Mutual Information-based State-Control for Intrinsically Motivated Reinforcement Learning

Open in new window