| 一个异步的PGD算法 |
| |
| 引用本文： | 孙莉,黄利国.一个异步的PGD算法[J].山东科技大学学报(自然科学版),2006,25(4):110-112. |
| |
| 作者姓名： | 孙莉 黄利国 |
| |
| 作者单位： | 1. 上海交通大学,数学系,上海,200240
2. 滨州学院,数学与信息科学系,山东,滨州,256600 |
| |
| 摘 要： | 分析了并行优化算法中同步运算与异步运算的优缺点。提出完全异步的PGD算法。并且在一定的条件下。给出了算法的收敛性证明。最后结合大规模分布式并行计算机系统曙光-2000做出数值试验。结果说明异步的并行优化算法的效率高于同步的算法。
|
| 关 键 词： | 无约束非线性最优化问题 并行梯度分配算法 加速比 |
| 文章编号： | 1672-3767（2006）04-0110-03 |
| 收稿时间： | 2005-12-05 |
| 修稿时间： | 2005年12月5日 |
Asynchronous Parallel Gradient Distribution Algorithm |
| |
| SUN Li,HUANG Li-guo.Asynchronous Parallel Gradient Distribution Algorithm[J].Journal of Shandong Univ of Sci and Technol: Nat Sci,2006,25(4):110-112. |
| |
| Authors: | SUN Li HUANG Li-guo |
| |
| Institution: | 1. Department of Mathematics, Shanghai Jaotong University, Shanghai 200240, China; 2. Dept. of Mathematics and Info Science, Binzhou College, Binzhou, Shandong 256600, China |
| |
| Abstract: | The paper analyzes the advantages and disadvantages of synchronous and asynchronous operations in parallel optimization algorithm, and presents the complete asynchronous PGD algorithm and under a certain condition, the proof of its global convergence is given. Finally, numerical tests are provided combined with the distributed large-scale parallel computer system, Dawning 2000, and its results show that the efficiency of asynchronous parallel optimization algorithm is better than that of its synchronous counterpart. |
| |
| Keywords: | unconstrained nonlinear optimization parallel gradient distribution speedup ratio |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
