摘要
1 前言从1985年Rumelhart提出BP算法以来,神经网络理论发展迅速,多层前向型网络更成为用途最为广泛的网络之一。探索高效的神经网络学习算法和优良的网络结构成为推动神经理论和应用的重要因素。在网络结构设计中经常遇到的两个问题是:所得网络缺少泛化能力和易产生干扰现象。
A new method for designing artificial neural network architectures, which is based on L-sys-tem and genetic algorithm,is presented in this paper. Production rules and parallel algorithm are used to solve traditional neural network design problems. The experiment results proved that the algorithm can improve network performance and converge speed.
出处
《计算机科学》
CSCD
北大核心
2000年第6期43-45,31,共4页
Computer Science
关键词
人工神经网络
进化设计
遗传算法
Genetic algorithm
Artificial neural network
L-system
Parallel algorithm
