Operational range bounding of spectroscopy models with anomaly detection
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866914975283412992 |
|---|---|
| author | Simões, Luís F. Casale, Pierluigi Felismino, Marília Yip, Kai Hou Waldmann, Ingo P. Tinetti, Giovanna Lueftinger, Theresa |
| author_facet | Simões, Luís F. Casale, Pierluigi Felismino, Marília Yip, Kai Hou Waldmann, Ingo P. Tinetti, Giovanna Lueftinger, Theresa |
| contents | Safe operation of machine learning models requires architectures that explicitly delimit their operational ranges. We evaluate the ability of anomaly detection algorithms to provide indicators correlated with degraded model performance. By placing acceptance thresholds over such indicators, hard boundaries are formed that define the model's coverage. As a use case, we consider the extraction of exoplanetary spectra from transit light curves, specifically within the context of ESA's upcoming Ariel mission. Isolation Forests are shown to effectively identify contexts where prediction models are likely to fail. Coverage/error trade-offs are evaluated under conditions of data and concept drift. The best performance is seen when Isolation Forests model projections of the prediction model's explainability SHAP values. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2408_02581 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Operational range bounding of spectroscopy models with anomaly detection Simões, Luís F. Casale, Pierluigi Felismino, Marília Yip, Kai Hou Waldmann, Ingo P. Tinetti, Giovanna Lueftinger, Theresa Machine Learning Instrumentation and Methods for Astrophysics I.2.6; I.5.1; I.6.4; J.2 Safe operation of machine learning models requires architectures that explicitly delimit their operational ranges. We evaluate the ability of anomaly detection algorithms to provide indicators correlated with degraded model performance. By placing acceptance thresholds over such indicators, hard boundaries are formed that define the model's coverage. As a use case, we consider the extraction of exoplanetary spectra from transit light curves, specifically within the context of ESA's upcoming Ariel mission. Isolation Forests are shown to effectively identify contexts where prediction models are likely to fail. Coverage/error trade-offs are evaluated under conditions of data and concept drift. The best performance is seen when Isolation Forests model projections of the prediction model's explainability SHAP values. |
| title | Operational range bounding of spectroscopy models with anomaly detection |
| topic | Machine Learning Instrumentation and Methods for Astrophysics I.2.6; I.5.1; I.6.4; J.2 |
| url | https://arxiv.org/abs/2408.02581 |