Gradient-Based Methods for Sparse Recovery

被引：37
|
作者：
Hager, William W. [1 ]
Phan, Dzung T. [2 ]
Zhang, Hongchao [3 ]
机构：
[1] Univ Florida, Dept Math, Gainesville, FL 32611 USA
[2] IBM Corp, Thomas J Watson Res Ctr, Dept Business Analyt & Math Sci, Yorktown Hts, NY 10598 USA
[3] Louisiana State Univ, Dept Math, Ctr Computat & Technol, Baton Rouge, LA 70803 USA
来源：
SIAM JOURNAL ON IMAGING SCIENCES | 2011年 / 4卷 / 01期
基金：
美国国家科学基金会;
关键词：
sparse reconstruction by separable approximation; iterative shrinkage thresholding algorithm; sparse recovery; sublinear convergence; linear convergence; image reconstruction; denoising; compressed sensing; nonsmooth optimization; nonmonotone convergence; BB method; THRESHOLDING ALGORITHM;
D O I：
10.1137/090775063
中图分类号：
TP18 [人工智能理论];
学科分类号：
081104 ; 0812 ; 0835 ; 1405 ;
摘要：
The convergence rate is analyzed for the sparse reconstruction by separable approximation (SpaRSA) algorithm for minimizing a sum f(x) + psi(x), where f is smooth and psi is convex, but possibly nonsmooth. It is shown that if f is convex, then the error in the objective function at iteration k is bounded by a/k for some a independent of k. Moreover, if the objective function is strongly convex, then the convergence is R-linear. An improved version of the algorithm based on a cyclic version of the BB iteration and an adaptive line search is given. The performance of the algorithm is investigated using applications in the areas of signal processing and image reconstruction.
引用
收藏
页码：146 / 165
页数：20
相关论文
共 50 条
  • [31] Gradient-Based Identification Methods for Hammerstein Nonlinear ARMAX Models
    Feng Ding
    Yang Shi
    Tongwen Chen
    Nonlinear Dynamics, 2006, 45 : 31 - 43
  • [32] Gradient-based identification methods for Hammerstein nonlinear ARMAX models
    Ding, Feng
    Shi, Yang
    Chen, Tongwen
    NONLINEAR DYNAMICS, 2006, 45 (1-2) : 31 - 43
  • [33] Gradient-Based Learning of Discrete Structured Measurement Operators for Signal Recovery
    Sauder J.
    Genzel M.
    Jung P.
    IEEE Journal on Selected Areas in Information Theory, 2022, 3 (03): : 481 - 492
  • [34] GRADIENT-FREE AND GRADIENT-BASED METHODS FOR SHAPE OPTIMIZATION OF WATER TURBINE BLADE
    Bastl, Bohumir
    Brandner, Marek
    Egermaier, Jiri
    Hornikova, Hana
    Michalkova, Kristyna
    Turnerova, Eva
    PROGRAMS AND ALGORITHMS OF NUMERICAL MATHEMATICS 19, 2019, : 15 - 26
  • [35] Gradient-based Resolution Enhancement Optimization Methods based on Vector Imaging Model
    Ma, Xu
    Li, Yanqiu
    Dong, Lisong
    OPTICAL MICROLITHOGRAPHY XXV, PTS 1AND 2, 2012, 8326
  • [36] Gradient-based defense methods for data leakage in vertical federated learning
    Chang, Wenhan
    Zhu, Tianqing
    COMPUTERS & SECURITY, 2024, 139
  • [37] Hybrid topological derivative and gradient-based methods for electrical impedance tomography
    Carpio, A.
    Rapun, M-L
    INVERSE PROBLEMS, 2012, 28 (09)
  • [38] Adaptive Meshless Methods in Electromagnetic Modeling: A Gradient-Based Refinement Strategy
    Kaufmann, Thomas
    Engstrom, Christian
    Fumeaux, Christophe
    2011 41ST EUROPEAN MICROWAVE CONFERENCE, 2011, : 559 - 562
  • [39] Using blade element momentum methods with gradient-based design optimization
    Andrew Ning
    Structural and Multidisciplinary Optimization, 2021, 64 : 991 - 1014
  • [40] Discrete optimization via gradient-based adaptive stochastic search methods
    Chen, Xi
    Zhou, Enlu
    Hu, Jiaqiao
    IISE TRANSACTIONS, 2018, 50 (09) : 789 - 805
12345