Timezone: »

Data-Efficient Policy Evaluation Through Behavior Policy Search
Josiah Hanna · Philip S. Thomas · Peter Stone · Scott Niekum

Mon Aug 07 05:30 PM -- 05:48 PM (PDT) @ C4.5

We consider the task of evaluating a policy for a Markov decision process (MDP). The standard unbiased technique for evaluating a policy is to deploy the policy and observe its performance. We show that the data collected from deploying a different policy, commonly called the behavior policy, can be used to produce unbiased estimates with lower mean squared error than this standard technique. We derive an analytic expression for the optimal behavior policy --- the behavior policy that minimizes the mean squared error of the resulting estimates. Because this expression depends on terms that are unknown in practice, we propose a novel policy evaluation sub-problem, behavior policy search: searching for a behavior policy that reduces mean squared error. We present a behavior policy search algorithm and empirically demonstrate its effectiveness in lowering the mean squared error of policy performance estimates.

Author Information

Josiah Hanna (University of Texas at Austin)
Philip S. Thomas (CMU)
Peter Stone (University of Texas at Austin)
Scott Niekum (University of Texas at Austin)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors