Statistical properties of Kernel Prinicipal Component Analysis - Archive ouverte HAL Accéder directement au contenu
Article Dans Une Revue Machine Learning Année : 2007

Statistical properties of Kernel Prinicipal Component Analysis

Résumé

We study the properties of the eigenvalues of Gram matrices in a non-asymptotic setting. Using local Rademacher averages, we provide data-dependent and tight bounds for their convergence towards eigenvalues of the corresponding kernel operator. We perform these computations in a functional analytic framework which allows to deal implicitly with reproducing kernel Hilbert spaces of infinite dimension. This can have applications to various kernel algorithms, such as Support Vector Machines (SVM). We focus on Kernel Principal Component Analysis (KPCA) and, using such techniques, we obtain sharp excess risk bounds for the reconstruction error. In these bounds, the dependence on the decay of the spectrum and on the closeness of successive eigenvalues is made explicit.
Fichier principal
Vignette du fichier
KPCA_versionlongue_1.pdf (317.85 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-00373789 , version 1 (07-04-2009)

Identifiants

Citer

Gilles Blanchard, Olivier Bousquet, Laurent Zwald. Statistical properties of Kernel Prinicipal Component Analysis. Machine Learning, 2007, 66 (2-3), pp.259-294. ⟨10.1007/s10994-006-6895-9⟩. ⟨hal-00373789⟩
337 Consultations
562 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More