Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2402.16388 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910836131364864 |
|---|---|
| author | Hennhöfer, Oliver Preisach, Christine |
| author_facet | Hennhöfer, Oliver Preisach, Christine |
| contents | The requirement of uncertainty quantification for anomaly detection systems has become increasingly important. In this context, effectively controlling Type I error rates ($α$) without compromising the statistical power ($1-β$) of these systems can build trust and reduce costs related to false discoveries. The field of conformal anomaly detection emerges as a promising approach for providing respective statistical guarantees by model calibration. However, the dependency on calibration data poses practical limitations - especially within low-data regimes. In this work, we formally define and evaluate leave-one-out-, bootstrap-, and cross-conformal methods for anomaly detection, incrementing on methods from the field of conformal prediction. Looking beyond the classical inductive conformal anomaly detection, we demonstrate that derived methods for calculating resampling-conformal $p$-values strike a practical compromise between statistical efficiency (full-conformal) and computational efficiency (split-conformal) as they make more efficient use of available data. We validate derived methods and quantify their improvements for a range of one-class classifiers and datasets. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_16388 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Leave-One-Out-, Bootstrap- and Cross-Conformal Anomaly Detectors Hennhöfer, Oliver Preisach, Christine Machine Learning The requirement of uncertainty quantification for anomaly detection systems has become increasingly important. In this context, effectively controlling Type I error rates ($α$) without compromising the statistical power ($1-β$) of these systems can build trust and reduce costs related to false discoveries. The field of conformal anomaly detection emerges as a promising approach for providing respective statistical guarantees by model calibration. However, the dependency on calibration data poses practical limitations - especially within low-data regimes. In this work, we formally define and evaluate leave-one-out-, bootstrap-, and cross-conformal methods for anomaly detection, incrementing on methods from the field of conformal prediction. Looking beyond the classical inductive conformal anomaly detection, we demonstrate that derived methods for calculating resampling-conformal $p$-values strike a practical compromise between statistical efficiency (full-conformal) and computational efficiency (split-conformal) as they make more efficient use of available data. We validate derived methods and quantify their improvements for a range of one-class classifiers and datasets. |
| title | Leave-One-Out-, Bootstrap- and Cross-Conformal Anomaly Detectors |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2402.16388 |