Drift Localization using Conformal Predictions
Abstract
Concept drift -- the change of the distribution over time -- poses significant challenges for learning systems and is of central interest for monitoring. Understanding drift is thus paramount, and drift localization -- determining which samples are affected by the drift -- is essential. While several approaches exist, most rely on local testing schemes, which tend to fail in high-dimensional, low-signal settings. In this work, we consider a fundamentally different approach based on conformal predictions. We discuss and show the shortcomings of common approaches and demonstrate the performance of our approach on state-of-the-art image datasets.
Cite
@article{arxiv.2602.19790,
title = {Drift Localization using Conformal Predictions},
author = {Fabian Hinder and Valerie Vaquet and Johannes Brinkrolf and Barbara Hammer},
journal= {arXiv preprint arXiv:2602.19790},
year = {2026}
}
Comments
Paper is an extended version; the original was published at the 34th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN) 2026