GenDICE: Generalized Offline Estimation of Stationary Values
This addresses a challenge in reinforcement learning and Monte Carlo methods for applications where data collection is limited, though it appears incremental as it builds on existing ratio estimation and constraint reformulation techniques.
The paper tackles the problem of estimating quantities defined by the stationary distribution of a Markov chain using only offline data, without additional environment interaction, and demonstrates strong empirical performance on benchmark problems like off-line PageRank and off-policy policy evaluation.
An important problem that arises in reinforcement learning and Monte Carlo methods is estimating quantities defined by the stationary distribution of a Markov chain. In many real-world applications, access to the underlying transition operator is limited to a fixed set of data that has already been collected, without additional interaction with the environment being available. We show that consistent estimation remains possible in this challenging scenario, and that effective estimation can still be achieved in important applications. Our approach is based on estimating a ratio that corrects for the discrepancy between the stationary and empirical distributions, derived from fundamental properties of the stationary distribution, and exploiting constraint reformulations based on variational divergence minimization. The resulting algorithm, GenDICE, is straightforward and effective. We prove its consistency under general conditions, provide an error analysis, and demonstrate strong empirical performance on benchmark problems, including off-line PageRank and off-policy policy evaluation.