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

arXiv:2310.17786v2 Announce Type: replace
Abstract: Recently, data augmentation (DA) has emerged as a method for leveraging domain knowledge to inexpensively generate additional data in reinforcement learning (RL) tasks, often yielding substantial improvements in data efficiency. While prior work has demonstrated the utility of incorporating augmented data directly into model-free RL updates, it is not well-understood when a particular DA strategy will improve data efficiency. In this paper, we seek to identify general aspects of DA responsible for observed learning improvements. …

abstract arxiv augmentation augmented data benefit cs.lg data domain domain knowledge dynamics efficiency free generate improvements knowledge prior reinforcement reinforcement learning tasks type understanding updates utility work