摘要
本文描述一种基于组合电路的Ｈｏｐｆｉｅｌｄ神经网络模型的测试生成系统，重点介绍了系统中实现神经网络学习并记忆基于电路拓扑的知识信息的学习策略，从而将基于电路拓扑的知识与数学计算结合起来．最后给出了实验结果．
A test generation system based on combinational circuit's Hopfield neural network model is described. In this system, we combine the circuit's topology information with mathematic computing, and propose learning strategies that can help neural network learn and memorize the circuit topology knowledge. Finally, the experimental results are given.
出处
《电子学报》
EI
CAS
CSCD
北大核心
1997年第8期55-58,共4页
Acta Electronica Sinica
基金
国家"八五"攻关
关键词
测试生成
学习策略
神经网络
数字电路
Automatic testing
Electric network topology
Neural networks
