Temporal Abstraction in Reinforcement Learning with the Successor Representation

Open in new window