Generalized SURE for optimal shrinkage of singular values in low-rank matrix denoising
From MaRDI portal
Publication:4637067
Abstract: We consider the problem of estimating a low-rank signal matrix from noisy measurements under the assumption that the distribution of the data matrix belongs to an exponential family. In this setting, we derive generalized Stein's unbiased risk estimation (SURE) formulas that hold for any spectral estimators which shrink or threshold the singular values of the data matrix. This leads to new data-driven spectral estimators, whose optimality is discussed using tools from random matrix theory and through numerical experiments. Under the spiked population model and in the asymptotic setting where the dimensions of the data matrix are let going to infinity, some theoretical properties of our approach are compared to recent results on asymptotically optimal shrinking rules for Gaussian noise. It also leads to new procedures for singular values shrinkage in finite-dimensional matrix denoising for Gamma-distributed and Poisson-distributed measurements.
Recommendations
- Adaptive shrinkage of singular values
- Minimax risk of matrix denoising by singular value thresholding
- Matrix denoising with partial noise statistics: optimal singular value shrinkage of spiked F-matrices
- Low-rank matrix completion and denoising under Poisson noise
- On Stein's unbiased risk estimate for reduced rank estimators
Cites work
- A fast ``Monte-Carlo cross-validation procedure for large least squares problems with noisy data
- A natural identity for exponential families with applications in multiparameter estimation
- A new look at the statistical model identification
- Adaptive shrinkage of singular values
- An introduction to random matrices
- Analysis of call centre arrival data using singular value decomposition
- Eigenvalues of large sample covariance matrices of spiked population models
- Empirical Bayes on vector observations: An extension of Stein's method
- Estimation of Kullback-Leibler losses for noisy recovery problems within the exponential family
- Estimation of the mean of a multivariate normal distribution
- Exact matrix completion via convex optimization
- Generalized low rank models
- Generalized SURE for Exponential Families: Applications to Regularization
- scientific article; zbMATH DE number 44577 (Why is no real title available?)
- scientific article; zbMATH DE number 194139 (Why is no real title available?)
- Minimax risk of matrix denoising by singular value thresholding
- Multivariate empirical Bayes and estimation of covariance matrices
- Nonlinear shrinkage estimation of large-dimensional covariance matrices
- On Kullback-Leibler loss and density estimation
- On Poisson signal estimation under Kullback-Leibler discrepancy and squared risk
- On the empirical distribution of eigenvalues of large dimensional information-plus-noise-type matrices
- Optimal Shrinkage of Singular Values
- OptShrink: An Algorithm for Improved Low-Rank Signal Matrix Denoising by Optimal, Data-Driven Singular Value Shrinkage
- Poisson Matrix Recovery and Completion
- Principal component analysis.
- Reconstruction of a low-rank matrix in the presence of Gaussian noise
- Selecting the number of principal components: estimation of the true rank of a noisy matrix
- Spectral analysis of large dimensional random matrices
- Séminaire de Probabilités XXXVI
- The Estimation of Prediction Error
- The Kullback-Leibler risk of the Stein estimator and the conditional MLE
- The Optimal Hard Threshold for Singular Values is <inline-formula> <tex-math notation="TeX">\(4/\sqrt {3}\) </tex-math></inline-formula>
- The singular values and vectors of low rank perturbations of large rectangular random matrices
- Twice differentiable spectral functions
- Unbiased Risk Estimates for Singular Value Thresholding and Spectral Estimators
Cited in
(17)- On Stein's unbiased risk estimate for reduced rank estimators
- ePCA: high dimensional exponential family PCA
- Low-rank matrix denoising for count data using unbiased Kullback-Leibler risk estimation
- Computing the degrees of freedom of rank-regularized estimators and cousins
- Adaptive higher-order spectral estimators
- Optimal singular value shrinkage for operator norm loss: extending to non-square matrices
- Adaptive shrinkage of singular values
- OptShrink: An Algorithm for Improved Low-Rank Signal Matrix Denoising by Optimal, Data-Driven Singular Value Shrinkage
- A general theory of singular values with applications to signal denoising
- Minimax risk of matrix denoising by singular value thresholding
- SURE shrinkage of Gaussian paths and signal identification
- Selecting Regularization Parameters for Nuclear Norm--Type Minimization Problems
- Matrix denoising for weighted loss functions and heterogeneous signals
- Biwhitening Reveals the Rank of a Count Matrix
- \textit{ScreeNOT}: exact MSE-optimal singular value thresholding in correlated noise
- Matrix denoising with partial noise statistics: optimal singular value shrinkage of spiked F-matrices
- Empirical Bayes Poisson matrix completion
This page was built for publication: Generalized SURE for optimal shrinkage of singular values in low-rank matrix denoising
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4637067)
