Timezone: »

 
Spotlight
State Relevance for Off-Policy Evaluation
Simon Shen · Jason Yecheng Ma · Omer Gottesman · Finale Doshi-Velez

Wed Jul 21 06:45 AM -- 06:50 AM (PDT) @ None
in Reinforcement Learning 12 »

Importance sampling-based estimators for off-policy evaluation (OPE) are valued for their simplicity, unbiasedness, and reliance on relatively few assumptions. However, the variance of these estimators is often high, especially when trajectories are of different lengths. In this work, we introduce Omitting-States-Irrelevant-to-Return Importance Sampling (OSIRIS), an estimator which reduces variance by strategically omitting likelihood ratios associated with certain states. We formalize the conditions under which OSIRIS is unbiased and has lower variance than ordinary importance sampling, and we demonstrate these properties empirically.

Keywords: Reinforcement Learning and Planning
[ Paper ]

Author Information

Simon Shen (Harvard University)
Jason Yecheng Ma (University of Pennsylvania)
Omer Gottesman (Harvard University)
Finale Doshi-Velez (Harvard University)

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

More from the Same Authors