Gradient projection Newton pursuit for sparsity constrained optimization

DOI10.1016/J.ACHA.2022.06.002zbMATH Open1501.65023arXiv2205.04580OpenAlexW4280614027MaRDI QIDQ2168680FDOQ2168680

Authors: Sheng-Long Zhou 

Publication date: 26 August 2022

Published in: Applied and Computational Harmonic Analysis (Search for Journal in Brave)

Abstract: Hard-thresholding-based algorithms have seen various advantages for sparse optimization in controlling the sparsity and allowing for fast computation. Recent research shows that when techniques of the Newton-type methods are integrated, their numerical performance can be improved surprisingly. This paper develops a gradient projection Newton pursuit algorithm that mainly adopts the hard-thresholding operator and employs the Newton pursuit only when certain conditions are satisfied. The proposed algorithm is capable of converging globally and quadratically under the standard assumptions. When it comes to compressive sensing problems, the imposed assumptions are much weaker than those for many state-of-the-art algorithms. Moreover, extensive numerical experiments have demonstrated its high performance in comparison with the other leading solvers.

Full work available at URL: https://arxiv.org/abs/2205.04580

Recommendations

• Iterative projection gradient hard thresholding pursuit algorithm for sparse optimization
• Optimal $k$-Thresholding Algorithms for Sparse Optimization Problems
• Partial gradient optimal thresholding algorithms for a class of sparse optimization problems
• A new conjugate gradient hard thresholding pursuit algorithm for sparse signal recovery
• Sparsity constrained nonlinear optimization: optimality conditions and algorithms

zbMATH Keywords

sparsity constrained optimizationglobal and quadratic convergencegradient projection Newton pursuit

Mathematics Subject Classification ID

Numerical optimization and variational techniques (65K10) Methods of reduced gradient type (90C52)

Cites Work

• Computing a Trust Region Step
• Sparse Optimization with Least-Squares Constraints
• Probing the Pareto frontier for basis pursuit solutions
• GESPAR: Efficient Phase Retrieval of Sparse Signals
• Near-Optimal Signal Recovery From Random Projections: Universal Encoding Strategies?
• A unified framework for high-dimensional analysis of \(M\)-estimators with decomposable regularizers
• Enhancing sparsity by reweighted \(\ell _{1}\) minimization
• Decoding by Linear Programming
• Stable signal recovery from incomplete and inaccurate measurements
• Robust uncertainty principles: exact signal reconstruction from highly incomplete frequency information
• Signal Recovery From Random Measurements Via Orthogonal Matching Pursuit
• From Sparse Solutions of Systems of Equations to Sparse Modeling of Signals and Images
• The \(L_1\) penalized LAD estimator for high dimensional linear regression
• Compressed sensing
• Iterative hard thresholding for compressed sensing
• CoSaMP: Iterative signal recovery from incomplete and inaccurate samples
• Improved iteratively reweighted least squares for unconstrained smoothed \(\ell_q\) minimization
• Sparsity constrained nonlinear optimization: optimality conditions and algorithms
• Alternating direction algorithms for \(\ell_1\)-problems in compressive sensing
• On solutions of sparsity constrained optimization
• Compressed Sensing With Nonlinear Observations and Related Nonlinear Optimization Problems
• Greedy Sparsity-Constrained Optimization
• Iterative thresholding for sparse approximations
• Trading accuracy for sparsity in optimization problems with sparsity constraints
• Compressive sensing and structured random matrices
• Subspace Pursuit for Compressive Sensing Signal Reconstruction
• Minimization of $\ell_{1-2}$ for Compressed Sensing
• Hard thresholding pursuit: an algorithm for compressive sensing
• Gradient Pursuits
• Fast global convergence of gradient methods for high-dimensional statistical recovery
• A Tight Bound of Hard Thresholding
• An extended Newton-type algorithm for \(\ell_2\)-regularized sparse logistic regression and its efficiency for classifying large-scale datasets
• Title not available (Why is that?)
• Title not available (Why is that?)
• A null-space-based weightedl₁minimization approach to compressed sensing

Cited In (4)

• Gradient projection Newton algorithm for sparse collaborative learning using synthetic and real datasets of applications
• Title not available (Why is that?)
• A greedy Newton-type method for multiple sparse constraint problem
• Projected Nesterov's Proximal-Gradient Algorithm for Sparse Signal Recovery

This page was built for publication: Gradient projection Newton pursuit for sparsity constrained optimization

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2168680)