A Forget-and-Grow Strategy for Deep Reinforcement Learning Scaling in Continuous Control

Open in new window