Confounding-Robust Policy Improvement

TL;DR

This paper introduces a confounding-robust method for learning personalized decision policies from observational data, ensuring safety and optimality despite unobserved confounding, validated through synthetic and real case studies.

Contribution

It develops a worst-case regret minimization approach that accounts for unobserved confounding, providing theoretical guarantees and efficient algorithms for safe policy learning.

Findings

Robust policies outperform traditional methods under confounding.

The approach guarantees safety and improves decision-making in practice.

Validated on synthetic and real observational data, including a hormone therapy case study.

Abstract

We study the problem of learning personalized decision policies from observational data while accounting for possible unobserved confounding. Previous approaches, which assume unconfoundedness, i.e., that no unobserved confounders affect both the treatment assignment as well as outcome, can lead to policies that introduce harm rather than benefit when some unobserved confounding is present, as is generally the case with observational data. Instead, since policy value and regret may not be point-identifiable, we study a method that minimizes the worst-case estimated regret of a candidate policy against a baseline policy over an uncertainty set for propensity weights that controls the extent of unobserved confounding. We prove generalization guarantees that ensure our policy will be safe when applied in practice and will in fact obtain the best-possible uniform control on the range of all possible population regrets that agree with the possible extent of confounding. We develop efficient algorithmic solutions to compute this confounding-robust policy. Finally, we assess and compare our methods on synthetic and semi-synthetic data. In particular, we consider a case study on personalizing hormone replacement therapy based on observational data, where we validate our results on a randomized experiment. We demonstrate that hidden confounding can hinder existing policy learning approaches and lead to unwarranted harm, while our robust approach guarantees safety and focuses on well-evidenced improvement, a necessity for making personalized treatment policies learned from observational data reliable in practice.