| 并行技术在约束凸规划化问题的对偶算法中的应用 |
| |
| 引用本文： | 高桂英,张立卫.并行技术在约束凸规划化问题的对偶算法中的应用[J].经济数学,2003,20(2):67-71. |
| |
| 作者姓名： | 高桂英 张立卫 |
| |
| 作者单位： | 大连理工大学应用数学系,大连,116024 |
| |
| 摘 要： | 用 Rosen(196 1)的投影梯度的方法求解约束凸规划化问题的对偶问题 ,在计算投影梯度方向时 ,涉及求关于原始变量的最小化问题的最优解 .我们用并行梯度分布算法 (PGD)计算出这一极小化问题的近似解 ,证明近似解可以达到任何给定的精度 ,并说明当精度选取合适时 ,Rosen方法仍然是收敛的
|
| 关 键 词： | 约束凸规划 并行算法 投影梯度 |
| 修稿时间： | 2002年9月25日 |
APPLICATION OF THE PARALLELIZATION TECHNIQUE TO A DUAL METHOD FOR A CONSTRAINED CONVEX PROGRAMMING PROBLEM |
| |
| Gao Guiying,Zhang Liwei.APPLICATION OF THE PARALLELIZATION TECHNIQUE TO A DUAL METHOD FOR A CONSTRAINED CONVEX PROGRAMMING PROBLEM[J].Mathematics in Economics,2003,20(2):67-71. |
| |
| Authors: | Gao Guiying Zhang Liwei |
| |
| Abstract: | Computing a projection gradient direction requires solving a minimization problem in original variables, if the projection gradient method due to Rosen(1961) is applied to solving the dual problem of a constrained convex programming problem. We use Parallel Gradient Distribution (PGD) algorithm to obtain an approximate solution with any given accuracy. It is explained that Rosen method is convergent when approximate solutions with proper accuracies are employed. |
| |
| Keywords: | Constrained convex programming parallel algorithm projection gradient |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
