A modified nonlinear Polak–Ribière–Polyak conjugate gradient method with sufficient descent property

被引:4
|
作者
Xiaoliang Dong
机构
[1] Xi’an Shiyou University,College of Science
[2] Nanjing Normal University,School of Mathematics Science
来源
Calcolo | 2020年 / 57卷
关键词
Polak–Ribière–Polyak conjugate gradient method; Sufficient descent condition; Global convergence; Numerical comparison; 90c30; 65K05;
D O I
暂无
中图分类号
学科分类号
摘要
In this paper, a small and necessary revision on an assumption condition of Aminifard and Babaie-Kafaki (Calcolo, 2019. https://doi.org/10.1007/s10092-019-0312-9) is made. By a little modification, a new conjugate gradient method is proposed, in which the search directions satisfy the sufficient descent condition with the strong Wolfe line search. The main difference between two algorithms is that the proposed method is globally convergent without boundedness assumption on the steplength. Comparative numerical results demonstrating efficiency of the proposed method are reported.
引用
收藏
相关论文
共 50 条