The spectrum of random kernel matrices: universality results for rough and varying kernels

DOI10.1142/S2010326313500056MaRDI QIDQ2853395zbMATH OpenOpenAlexFDO

Authors Yen Do, Van Vu

Publication date 21 October 2013

Published in Random Matrices: Theory and Applications (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1206.3763

zbMATH Keywords

rough kernelsspectrumuniversalityrandom kernel matricesvarying kernels

Mathematics Subject Classification ID

Random matrices (algebraic aspects) (15B52)

Abstract: We consider random matrices whose entries are f(<Xi,Xj>) or f(||Xi-Xj||^2) for iid vectors Xi in R^p with normalized distribution. Assuming that f is sufficiently smooth and the distribution of Xi's is sufficiently nice, El Karoui [17] showed that the spectral distributions of these matrices behave as if f is linear in the Marchenko--Pastur limit. When Xi's are Gaussian vectors, variants of this phenomenon were recently proved for varying kernels, i.e. when f may depend on p, by Cheng and Singer [13]. Two results are shown in this paper: first it is shown that for a large class of distributions the regularity assumptions on f in El Karoui's results can be reduced to minimal; and secondly it is shown that the Gaussian assumptions in Cheng--Singer's result can be removed, answering a question posed in [13] about the universality of the limiting spectral distribution.

Recommendations

• The spectrum of kernel random matrices
• The spectrum of random inner-product kernel matrices
• Spectral distribution of large generalized random kernel matrices
• The spectral norm of random inner-product kernel matrices
• Spectrum of large random inner-product kernel matrices generated from l_p ellipsoids
• On spectral distribution of kernel matrices related to radial basis functions
• scientific article; zbMATH DE number 3336423
• A note on the large random inner-product kernel matrices
• Spectral properties of kernel matrices in the flat limit
• Correlation kernels for sums and products of random matrices

Cites work

• A generalization of the Lindeberg principle
• Circular law theorem for random Markov matrices
• Concentration of mass on convex bodies
• Concentration of measure and spectra of random matrices: applications to correlation matrices, elliptical distributions and beyond
• Eigenvalues of Euclidean random matrices
• Interpolating thin-shell and sharp large-deviation estimates for isotropic log-concave measures
• On information plus noise kernel random matrices
• On the Convergence of Moments in the Central Limit Theorem
• On the Marchenko-Pastur and circular laws for some classes of random matrices with dependent entries
• On the limiting empirical measure of eigenvalues of the sum of rank one matrices with log-concave distribution
• Operator norm consistent estimation of large-dimensional sparse covariance matrices
• Random matrices: universality of local eigenvalue statistics
• Random matrix approximation of spectra of integral operators
• Random points in the unit ball of \(\ell^{ n }_{ p }\)
• Spectra of Euclidean random matrices
• Spectral analysis of large dimensional random matrices
• Spectral properties of distance matrices
• Strong convergence of the empirical distribution of eigenvalues of large dimensional random matrices
• The Dirichlet Markov ensemble
• The spectrum of kernel random matrices
• The strong limits of random matrix spectra for sample matrices of independent elements

Cited in

(16)

• Spectral distribution of large generalized random kernel matrices
• On spectral distribution of kernel matrices related to radial basis functions
• The smallest eigenvalues of random kernel matrices: asymptotic results on the min kernel
• A note on the large random inner-product kernel matrices
• The spectral norm of random inner-product kernel matrices
• Learning low-dimensional nonlinear structures from high-dimensional noisy data: an integral operator approach
• Local and global expansion in random geometric graphs
• Concentration of kernel matrices with application to kernel spectral clustering
• The spectrum of kernel random matrices
• Deep learning: a statistical viewpoint
• Spectrum of large Euclidean random matrices generated from \(l_p\) ellipsoids
• Graph connection Laplacian and random matrices with random blocks
• Covariance discriminative power of kernel clustering methods
• Spectrum of large random inner-product kernel matrices generated from l_p ellipsoids
• On information plus noise kernel random matrices
• Spectral properties of kernel matrices in the flat limit

This page was built for publication: The spectrum of random kernel matrices: universality results for rough and varying kernels

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2853395)