摘 要:
在wolfe步长搜索下,对解无约束最优化问题的共轭梯度法的迭代参数做出改进,扩大了它的选取范围,并在目标函数可微的条件下,证明了算法的全局收敛性.[著者文摘]

文章出处:
《平顶山学院学报》-2007年22卷5期 -63-66页
Journal of Pingdingshan University
栏目信息:
分 类 号:
文献标识码:
A
文章编号:
1673-1670(2007)05-0063-04
Global Convergence of Modified Conjugate Gradient Methods
JING Shu - jie, ZHANG Zhi - rong ( Henan Polytechnic University, Jiaozuo Henan 454003, China)
Abstract:
This paper presents a new conj sing Wolfe inexact line search and extending ugate gradient method for unconstrained optimization problem by u- the range of the iterative parajueter. Global convergence of such a method is proved under the condition that the objective function is continuously differentiable.[著者文摘]
Key words:
conjugate gradient method; Wolfe step size rule; global convergence
收稿日期: 2007-01-14
基金资助:
国家自然科学基金(10671057)

更多评论>>文章评论
学术















cqvip.com