Rate Optimal Denoising of Simultaneously Sparse and Low Rank Matrices

MaRDI QIDQ3188017zbMATH OpenFDO

Authors Dan Yang, Zongming Ma, Andreas Buja

Publication date 16 August 2016

Full work available at URL https://arxiv.org/abs/1405.0338, http://jmlr.csail.mit.edu/papers/v17/15-134.html

zbMATH Keywords

denoisinghigh dimensionalitylow-rank matricesminimax ratessimultaneously structured matricessparse SVDsparsity

Mathematics Subject Classification ID

Estimation in multivariate analysis (62H12) Time series, auto-correlation, regression, etc. in statistics (GARCH) (62M10)

Abstract: We study minimax rates for denoising simultaneously sparse and low rank matrices in high dimensions. We show that an iterative thresholding algorithm achieves (near) optimal rates adaptively under mild conditions for a large class of loss functions. Numerical experiments on synthetic datasets also demonstrate the competitive performance of the proposed method.

Recommendations

• Near-optimal estimation of simultaneously sparse and low-rank matrices from nested linear measurements
• Low-rank matrix completion and denoising under Poisson noise
• Compressed sensing of low-rank plus sparse matrices
• Matrix denoising with partial noise statistics: optimal singular value shrinkage of spiked F-matrices
• Low rank matrix recovery with adversarial sparse noise
• Denoising by sparse approximation: error bounds based on rate-distortion theory
• Nuclear-norm penalization and optimal rates for noisy low-rank matrix completion
• Optimal rank-sparsity decomposition
• Simultaneously Sparse and Low-Rank Matrix Reconstruction via Nonconvex and Nonseparable Regularization
• Noisy matrix decomposition via convex relaxation: optimal rates in high dimensions

Cited in

(20)

• Rate-optimal perturbation bounds for singular subspaces with applications to high-dimensional statistics
• Perturbation expansions and error bounds for the truncated singular value decomposition
• Edge statistics of large dimensional deformed rectangular matrices
• Statistical and computational limits for sparse matrix detection
• High dimensional deformed rectangular matrices with applications in matrix denoising
• Singular vector and singular subspace distribution for the matrix denoising model
• OptShrink: An Algorithm for Improved Low-Rank Signal Matrix Denoising by Optimal, Data-Driven Singular Value Shrinkage
• Estimation of matrices with row sparsity
• Near-optimal estimation of simultaneously sparse and low-rank matrices from nested linear measurements
• scientific article; zbMATH DE number 7370563 (Why is no real title available?)
• Low-rank matrix completion and denoising under Poisson noise
• Matrix denoising for weighted loss functions and heterogeneous signals
• Optimal sparse singular value decomposition for high-dimensional high-order data
• Generalized Liquid Association Analysis for Multimodal Data Integration
• Tracy-Widom law for the extreme eigenvalues of large signal-plus-noise matrices
• Matrix denoising with partial noise statistics: optimal singular value shrinkage of spiked F-matrices
• Beyond symmetry: best submatrix selection for the sparse truncated SVD
• Asymptotic limits of spiked eigenvalues and eigenvectors of signal-plus-noise matrices with weak signals and heteroskedastic noise
• The Tucker low-rank classification model for tensor data
• Network Regression and Supervised Centrality Estimation

This page was built for publication: Rate Optimal Denoising of Simultaneously Sparse and Low Rank Matrices

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3188017)