English

应用科学学报 ›› 2004, Vol. 22 ›› Issue (3): 384-387.

• 论文 • 上一篇    下一篇

一种改进的反向传播神经网络算法

邱浩1, 王道波2, 张焕春1   

  1. 1 南京航空航天大学自动化学院测试工程系 江苏南京 210016;
    2 南京航空航天大学仿真与控制实验室 江苏南京 210016
  • 收稿日期:2003-04-21 修回日期:2003-06-10 出版日期:2004-09-30 发布日期:2004-09-30
  • 作者简介:邱浩(1976-),男,江苏扬州人,博士生;王道波(1957-),男,河北易县人,教授,博导;张焕春(1940-),男,江苏常熟人,教授,博导.

An Improved Back-Propagation Neural Network Algorithm

QIU Hao1, WANG Dao-bo2, ZHANG Huan-chun1   

  1. 1. Department of Testing and Measurement Engineering, College of Automation, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China;
    2. Aeronautic Simulation and Control Lab, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China
  • Received:2003-04-21 Revised:2003-06-10 Online:2004-09-30 Published:2004-09-30

PDF

232

被引次数

7

摘要： 在标准反向传播神经网络算法的基础上,提出了一种改进的反向传播神经网络算法.通过对每个处理单元增加3个参数来增强作用函数,且3个参数与连接权一样,在学习过程中进行实时更新.此算法提高了学习速度,且减少了进入局部最小点的可能性.通过XOR问题的仿真证明了改进算法的有效性.

关键词: 神经网络, 算法, 仿真, 反向传播

Abstract: Based on the idea of standard back-propagation(BP) learning algorithm, an improved BP learning algorithm is presented. Three parameters are incorporated into each processing unit(PU) to enhance the output function. The improved BP learning algorithm is developed for updating the three parameters as well as the connection weights. It not only improves the learning speed, but also reduces the occurrence of local minima. Finally, the algorithm is tested on the XOR problem to verify the validity of the improved BP.

Key words: neural networks, back-propagation, algorithm, simulation

中图分类号: