| 求解无约束优化问题的一类新的下降算法 |
| |
| 引用本文： | 潘翠英,陈兰平.求解无约束优化问题的一类新的下降算法[J].应用数学学报,2007,30(1):88-98. |
| |
| 作者姓名： | 潘翠英 陈兰平 |
| |
| 作者单位： | 首都师范大学数学科学学院,北京,100037 |
| |
| 基金项目： | 国家自然科学基金;北京市教委科研项目 |
| |
| 摘 要： | 本文对求解无约束优化问题提出了一类新的下降算法,并且给出了HS算法与其相结合的两类杂交算法．在Wolfe线搜索下不需给定下降条件,即证明了它们的全局收敛性．数值实验表明新的算法十分有效,尤其是对求解大规模问题而言．
|
| 关 键 词： | 无约束最优化 下降类算法 Wolfe线搜索 全局收敛性 |
| 修稿时间： | 2005-04-282006-03-26 |
A Class of Efficient New Descent Methods |
| |
| PAN CUIYING,CHEN LANPING.A Class of Efficient New Descent Methods[J].Acta Mathematicae Applicatae Sinica,2007,30(1):88-98. |
| |
| Authors: | PAN CUIYING CHEN LANPING |
| |
| Institution: | School of Mathematical Sciences, Capital Normal University, Beijing 100037;School of Mathematical Sciences, Capital Normal University, Beijing 100037 |
| |
| Abstract: | In this paper, we propose a class of new descent methods and also we give two hybrids methods based on Hestenes-Stiefel and our new methods. And, we proved their global congenvence in Wolfe line search without descent condition. Numerical experiments shows that our methods are very efficient, especially for large scale problems. |
| |
| Keywords: | unconstrained optimization descent method wolfe line search global convergence |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
