Learning Optimal Individualized Decision Rules with Conditional Demographic Parity

TL;DR

This paper introduces a framework for learning optimal individualized decision rules that incorporate demographic parity and conditional demographic parity constraints to promote fairness in societal applications.

Contribution

It proposes a novel, computationally efficient method to incorporate DP and CDP constraints into IDRs, with theoretical convergence guarantees and empirical validation.

Findings

Optimal IDRs under DP and CDP can be obtained via perturbations of unconstrained solutions.

The proposed methods achieve convergence in policy value and fairness constraints.

Empirical results demonstrate effectiveness in simulations and real-world data.

Abstract

Individualized decision rules (IDRs) have become increasingly prevalent in societal applications such as personalized marketing, healthcare, and public policy design. However, a critical ethical concern arises from the potential discriminatory effects of IDRs trained on biased data. These algorithms may disproportionately harm individuals from minority subgroups defined by sensitive attributes like gender, race, or language. To address this issue, we propose a novel framework that incorporates demographic parity (DP) and conditional demographic parity (CDP) constraints into the estimation of optimal IDRs. We show that the theoretically optimal IDRs under DP and CDP constraints can be obtained by applying perturbations to the unconstrained optimal IDRs, enabling a computationally efficient solution. Theoretically, we derive convergence rates for both policy value and the fairness constraint term. The effectiveness of our methods is illustrated through comprehensive simulation studies and an empirical application to the Oregon Health Insurance Experiment.