节点文献

改进的神经网络快速学习算法

Improvement on the Fast Learning Algorithm for the Neural Network

  • 推荐 CAJ下载
  • PDF下载
  • 不支持迅雷等下载工具,请取消加速工具后下载。

【作者】 冯昭志黄载禄杨叔子

【Author】 Feng Zhaozhi Dept. of Electronics & Information Eng., HUST, Wuhan 430074, China . Huang Zailu Yang Shuzi

【机构】 华中理工大学电子与信息工程系

【摘要】 提出了一种新颖的神经元模型和用于神经网络训练的推广准则,给出了多层神经网络的快速学习算法,研究了一种用于语音识别的时变多层神经网络及其相应的快速学习算法.语音识别实验表明:所给出的快速学习算法能有效地加速网络训练进程.

【Abstract】 A new neuron model for the neural network is developed and a generalized criterion for the training of neural network is proposed. A fast learning algorithm for a multilayer neural network is described. A second order time variant neural network applied to speech recognition is presented and its fast BP algorithm is discussed. Experiments on speech recognition show that the fast BP algorithm proposed can speed up the training of neural network.

  • 【文献出处】 华中理工大学学报 ,JOURNAL OF HUAZHONG UNIVERSITY OF SCIENCE AND TECHNOLOGY , 编辑部邮箱 ,1996年S2期
  • 【分类号】TP301.6
  • 【被引频次】5
  • 【下载频次】93
节点文献中: