摘要
阐明了遗传算法和神经网络结合的可行性,提出了一种改进的面向神经网络权值学习的遗传算法。通过对XOR问题的实验,显示出其快速学习网络权值的能力,且能摆脱局部极值的困扰和初始权值的限制从各方面都表现出优于标准遗传算法和BP算法的性能。
The paper demonstrates the possibility of combining neural network with genetic algorithm. An improved genetic algorithm for the learning of neural network's connection weights is presented. According to The XOR problem, it indicates that the new method has the capability in fast learning of neural network and the capability in escaping the limitation of local extreme value and initial weights. The algorithm gets far superior performance to traditional genetic algorithm and BP algorithm in all sides.
出处
《计算机工程》
CAS
CSCD
北大核心
2002年第4期120-121,129,共3页
Computer Engineering
