An Intrinsically-Motivated Approach for Learning Highly Exploring and Fast Mixing Policies

Open in new window