| 一种改进的RBF神经网络混合学习算法 |
| |
| 引用本文： | 孙丹,万里明,孙延风,梁艳春.一种改进的RBF神经网络混合学习算法[J].吉林大学学报(理学版),2010,48(5):817-822. |
| |
| 作者姓名： | 孙丹 万里明 孙延风 梁艳春 |
| |
| 作者单位： | 1. 吉林大学 计算机科学与技术学院, 长春 130012；2. 中国人民解放军空军装备研究院 装备总体论证研究所, 北京 100076 |
| |
| 基金项目： | 国家自然科学基金，国家高技术研究发展计划863项目基金，吉林省科技发展计划项目 |
| |
| 摘 要： | 提出一种基于粒子群优化算法、K-means算法及减聚类算法的径向基函数(RBF)神经网络混合学习算法.该算法使用减聚类方法确定隐层节点数,具有自适应确定隐层节点的能力,避免了调整隐层节点的人为干预.通过K-means算法形成粒子群优化(PSO)算法初始粒子群,避免了初始粒子群的随机性,提高了粒子群优化算法的优选能力;采用PSO算法训练RBF神经网络中的所有参数.数值结果表明,改进的混合算法具有更高的分类准确率。
|
| 关 键 词： | 聚类 粒子群优化算法 径向基函数（RBF）神经网络 |
| 收稿时间： | 2009-12-08 |
An Improved Hybrid Learning Algorithm for RBF Neural Network |
| |
| SUN Dan,WAN Li-ming,SUN Yan-feng,LIANG Yan-chun.An Improved Hybrid Learning Algorithm for RBF Neural Network[J].Journal of Jilin University: Sci Ed,2010,48(5):817-822. |
| |
| Authors: | SUN Dan WAN Li-ming SUN Yan-feng LIANG Yan-chun |
| |
| Affiliation: | 1. College of Computer Science and Technology, Jilin University, Changchun 130012, China；2. Research Institute on General Development and Evaluation of Equipment, EAAF of PLA, Beijing 100076, China |
| |
| Abstract: | We presented a hybrid learning algorithm for radial basis function neural network, which is based on particle swarm optimization, K-means clustering and subtractive clustering algorithm. The algorithm can be used to determine the number of hidden layer nodes adaptively by using subtractiveclustering, avoiding the man made interference for adjusting the hidden layer
nodes. The initial particle swarm of particle swarm optimization can be formed by K-means clustering algorithm to avoid the randomness of the initial particle swarm. In this way, the optimization capability of particle swarm optimization is improved. Particle swarm optimization algorithm is used to train all the parameters. Numerical results show that the accuracy of the improved hybrid algorithm is superior to two existing popular methods. |
| |
| Keywords: | clustering particle swarm optimization radial basis function(RBF) neural network |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
| 点击此处可从《吉林大学学报(理学版)》浏览原始摘要信息 |
|
点击此处可从《吉林大学学报(理学版)》下载全文 |
|
