DOE PAGES title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Accelerated Sparse Recovery via Gradient Descent with Nonlinear Conjugate Gradient Momentum

Journal Article · · Journal of Scientific Computing
 [1]; ORCiD logo [2];  [3];  [4];  [1];  [5]
  1. Lehigh University, Bethlehem, PA (United States)
  2. University of Texas at Dallas, Richardson, TX (United States)
  3. University of Utah, Salt Lake City, UT (United States)
  4. Chinese University of Hong Kong, Guangdong (China); Michigan State University, East Lansing, MI (United States)
  5. University of Kentucky, Lexington, KY (United States)

This paper applies an idea of adaptive momentum for the nonlinear conjugate gradient to accelerate optimization problems in sparse recovery. Specifically, we consider two types of minimization problems: a (single) differentiable function and the sum of a non-smooth function and a differentiable function. In the first case, we adopt a fixed step size to avoid the traditional line search and establish the convergence analysis of the proposed algorithm for a quadratic problem. This acceleration is further incorporated with an operator splitting technique to deal with the non-smooth function in the second case. As a result, we use the convex ι1 and the nonconvex ι1 – ι2 functionals as two case studies to demonstrate the efficiency of the proposed approaches over traditional methods.

Research Organization:
Univ. of Utah, Salt Lake City, UT (United States)
Sponsoring Organization:
USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR); National Science Foundation (NSF)
Grant/Contract Number:
SC0021142; SC0002722; DMS-1821144; DMS-1924935; DMS-1952339; DMS-2152762; DMS-2208361; DMS-2012439
OSTI ID:
2280650
Journal Information:
Journal of Scientific Computing, Vol. 95, Issue 1; ISSN 0885-7474
Publisher:
SpringerCopyright Statement
Country of Publication:
United States
Language:
English

References (54)

Signal Recovery by Proximal Forward-Backward Splitting journal January 2005
A fast iterative thresholding algorithm for wavelet-regularized deconvolution conference September 2007
Sparse Reconstruction by Separable Approximation journal July 2009
Computing Sparse Representation in a Highly Coherent Dictionary Based on Difference of $$L_1$$ L 1 and $$L_2$$ L 2 journal October 2014
Another hybrid conjugate gradient algorithm for unconstrained optimization journal January 2008
An iterative thresholding algorithm for linear inverse problems with a sparsity constraint journal January 2004
Nonlinear wavelet image processing: variational problems, compression, and noise removal through wavelet shrinkage journal March 1998
Inexact Preconditioned Conjugate Gradient Method with Inner-Outer Iteration journal January 1999
Compressed sensing journal April 2006
Fast L1–L2 Minimization via a Proximal Operator journal May 2017
Iterative Methods for Sparse Linear Systems book January 2003
An EM algorithm for wavelet-based image restoration journal August 2003
Sparse Approximate Solutions to Linear Systems journal April 1995
A Scale-Invariant Approach for Sparse Signal Recovery journal January 2019
IDENT: Identifying Differential Equations with Numerical Time Evolution journal February 2021
Point Source Super-resolution Via Non-convex $$L_1$$ Based Methods journal January 2016
A Fast Iterative Shrinkage-Thresholding Algorithm for Linear Inverse Problems journal January 2009
Restart procedures for the conjugate gradient method journal December 1977
Accelerated Schemes for the $L_1/L_2$ Minimization journal January 2020
Ratio and difference of $l_1$ and $l_2$ norms and sparse representation with coherent dictionaries journal January 2014
Distributed Optimization and Statistical Learning via the Alternating Direction Method of Multipliers book January 2011
Fitting Data with Errors in All Variables Using the Huber M-estimator journal January 1999
Methods of conjugate gradients for solving linear systems journal December 1952
Nonconvex Sorted $$\ell _1$$ ℓ 1 Minimization for Sparse Approximation journal February 2015
Function minimization by conjugate gradients journal February 1964
Descent Property and Global Convergence of the Fletcher—Reeves Method with Inexact Line Search journal January 1985
Global Convergence Properties of Conjugate Gradient Methods for Optimization journal February 1992
Minimization of $\ell_{1-2}$ for Compressed Sensing journal January 2015
A smoothing conjugate gradient method for solving systems of nonsmooth equations journal April 2013
Some methods of speeding up the convergence of iteration methods journal January 1964
Likelihood-Based Selection and Sharp Parameter Estimation journal March 2012
The smoothing Fletcher-Reeves conjugate gradient method for solving finite minimax problems journal January 2016
Sharpness, Restart, and Acceleration journal January 2020
Proximal Algorithms journal January 2014
A Generalized Krylov Subspace Method for $\ell_p$-$\ell_q$ Minimization journal January 2015
Stochastic Collocation Methods via Minimisation of the Transformed L1-Penalty journal June 2018
Monotonicity and restart in fast gradient methods conference December 2014
A D.C. Optimization Algorithm for Solving the Trust-Region Subproblem journal May 1998
Iterative reweighted minimization methods for $$l_p$$ l p regularized unconstrained nonlinear programming journal November 2013
Minimization of transformed $$L_1$$ L 1 penalty: theory, difference of convex function algorithm, and robust application in compressed sensing journal March 2018
Half-Quadratic Algorithm for $ℓ_p-ℓ_q$ Problems with Applications to TV-$ℓ_1$ Image Restoration and Compressive Sensing book January 2014
Minimization of transformed $l_1$ penalty: Closed form representation and iterative thresholding algorithms journal January 2017
Majorization–minimization generalized Krylov subspace methods for $${\ell _p}$$–$${\ell _q}$$ optimization applied to image restoration journal January 2017
A new class of nonlinear conjugate gradient coefficients with exact and inexact line searches journal October 2015
A New Conjugate Gradient Method with Guaranteed Descent and an Efficient Line Search journal January 2005
A Nonlinear Conjugate Gradient Method with a Strong Global Convergence Property journal January 1999
The place of the in robust estimation journal September 1987
Constructing Test Instances for Basis Pursuit Denoising journal March 2013
Sampling-50 years after Shannon journal April 2000
Enhancing Sparsity by Reweighted ℓ 1 Minimization journal October 2008
Adaptive Huber Regression journal April 2019
Analysis of the finite precision bi-conjugate gradient algorithm for nonsymmetric linear systems journal August 1999
Minimization of functions having Lipschitz continuous first partial derivatives journal January 1966
Smoothing Nonlinear Conjugate Gradient Method for Image Restoration Using Nonsmooth Nonconvex Minimization journal January 2010

Similar Records

Improving Deep Neural Networks’ Training for Image Classification With Nonlinear Conjugate Gradient-Style Adaptive Momentum
Journal Article · 2023 · IEEE Transactions on Neural Networks and Learning Systems · OSTI ID:2280651

Bayesian sparse learning with preconditioned stochastic gradient MCMC and its applications
Journal Article · 2021 · Journal of Computational Physics · OSTI ID:1853726

Glassoformer: A Query-Sparse Transformer for Post-Fault Power Grid Voltage Prediction
Journal Article · 2022 · Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing · OSTI ID:1866753