| 改进的 BP 算法及其应用研究 |
| |
| 引用本文： | 吴焱明,王纯贤,王治森.改进的 BP 算法及其应用研究[J].合肥工业大学学报(自然科学版),1998(4). |
| |
| 作者姓名： | 吴焱明 王纯贤 王治森 |
| |
| 作者单位： | 合肥工业大学ＣＩＭＳ所 |
| |
| 摘 要： | ＢＰ网络广泛应用于函数逼近、模式识别和系统辨识，但ＢＰ算法收敛速度很慢。为此提出了ＢＰ算法的一种新的改进方式，即在误差反向传播时，不仅改变网络的联接权值，也改变神经元模型参数。详细推导了改进ＢＰ算法的迭代公式。仿真研究表明，与传统ＢＰ算法相比，该算法具有收敛速度快，函数逼近精度高的优点。
|
| 关 键 词： | BP算法 收敛速度 函数逼近 |
ON ALGORITHM AND ITS APPLICATION |
| |
| Wu Yanming,Wang Chunxian,Wang Zhisen.ON ALGORITHM AND ITS APPLICATION[J].Journal of Hefei University of Technology(Natural Science),1998(4). |
| |
| Authors: | Wu Yanming Wang Chunxian Wang Zhisen |
| |
| Institution: | Wu Yanming Wang Chunxian Wang Zhisen |
| |
| Abstract: | Back Propagation (BP) network is widely used in function impending,pattern recognition and system identification.But the Back Propagation algorithms convergence very slowly.A new improved BP algorithm is given in this paper, which modifies not only the weights between neurons but also the parameters of neuron model while error is back propagating. The iterative formulation of the improved BP algorithm is derived in detail. The simulation results show that this BP algorithm is more efficient than the conventional BP algorthm on the aspects of the convergence rate and the ability of function approximation. |
| |
| Keywords: | BP networks rate of convergence function approximation |
| 本文献已被 CNKI 等数据库收录！ |
