Joint quantile regression in vector-valued RKHSs - IMT - Institut Mines-Télécom Accéder directement au contenu
Communication Dans Un Congrès Année : 2016

Joint quantile regression in vector-valued RKHSs

Résumé

Addressing the will to give a more complete picture than an average relationship provided by standard regression, a novel framework for estimating and predicting simultaneously several conditional quantiles is introduced. The proposed methodology leverages kernel-based multi-task learning to curb the embarrassing phenomenon of quantile crossing, with a one-step estimation procedure and no post-processing. Moreover, this framework comes along with theoretical guarantees and an efficient coordinate descent learning algorithm. Numerical experiments on benchmark and real datasets highlight the enhancements of our approach regarding the prediction error, the crossing occurrences and the training time.
Fichier principal
Vignette du fichier
paper_hal.pdf (1.02 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01272327 , version 1 (10-02-2016)
hal-01272327 , version 2 (26-09-2017)

Identifiants

  • HAL Id : hal-01272327 , version 2

Citer

Maxime Sangnier, Olivier Fercoq, Florence d'Alché-Buc. Joint quantile regression in vector-valued RKHSs. Neural Information Processing Systems, Dec 2016, Barcelona, France. ⟨hal-01272327v2⟩
401 Consultations
665 Téléchargements

Partager

Gmail Facebook X LinkedIn More