Using Imagery to Simplify Perceptual Abstraction in Reinforcement Learning Agents

Open in new window