Adaptive shrinkage of singular values
From MaRDI portal
Publication:294253
DOI10.1007/S11222-015-9554-9zbMATH Open1505.62207arXiv1310.6602OpenAlexW2101797518MaRDI QIDQ294253FDOQ294253
Publication date: 10 June 2016
Published in: Statistics and Computing (Search for Journal in Brave)
Abstract: To recover a low rank structure from a noisy matrix, truncated singular value decomposition has been extensively used and studied. Recent studies suggested that the signal can be better estimated by shrinking the singular values. We pursue this line of research and propose a new estimator offering a continuum of thresholding and shrinking functions. To avoid an unstable and costly cross-validation search, we propose new rules to select two thresholding and shrinking parameters from the data. In particular we propose a generalized Stein unbiased risk estimation criterion that does not require knowledge of the variance of the noise and that is computationally fast. A Monte Carlo simulation reveals that our estimator outperforms the tested methods in terms of mean squared error on both low-rank and general signal matrices across different signal to noise ratio regimes. In addition, it accurately estimates the rank of the signal when it is detectable.
Full work available at URL: https://arxiv.org/abs/1310.6602
Recommendations
- Generalized SURE for optimal shrinkage of singular values in low-rank matrix denoising
- Adaptive higher-order spectral estimators
- A singular value \(p\)-shrinkage thresholding algorithm for low rank matrix recovery
- Matrix estimation by universal singular value thresholding
- Bootstrap-based regularization for low-rank matrix estimation
denoisingrank estimationadaptive trace normsingular values shrinking and thresholdingStein's unbiased risk estimate
Computational methods for problems pertaining to statistics (62-08) Factor analysis and principal components; correspondence analysis (62H25)
Cites Work
- Spectral regularization algorithms for learning large incomplete matrices
- Bi-cross-validation of the SVD and the nonnegative matrix factorization
- Reduced rank regression via adaptive nuclear norm penalization
- The Adaptive Lasso and Its Oracle Properties
- Ideal spatial adaptation by wavelet shrinkage
- Title not available (Why is that?)
- Estimation of the mean of a multivariate normal distribution
- Smoothing noisy data with spline functions: Estimating the correct degree of smoothing by the method of generalized cross-validation
- Global Image Denoising
- Robust Statistics
- Nonlinear shrinkage estimation of large-dimensional covariance matrices
- On the distribution of the largest eigenvalue in principal components analysis
- Matrix estimation by universal singular value thresholding
- Eigenvalues of large sample covariance matrices of spiked population models
- Robust principal component analysis?
- A Singular Value Thresholding Algorithm for Matrix Completion
- Title not available (Why is that?)
- Asymptotics of sample eigenstructure for a large dimensional spiked covariance model
- The sparsity and bias of the LASSO selection in high-dimensional linear regression
- The Optimal Hard Threshold for Singular Values is <inline-formula> <tex-math notation="TeX">\(4/\sqrt {3}\) </tex-math></inline-formula>
- Equivariant and scale-free Tucker decomposition models
- Model Averaging and Dimension Selection for the Singular Value Decomposition
- Selecting the number of components in principal component analysis using cross-validation approximations
- Practical approaches to principal component analysis in the presence of missing values
- Reconstruction of a low-rank matrix in the presence of Gaussian noise
- Unbiased Risk Estimates for Singular Value Thresholding and Spectral Estimators
- Minimax risk of matrix denoising by singular value thresholding
- Smooth Blockwise Iterative Thresholding: A Smooth Fixed Point Estimator Based on the Likelihood’s Block Gradient
- Sharp Oracle Inequalities for High-Dimensional Matrix Prediction
- Optimal Shrinkage of Singular Values
- Partitioning of interaction in analysis of variance
- Regularised PCA to denoise and visualise data
Cited In (14)
- Adaptive singular value shrinkage estimate for low rank tensor denoising
- Selecting Regularization Parameters for Nuclear Norm--Type Minimization Problems
- Ridge-type linear shrinkage estimation of the mean matrix of a high-dimensional normal distribution
- An adaptive singular value shrinkage for estimation problem of low-rank matrix mean with unknown covariance matrix
- Hierarchical nuclear norm penalization for multi-view data integration
- Accelerating patch-based low-rank image restoration using kd-forest and Lanczos approximation
- Imputation of Mixed Data With Multilevel Singular Value Decomposition
- Generalized SURE for optimal shrinkage of singular values in low-rank matrix denoising
- Optimal singular value shrinkage for operator norm loss: extending to non-square matrices
- Multiple imputation for continuous variables using a Bayesian principal component analysis
- Channel estimation for finite scatterers massive multi-user MIMO system
- MIMCA: multiple imputation for categorical variables with multiple correspondence analysis
- Sparse Single Index Models for Multivariate Responses
- Smooth singular value thresholding algorithm for low-rank matrix completion problem
Uses Software
This page was built for publication: Adaptive shrinkage of singular values
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q294253)
