Conformal Prediction for Long-Tailed Classification

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Ding, Tiffany, Fermanian, Jean-Baptiste, Salmon, Joseph
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866908854966550528
author Ding, Tiffany
Fermanian, Jean-Baptiste
Salmon, Joseph
author_facet Ding, Tiffany
Fermanian, Jean-Baptiste
Salmon, Joseph
contents Many real-world classification problems, such as plant identification, have extremely long-tailed class distributions. In order for prediction sets to be useful in such settings, they should (i) provide good class-conditional coverage, ensuring that rare classes are not systematically omitted from the prediction sets, and (ii) be a reasonable size, allowing users to easily verify candidate labels. Unfortunately, existing conformal prediction methods, when applied to the long-tailed setting, force practitioners to make a binary choice between small sets with poor class-conditional coverage or sets that have very good class-conditional coverage but are extremely large. We propose methods with marginal coverage guarantees that smoothly trade off set size and class-conditional coverage. First, we introduce a new conformal score function called prevalence-adjusted softmax that optimizes for macro-coverage, defined as the average class-conditional coverage across classes. Second, we propose a new procedure that interpolates between marginal and class-conditional conformal prediction by linearly interpolating their conformal score thresholds. We demonstrate our methods on Pl@ntNet-300K and iNaturalist-2018, two long-tailed image datasets with 1,081 and 8,142 classes, respectively.
format Preprint
id arxiv_https___arxiv_org_abs_2507_06867
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Conformal Prediction for Long-Tailed Classification
Ding, Tiffany
Fermanian, Jean-Baptiste
Salmon, Joseph
Machine Learning
Computer Vision and Pattern Recognition
Methodology
Many real-world classification problems, such as plant identification, have extremely long-tailed class distributions. In order for prediction sets to be useful in such settings, they should (i) provide good class-conditional coverage, ensuring that rare classes are not systematically omitted from the prediction sets, and (ii) be a reasonable size, allowing users to easily verify candidate labels. Unfortunately, existing conformal prediction methods, when applied to the long-tailed setting, force practitioners to make a binary choice between small sets with poor class-conditional coverage or sets that have very good class-conditional coverage but are extremely large. We propose methods with marginal coverage guarantees that smoothly trade off set size and class-conditional coverage. First, we introduce a new conformal score function called prevalence-adjusted softmax that optimizes for macro-coverage, defined as the average class-conditional coverage across classes. Second, we propose a new procedure that interpolates between marginal and class-conditional conformal prediction by linearly interpolating their conformal score thresholds. We demonstrate our methods on Pl@ntNet-300K and iNaturalist-2018, two long-tailed image datasets with 1,081 and 8,142 classes, respectively.
title Conformal Prediction for Long-Tailed Classification
topic Machine Learning
Computer Vision and Pattern Recognition
Methodology
url https://arxiv.org/abs/2507.06867