Another modified dprp conjugate gradient method with global convergent properties

Mustafa, Mamat and Kamilu Uba, Kamfa and Mohd, Rivaie (2015) Another modified dprp conjugate gradient method with global convergent properties. Far East Journal of Mathematical Sciences, 98 (5). pp. 563-577. ISSN 09720871

[img] Image
FH02-FIK-15-03927.jpg
Restricted to Registered users only

Download (145kB)

Abstract

Conjugate gradient (CG) methods play significant role in solving large scale unconstrained optimization problem, due to their low memory requirement and global convergent properties. For many years, different studies and modification have been carried out to improve this method. In this paper, we present a new CG method based on modifying the Abashar et al. [4] method (ADPRP). This new method possesses the global convergent properties and sufficient descent condition under exact line searches. Numerical result based on number of iteration and CPU time shows that the proposed coefficient is efficient and effective when compared with other CG formulas.

Item Type: Article
Uncontrolled Keywords: unconstrained optimization, exact line search, global convergence, sufficient descent
Subjects: Q Science > QA Mathematics
Divisions: Faculty of Informatics & Computing
Depositing User: Syahmi Manaf
Date Deposited: 13 Sep 2022 04:27
Last Modified: 13 Sep 2022 04:27
URI: http://eprints.unisza.edu.my/id/eprint/6694

Actions (login required)

View Item View Item