Nonlinear conjugate gradient methods with sufficient descent properties for unconstrained optimization

Wataru Nakamura, Yasushi Narushima, Hiroshi Yabe

研究成果: Article

9 引用 (Scopus)

抄録

It is very important to generate a descent search direction independent of line searches in showing the global convergence of conjugate gradient methods. The method of Hager and Zhang (2005) satisfies the sufficient descent condition. In this paper, we treat two subjects. We first consider a unified formula of parameters which establishes the sufficient descent condition and follows the modification technique of Hager and Zhang. In order to show the global convergence of the conjugate gradient method with the unified formula of parameters, we define some property (say Property A). We prove the global convergence of the method with Property A. Next, we apply the unified formula to a scaled conjugate gradient method and show its global convergence property. Finally numerical results are given.

元の言語English
ページ(範囲)595-619
ページ数25
ジャーナルJournal of Industrial and Management Optimization
9
発行部数3
DOI
出版物ステータスPublished - 2013 1 1
外部発表Yes

Fingerprint

Conjugate gradient method
Unconstrained Optimization
Conjugate Gradient Method
Descent
Global Convergence
Sufficient
Property A
Line Search
Convergence Properties
Numerical Results
Gradient
Global convergence

ASJC Scopus subject areas

  • Business and International Management
  • Strategy and Management
  • Control and Optimization
  • Applied Mathematics

これを引用

@article{707b8879c5f345259a65cb81b19cf22b,
title = "Nonlinear conjugate gradient methods with sufficient descent properties for unconstrained optimization",
abstract = "It is very important to generate a descent search direction independent of line searches in showing the global convergence of conjugate gradient methods. The method of Hager and Zhang (2005) satisfies the sufficient descent condition. In this paper, we treat two subjects. We first consider a unified formula of parameters which establishes the sufficient descent condition and follows the modification technique of Hager and Zhang. In order to show the global convergence of the conjugate gradient method with the unified formula of parameters, we define some property (say Property A). We prove the global convergence of the method with Property A. Next, we apply the unified formula to a scaled conjugate gradient method and show its global convergence property. Finally numerical results are given.",
keywords = "Global convergence, Nonlinear conjugate gradient method, Scaled conjugate gradient method, Sufficient descent property, Unconstrained optimization",
author = "Wataru Nakamura and Yasushi Narushima and Hiroshi Yabe",
year = "2013",
month = "1",
day = "1",
doi = "10.3934/jimo.2013.9.595",
language = "English",
volume = "9",
pages = "595--619",
journal = "Journal of Industrial and Management Optimization",
issn = "1547-5816",
publisher = "American Institute of Mathematical Sciences",
number = "3",

}

TY - JOUR

T1 - Nonlinear conjugate gradient methods with sufficient descent properties for unconstrained optimization

AU - Nakamura, Wataru

AU - Narushima, Yasushi

AU - Yabe, Hiroshi

PY - 2013/1/1

Y1 - 2013/1/1

N2 - It is very important to generate a descent search direction independent of line searches in showing the global convergence of conjugate gradient methods. The method of Hager and Zhang (2005) satisfies the sufficient descent condition. In this paper, we treat two subjects. We first consider a unified formula of parameters which establishes the sufficient descent condition and follows the modification technique of Hager and Zhang. In order to show the global convergence of the conjugate gradient method with the unified formula of parameters, we define some property (say Property A). We prove the global convergence of the method with Property A. Next, we apply the unified formula to a scaled conjugate gradient method and show its global convergence property. Finally numerical results are given.

AB - It is very important to generate a descent search direction independent of line searches in showing the global convergence of conjugate gradient methods. The method of Hager and Zhang (2005) satisfies the sufficient descent condition. In this paper, we treat two subjects. We first consider a unified formula of parameters which establishes the sufficient descent condition and follows the modification technique of Hager and Zhang. In order to show the global convergence of the conjugate gradient method with the unified formula of parameters, we define some property (say Property A). We prove the global convergence of the method with Property A. Next, we apply the unified formula to a scaled conjugate gradient method and show its global convergence property. Finally numerical results are given.

KW - Global convergence

KW - Nonlinear conjugate gradient method

KW - Scaled conjugate gradient method

KW - Sufficient descent property

KW - Unconstrained optimization

UR - http://www.scopus.com/inward/record.url?scp=84890017726&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84890017726&partnerID=8YFLogxK

U2 - 10.3934/jimo.2013.9.595

DO - 10.3934/jimo.2013.9.595

M3 - Article

AN - SCOPUS:84890017726

VL - 9

SP - 595

EP - 619

JO - Journal of Industrial and Management Optimization

JF - Journal of Industrial and Management Optimization

SN - 1547-5816

IS - 3

ER -