Timely detection of concerning events is an important problem in clinical practice. In this paper, we consider the problem of conditional anomaly detection that aims to identify data instances with an unusual response, such as the omission of an important lab test. We develop a new non-parametric approach for conditional anomaly detection based on the soft harmonic solution, with which we estimate the confidence of the label to detect anomalous mislabeling. We further regularize the solution to avoid the detection of isolated examples and examples on the boundary of the distribution support. We demonstrate the efficacy of the proposed method in detecting unusual labels on a real-world electronic health record dataset and compare it to several baseline approaches.
@article{arxiv.2604.21956,
title = {Conditional anomaly detection using soft harmonic functions: An application to clinical alerting},
author = {Michal Valko and Hamed Valizadegan and Branislav Kveton and Gregory F. Cooper and Milos Hauskrecht},
journal= {arXiv preprint arXiv:2604.21956},
year = {2026}
}
Comments
ICML 2011 Workshop on Machine Learning for Global Challenges. arXiv admin note: substantial text overlap with arXiv:2604.21462. substantial text overlap with arXiv:2604.21462