Infinite-horizon Off-Policy Policy Evaluation with Multiple Behavior Policies
This work addresses a key challenge in reinforcement learning for scenarios with diverse data sources, offering incremental improvements in estimation accuracy.
The paper tackles off-policy policy evaluation with data from multiple behavior policies by proposing the estimated mixture policy (EMP) method to estimate stationary distribution corrections, resulting in significantly improved accuracy in experiments with continuous and discrete environments.
We consider off-policy policy evaluation when the trajectory data are generated by multiple behavior policies. Recent work has shown the key role played by the state or state-action stationary distribution corrections in the infinite horizon context for off-policy policy evaluation. We propose estimated mixture policy (EMP), a novel class of partially policy-agnostic methods to accurately estimate those quantities. With careful analysis, we show that EMP gives rise to estimates with reduced variance for estimating the state stationary distribution correction while it also offers a useful induction bias for estimating the state-action stationary distribution correction. In extensive experiments with both continuous and discrete environments, we demonstrate that our algorithm offers significantly improved accuracy compared to the state-of-the-art methods.