Cross-validation failure: small sample sizes lead to large error bars

Gaël Varoquaux 1
1 PARIETAL - Modelling brain structure, function and variability based on high-field MRI data
NEUROSPIN - Service NEUROSPIN, Inria Saclay - Ile de France
Abstract : Predictive models ground many state-of-the-art developments in statistical brain image analysis: decoding, MVPA, searchlight, or extraction of biomarkers. The principled approach to establish their validity and usefulness is cross-validation, testing prediction on unseen data. Here, I would like to raise awareness on error bars of cross-validation, which are often underestimated. Simple experiments show that sample sizes of many neuroimaging studies inherently lead to large error bars, eg ±10% for 100 samples. The standard error across folds strongly underestimates them. These large error bars compromise the reliability of conclusions drawn with predictive models, such as biomarkers or methods developments where, unlike with cognitive neuroimaging MVPA approaches, more samples cannot be acquired by repeating the experiment across many subjects. Solutions to increase sample size must be investigated, tackling possible increases in heterogeneity of the data.
Liste complète des métadonnées
Contributeur : Gaël Varoquaux <>
Soumis le : jeudi 22 juin 2017 - 13:28:59
Dernière modification le : vendredi 22 juin 2018 - 01:20:24
Document(s) archivé(s) le : mercredi 10 janvier 2018 - 14:43:37


Fichiers produits par l'(les) auteur(s)



Gaël Varoquaux. Cross-validation failure: small sample sizes lead to large error bars. NeuroImage, Elsevier, 2017, 〈10.1016/j.neuroimage.2017.06.061〉. 〈hal-01545002〉



Consultations de la notice


Téléchargements de fichiers