Adjusting regression models for conditional uncertainty calibration
Machine Learning · 2024-12 · 113 (11-12) · 8347-8370
DOI: 10.1007/s10994-024-06627-7 ↗Kurz gefasst
Fachartikel in der Zeitschrift Machine Learning, online erschienen am 11.11.2024. Verfasst von Ruijiang Gao und 3 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 34 Literaturangaben.
Abstract (Originalsprache)
Conformal Prediction methods have finite-sample distributional-free marginal coverage guarantees. However, they generally do not offer conditional coverage guarantees, which can be important for high-stakes decisions. In this paper, we propose a novel algorithm to train a regression function to improve the conditional coverage after applying the split conformal prediction procedure. We establish an upper bound for the miscoverage gap between the conditional coverage and the nominal coverage rate and propose an end-to-end algorithm to control this upper bound. We demonstrate the efficacy of our method empirically on synthetic and real-world datasets.
Zitieren
APA
Gao, R., Yin, M., Mcinerney, J., & Kallus, N. (2024). Adjusting regression models for conditional uncertainty calibration. Machine Learning, 113(11-12), 8347-8370. https://doi.org/10.1007/s10994-024-06627-7
Vancouver
Gao R, Yin M, Mcinerney J, Kallus N. Adjusting regression models for conditional uncertainty calibration. Machine Learning. 2024;113(11-12):8347-8370. doi:10.1007/s10994-024-06627-7
BibTeX
@article{Gao2024,
author = {Gao, Ruijiang and Yin, Mingzhang and Mcinerney, James and Kallus, Nathan},
title = {Adjusting regression models for conditional uncertainty calibration},
journal = {Machine Learning},
year = {2024},
volume = {113},
number = {11-12},
pages = {8347--8370},
doi = {10.1007/s10994-024-06627-7}
}
EUVIMED