| 一种构造径向基神经网络的新型IPL算法 |
| |
| 引用本文： | 游培寒,王振家.一种构造径向基神经网络的新型IPL算法[J].系统工程与电子技术,2003,25(9):1151-1153. |
| |
| 作者姓名： | 游培寒 王振家 |
| |
| 作者单位： | 空军工程大学工程学院,陕西,西安,710038 |
| |
| 摘 要： | 提出了一种新型的IPL(incremental projection learning)算法。该算法通过调整网络参数,达到修正IPL 算子A的目的。与原来的IPL算法相比,新算法所需训练样本数较少,步骤简单,训练速度快。仿真结果表明,由 经过改进的。IPL算法得到的径向基网络结构相对简单,结果也较精确。
|
| 关 键 词： | IPL算法 RBF神经网络 三段训练法 |
| 文章编号： | 1001-506X(2003)09-1151-03 |
| 修稿时间： | 2002年4月10日 |
A New IPL Algorithm for Constructing an RBF Network |
| |
| YOU Pei-han,WANG Zhen-jia.A New IPL Algorithm for Constructing an RBF Network[J].System Engineering and Electronics,2003,25(9):1151-1153. |
| |
| Authors: | YOU Pei-han WANG Zhen-jia |
| |
| Abstract: | In this paper a new incremental projection leaming( IPL) algorithm is presented. Through the network parameter adjustment, this new algorithm can adjust the operator A of IPL. It needs a small number of sampling data, and its training speed is higher than the traditional one. The new IPL algorithm is utilized to produce an RBF network. The computer simulation results show that the structure of the network constructed by the new algorithm is simple and the output of the system is accurate. |
| |
| Keywords: | IPL algorithm Radial basis function neural network Three-phase training algorithm |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
