Hanson-Wright inequality and sub-Gaussian concentration

DOI10.1214/ECP.V18-2865zbMATH Open1329.60056arXiv1306.2872MaRDI QIDQ743056FDOQ743056

Authors: Mark Rudelson, Roman Vershynin 

Publication date: 22 September 2014

Published in: Electronic Communications in Probability (Search for Journal in Brave)

Abstract: In this expository note, we give a modern proof of Hanson-Wright inequality for quadratic forms in sub-gaussian random variables. We deduce a useful concentration inequality for sub-gaussian random vectors. Two examples are given to illustrate these results: a concentration of distances between random vectors and subspaces, and a bound on the norms of products of random and deterministic matrices.

Full work available at URL: https://arxiv.org/abs/1306.2872

Recommendations

• Sparse Hanson-Wright inequalities for subgaussian quadratic forms
• Hanson-Wright inequality in Banach spaces
• A note on the Hanson-Wright inequality for random vectors with dependencies
• Concentration inequalities for polynomials in \(\alpha\)-sub-exponential random variables
• Uniform Hanson-Wright type concentration inequalities for unbounded entries via the entropy method

zbMATH Keywords

concentration inequalitiesrandom matricesquadratic formsHanson-Wright inequalitysub-Gaussian random variables

Mathematics Subject Classification ID

Large deviations (60F10) Random matrices (probabilistic aspects) (60B20)

Cited In (only showing first 100 items - show all)

