Solving unconstrained optimization problems via hybrid CD-DY conjugate gradient methods with applications

Jitsupa Deepho, Auwal Bala Abubakar, Maulana Malik, Ioannis K. Argyros

Research output: Contribution to journalArticlepeer-review

Abstract

In this work, a new hybrid conjugate gradient (CG) algorithm is developed for finding solutions to unconstrained optimization problems. The search direction of the algorithm consists of a combination of conjugate descent (CD) and Dai–Yuan (DY) CG parameters. The search direction is also close to the direction of the memoryless Broyden–Fletcher–Goldfarb–Shanno (BFGS) quasi-Newton algorithm. Moreover, the search direction is bounded and satisfies the descent condition independent of the line search. The global convergence of the algorithm under the Wolfe-type is proved with the help of some proper assumptions. Numerical experiments on some benchmark test problems are reported to show the efficiency of the new algorithm compared with other existing schemes. Finally, application of the algorithm in risk optimization completes the work.

Original languageEnglish
Article number113823
JournalJournal of Computational and Applied Mathematics
Volume405
DOIs
Publication statusPublished - 15 May 2022

Keywords

  • Global convergence
  • Hybrid conjugate gradient method
  • Line search
  • Unconstrained optimization

Fingerprint

Dive into the research topics of 'Solving unconstrained optimization problems via hybrid CD-DY conjugate gradient methods with applications'. Together they form a unique fingerprint.

Cite this