Group Fairness with Uncertainty in Sensitive Attributes

TL;DR

This paper introduces a bootstrap-based method to ensure group fairness in predictive models despite uncertainty in sensitive attributes, applicable to classification and regression tasks with discrete or continuous attributes.

Contribution

It proposes a robust quadratic programming approach guided by Gaussian analysis to achieve fairness guarantees under uncertain sensitive attributes.

Findings

Effective in real-world classification tasks

Ensures strict fairness guarantees despite uncertainty

Applicable to both discrete and continuous sensitive attributes

Abstract

Learning a fair predictive model is crucial to mitigate biased decisions against minority groups in high-stakes applications. A common approach to learn such a model involves solving an optimization problem that maximizes the predictive power of the model under an appropriate group fairness constraint. However, in practice, sensitive attributes are often missing or noisy resulting in uncertainty. We demonstrate that solely enforcing fairness constraints on uncertain sensitive attributes can fall significantly short in achieving the level of fairness of models trained without uncertainty. To overcome this limitation, we propose a bootstrap-based algorithm that achieves the target level of fairness despite the uncertainty in sensitive attributes. The algorithm is guided by a Gaussian analysis for the independence notion of fairness where we propose a robust quadratically constrained quadratic problem to ensure a strict fairness guarantee with uncertain sensitive attributes. Our algorithm is applicable to both discrete and continuous sensitive attributes and is effective in real-world classification and regression tasks for various group fairness notions, e.g., independence and separation.