Conditional anomaly detection using soft harmonic functions: An application to clinical alerting
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866917431651336192 |
|---|---|
| author | Valko, Michal Valizadegan, Hamed Kveton, Branislav Cooper, Gregory F. Hauskrecht, Milos |
| author_facet | Valko, Michal Valizadegan, Hamed Kveton, Branislav Cooper, Gregory F. Hauskrecht, Milos |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_21956 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Conditional anomaly detection using soft harmonic functions: An application to clinical alerting Valko, Michal Valizadegan, Hamed Kveton, Branislav Cooper, Gregory F. Hauskrecht, Milos Machine Learning 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. |
| title | Conditional anomaly detection using soft harmonic functions: An application to clinical alerting |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2604.21956 |