The limits of distribution-free conditional predictive inference
Information and Inference: A Journal of the IMA · 15.06.2021 · 10 (2) · 455-482
DOI: 10.1093/imaiai/iaaa017 ↗Kurz gefasst
Fachartikel in der Zeitschrift Information and Inference: A Journal of the IMA, online erschienen am 25.08.2020. Verfasst von Rina Foygel Barber und 3 weiteren Autor:innen. Die Arbeit wurde bisher 107-mal zitiert (Crossref) und nennt 14 Literaturangaben. Gefördert von: National Science Foundation, Office of Naval Research.
Abstract (Originalsprache)
We consider the problem of distribution-free predictive inference, with the goal of producing predictive coverage guarantees that hold conditionally rather than marginally. Existing methods such as conformal prediction offer marginal coverage guarantees, where predictive coverage holds on average over all possible test points, but this is not sufficient for many practical applications where we would like to know that our predictions are valid for a given individual, not merely on average over a population. On the other hand, exact conditional inference guarantees are known to be impossible without imposing assumptions on the underlying distribution. In this work, we aim to explore the space in between these two and examine what types of relaxations of the conditional coverage property would alleviate some of the practical concerns with marginal coverage guarantees while still being possible to achieve in a distribution-free setting.
Förderung
- National Science Foundation
- Office of Naval Research
Zitieren
APA
Barber, R. F., Candès, E. J., Ramdas, A., & Tibshirani, R. J. (2021). The limits of distribution-free conditional predictive inference. Information and Inference: A Journal of the IMA, 10(2), 455-482. https://doi.org/10.1093/imaiai/iaaa017
Vancouver
Barber RF, Candès EJ, Ramdas A, Tibshirani RJ. The limits of distribution-free conditional predictive inference. Information and Inference: A Journal of the IMA. 2021;10(2):455-482. doi:10.1093/imaiai/iaaa017
BibTeX
@article{Barber2021,
author = {Barber, Rina Foygel and Candès, Emmanuel J and Ramdas, Aaditya and Tibshirani, Ryan J},
title = {The limits of distribution-free conditional predictive inference},
journal = {Information and Inference: A Journal of the IMA},
year = {2021},
volume = {10},
number = {2},
pages = {455--482},
doi = {10.1093/imaiai/iaaa017}
}
EUVIMED