Fairness seen as Global Sensitivity Analysis

Clément, Bénesse,; Gamboa, Fabrice; Loubes, Jean–Michel; Boissin, Thibaut

doi:10.48550/arxiv.2103.04613

Cited by 1 publication

(1 citation statement)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To the best of our knowledge, this is the first work to do the both. Among other related works, [4] link GSA measures such as Sobol and Cramér-von-Mises indices to different fairness metrics. While their approach relates the GSA of sensitive features on the resulting bias, we focus on applying GSA to all features to compute FIFs.…”

Section: Related Workmentioning

confidence: 99%

How Biased is Your Feature?: Computing Fairness Influence Functions with Global Sensitivity Analysis

Ghosh¹,

Basu²,

Meel³

2022

Preprint

View full text Add to dashboard Cite

Fairness in machine learning has attained significant focus due to the widespread application of machine learning in high-stake decision-making tasks. Unless regulated with a fairness objective, machine learning classifiers might demonstrate unfairness/bias towards certain demographic populations in the data. Thus, the quantification and mitigation of the bias induced by classifiers have become a central concern. In this paper, we aim to quantify the influence of different features on the bias of a classifier. To this end, we propose a framework of Fairness Influence Function (FIF), and compute it as a scaled difference of conditional variances in the classifier's prediction. We also instantiate an algorithm, FairXplainer, that uses variance decomposition among the subset of features and a local regressor to compute FIFs accurately, while also capturing the intersectional effects of the features. Our experimental analysis validates that FairXplainer captures the influences of both individual features and higher-order feature interactions, estimates the bias more accurately than existing local explanation methods, and detects the increase/decrease in bias due to affirmative/punitive actions in the classifier.Preprint. Under review.

show abstract

Section: Related Workmentioning

confidence: 99%