Auxiliary data enhanced conformal prediction
Random Matrices: Theory and Applications · 2026-04 · 15 (02)
DOI: 10.1142/s2010326326500048 ↗Kurz gefasst
Fachartikel in der Zeitschrift Random Matrices: Theory and Applications, online erschienen am 19.03.2026. Verfasst von Congbin Xu und 3 weiteren Autor:innen. Die Arbeit nennt 23 Literaturangaben. Gefördert von: National Key R&D Program of China, National Natural Science Foundation of China.
Abstract (Originalsprache)
Conformal prediction (CP) offers a principled framework for quantifying predictive uncertainty with finite-sample coverage guarantees. However, when the calibration data are limited, the coverage of CP sets can deviate substantially from the nominal target. This paper introduces Enhanced Conformal Prediction (ECP), a new framework that incorporates abundant but potentially corrupted auxiliary data to recalibrate prediction sets by left-shifting the score thresholds derived from the clean calibration set, provably reducing set size while preserving finite-sample coverage guarantees. Further theoretical analysis demonstrates that ECP achieves a higher breakdown point than existing methods. Extensive experiments confirm the robustness and efficiency of ECP across a variety of settings.
Förderung
- National Key R&D Program of China
- National Natural Science Foundation of China
Zitieren
APA
Xu, C., Yu, Y., Wang, Z., & Zou, C. (2026). Auxiliary data enhanced conformal prediction. Random Matrices: Theory and Applications, 15(02). https://doi.org/10.1142/s2010326326500048
Vancouver
Xu C, Yu Y, Wang Z, Zou C. Auxiliary data enhanced conformal prediction. Random Matrices: Theory and Applications. 2026;15(02). doi:10.1142/s2010326326500048
BibTeX
@article{Xu2026,
author = {Xu, Congbin and Yu, Yue and Wang, Zhaojun and Zou, Changliang},
title = {Auxiliary data enhanced conformal prediction},
journal = {Random Matrices: Theory and Applications},
year = {2026},
volume = {15},
number = {02},
doi = {10.1142/s2010326326500048}
}
EUVIMED