Tackling Small Eigen-Gaps: Fine-Grained Eigenvector Estimation and Inference Under Heteroscedastic Noise

DOI10.1109/TIT.2021.3111828MaRDI QIDQ5032583zbMATH OpenOpenAlexFDO

Authors Chen Cheng, Yuting Wei, Yuxin Chen

Publication date 17 February 2022

Published in IEEE Transactions on Information Theory (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2001.04620

zbMATH Keywords

eigenvector estimationinferencelow-rank matrix from noisy observations

Mathematics Subject Classification ID

Eigenvalues, singular values, and eigenvectors (15A18) Inequalities involving eigenvalues and eigenvectors (15A42) Signal theory (characterization, reconstruction, filtering, etc.) (94A12)

Abstract: This paper aims to address two fundamental challenges arising in eigenvector estimation and inference for a low-rank matrix from noisy observations: (1) how to estimate an unknown eigenvector when the eigen-gap (i.e. the spacing between the associated eigenvalue and the rest of the spectrum) is particularly small; (2) how to perform estimation and inference on linear functionals of an eigenvector -- a sort of "fine-grained" statistical reasoning that goes far beyond the usual $el{l}_2$ analysis. We investigate how to address these challenges in a setting where the unknown $nimesn$ matrix is symmetric and the additive noise matrix contains independent (and non-symmetric) entries. Based on eigen-decomposition of the asymmetric data matrix, we propose estimation and uncertainty quantification procedures for an unknown eigenvector, which further allow us to reason about linear functionals of an unknown eigenvector. The proposed procedures and the accompanying theory enjoy several important features: (1) distribution-free (i.e. prior knowledge about the noise distributions is not needed); (2) adaptive to heteroscedastic noise; (3) minimax optimal under Gaussian noise. Along the way, we establish optimal procedures to construct confidence intervals for the unknown eigenvalues. All this is guaranteed even in the presence of a small eigen-gap (up to $O(sqrtn/mathrmpolylog(n))$ times smaller than the requirement in prior theory), which goes significantly beyond what generic matrix perturbation theory has to offer.

Recommendations

• Inferring the eigenvalues of covariance matrices from limited, noisy data
• EigenPrism: inference for high dimensional signal-to-noise ratios
• Estimation of (near) low-rank matrices with noise and high-dimensional scaling
• Small-sample statistical condition estimation of large-scale generalized eigenvalue problems
• On estimators for eigenvalue/eigenvector approximations
• Eigenvalue Estimation of Parameterized Covariance Matrices of Large Dimensional Data
• Improving the Estimation of Eigenvectors Under Quadratic Loss
• Small-Sample Statistical Estimates for the Sensitivity of Eigenvalue Problems
• Nonparametric eigenvalue-regularized precision or covariance matrix estimator

Cited in

(4)

• Optimal estimation and computational limit of low-rank Gaussian mixtures
• Inference for heteroskedastic PCA with missing data
• Analysis of singular subspaces under random perturbations
• Rank and Factor Loadings Estimation in Time Series Tensor Factor Model by Pre-averaging

This page was built for publication: Tackling Small Eigen-Gaps: Fine-Grained Eigenvector Estimation and Inference Under Heteroscedastic Noise

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5032583)