摘要
提出一种用基因表达式编程(GEP)自动设计神经网络的算法。针对标准GEP算法在优化神经网络过程中的早熟现象和变异率低问题,对算法进行了改进,并给出算法的具体应用实例。与其它优化算法的对比实验表明,GEP是一种有效的神经网络设计方法,并且改进的GEP算法比标准GEP算法进化效率高,将收敛率提高了37个百分点,收敛速度快,进化代数仅是标准算法的58%。
An algorithm for automatic design of neural networks using gene expression programming（GEP） is presented.The standard GEP is improved to solve the problem of prematurity and lower mutational rate in optimizing neural networks.An application of designing neural networks is formulated and compared with others.The results demonstrated that the performance of modified algorithm is much better than that of standard GEP in that it not only has higher evolution efficiency,improving convergence rate by 37 percentage point but has faster convergence speed with only 58% evolutionary generations of standard GEP algorithm.
出处
《计算机工程与设计》
CSCD
北大核心
2010年第11期2554-2556,2560,共4页
Computer Engineering and Design
基金
国家自然科学基金项目(30471138)
关键词
基因表达式编程
神经网络
网络结构
权值
进化
gene expression programming
neural networks
network architecture
weights
evolve
