| 一种用GEP进化神经网络结构和权值的方法 |
| |
| 引用本文： | 王艳春,何东健,王守志. 一种用GEP进化神经网络结构和权值的方法[J]. 重庆大学学报(自然科学版), 2008, 31(12): 1395-1399 |
| |
| 作者姓名： | 王艳春 何东健 王守志 |
| |
| 作者单位： | 西北农林科技大学机械与电子工程学院;青岛农业大学信息科学与工程学院;西北农林科技大学信息工程学院;威海职业技术学院机电工程系; |
| |
| 基金项目： | 国家自然科学基金资助项目 |
| |
| 摘 要： | 提出一种用基因表达式编程(gene expression programming,GEP)自动设计神经网络结构和权值的算法.论述算法的基本思想和基本操作,针对算法的早熟现象和变异率低问题进行了相应的改进,给出这种算法的应用实例.实验结果表明,GEP可以自动设计神经网络的结构,并能给出优化的网络权值,与其他优化算法相比,收敛速度更快.
|
| 关 键 词： | 基因表达式 神经网络 网络结构 |
A method for evolving the architecture and weights of neural networks via gene expression programming |
| |
| WANG Yan-chuna,,HE Dong-jianb,WANG Shou-zhia. A method for evolving the architecture and weights of neural networks via gene expression programming[J]. Journal of Chongqing University(Natural Science Edition), 2008, 31(12): 1395-1399 |
| |
| Authors: | WANG Yan-chuna HE Dong-jianb WANG Shou-zhia |
| |
| Affiliation: | WANG Yan-chun1a,2,HE Dong-jian1b,WANG Shou-zhi1a,3(1a.College of Mechanical , Electronic Engineering,b.College of Information Engineering,Northwest A & F University,Shannxi 712100,P.R.China,2.College of Information Science , Engineering,Qingdao Agricultural University,Sh,ong 266109,3.Department of Mechanical , Electronic Engineering,Weihai Vocational Technical College,Sh,ong 264210,P.R.China) |
| |
| Abstract: | An algorithm for automatic designation of the architecture and the weights of neural networks using gene expression programming(GEP) was presented.The fundamental ideas and procedures of the algorithm were discussed.The algorithm was improved to solve the problems of prematurity and lower variance rate.An application for neural networks designation was given.The experimental results indicate that the proposed GEP approach may evolve the architecture of neural network,and can obtain the weights more precisel... |
| |
| Keywords: | gene expression neural networks network architecture |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
| 点击此处可从《重庆大学学报(自然科学版)》浏览原始摘要信息 |
|
点击此处可从《重庆大学学报(自然科学版)》下载全文 |
|
