Policy Finetuning: Bridging Sample-Efficient Offline and Online Reinforcement Learning

Tengyang Xie, Nan Jiang, Huan Wang, Caiming Xiong, Yu Bai

Keywords: reinforcement learning theory, offline RL, online RL, policy finetuning

Abstract: Recent theoretical work studies sample-efficient reinforcement learning (RL) extensively in two settings: learning interactively in the environment (online RL), or learning from an offline dataset (offline RL). However, existing algorithms and theories for learning near-optimal policies in these two settings are rather different and disconnected. Towards bridging this gap, this paper initiates the theoretical study of *policy finetuning*, that is, online RL where the learner has additional access to a "reference policy" $\mu$ close to the optimal policy $\pi_\star$ in a certain sense. We consider the policy finetuning problem in episodic Markov Decision Processes (MDPs) with $S$ states, $A$ actions, and horizon length $H$. We first design a sharp *offline reduction* algorithm---which simply executes $\mu$ and runs offline policy optimization on the collected dataset---that finds an $\varepsilon$ near-optimal policy within $\widetilde{O}(H^3SC^\star/\varepsilon^2)$ episodes, where $C^\star$ is the single-policy concentrability coefficient between $\mu$ and $\pi_\star$. This offline result is the first that matches the sample complexity lower bound in this setting, and resolves a recent open question in offline RL. We then establish an $\Omega(H^3S\min\{C^\star, A\}/\varepsilon^2)$ sample complexity lower bound for *any* policy finetuning algorithm, including those that can adaptively explore the environment. This implies that---perhaps surprisingly---the optimal policy finetuning algorithm is either offline reduction or a purely online RL algorithm that does not use $\mu$. Finally, we design a new hybrid offline/online algorithm for policy finetuning that achieves better sample complexity than both vanilla offline reduction and purely online RL algorithms, in a relaxed setting where $\mu$ only satisfies concentrability partially up to a certain time step. Overall, our results offer a quantitative understanding on the benefit of a good reference policy, and make a step towards bridging offline and online RL.

Code Of Conduct: I certify that all co-authors of this work have read and commit to adhering to the NeurIPS Statement on Ethics, Fairness, Inclusivity, and Code of Conduct.

TL;DR: We establish sharp upper and lower bounds for sample-efficient RL in policy finetuning, a new RL setting that bridges online and offline reinforcement learning.