Skip to yearly menu bar Skip to main content


Poster

Two-way Deconfounder for Off-policy Evaluation in Causal Reinforcement Learning

Shuguang Yu · Shuxing Fang · Ruixin Peng · Zhengling Qi · Fan Zhou · Chengchun Shi


Abstract:

This paper studies off-policy evaluation (OPE) in the presence of unmeasured confounders. Inspired by the two-way fixed effects regression model widely used in the panel data literature, we propose a two-way unmeasured confounding assumption to model the system dynamics in causal reinforcement learning and develop a two-way deconfounder algorithm that devises a neural tensor network to simultaneously learn both the unmeasured confounders and the system dynamics, based on which a model-based estimator can be constructed for consistent policy value estimation. We illustrate the effectiveness of the proposed estimator through theoretical results and numerical experiments.

Live content is unavailable. Log in and register to view live content