Understanding when Dynamics-Invariant Data Augmentations Benefit Model-Free Reinforcement Learning Updates

Open in new window