Transductive versions of the LASSO and the Dantzig Selector

Abstract : Transductive methods are useful in prediction problems when the training dataset is composed of a large number of unlabeled observations and a smaller number of labeled observations. In this paper, we propose an approach for developing transductive prediction procedures that are able to take advantage of the sparsity in the high dimensional linear regression. More precisely, we define transductive versions of the LASSO and the Dantzig Selector . These procedures combine labeled and unlabeled observations of the training dataset to produce a prediction for the unlabeled observations. We propose an experimental study of the transductive estimators, that shows that they improve the LASSO and Dantzig Selector in many situations, and particularly in high dimensional problems when the predictors are correlated. We then provide non-asymptotic theoretical guarantees for these estimation methods. Interestingly, our theoretical results show that the Transductive LASSO and Dantzig Selector satisfy sparsity inequalities under weaker assumptions than those required for the "original" LASSO.
Type de document :
Pré-publication, Document de travail
Liste complète des métadonnées

Littérature citée [43 références]  Voir  Masquer  Télécharger
Contributeur : Pierre Alquier <>
Soumis le : mardi 4 mai 2010 - 15:54:39
Dernière modification le : jeudi 10 mai 2018 - 01:33:37
Document(s) archivé(s) le : jeudi 23 septembre 2010 - 18:22:31


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


  • HAL Id : hal-00390851, version 3
  • ARXIV : 1005.0829



Pierre Alquier, Mohamed Hebiri. Transductive versions of the LASSO and the Dantzig Selector. 2009. 〈hal-00390851v3〉



Consultations de la notice


Téléchargements de fichiers