| 摘要: |
| 非线性共轭梯度方法是解决大规模无约束问题最有效的方法之一，提出了一类新的修正共轭梯度算法，新算法推广了黄海东等的共轭梯度参数算法,不依赖任何线搜索且具有充分下降性；然后,在标准 Wolfe非精确线搜索下,得到了新算法的全局收敛性. |
| 关键词: 无约束优化问题 共轭梯度法 充分下降性 全局收敛性 |
| DOI： |
| 分类号: |
| 基金项目: |
|
| An Improved NVPRP Method |
|
WU Su hua
|
| Abstract: |
| Nonlinear conjugate gradient method is one of the most effective methods to solve large scale unconstrained problems. This paper presents a class of modified conjugate gradient algorithms. The new algorithm generalizes the conjugate gradient parameter algorithm of Huang Haidong et al, and has sufficient descent property without depending on any line search. Then, under Wolfe non accurate line search, the global convergence of the new algorithm is obtained. |
| Key words: unconstrained optimization conjugate gradient algorithm sufficient descent property global convergence |
