Statistical properties of kernel principal component analysis

DOI10.1007/s10994-006-6895-9zbMath1470.62077MaRDI QIDQ2384134

Olivier Bousquet, Laurent Zwald, Gilles Blanchard

Publication date: 20 September 2007

Published in: Machine Learning (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1007/s10994-006-6895-9

zbMATH Keywords

covariance operator; fast convergence rates; kernel principal components analysis; kernel integral operator; kernel spectrum estimation

Mathematics Subject Classification ID

62H25: Factor analysis and principal components; correspondence analysis

46E22: Hilbert spaces with reproducing kernels (= (proper) functional Hilbert spaces, including de Branges-Rovnyak and other structured spaces)

Related Items

Regularization: From Inverse Problems to Large-Scale Machine Learning, Certified dimension reduction in nonlinear Bayesian inverse problems, Learning noisy linear classifiers via adaptive and selective sampling, Geometry on probability spaces, Estimates of the approximation error using Rademacher complexity: Learning vector-valued functions, Kernel sliced inverse regression: regularization and consistency, Estimation of the number of components of nonparametric multivariate finite mixture models, Statistical analysis of Mapper for stochastic and multivariate filters, An \({\ell_p}\) theory of PCA and spectral clustering, Approximate kernel PCA: computational versus statistical trade-off, Relative perturbation bounds with applications to empirical covariance operators, High-probability bounds for the reconstruction error of PCA, Random discretization of the finite Fourier transform and related kernel random matrices, Detecting influential observations in kernel PCA, High-dimensional principal projections, Compressive statistical learning with random feature moments, Stochastic low-dimensional modelling of a random laminar wake past a circular cylinder

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• A Bennett concentration inequality and its application to suprema of empirical processes
• Random matrix approximation of spectra of integral operators
• Local Rademacher complexities and oracle inequalities in risk minimization. (2004 IMS Medallion Lecture). (With discussions and rejoinder)
• Local Rademacher complexities
• 10.1162/153244303768966085
• On the Eigenspectrum of the Gram Matrix and the Generalization Error of Kernel-PCA
• Approximation spline de l'analyse en composantes principales d'une variable aléatoire hilbertienne
• Gaussian Measures on Function Spaces
• Generalization performance of regularization networks and support vector machines via entropy numbers of compact operators
• Probability Inequalities for Sums of Bounded Random Variables
• Learning Theory
• Asymptotic Theory for Principal Component Analysis
• Convexity, Classification, and Risk Bounds
• Theory of Reproducing Kernels
• Some applications of concentration inequalities to statistics