Guided Data Augmentation for Offline Reinforcement Learning and Imitation Learning

arXiv:2310.18247v2 Announce Type: replace
Abstract: In offline reinforcement learning (RL), an RL agent learns to solve a task using only a fixed dataset of previously collected data. While offline RL has been successful in learning real-world robot control policies, it typically requires large amounts of expert-quality data to learn effective policies that generalize to out-of-distribution states. Unfortunately, such data is often difficult and expensive to acquire in real-world tasks. Several recent works have leveraged data augmentation (DA) to inexpensively generate …

abstract agent arxiv augmentation control cs.lg cs.ro data dataset expert imitation learning learn offline quality quality data reinforcement reinforcement learning robot solve type world