Average performance analysis of the projected gradient method for online PCA - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2018

Average performance analysis of the projected gradient method for online PCA

Résumé

This paper studies the complexity of the stochastic gradient algorithm for PCA when the data are observed in a streaming setting. We also propose an online approach for selecting the learning rate. Simulation experiments confirm the practical relevance of the plain stochastic gradient approach and that drastic improvements can be achieved by learning the learning rate.
Fichier principal
Vignette du fichier
00673262-12f2-4daa-af24-a3af51d5d081-author.pdf (433.62 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-02182816 , version 1 (13-07-2019)

Identifiants

  • HAL Id : hal-02182816 , version 1

Citer

Stéphane Chrétien, Christophe Guyeux, Zhen Ho. Average performance analysis of the projected gradient method for online PCA. Annual Conference on machine Learning, Optimization and Data science, Sep 2018, Volterra, Italy. pp.231 - 242. ⟨hal-02182816⟩
23 Consultations
34 Téléchargements

Partager

Gmail Mastodon Facebook X LinkedIn More