Rate-optimal perturbation bounds for singular subspaces with applications to high-dimensional statistics

DOI10.1214/17-AOS1541zbMath1395.62122arXiv1605.00353OpenAlexW2964066048MaRDI QIDQ1747733

Anru Zhang, T. Tony Cai

Publication date: 27 April 2018

Published in: The Annals of Statistics (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/1605.00353

zbMATH Keywords

clusteringsingular value decompositioncanonical correlation analysisspectral methodperturbation boundhigh-dimensional statistics\(\sin\Theta\) distanceslow-rank matrix denoising

Mathematics Subject Classification ID

Factor analysis and principal components; correspondence analysis (62H25) Estimation in multivariate analysis (62H12) Classification and discrimination; cluster analysis (statistical aspects) (62H30) Random matrices (algebraic aspects) (15B52)

Related Items

Heteroskedastic PCA: algorithm, optimality, and applications, On the non-asymptotic concentration of heteroskedastic Wishart-type matrix, A Schatten-\(q\) low-rank matrix perturbation analysis via perturbation projection error bound, A Projected Subgradient Method for the Computation of Adapted Metrics for Dynamical Systems, Lower bounds for invariant statistical models with applications to principal component analysis, Singular vector and singular subspace distribution for the matrix denoising model, Asymmetry helps: eigenvalue and eigenvector analyses of asymmetrically perturbed low-rank matrices, Optimal estimation for lower bound of the packing number, Optimal Permutation Recovery in Permuted Monotone Matrix Model, Random perturbation of low rank matrices: improving classical bounds, Perturbation upper bounds for singular subspaces with a kind of heteroskedastic noise and its application in clustering, Distributed Estimation for Principal Component Analysis: An Enlarged Eigenspace Analysis, Van Trees inequality, group equivariance, and estimation of principal subspaces, Unnamed Item, Hybrid Kronecker Product Decomposition and Approximation, Covariance Estimation for Matrix-valued Data, Efficient kernel canonical correlation analysis using Nyström approximation, Euclidean Representation of Low-Rank Matrices and Its Geometric Properties, Optimal estimation and computational limit of low-rank Gaussian mixtures, Estimation of misclassification rate in the Asymptotic Rare and Weak model with sub-Gaussian noises, Entrywise limit theorems for eigenvectors of signal-plus-noise matrix models with weak signals, Latent Space Model for Higher-Order Networks and Generalized Tensor Decomposition, Lower bound estimation for a family of high-dimensional sparse covariance matrices, Rejoinder, Statistical Inference for High-Dimensional Matrix-Variate Factor Models, Normal approximation and confidence region of singular subspaces, Estimation of canonical correlation directions: from Gaussian to sub-Gaussian population, An $\ell_{\infty}$ Eigenvector Perturbation Bound and Its Application to Robust Covariance Estimation, Subspace perspective on canonical correlation analysis: dimension reduction and minimax rates, The Sup-norm Perturbation of HOSVD and Low Rank Tensor Denoising, Angle-based joint and individual variation explained, Optimal Sparse Singular Value Decomposition for High-Dimensional High-Order Data, Unnamed Item, Perturbation expansions and error bounds for the truncated singular value decomposition, Two lower bounds about singular subspaces, The two-to-infinity norm and singular subspace geometry with applications to high-dimensional statistics, An \({\ell_p}\) theory of PCA and spectral clustering, Unnamed Item

Uses Software

• PMA

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• A penalized matrix decomposition, with applications to sparse principal components and canonical correlation analysis
• Optimal rates of convergence for noisy sparse phase retrieval via thresholded Wirtinger flow
• Reconstruction of a low-rank matrix in the presence of Gaussian noise
• The singular values and vectors of low rank perturbations of large rectangular random matrices
• Minimax risk of matrix denoising by singular value thresholding
• Influential features PCA for high dimensional clustering
• Spectral clustering and the high-dimensional stochastic blockmodel
• Perturbation of the SVD in the presence of small singular values
• Minimax estimation in sparse canonical correlation analysis
• On the distribution of the left singular vectors of a random matrix and its applications
• Convex multi-task feature learning
• The largest eigenvalues of finite rank deformation of large Wigner matrices: Convergence and nonuniversality of the fluctuations
• A note on \(\sin\Theta\) theorems for singular subspace variations
• Sparse CCA: adaptive estimation and computational barriers
• Phase transitions for high dimensional clustering and related problems
• Random perturbation of low rank matrices: improving classical bounds
• Matrix estimation by universal singular value thresholding
• Consistency of spectral clustering in stochastic block models
• Optimal estimation and rank detection for sparse spiked covariance matrices
• Consistency of spectral clustering
• Sparse PCA: optimal rates and adaptive estimation
• Modern multidimensional scaling. Theory and applications.
• Exact matrix completion via convex optimization
• Randomized Dimensionality Reduction for <inline-formula> <tex-math notation="LaTeX">$k$ </tex-math></inline-formula>-Means Clustering
• The Optimal Hard Threshold for Singular Values is <inline-formula> <tex-math notation="TeX">\(4/\sqrt {3}\) </tex-math></inline-formula>
• Uniqueness of Low-Rank Matrix Completion by Rigidity Theory
• Rate Optimal Denoising of Simultaneously Sparse and Low Rank Matrices
• Interior-Point Method for Nuclear Norm Approximation with Application to System Identification
• Unbiased Risk Estimates for Singular Value Thresholding and Spectral Estimators
• Asymptotic theory for estimating the singular vectors and values of a partially-observed low rank matrix with noise
• Canonical Correlation Analysis: An Overview with Application to Learning Methods
• Large-Scale Simultaneous Testing of Cross-Covariance Matrices with Applications to PheWAS
• Sparse CCA via Precision Adjusted Iterative Thresholding
• Singular Vector Perturbation Under Gaussian Noise
• On Consistency and Sparsity for Principal Components Analysis in High Dimensions
• Recovering Low-Rank Matrices From Few Coefficients in Any Basis
• The Power of Convex Relaxation: Near-Optimal Matrix Completion
• Singular vectors under random perturbation
• The Rotation of Eigenvectors by a Perturbation. III
• Perturbation bounds in connection with singular value decomposition
• Turning Big data into tiny data: Constant-size coresets for k-means, PCA and projective clustering
• RELATIONS BETWEEN TWO SETS OF VARIATES