[2505.06123] Wasserstein Distances Made Explainable: Insights Into Dataset Shifts and Transport Phenomena

Computer Science > Machine Learning arXiv:2505.06123 (cs) [Submitted on 9 May 2025 (v1), last revised 1 Mar 2026 (this version, v2)] Title:Wasserstein Distances Made Explainable: Insights Into Dataset Shifts and Transport Phenomena Authors:Philip Naumann, Jacob Kauffmann, Grégoire Montavon View a PDF of the paper titled Wasserstein Distances Made Explainable: Insights Into Dataset Shifts and Transport Phenomena, by Philip Naumann and 2 other authors View PDF HTML (experimental) Abstract:Wasserstein distances provide a powerful framework for comparing data distributions. They can be used to analyze processes over time or to detect inhomogeneities within data. However, simply calculating the Wasserstein distance or analyzing the corresponding transport plan (or coupling) may not be sufficient for understanding what factors contribute to a high or low Wasserstein distance. In this work, we propose a novel solution based on Explainable AI that allows us to efficiently and accurately attribute Wasserstein distances to various data components, including data subgroups, input features, or interpretable subspaces. Our method achieves high accuracy across diverse datasets and Wasserstein distance specifications, and its practical utility is demonstrated in three use cases. Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV) Cite as: arXiv:2505.06123 [cs.LG]   (or arXiv:2505.06123v2 [cs.LG] for this version)   https://...