From Cross-Validation to SURE: Asymptotic Risk of Tuned Regularized Estimators

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Adusumilli, Karun, Kasy, Maximilian, Wilson, Ashia
Format: Preprint
Publié: 2026
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866912976501473280
author Adusumilli, Karun
Kasy, Maximilian
Wilson, Ashia
author_facet Adusumilli, Karun
Kasy, Maximilian
Wilson, Ashia
contents We derive the asymptotic risk function of regularized empirical risk minimization (ERM) estimators tuned by $n$-fold cross-validation (CV). The out-of-sample prediction loss of such estimators converges in distribution to the squared-error loss (risk function) of shrinkage estimators in the normal means model, tuned by Stein's unbiased risk estimate (SURE). This risk function provides a more fine-grained picture of predictive performance than uniform bounds on worst-case regret, which are common in learning theory: it quantifies how risk varies with the true parameter. As key intermediate steps, we show that (i) $n$-fold CV converges uniformly to SURE, and (ii) while SURE typically has multiple local minima, its global minimum is generically well separated. Well-separation ensures that uniform convergence of CV to SURE translates into convergence of the tuning parameter chosen by CV to that chosen by SURE.
format Preprint
id arxiv_https___arxiv_org_abs_2603_20388
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle From Cross-Validation to SURE: Asymptotic Risk of Tuned Regularized Estimators
Adusumilli, Karun
Kasy, Maximilian
Wilson, Ashia
Statistics Theory
Machine Learning
Econometrics
We derive the asymptotic risk function of regularized empirical risk minimization (ERM) estimators tuned by $n$-fold cross-validation (CV). The out-of-sample prediction loss of such estimators converges in distribution to the squared-error loss (risk function) of shrinkage estimators in the normal means model, tuned by Stein's unbiased risk estimate (SURE). This risk function provides a more fine-grained picture of predictive performance than uniform bounds on worst-case regret, which are common in learning theory: it quantifies how risk varies with the true parameter. As key intermediate steps, we show that (i) $n$-fold CV converges uniformly to SURE, and (ii) while SURE typically has multiple local minima, its global minimum is generically well separated. Well-separation ensures that uniform convergence of CV to SURE translates into convergence of the tuning parameter chosen by CV to that chosen by SURE.
title From Cross-Validation to SURE: Asymptotic Risk of Tuned Regularized Estimators
topic Statistics Theory
Machine Learning
Econometrics
url https://arxiv.org/abs/2603.20388