Keywords: reinforcement learning theory, online RL, offline RL, learnability, general function approximation
TL;DR: This paper shows surprising connections between online and offline learnability, in particular, how coverage in offline RL enables exploration in online RL.
Abstract: Coverage conditions---which assert that the data logging distribution adequately covers the state space---play a fundamental role in determining the sample complexity of offline reinforcement learning. While such conditions might seem irrelevant to online reinforcement learning at first glance, we establish a new connection by showing---somewhat surprisingly---that the mere existence of a data distribution with good coverage can enable sample-efficient online RL. Concretely, we show that coverability---that is, existence of a data distribution that satisfies a ubiquitous coverage condition called concentrability---can be viewed as a structural property of the underlying MDP, and can be exploited by standard algorithms for sample-efficient exploration, even when the agent does not know said distribution. We complement this result by proving that several weaker notions of coverage, despite being sufficient for offline RL, are insufficient for online RL. We also show that existing complexity measures for online RL, including Bellman rank and Bellman-Eluder dimension, fail to optimally capture coverability, and propose a new complexity measure, the self-normalized coefficient, to provide a unification.
Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.
No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics
Submission Guidelines: Yes
Please Choose The Closest Area That Your Submission Falls Into: Theory (eg, control theory, learning theory, algorithmic game theory)