Comprehensive Validation on Reweighting Samples for Bias Mitigation via AIF360

Christina Hastings Blow, Lijun Qian, Camille Gibson, Pamela Obiomon, Xishuang Dong

Christina Hastings Blow et al., Applied Sciences, 2024 | | https://doi.org/10.3390/app14093826

PDF DOI: 10.3390/app14093826

Abstract

Fairness in Artificial Intelligence (AI) aims to identify and mitigate bias throughout the AI development lifecycle. Tackling data bias through techniques like reweighting samples proves effective for promoting fairness. This paper undertakes a systematic exploration of reweighting samples for conventional Machine-Learning models using IBM’s AI Fairness 360 (AIF360). Experimental results show that reweighting samples effectively reduces bias in traditional ML methods across multiple fairness metrics while maintaining balanced predictive accuracy.

Back to Papers