基于BP网络的权值更新快速收敛算法

计算机应用

基于BP网络的权值更新快速收敛算法

周昌能余雪丽

太原理工大学计算机学院太原理工大学计算机学院

收稿日期:2006-02-13 修回日期:1900-01-01 出版日期:2006-08-01 发布日期:2006-08-01
通讯作者: 周昌能

Rapid convergence algorithms for weight values updating based on BP network

Received:2006-02-13 Revised:1900-01-01 Online:2006-08-01 Published:2006-08-01

摘要/Abstract

摘要： 针对标准BP网络学习算法收敛慢的问题，提出了两种权值更新的快速收敛算法，即基于梯度变化率的快速传递算法和基于梯度方向的弹性传递算法，并在煤矿事故救援游戏式训练系统中进行仿真和比较，让游戏角色根据井下空气成分学习判断危险程度，以便受训人员或仿生机器人采取相应的措施。仿真结果表明，所提算法的收敛时间比标准算法有一定改善。

关键词: 快速收敛算法, 游戏式训练, BP人工神经网络

Abstract: To solve the slow convergence of standard learning algorithm in BP network,two rapid convergence algorithms were suggested for weight values updating. One is rapid transmission algorithm based on gradient change rate. The other is flexible transmission algorithm based on gradient orientation. The two algorithms were simulated and compared in Game Style Training System for Mine Accident Rescuing. Here the algorithms would help game roles learn to estimate the danger degree according to ingredients of mine air, and then help trainees or biorobots take corresponding actions. The simulating results show that shorter convergence time is taken for the two algorithms than the standard algorithm.

Key words: quick convergence algorithms, game style training, BP artificial neural network

周昌能余雪丽 . 基于BP网络的权值更新快速收敛算法[J]. 计算机应用.

. Rapid convergence algorithms for weight values updating based on BP network[J]. Journal of Computer Applications.