Off-Policy Confidence Interval Estimation with Confounded Markov Decision Process

Chengchun Shi

doi:10.1080/01621459.2022.2110878

Off-Policy Confidence Interval Estimation with Confounded Markov Decision Process

2022-10-05

5

OA

AI

DOI：10.1080/01621459.2022.2110878

Original

Original request for help

Save

Abstract

En 中文

This article is concerned with constructing a confidence interval for a target policy's value offline based on a pre-collected observational data in infinite horizon settings. Most of the existing works assume no unmeasured variables exist that confound the observed actions. This assumption, however, is likely to be violated in real applications such as healthcare and technological industries. In this article, we show that with some auxiliary variables that mediate the effect of actions on the system dynamics, the target policy's value is identifiable in a confounded Markov decision process. Based on this result, we develop an efficient off policy value estimator that is robust to potential model misspecification and provide rigorous uncertainty quantification. Our method is justified by theoretical results, simulated and real datasets obtained from ridesharing companies. A Python implementation of the proposed procedure is available at https://github.com/Mamba413/cope. Supplementary materials for this article are available online.

Keywords:

Infinite horizons

Off-policy evaluation

Reinforcement learning

Ridesourcing platforms

Statistical inference

Unmeasured confounders

Off-Policy Confidence Interval Estimation with Confounded Markov Decision Process

Abstract

Journal

Organization