Offline Reinforcement Learning Under Value And Density-ratio Realizability: The Power Of Gaps

Abstract

We consider a challenging theoretical problem in offline reinforcement learning (RL): obtaining sample-efficiency guarantees with a dataset lacking sufficient coverage, under only realizability-type assumptions for the function approximators. While the existing theory has addressed learning under realizability and under non-exploratory data separately, no work has been able to address both simultaneously (except for a concurrent work which we compare in detail). Under an additional gap assumption, we provide guarantees to a simple pessimistic algorithm based on a version space formed by marginalized importance sampling (MIS), and the guarantee only requires the data to cover the optimal policy and the function classes to realize the optimal value and density-ratio functions. While similar gap assumptions have been used in other areas of RL theory, our work is the first to identify the utility and the novel mechanism of gap assumptions in offline RL with weak function approximation.

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keychen2022offline

Related papers

• Offline Reinforcement Learning With Realizability And Single-policy Concentrability (2022)0.00
• Offline Reinforcement Learning: Fundamental Barriers For Value Function Approximation (2021)0.00
• Distributionally Robust Offline Reinforcement Learning With Linear Function Approximation (2022)0.00
• Sample-efficient Reinforcement Learning Is Feasible For Linearly Realizable Mdps With Limited Revisiting (2021)0.00
• Optimal Conservative Offline RL With General Function Approximation Via Augmented Lagrangian (2022)0.00
• What Can Online Reinforcement Learning With Function Approximation Benefit From General Coverage Conditions? (2023)0.00
• Distributionally Robust Online Markov Game With Linear Function Approximation (2025)0.00
• Bridging Distributionally Robust Learning And Offline RL: An Approach To Mitigate Distribution Shift And Partial Data Coverage (2023)0.00