Mitigating Unfairness via Evolutionary Multi-objective Ensemble Learning

TL;DR

This paper introduces a multi-objective evolutionary ensemble learning framework to mitigate unfairness in machine learning by balancing accuracy and multiple fairness measures, outperforming existing methods on several datasets.

Contribution

It proposes a novel multi-objective evolutionary approach to optimize accuracy and fairness simultaneously, and constructs ensembles to better balance these metrics.

Findings

Outperforms state-of-the-art unfairness mitigation methods.

Provides better tradeoffs among accuracy and multiple fairness measures.

Ensembles improve fairness-accuracy balance.

Abstract

In the literature of mitigating unfairness in machine learning, many fairness measures are designed to evaluate predictions of learning models and also utilised to guide the training of fair models. It has been theoretically and empirically shown that there exist conflicts and inconsistencies among accuracy and multiple fairness measures. Optimising one or several fairness measures may sacrifice or deteriorate other measures. Two key questions should be considered, how to simultaneously optimise accuracy and multiple fairness measures, and how to optimise all the considered fairness measures more effectively. In this paper, we view the mitigating unfairness problem as a multi-objective learning problem considering the conflicts among fairness measures. A multi-objective evolutionary learning framework is used to simultaneously optimise several metrics (including accuracy and multiple fairness measures) of machine learning models. Then, ensembles are constructed based on the learning models in order to automatically balance different metrics. Empirical results on eight well-known datasets demonstrate that compared with the state-of-the-art approaches for mitigating unfairness, our proposed algorithm can provide decision-makers with better tradeoffs among accuracy and multiple fairness metrics. Furthermore, the high-quality models generated by the framework can be used to construct an ensemble to automatically achieve a better tradeoff among all the considered fairness metrics than other ensemble methods. Our code is publicly available at https://github.com/qingquan63/FairEMOL