Mitigating Plasticity Loss in Continual Reinforcement Learning by Reducing Churn

Open in new window