\textit{ScreeNOT}: exact MSE-optimal singular value thresholding in correlated noise

DOI10.1214/22-aos2232arXiv2009.12297MaRDI QIDQ6046305

Elad Romanov, Matan Gavish, David L. Donoho

Publication date: 10 May 2023

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

Full work available at URL: https://arxiv.org/abs/2009.12297

zbMATH Keywords

optimal thresholdsingular value thresholdinghigh-dimensional asymptoticslow-rank matrix denoisingscree plot

Mathematics Subject Classification ID

Factor analysis and principal components; correspondence analysis (62H25) Minimax procedures in statistical decision theory (62C20) Semidefinite programming (90C22) Convex programming (90C25)

Cites Work

• Unnamed Item
• Bi-cross-validation of the SVD and the nonnegative matrix factorization
• 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
• Central limit theorems for eigenvalues in a spiked population model
• Covariance regularization by thresholding
• Spectral analysis of large dimensional random matrices
• No eigenvalues outside the support of the limiting spectral distribution of large-dimensional sample covariance matrices
• Asymptotic performance of PCA for high-dimensional heteroscedastic data
• On the distribution of the largest eigenvalue in principal components analysis
• CLT for linear spectral statistics of large-dimensional sample covariance matrices.
• Analysis of the limiting spectral distribution of large dimensional random matrices
• Matrix estimation by universal singular value thresholding
• Eigenvalues of large sample covariance matrices of spiked population models
• Phase transition of the largest eigenvalue for nonnull complex sample covariance matrices
• The Optimal Hard Threshold for Singular Values is <inline-formula> <tex-math notation="TeX">\(4/\sqrt {3}\) </tex-math></inline-formula>
• OptShrink: An Algorithm for Improved Low-Rank Signal Matrix Denoising by Optimal, Data-Driven Singular Value Shrinkage
• Deterministic Parallel Analysis: An Improved Method for Selecting Factors and Principal Components
• Cross-Validatory Estimation of the Number of Components in Factor and Principal Components Models
• Optimal Spectral Shrinkage and PCA With Heteroscedastic Noise
• Fast computation of low rank matrix approximations
• Spectral analysis of data
• Model Averaging and Dimension Selection for the Singular Value Decomposition

This page was built for publication: \textit{ScreeNOT}: exact MSE-optimal singular value thresholding in correlated noise