Deep Reinforcement Learning with Explicitly Represented Knowledge and Variable State and Action Spaces

Open in new window