Individual Fairness Revisited: Transferring Techniques from Adversarial Robustness

TL;DR

This paper proposes a novel approach to individual fairness by finding suitable metrics for models rather than assuming predefined ones, using techniques from adversarial robustness to improve fairness guarantees.

Contribution

It introduces the concept of minimal metrics for models and applies randomized smoothing to enhance fairness in complex neural networks.

Findings

Minimal metrics effectively characterize model behavior.

Randomized smoothing improves fairness with minimal utility loss.

Fairness guarantees are interpretable and meaningful.

Abstract

We turn the definition of individual fairness on its head---rather than ascertaining the fairness of a model given a predetermined metric, we find a metric for a given model that satisfies individual fairness. This can facilitate the discussion on the fairness of a model, addressing the issue that it may be difficult to specify a priori a suitable metric. Our contributions are twofold: First, we introduce the definition of a minimal metric and characterize the behavior of models in terms of minimal metrics. Second, for more complicated models, we apply the mechanism of randomized smoothing from adversarial robustness to make them individually fair under a given weighted $L^p$ metric. Our experiments show that adapting the minimal metrics of linear models to more complicated neural networks can lead to meaningful and interpretable fairness guarantees at little cost to utility.