[edit]

# On the Maximal Local Disparity of Fairness-Aware Classifiers

*Proceedings of the 41st International Conference on Machine Learning*, PMLR 235:22115-22144, 2024.

#### Abstract

Fairness has become a crucial aspect in the development of trustworthy machine learning algorithms. Current fairness metrics to measure the violation of demographic parity have the following drawbacks: (i) the

*average difference*of model predictions on two groups cannot reflect their*distribution disparity*, and (ii) the*overall*calculation along all possible predictions conceals the*extreme local disparity*at or around certain predictions. In this work, we propose a novel fairness metric called**M**aximal**C**umulative ratio**D**isparity along varying**P**redictionsâ€™ neighborhood (MCDP), for measuring the maximal local disparity of the fairness-aware classifiers. To accurately and efficiently calculate the MCDP, we develop a provably exact and an approximate calculation algorithm that greatly reduces the computational complexity with low estimation error. We further propose a bi-level optimization algorithm using a differentiable approximation of the MCDP for improving the algorithmic fairness. Extensive experiments on both tabular and image datasets validate that our fair training algorithm can achieve superior fairness-accuracy trade-offs.