Bias Detection via Maximum Subgroup Discrepancy

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Němeček, Jiří, Kozdoba, Mark, Kryvoviaz, Illia, Pevný, Tomáš, Mareček, Jakub
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913888875839488
author Němeček, Jiří
Kozdoba, Mark
Kryvoviaz, Illia
Pevný, Tomáš
Mareček, Jakub
author_facet Němeček, Jiří
Kozdoba, Mark
Kryvoviaz, Illia
Pevný, Tomáš
Mareček, Jakub
contents Bias evaluation is fundamental to trustworthy AI, both in terms of checking data quality and in terms of checking the outputs of AI systems. In testing data quality, for example, one may study the distance of a given dataset, viewed as a distribution, to a given ground-truth reference dataset. However, classical metrics, such as the Total Variation and the Wasserstein distances, are known to have high sample complexities and, therefore, may fail to provide a meaningful distinction in many practical scenarios. In this paper, we propose a new notion of distance, the Maximum Subgroup Discrepancy (MSD). In this metric, two distributions are close if, roughly, discrepancies are low for all feature subgroups. While the number of subgroups may be exponential, we show that the sample complexity is linear in the number of features, thus making it feasible for practical applications. Moreover, we provide a practical algorithm for evaluating the distance based on Mixed-integer optimization (MIO). We also note that the proposed distance is easily interpretable, thus providing clearer paths to fixing the biases once they have been identified. Finally, we describe a natural general bias detection framework, termed MSDD distances, and show that MSD aligns well with this framework. We empirically evaluate MSD by comparing it with other metrics and by demonstrating the above properties of MSD on real-world datasets.
format Preprint
id arxiv_https___arxiv_org_abs_2502_02221
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Bias Detection via Maximum Subgroup Discrepancy
Němeček, Jiří
Kozdoba, Mark
Kryvoviaz, Illia
Pevný, Tomáš
Mareček, Jakub
Machine Learning
Artificial Intelligence
Bias evaluation is fundamental to trustworthy AI, both in terms of checking data quality and in terms of checking the outputs of AI systems. In testing data quality, for example, one may study the distance of a given dataset, viewed as a distribution, to a given ground-truth reference dataset. However, classical metrics, such as the Total Variation and the Wasserstein distances, are known to have high sample complexities and, therefore, may fail to provide a meaningful distinction in many practical scenarios. In this paper, we propose a new notion of distance, the Maximum Subgroup Discrepancy (MSD). In this metric, two distributions are close if, roughly, discrepancies are low for all feature subgroups. While the number of subgroups may be exponential, we show that the sample complexity is linear in the number of features, thus making it feasible for practical applications. Moreover, we provide a practical algorithm for evaluating the distance based on Mixed-integer optimization (MIO). We also note that the proposed distance is easily interpretable, thus providing clearer paths to fixing the biases once they have been identified. Finally, we describe a natural general bias detection framework, termed MSDD distances, and show that MSD aligns well with this framework. We empirically evaluate MSD by comparing it with other metrics and by demonstrating the above properties of MSD on real-world datasets.
title Bias Detection via Maximum Subgroup Discrepancy
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2502.02221