| 一种前馈网络的新型混合算法 |
| |
| 引用本文： | 徐翔,黄道.一种前馈网络的新型混合算法[J].华东理工大学学报(自然科学版),2004,30(2):175-178. |
| |
| 作者姓名： | 徐翔 黄道 |
| |
| 作者单位： | 华东理工大学自动化系,上海,200237 |
| |
| 摘 要： | 提出了一种针对前馈神经网络的混合算法，该算法将最速下降法与共轭梯度法相结合，有效地改善了传统BP算法收敛速度慢、可能陷入局部极小等缺点。两个仿真结果表明，该算法是有效的。
|
| 关 键 词： | 前馈神经网络 最速下降法 共轭梯度法 混合法 |
| 文章编号： | 1006-3080(2004)02-0175-04 |
| 修稿时间： | 2003年4月30日 |
A New Mixed Algorithm Based on Feedforward Neural Networks |
| |
| XU Xiang,HUANG Dao.A New Mixed Algorithm Based on Feedforward Neural Networks[J].Journal of East China University of Science and Technology,2004,30(2):175-178. |
| |
| Authors: | XU Xiang HUANG Dao |
| |
| Institution: | XU Xiang,HUANG Dao~* |
| |
| Abstract: | In this paper, a mixed algorithm based on feedforward neural networks is introduced. This algorithm combines rapidly descent method with conjugate gradient method to overcome shortcomings of the traditional back-propagation algorithm, such as slow convergence and possible running into local optimum being affected by poor initial weights and setup parameters. The result of simulations shows that the mixed algorithm can be used effectively. |
| |
| Keywords: | feedforward neural networks rapidly descent method conjugate gradient method mixed algorithm |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
