首页 | 本学科首页   官方微博 | 高级检索  
     

神经网络的规划学习算法
引用本文:张铃,张钹.神经网络的规划学习算法[J].计算机学报,1994,17(9):669-675.
作者姓名:张铃  张钹
作者单位:安徽大学计算机科学系,清华大学计算机科学与技术系
基金项目:国家攀登计划资助,国家863高技术计划资助
摘    要:本文利用二次规划方法,讨论对神经网络训练样本的吸引半径的优化问题,并借用二交规划中的RPA算法求该优化解,得到一种新的神经网络基于规划的学习算法,其次,将规划学习算法与现有的几种常见算法进行比较,指出该算法的特点。

关 键 词:神经网络  学习算法  规划

PROGRAMMING BASED LEARNING ALGORITHM OF NEURAL NETWORKS
Zhang Ling.PROGRAMMING BASED LEARNING ALGORITHM OF NEURAL NETWORKS[J].Chinese Journal of Computers,1994,17(9):669-675.
Authors:Zhang Ling
Abstract:In this paper,the optimization of the attractive radius of training samples in neural networks is discussed using quadratic programming technique.A kind of quatratic programming technique,Potential Reduction Algorithm(PRA ),is used for solving the optimization problem.Then,A new programming based learning algorithm of neural networks with self-feedback connections is presented. The comparison is made between the given learning algorithm and some other known algorithms,in order to show the characteristics of the new algorithm.
Keywords:Neural network  learning algorithm  attractive radius  quadratic programming  
本文献已被 CNKI 维普 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号