• Strong selection consistency of Bayesian vector autoregressive models based on a pseudo-likelihood approach
• Sparse principal component analysis for high‐dimensional stationary time series
• Clustering multivariate count data via Dirichlet-multinomial network fusion
• On binary embedding using circulant matrices
• On Dvoretzky's theorem for subspaces of \(L_p\)
• Convergence of covariance and spectral density estimates for high-dimensional locally stationary processes
• Lifting for blind deconvolution in random mask imaging: identifiability and convex relaxation
• On the volume of non-central sections of a cube
• Lasso guarantees for \(\beta \)-mixing heavy-tailed time series
• Mass equidistribution for random polynomials
• A bootstrap method for error estimation in randomized matrix multiplication
• Posterior graph selection and estimation consistency for high-dimensional Bayesian DAG models
• On the asymptotic variance of the debiased Lasso
• Comparison and anti-concentration bounds for maxima of Gaussian random vectors
• Implicit regularization in nonconvex statistical estimation: gradient descent converges linearly for phase retrieval, matrix completion, and blind deconvolution
• The likelihood ratio test in high-dimensional logistic regression is asymptotically a rescaled Chi-square
• Asymptotics for high dimensional regression \(M\)-estimates: fixed design results
• Quantization and compressive sensing
• Testing for high-dimensional network parameters in auto-regressive models
• Regularized estimation and testing for high-dimensional multi-block vector-autoregressive models
• A note on the Hanson-Wright inequality for random vectors with dependencies
• Optimal bounds for aggregation of affine estimators
• Consistent Bayesian sparsity selection for high-dimensional Gaussian DAG models with multiplicative and beta-mixture priors
• Kernel spectral clustering of large dimensional data
• Noise-shaping quantization methods for frame-based and compressive sampling systems
• Hanson-Wright inequality in Banach spaces
• Estimating high-dimensional covariance and precision matrices under general missing dependence
• On delocalization of eigenvectors of random non-Hermitian matrices
• Resonances for random highly oscillatory potentials
• Eigenvectors of random matrices: A survey
• Minimax rate-optimal estimation of high-dimensional covariance matrices with incomplete data
• On the consistency of inversion-free parameter estimation for Gaussian random fields
• Analysis of sparse MIMO radar
• Tyler's and Maronna's M-estimators: non-asymptotic concentration results
• Upper bound for intermediate singular values of random matrices
• A random matrix approach to neural networks
• A note on moment inequality for quadratic forms
• Finite impulse response models: a non-asymptotic analysis of the least squares estimator
• Concentration inequalities for bounded functionals via log-Sobolev-type inequalities
• Circular law for sparse random regular digraphs
• Bounds on tail probabilities for quadratic forms in dependent sub-Gaussian random variables
• Estimation of low-rank covariance function
• Performance analysis of local ensemble Kalman filter
• Delocalization of eigenvectors of random matrices with independent entries
• Regularized estimation in sparse high-dimensional time series models
• A two-stage penalized least squares method for constructing large systems of structural equations
• Gibbs measures associated to the integrals of motion of the periodic derivative nonlinear Schrödinger equation
• Mixing of Hamiltonian Monte Carlo on strongly log-concave distributions: continuous dynamics
• Fermionic eigenvector moment flow
• Super-resolution radar
• Concentration of distances in Wigner matrices
• On the complexity of the set of unconditional convex bodies
• The benefit of group sparsity in group inference with de-biased scaled group Lasso
• Local inversion-free estimation of spatial Gaussian processes
• Localising change points in piecewise polynomials of general degrees
• Local semicircle law under fourth moment condition
• The sub-Gaussian constant and concentration inequalities
• Multiple change-points detection in high dimension
• Spectral Methods for Passive Imaging: Nonasymptotic Performance and Robustness
• A local limit law for the empirical spectral distribution of the anticommutator of independent Wigner matrices
• Concentration inequalities for statistical inference
• On the tightness of Gaussian concentration for convex functions
• Bias-Adjusted Spectral Clustering in Multi-Layer Stochastic Block Models
• Random weighted projections, random quadratic forms and random eigenvectors
• Isotonic regression meets Lasso
• Heterogeneity adjustment with applications to graphical model inference
• Multichannel deconvolution with long-range dependence: a minimax study
• Optimal multiple change-point detection for high-dimensional data
• A simple tool for bounding the deviation of random matrices on geometric sets
• Polynomial threshold functions, hyperplane arrangements, and random tensors
• Logarithmic Sobolev inequalities for finite spin systems and applications
• Robust matrix completion
• Estimation of binary time-frequency masks from ambient noise
• Learning particle swarming models from data with Gaussian processes
• Model selection for high-dimensional linear regression with dependent observations
• On the asymptotic behavior of the eigenvalue distribution of block correlation matrices of high-dimensional time series
• Guaranteed Functional Tensor Singular Value Decomposition
• A robust bootstrap change point test for high-dimensional location parameter
• Modified log-Sobolev inequalities and two-level concentration
• Fixed-Domain Posterior Contraction Rates for Spatial Gaussian Process Model with Nugget
• Norms of sub-exponential random vectors
• Complex phase retrieval from subgaussian measurements
• Matrix denoising for weighted loss functions and heterogeneous signals
• IFF: A Superresolution Algorithm for Multiple Measurements
• On the asymptotic distribution of the maximum sample spectral coherence of Gaussian time series in the high dimensional regime
• Uniform Hanson-Wright type concentration inequalities for unbounded entries via the entropy method
• An autocovariance-based learning framework for high-dimensional functional time series
• Some notes on concentration for \(\alpha\)-subexponential random variables
• High-Dimensional Analysis of Double Descent for Linear Regression with Random Projections
• Deformed semicircle law and concentration of nonlinear random matrices for ultra-wide neural networks
• Spectrahedral Regression
• The Hanson-Wright inequality for random tensors
• Higher order concentration on Stiefel and Grassmann manifolds
• Subspace perspective on canonical correlation analysis: dimension reduction and minimax rates
• Strong consistency guarantees for clustering high-dimensional bipartite graphs with the spectral method
• Finite sample theory for high-dimensional functional/scalar time series with applications
• Penalized estimation of threshold auto-regressive models with many components and thresholds
• On the smoothed analysis of the smallest singular value with discrete noise
• An exponential rarefaction result for sub-Gaussian real algebraic maximal curves
• Concentration of the collision estimator

This page was built for publication: Hanson-Wright inequality and sub-Gaussian concentration

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q743056)