Benchmarking solvers for TV-l1 least-squares and logistic regression in brain imaging

Abstract : Learning predictive models from brain imaging data, as in decoding cognitive states from fMRI (functional Magnetic Resonance Imaging), is typically an ill-posed problem as it entails estimating many more parameters than available sample points. This estimation problem thus requires regularization. Total variation regularization, combined with sparse models, has been shown to yield good predictive performance, as well as stable and interpretable maps. However, the corresponding optimization problem is very challenging: it is non-smooth, non-separable and heavily ill-conditioned. For the penalty to fully exercise its structuring effect on the maps, this optimization problem must be solved to a good tolerance resulting in a computational challenge. Here we explore a wide variety of solvers and exhibit their convergence properties on fMRI data. We introduce a variant of smooth solvers and show that it is a promising approach in these settings. Our findings show that care must be taken in solving TV-l1 estimation in brain imaging and highlight the successful strategies.
Type de document :
Communication dans un congrès
Pattern Recoginition in Neuroimaging (PRNI), Jun 2014, Tübingen, Germany. IEEE, 2014, <IEEE>
Liste complète des métadonnées

https://hal.inria.fr/hal-00991743
Contributeur : Elvis Dohmatob <>
Soumis le : jeudi 15 mai 2014 - 19:51:21
Dernière modification le : jeudi 9 février 2017 - 15:50:02
Document(s) archivé(s) le : vendredi 15 août 2014 - 11:31:09

Fichier

PRNI2014_TVl1.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-00991743, version 1

Collections

Citation

Elvis Dohmatob, Alexandre Gramfort, Bertrand Thirion, Gaël Varoquaux. Benchmarking solvers for TV-l1 least-squares and logistic regression in brain imaging. Pattern Recoginition in Neuroimaging (PRNI), Jun 2014, Tübingen, Germany. IEEE, 2014, <IEEE>. <hal-00991743>

Partager

Métriques

Consultations de
la notice

2568

Téléchargements du document

769