Addressing Strategic Manipulation Disparities in Fair Classification

TL;DR

This paper investigates how fairness constraints in classification can fail to reduce disparities in strategic manipulation costs across groups and proposes a new constrained optimization framework to address this issue.

Contribution

It introduces a novel framework that explicitly reduces strategic manipulation cost disparities, improving fairness in strategic classification settings.

Findings

Standard fairness constraints do not reduce strategic cost disparities.

The proposed framework lowers manipulation costs for minority groups.

Empirical results demonstrate improved fairness on real-world datasets.

Abstract

In real-world classification settings, such as loan application evaluation or content moderation on online platforms, individuals respond to classifier predictions by strategically updating their features to increase their likelihood of receiving a particular (positive) decision (at a certain cost). Yet, when different demographic groups have different feature distributions or pay different update costs, prior work has shown that individuals from minority groups often pay a higher cost to update their features. Fair classification aims to address such classifier performance disparities by constraining the classifiers to satisfy statistical fairness properties. However, we show that standard fairness constraints do not guarantee that the constrained classifier reduces the disparity in strategic manipulation cost. To address such biases in strategic settings and provide equal opportunities for strategic manipulation, we propose a constrained optimization framework that constructs classifiers that lower the strategic manipulation cost for minority groups. We develop our framework by studying theoretical connections between group-specific strategic cost disparity and standard selection rate fairness metrics (e.g., statistical rate and true positive rate). Empirically, we show the efficacy of this approach over multiple real-world datasets.