Descent three-term DY-type conjugate gradient methods for constrained monotone equations with application

As it is known that, not all conjugate gradient (CG) methods satisfy descent property, a necessary condition for attaining global convergence result. In this article, we propose three different sufficient-descent conjugate gradient projection algorithms for constrained monotone equations. Using Dai–...

Celý popis

Uloženo v:

Podrobná bibliografie
Vydáno v:	Computational & applied mathematics Ročník 41; číslo 1
Hlavní autoři:	Abdullahi, Habibu, Awasthi, A. K., Waziri, Mohammed Yusuf, Halilu, Abubakar Sani
Médium:	Journal Article
Jazyk:	angličtina
Vydáno:	Cham Springer International Publishing 01.02.2022 Springer Nature B.V
Témata:	Algorithms Applications of Mathematics Applied physics Computational mathematics Computational Mathematics and Numerical Analysis Conjugate gradient method Convergence Mathematical Applications in Computer Science Mathematical Applications in the Physical Sciences Mathematics Mathematics and Statistics Nonlinear equations Performance assessment Projection method Constrained-monotone nonlinear equations Compressive sensing 65K05 34G20H18 65D32 Descent conjugate gradient 90C05
ISSN:	2238-3603, 1807-0302
On-line přístup:	Získat plný text
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

Popis
Shrnutí:	As it is known that, not all conjugate gradient (CG) methods satisfy descent property, a necessary condition for attaining global convergence result. In this article, we propose three different sufficient-descent conjugate gradient projection algorithms for constrained monotone equations. Using Dai–Yuan (DY) conjugate gradient parameter, we generate three satisfied sufficient-descent directions. Under suitable conditions, global convergence of the algorithms is established. Numerical examples using benchmark test functions indicate that the algorithms are effective for solving constrained monotone nonlinear equations. Moreover, we also extend the method to solve ℓ 1 -norm regularized problems to decode a sparse signal in compressive sensing. Performance comparisons show that the proposed methods are practical, efficient and competitive with the compared methods.
Bibliografie:	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14
ISSN:	2238-3603 1807-0302
DOI:	10.1007/s40314-021-01724-y