Intersectional Fairness via Mixed-Integer Optimization
This work provides a robust solution for regulated industries like finance and healthcare, addressing the need for fair and transparent AI models as mandated by frameworks such as the EU's AI Act, though it is incremental in building on existing intersectional fairness research.
The paper tackles the problem of ensuring fairness in AI models for high-risk domains by addressing bias at the intersections of protected groups, proposing a Mixed-Integer Optimization framework that trains interpretable classifiers with bounded intersectional bias and empirically improves bias detection performance.
The deployment of Artificial Intelligence in high-risk domains, such as finance and healthcare, necessitates models that are both fair and transparent. While regulatory frameworks, including the EU's AI Act, mandate bias mitigation, they are deliberately vague about the definition of bias. In line with existing research, we argue that true fairness requires addressing bias at the intersections of protected groups. We propose a unified framework that leverages Mixed-Integer Optimization (MIO) to train intersectionally fair and intrinsically interpretable classifiers. We prove the equivalence of two measures of intersectional fairness (MSD and SPSF) in detecting the most unfair subgroup and empirically demonstrate that our MIO-based algorithm improves performance in finding bias. We train high-performing, interpretable classifiers that bound intersectional bias below an acceptable threshold, offering a robust solution for regulated industries and beyond.