Fairness in multi-class multi-group classification problems via contextial coherent risk measures
2026-08-31 • Machine Learning
Machine Learning
AI summaryⓘ
The authors present a new way to build fair classifiers when sensitive information has multiple overlapping groups, like race and gender combined. Their method uses a mathematical concept called coherent risk measures to handle fairness without harming individual rights. They also develop a solution technique that works well even with large datasets or imperfect data. The authors show that their approach performs better compared to traditional methods like support-vector machines when fairness is important.
multi-class classificationsensitive attributesfairness in machine learningcoherent risk measuresoverlapping groupsoptimization algorithmsrobustnesssupport-vector machinesgroup fairnessscarce data
Authors
Darinka Dentcheva, Xiangyu Tian
Abstract
We propose a new design of fair classifiers for multi-class classification problems in the presence of vector-valued sensitive attributes. In that scenario each sensitive attribute has multiple values and forms several groups relevant to the fairness consideration. Naturally those groups are overlapping and one should also analyze the interaction of factors. Additionally, the decision makers aided by the classification should not violate individual rights at the expense of satisfying fairness metrics at the group level. We propose an approach using the theory and methods of coherent measures of risk aiming at resolving the fairness challenges. Further, we propose a specialized numerical method for solving the resulting optimization problem. The method scales well with the increase of the number of observations. Additionally, we note that the obtained classifier is robust with respect to corrupted data or to situation when data is scarce. We demonstrate the advantages of the proposed framework in comparison to the support-vector machine framework and other methods handling fairness.