摘要： 介绍了BP网络的基本结构模型与常见改进算法,在此基础上提出了一种新型的结构自组织BP网络算法。该自组织BP网络算法能够根据当前收敛状态自动调整学习率,使得网络收敛速度与学习率变化保持一致。同时该BP网络算法还能根据神经元在学习过程中的变化情况及其相互关系进行网络结构自适应调整,以保证网络结构逐渐趋于合理性。

关键词: 神经网络, BP算法, 结构优化

Abstract: Based on expatiated the basic structure model and some general improved algorithms of BP neural network, this paper brings forward a new self-organization learning algorithm. The algorithm can change network’s learning rate followed by network’s convergence state, and can adjust network’s structure based on the neurons’ change and their relationship. Because of these improvements, the algorithm can keep consistency between the network’s converging speed and the learning rate change, also it can ensure that the network’s structure tends to validity progressively.

中图分类号: 

• TP183