EUVIMEDEUVIMED entdecken
Crossref

Conformal prediction with conditional guarantees

Isaac Gibbs, John J Cherian, Emmanuel J Candès

Journal of the Royal Statistical Society Series B: Statistical Methodology · 15.09.2025 · 87 (4) · 1100-1126

DOI: 10.1093/jrsssb/qkaf008 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of the Royal Statistical Society Series B: Statistical Methodology, online erschienen am 06.03.2025. Verfasst von Isaac Gibbs und 2 weiteren Autor:innen. Die Arbeit wurde bisher 49-mal zitiert (Crossref) und nennt 37 Literaturangaben. Gefördert von: Office of Naval Research, National Science Foundation, Simons Foundation.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

We consider the problem of constructing distribution-free prediction sets with finite-sample conditional guarantees. Prior work has shown that it is impossible to provide exact conditional coverage universally in finite samples. Thus, most popular methods only guarantee marginal coverage over the covariates or are restricted to a limited set of conditional targets, e.g. coverage over a finite set of prespecified subgroups. This paper bridges this gap by defining a spectrum of problems that interpolate between marginal and conditional validity. We motivate these problems by reformulating conditional coverage as coverage over a class of covariate shifts. When the target class of shifts is finite-dimensional, we show how to simultaneously obtain exact finite-sample coverage over all possible shifts. For example, given a collection of subgroups, our prediction sets guarantee coverage over each group. For more flexible, infinite-dimensional classes where exact coverage is impossible, we provide a procedure for quantifying the coverage errors of our algorithm. Moreover, by tuning interpretable hyperparameters, we allow the practitioner to control the size of these errors across shifts of interest. Our methods can be incorporated into existing split conformal inference pipelines, and thus can be used to quantify the uncertainty of modern black-box algorithms without distributional assumptions.

Förderung

  • Office of Naval Research
  • National Science Foundation
  • Simons Foundation
  • ARO

Zitieren

APA

Gibbs, I., Cherian, J. J., & Candès, E. J. (2025). Conformal prediction with conditional guarantees. Journal of the Royal Statistical Society Series B: Statistical Methodology, 87(4), 1100-1126. https://doi.org/10.1093/jrsssb/qkaf008

Vancouver

Gibbs I, Cherian JJ, Candès EJ. Conformal prediction with conditional guarantees. Journal of the Royal Statistical Society Series B: Statistical Methodology. 2025;87(4):1100-1126. doi:10.1093/jrsssb/qkaf008

BibTeX
@article{Gibbs2025,
  author = {Gibbs, Isaac and Cherian, John J and Candès, Emmanuel J},
  title = {Conformal prediction with conditional guarantees},
  journal = {Journal of the Royal Statistical Society Series B: Statistical Methodology},
  year = {2025},
  volume = {87},
  number = {4},
  pages = {1100--1126},
  doi = {10.1093/jrsssb/qkaf008}
}
Kostenlos starten →