Statistical Guarantees for Fairness Aware Plug-In Algorithms

TL;DR

This paper establishes the statistical consistency and finite sample guarantees of a fairness-aware plug-in algorithm for binary classification, and introduces a protocol ensuring fairness and differential privacy for sensitive features.

Contribution

It proves the statistical efficacy of a previously proposed fairness-aware plug-in classifier and proposes a new protocol combining fairness with differential privacy.

Findings

The plug-in algorithm is statistically consistent.

Finite sample guarantees are derived for the algorithm.

A protocol is proposed to ensure fairness and differential privacy.

Abstract

A plug-in algorithm to estimate Bayes Optimal Classifiers for fairness-aware binary classification has been proposed in (Menon & Williamson, 2018). However, the statistical efficacy of their approach has not been established. We prove that the plug-in algorithm is statistically consistent. We also derive finite sample guarantees associated with learning the Bayes Optimal Classifiers via the plug-in algorithm. Finally, we propose a protocol that modifies the plug-in approach, so as to simultaneously guarantee fairness and differential privacy with respect to a binary feature deemed sensitive.