Nonmonotone spectral gradient method based on memoryless symmetric rank-one update for large-scale unconstrained optimization

This paper proposes a nonmonotone spectral gradient method for solving large-scale unconstrained optimization problems. The spectral parameter is derived from the eigenvalues of an optimally sized memoryless symmetric rank-one matrix obtained under the measure defined as a ratio of the determinant...

Full description

Bibliographic Details
Main Authors: Hong, Seng Sim, Chuei, Yee Chen, Wah, June Leong, Jiao, Li
Format: Article
Published: American Institute of Mathematical Sciences 2021
Online Access:http://psasir.upm.edu.my/id/eprint/94373/
_version_ 1848861979390771200
author Hong, Seng Sim
Chuei, Yee Chen
Wah, June Leong
Jiao, Li
author_facet Hong, Seng Sim
Chuei, Yee Chen
Wah, June Leong
Jiao, Li
author_sort Hong, Seng Sim
building UPM Institutional Repository
collection Online Access
description This paper proposes a nonmonotone spectral gradient method for solving large-scale unconstrained optimization problems. The spectral parameter is derived from the eigenvalues of an optimally sized memoryless symmetric rank-one matrix obtained under the measure defined as a ratio of the determinant of updating matrix over its largest eigenvalue. Coupled with a nonmonotone line search strategy where backtracking-type line search is applied selectively, the spectral parameter acts as a stepsize during iterations when no line search is performed and as a milder form of quasi-Newton update when backtracking line search is employed. Convergence properties of the proposed method are established for uniformly convex functions. Extensive numerical experiments are conducted and the results indicate that our proposed spectral gradient method outperforms some standard conjugate-gradient methods.
first_indexed 2025-11-15T13:09:44Z
format Article
id upm-94373
institution Universiti Putra Malaysia
institution_category Local University
last_indexed 2025-11-15T13:09:44Z
publishDate 2021
publisher American Institute of Mathematical Sciences
recordtype eprints
repository_type Digital Repository
spelling upm-943732023-04-04T04:25:06Z http://psasir.upm.edu.my/id/eprint/94373/ Nonmonotone spectral gradient method based on memoryless symmetric rank-one update for large-scale unconstrained optimization Hong, Seng Sim Chuei, Yee Chen Wah, June Leong Jiao, Li This paper proposes a nonmonotone spectral gradient method for solving large-scale unconstrained optimization problems. The spectral parameter is derived from the eigenvalues of an optimally sized memoryless symmetric rank-one matrix obtained under the measure defined as a ratio of the determinant of updating matrix over its largest eigenvalue. Coupled with a nonmonotone line search strategy where backtracking-type line search is applied selectively, the spectral parameter acts as a stepsize during iterations when no line search is performed and as a milder form of quasi-Newton update when backtracking line search is employed. Convergence properties of the proposed method are established for uniformly convex functions. Extensive numerical experiments are conducted and the results indicate that our proposed spectral gradient method outperforms some standard conjugate-gradient methods. American Institute of Mathematical Sciences 2021-09 Article PeerReviewed Hong, Seng Sim and Chuei, Yee Chen and Wah, June Leong and Jiao, Li (2021) Nonmonotone spectral gradient method based on memoryless symmetric rank-one update for large-scale unconstrained optimization. Journal of Industrial and Management Optimization, 18 (6). pp. 3975-3988. ISSN 1547-5816; ESSN: 1553-166X https://www.aimsciences.org/article/doi/10.3934/jimo.2021143 10.3934/jimo.2021143
spellingShingle Hong, Seng Sim
Chuei, Yee Chen
Wah, June Leong
Jiao, Li
Nonmonotone spectral gradient method based on memoryless symmetric rank-one update for large-scale unconstrained optimization
title Nonmonotone spectral gradient method based on memoryless symmetric rank-one update for large-scale unconstrained optimization
title_full Nonmonotone spectral gradient method based on memoryless symmetric rank-one update for large-scale unconstrained optimization
title_fullStr Nonmonotone spectral gradient method based on memoryless symmetric rank-one update for large-scale unconstrained optimization
title_full_unstemmed Nonmonotone spectral gradient method based on memoryless symmetric rank-one update for large-scale unconstrained optimization
title_short Nonmonotone spectral gradient method based on memoryless symmetric rank-one update for large-scale unconstrained optimization
title_sort nonmonotone spectral gradient method based on memoryless symmetric rank-one update for large-scale unconstrained optimization
url http://psasir.upm.edu.my/id/eprint/94373/
http://psasir.upm.edu.my/id/eprint/94373/
http://psasir.upm.edu.my/id/eprint/94373/