Skip to Main content Skip to Navigation
Book sections

Validation croisée

Sylvain Arlot 1, 2 
1 SELECT - Model selection in statistical learning
Inria Saclay - Ile de France, LMO - Laboratoire de Mathématiques d'Orsay
Abstract : This text is a survey on cross-validation. We define all classical cross-validation procedures, and we study their properties for two different goals: estimating the risk of a given estimator, and selecting the best estimator among a given family. For the risk estimation problem, we compute the bias (which can also be corrected) and the variance of cross-validation methods. For estimator selection, we first provide a first-order analysis (based on expectations). Then, we explain how to take into account second-order terms (from variance computations, and by taking into account the usefulness of overpenalization). This allows, in the end, to provide some guidelines for choosing the best cross-validation method for a given learning problem.
Complete list of metadata
Contributor : Sylvain Arlot Connect in order to contact the contributor
Submitted on : Wednesday, March 8, 2017 - 10:31:25 PM
Last modification on : Saturday, June 25, 2022 - 10:23:54 PM
Long-term archiving on: : Friday, June 9, 2017 - 2:31:26 PM


Files produced by the author(s)


  • HAL Id : hal-01485508, version 1
  • ARXIV : 1703.03167


Sylvain Arlot. Validation croisée. Myriam Maumy-Bertrand; Gilbert Saporta; Christine Thomas-Agnan. Apprentissage statistique et donn\'ees massives, Editions Technip, 2018, 9782710811824. ⟨hal-01485508⟩



Record views


Files downloads