| MADALINES人工神经网络及其快速学习算法研究 |
| |
| 引用本文： | 胡瑞敏,姚天任,黄铁侠.MADALINES人工神经网络及其快速学习算法研究[J].华中科技大学学报(自然科学版),1993(6). |
| |
| 作者姓名： | 胡瑞敏 姚天任 黄铁侠 |
| |
| 作者单位： | 华中理工大学电子与信息工程系
(胡瑞敏，姚天任)，华中理工大学电子与信息工程系(黄铁侠) |
| |
| 摘 要： | 在研究MADALINES人工神经网络中心差梯度学习算法(CDG)的基础上,提出了一种加速中心差梯度学习算法(SCDG),给出了该算法收敛性的证明,在计算机上模拟实现了SCDG算法且分析了实验结果,理论分析与实验均表明,SCDG算法较之传统的CDG算法,收敛速度提高了二个数量级以上.
|
| 关 键 词： | 神经网络 学习算法 模式识别 语音处理 |
On Madalines Artificial Neural Networks and a Fast Learning Algorithm |
| |
| Hu Ruimin Yao Tianren Huang Tiexia.On Madalines Artificial Neural Networks and a Fast Learning Algorithm[J].JOURNAL OF HUAZHONG UNIVERSITY OF SCIENCE AND TECHNOLOGY.NATURE SCIENCE,1993(6). |
| |
| Authors: | Hu Ruimin Yao Tianren Huang Tiexia |
| |
| Institution: | Hu Ruimin Yao Tianren Huang Tiexia |
| |
| Abstract: | Based on the Central-Difference-Gradient Learning Algorithm (CDG) used in MADA-LINES artificial neural networks, a speed Central-Difference-Gradient Learning Algorithm (SCDG) is proposed and the convergency demonstrated. The simulation of this algorithm is performed on a type T&W386 computer and the experimental results are discussed. It has been proved that the convergence rate of SCDG algorithm can be improved by more than two orders of magnitude as compared with that of the conventional one. |
| |
| Keywords: | ANN learning algorithm pattern recognition speech processing |
| 本文献已被 CNKI 等数据库收录！ |
