Overcoming Catastrophic Interference in Online Reinforcement Learning with Dynamic Self-Organizing Maps

Open in new window