Data-Efficient Off-Policy Policy Evaluation for Reinforcement Learning
This addresses the need for safe and cost-effective policy evaluation in applications where deploying bad policies is risky, though it appears incremental as it builds on prior work like the doubly robust estimator.
The paper tackles the problem of evaluating reinforcement learning policies from historical data generated by different policies, presenting a new estimator that achieves orders of magnitude lower mean squared error than existing methods.
In this paper we present a new way of predicting the performance of a reinforcement learning policy given historical data that may have been generated by a different policy. The ability to evaluate a policy from historical data is important for applications where the deployment of a bad policy can be dangerous or costly. We show empirically that our algorithm produces estimates that often have orders of magnitude lower mean squared error than existing methods---it makes more efficient use of the available data. Our new estimator is based on two advances: an extension of the doubly robust estimator (Jiang and Li, 2015), and a new way to mix between model based estimates and importance sampling based estimates.