Plug in estimation in high dimensional linear inverse problems a rigorous analysis

DOI10.1088/1742-5468/AB321AMaRDI QIDQ5854126zbMATH OpenOpenAlexWikidataFDO

Authors Alyson K. Fletcher, Parthe Pandit, Sundeep Rangan, Subrata Sarkar, Philip Schniter

Publication date 16 March 2021

Published in Journal of Statistical Mechanics: Theory and Experiment (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1806.10466

Mathematics Subject Classification ID

Linear regression; mixed models (62J05) Numerical solution to inverse problems in abstract spaces (65J22)

Abstract: Estimating a vector $mathbfx$ from noisy linear measurements $mathbfAx+mathbfw$ often requires use of prior knowledge or structural constraints on $mathbfx$ for accurate reconstruction. Several recent works have considered combining linear least-squares estimation with a generic or "plug-in" denoiser function that can be designed in a modular manner based on the prior knowledge about $mathbfx$. While these methods have shown excellent performance, it has been difficult to obtain rigorous performance guarantees. This work considers plug-in denoising combined with the recently-developed Vector Approximate Message Passing (VAMP) algorithm, which is itself derived via Expectation Propagation techniques. It shown that the mean squared error of this "plug-and-play" VAMP can be exactly predicted for high-dimensional right-rotationally invariant random $mathbfA$ and Lipschitz denoisers. The method is demonstrated on applications in image recovery and parametric bilinear estimation.

Recommendations

• High dimensional inverse covariance matrix estimation via linear programming
• Statistical multiresolution estimators in linear inverse problems -- foundations and algorithmic aspects
• Maximum likelihood estimation of regularization parameters in high-dimensional inverse problems: an empirical Bayesian approach. II: Theoretical analysis
• Geometric inference for general high-dimensional linear inverse problems
• Solving Inverse Problems With Piecewise Linear Estimators: From Gaussian Mixture Models to Structured Sparsity
• Nonlinear estimation for linear inverse problems with error in the operator
• Ridge estimation of inverse covariance matrices from high-dimensional data
• Total variation multiscale estimators for linear inverse problems
• Maximum likelihood estimation of regularization parameters in high-dimensional inverse problems: an empirical Bayesian approach. I: Methodology and experiments
• Penalized estimators for non linear inverse problems

Cites work

• A Singular Value Thresholding Algorithm for Matrix Completion
• Accurate Prediction of Phase Transitions in Compressed Sensing via a Connection to Minimax Denoising
• AMP-Inspired Deep Networks for Sparse Linear Inverse Problems
• Approximate Message Passing With Consistent Parameter Estimation and Applications to Sparse Learning
• Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising
• Blind Deconvolution Using Convex Programming
• Expectation consistent approximate inference
• From Denoising to Compressed Sensing
• scientific article; zbMATH DE number 845714 (Why is no real title available?)
• Hybrid Approximate Message Passing
• Joint Channel-Estimation and Equalization of Single-Carrier Systems via Bilinear AMP
• Phaselift: exact and stable signal recovery from magnitude measurements via convex programming
• Self-calibration and biconvex compressive sensing
• Sparsity-Cognizant Total Least-Squares for Perturbed Compressive Sampling
• State evolution for general approximate message passing algorithms, with applications to spatial coupling
• The Dynamics of Message Passing on Dense Graphs, with Applications to Compressed Sensing

Cited in

(2)

• Bayesian Imaging Using Plug & Play Priors: When Langevin Meets Tweedie
• Bayesian Imaging with Data-Driven Priors Encoded by Neural Networks

Describes a project that uses

Uses Software

• DnCNN
• PhaseLift

This page was built for publication: Plug in estimation in high dimensional linear inverse problems a rigorous analysis

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5854126)