On the Complexity of Offline Reinforcement Learning with $Q^\star$-Approximation and Partial Coverage

TL;DR

This paper investigates the theoretical limits of offline reinforcement learning with $Q^ ext{*}$-approximation and partial coverage, introducing a new complexity framework and improving sample complexity bounds for certain algorithms.

Contribution

It introduces a general complexity framework inspired by decision-estimation coefficients, providing new insights and bounds for offline RL under partial coverage and $Q^ ext{*}$-realizability.

Findings

Established an information-theoretic lower bound for the setting.

Achieved an $oldsymbol{oldsymbol{ ext{ extonehalf}}}$-order sample complexity for soft Q-learning.

Provided the first characterization of offline learnability for low-Bellman-rank MDPs without Bellman completeness.

Abstract

We study offline reinforcement learning under $Q^\star$-approximation and partial coverage, a setting that motivates practical algorithms such as Conservative $Q$-Learning (CQL; Kumar et al., 2020) but has received limited theoretical attention. Our work is inspired by the following open question: "Are $Q^\star$-realizability and Bellman completeness sufficient for sample-efficient offline RL under partial coverage?" We answer in the negative by establishing an information-theoretic lower bound. Going substantially beyond this, we introduce a general framework that characterizes the intrinsic complexity of a given $Q^\star$ function class, inspired by model-free decision-estimation coefficients (DEC) for online RL (Foster et al., 2023b; Liu et al., 2025b). This complexity recovers and improves the quantities underlying the guarantees of Chen and Jiang (2022) and Uehara et al. (2023), and extends to broader settings. Our decision-estimation decomposition can be combined with a wide range of $Q^\star$ estimation procedures, modularizing and generalizing existing approaches. Beyond the general framework, we make further contributions: By developing a novel second-order performance difference lemma, we obtain the first $\epsilon^{-2}$ sample complexity under partial coverage for soft $Q$-learning, improving the $\epsilon^{-4}$ bound of Uehara et al. (2023). We remove Chen and Jiang's (2022) need for additional online interaction when the value gap of $Q^\star$ is unknown. We also give the first characterization of offline learnability for general low-Bellman-rank MDPs without Bellman completeness (Jiang et al., 2017; Du et al., 2021; Jin et al., 2021), a canonical setting in online RL that remains unexplored in offline RL except for special cases. Finally, we provide the first analysis for CQL under $Q^\star$-realizability and Bellman completeness beyond the tabular case.