning