Stochastic Methods for AUC Optimization subject to AUC-based Fairness Constraints

TL;DR

This paper introduces a stochastic optimization approach for training fair machine learning models that optimize AUC while satisfying AUC-based fairness constraints, addressing fairness-performance trade-offs.

Contribution

It formulates AUC optimization with fairness constraints as a min-max problem and proposes a novel stochastic method using Bregman divergence for efficient solutions.

Findings

Effective in real-world datasets

Balances fairness and performance

Outperforms baseline methods

Abstract

As machine learning being used increasingly in making high-stakes decisions, an arising challenge is to avoid unfair AI systems that lead to discriminatory decisions for protected population. A direct approach for obtaining a fair predictive model is to train the model through optimizing its prediction performance subject to fairness constraints, which achieves Pareto efficiency when trading off performance against fairness. Among various fairness metrics, the ones based on the area under the ROC curve (AUC) are emerging recently because they are threshold-agnostic and effective for unbalanced data. In this work, we formulate the training problem of a fairness-aware machine learning model as an AUC optimization problem subject to a class of AUC-based fairness constraints. This problem can be reformulated as a min-max optimization problem with min-max constraints, which we solve by stochastic first-order methods based on a new Bregman divergence designed for the special structure of the problem. We numerically demonstrate the effectiveness of our approach on real-world data under different fairness metrics.