Lifted coordinate descent for learning with trace-norm regularization - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Communication Dans Un Congrès Année : 2012

Lifted coordinate descent for learning with trace-norm regularization

Résumé

We consider the minimization of a smooth loss with trace-norm regularization, which is a natural objective in multi-class and multi-task learning. Even though the problem is convex, existing approaches rely on optimizing a non-convex variational bound, which is not guaranteed to converge, or repeatedly perform singular-value decomposition, which prevents scaling beyond moderate matrix sizes. We lift the non-smooth convex problem into an infinitely dimensional smooth problem and apply coordinate descent to solve it. We prove that our approach converges to the optimum, and is competitive or outperforms state of the art.
Fichier principal
Vignette du fichier
dhm_2012_rod_atomdescent.pdf (341.23 Ko) Télécharger le fichier
Vignette du fichier
gauge.png (37.74 Ko) Télécharger le fichier
Origine : Accord explicite pour ce dépôt
Format : Figure, Image
Loading...

Dates et versions

hal-00756802 , version 1 (23-11-2012)

Identifiants

  • HAL Id : hal-00756802 , version 1

Citer

Miro Dudik, Zaid Harchaoui, Jérôme Malick. Lifted coordinate descent for learning with trace-norm regularization. AISTATS - Proceedings of the Fifteenth International Conference on Artificial Intelligence and Statistics - 2012, Apr 2012, La Palma, Spain. pp.327-336. ⟨hal-00756802⟩
771 Consultations
530 Téléchargements

Partager

Gmail Facebook X LinkedIn More