On generic chaining and the smallest singular value of random matrices with heavy tails

DOI10.1016/J.JFA.2012.01.027zbMATH Open1242.60008arXiv1108.3886OpenAlexW2018971882MaRDI QIDQ418699FDOQ418699

Authors: Shahar Mendelson, Grigoris Paouris 

Publication date: 30 May 2012

Published in: Journal of Functional Analysis (Search for Journal in Brave)

Abstract: We present a very general chaining method which allows one to control the supremum of the empirical process $su{p}_{hinH}|N^{-1}su{m}_{i=1}^N{h}^2(X_i)-E{h}^2|$ in rather general situations. We use this method to establish two main results. First, a quantitative (non asymptotic) version of the classical Bai-Yin Theorem on the singular values of a random matrix with i.i.d entries that have heavy tails, and second, a sharp estimate on the quadratic empirical process when $H=inrt,cdot:tinT$, $Tsubset{R}^n$ and $mu$ is an isotropic, unconditional, log-concave measure.

Full work available at URL: https://arxiv.org/abs/1108.3886

Recommendations

• On the singular values of random matrices
• Tail bounds via generic chaining
• Chaining, interpolation and convexity II: the contraction principle
• The probabilistic estimates on the largest and smallest \(q\)-singular values of random matrices
• The smallest singular value of heavy-tailed not necessarily i.i.d. random matrices via random rounding

zbMATH Keywords

empirical processesrandom matricesgeneric chainingBai-Yin theorem

Mathematics Subject Classification ID

Random matrices (probabilistic aspects) (60B20) Foundations of stochastic processes (60G05)

Cites Work

• Weak convergence and empirical processes. With applications to statistics
• Title not available (Why is that?)
• Title not available (Why is that?)
• How close is the sample covariance matrix to the actual covariance matrix?
• Uniform Central Limit Theorems
• Limit of the smallest eigenvalue of a large dimensional sample covariance matrix
• Global versus local asymptotic theories of finite-dimensional normed spaces
• Quantitative estimates of the convergence of the empirical covariance matrix in log-concave ensembles
• Title not available (Why is that?)
• Title not available (Why is that?)
• The Generic Chaining
• Concentration of mass on convex bodies
• The Brunn-Minkowski inequality in Gauss space
• Some limit theorems for empirical processes (with discussion)
• Random vectors in the isotropic position
• Title not available (Why is that?)
• Non-asymptotic theory of random matrices: extreme singular values
• Title not available (Why is that?)
• Regularity of Gaussian processes
• Sampling convex bodies: a random matrix approach
• Estimation of moments of sums of independent real random variables
• Reconstruction and subgaussian operators in asymptotic geometric analysis
• The Supremum of Some Canonical Processes
• Tail and moment estimates for sums of independent random variables with logarithmically concave tails
• Empirical processes with a bounded \(\psi_1\) diameter
• Sharp bounds on the rate of convergence of the empirical covariance matrix
• Chevet type inequality and norms of submatrices
• On Weak Tail Domination of Random Vectors
• Order statistics and concentration of \(l_r\) norms for log-concave vectors
• RANDOM POINTS IN ISOTROPIC UNCONDITIONAL CONVEX BODIES

Cited In (17)

• Tail bounds via generic chaining
• On the interval of fluctuation of the singular values of random matrices
• Bounds for stochastic processes on product index spaces
• On a contraction property of Bernoulli canonical processes
• Covariance estimation under missing observations and \(L_4 - L_2\) moment equivalence
• Asymptotic geometric analysis: achievements and perspective
• Chaining, interpolation and convexity II: the contraction principle
• On multiplier processes under weak moment assumptions
• Upper bounds on product and multiplier empirical processes
• Suprema of canonical Weibull processes
• Quantitative Version of a Silverstein’s Result
• Sampling discretization and related problems
• Learning without concentration
• General nonexact oracle inequalities for classes with a subexponential envelope
• The smallest singular value of heavy-tailed not necessarily i.i.d. random matrices via random rounding
• Empirical approximation of the Gaussian distribution in \(\mathbb{R}^d\)
• Phase retrieval: stability and recovery guarantees

This page was built for publication: On generic chaining and the smallest singular value of random matrices with heavy tails

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q418699)