Saved in:
Bibliographic Details
Main Authors: Xu, Sascha, Vreeken, Jilles
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2604.27741
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914520627150848
author Xu, Sascha
Vreeken, Jilles
author_facet Xu, Sascha
Vreeken, Jilles
contents We study the problem of understanding where two populations differ within a feature space, which we formalize in the concept of a differential subgroup: a subset of individuals from both populations who, despite sharing similar characteristics, exhibit exceptional differences in a target outcome. Differential subgroups reveal the regions of the feature space where population-level gaps are most pronounced and can help practitioners identify the covariate combinations that are structurally responsible for these differences, e.g.~in clinical analysis, model diagnostics, or treatment-effect studies. We introduce a general optimization objective for discovering differential subgroups and establish conditions under which the resulting subgroups admit a causal interpretation of population differences. We propose DiffSub, a gradient-based approach that discovers interpretable differential subgroups in tabular data. Across synthetic benchmarks, medical case studies, model-error analyses, and treatment-effect settings, DiffSub identifies informative subgroups that reveal where population differences arise and why.
format Preprint
id arxiv_https___arxiv_org_abs_2604_27741
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Differential Subgroup Discovery: Characterizing Where Two Populations Differ, and Why
Xu, Sascha
Vreeken, Jilles
Machine Learning
We study the problem of understanding where two populations differ within a feature space, which we formalize in the concept of a differential subgroup: a subset of individuals from both populations who, despite sharing similar characteristics, exhibit exceptional differences in a target outcome. Differential subgroups reveal the regions of the feature space where population-level gaps are most pronounced and can help practitioners identify the covariate combinations that are structurally responsible for these differences, e.g.~in clinical analysis, model diagnostics, or treatment-effect studies. We introduce a general optimization objective for discovering differential subgroups and establish conditions under which the resulting subgroups admit a causal interpretation of population differences. We propose DiffSub, a gradient-based approach that discovers interpretable differential subgroups in tabular data. Across synthetic benchmarks, medical case studies, model-error analyses, and treatment-effect settings, DiffSub identifies informative subgroups that reveal where population differences arise and why.
title Differential Subgroup Discovery: Characterizing Where Two Populations Differ, and Why
topic Machine Learning
url https://arxiv.org/abs/2604.27741