Reinforcement Learning Journal, vol. 3, 2024, pp. 1377–1409.
Presented at the Reinforcement Learning Conference (RLC), Amherst Massachusetts, August 9–12, 2024.
There has been significant recent progress in the area of unsupervised skill discovery, utilizing various information-theoretic objectives as measures of diversity. Despite these advances, challenges remain: current methods require significant online interaction, fail to leverage vast amounts of available task-agnostic data and typically lack a quantitative measure of skill utility. We address these challenges by proposing a principled offline algorithm for unsupervised skill discovery that, in addition to maximizing diversity, ensures that each learned skill imitates state-only expert demonstrations to a certain degree. Our main analytical contribution is to connect Fenchel duality, reinforcement learning, and unsupervised skill discovery to maximize a mutual information objective subject to KL-divergence state occupancy constraints. Furthermore, we demonstrate the effectiveness of our method on the standard offline benchmark D4RL and on a custom offline dataset collected from a 12-DoF quadruped robot for which the policies trained in simulation transfer well to the real robotic system.
Marin Vlastelica, Jin Cheng, Georg Martius, and Pavel Kolev. "Offline Diversity Maximization under Imitation Constraints." Reinforcement Learning Journal, vol. 3, 2024, pp. 1377–1409.
BibTeX:@article{vlastelica2024offline,
title={Offline Diversity Maximization under Imitation Constraints},
author={Vlastelica, Marin and Cheng, Jin and Martius, Georg and Kolev, Pavel},
journal={Reinforcement Learning Journal},
volume={3},
pages={1377--1409},
year={2024}
}