基于LabWindows/CVI的人工神经网络BP算法的实现

提出了一种BP学习算法的改进算法，与传统BP算法相比，本算法针对传统BP算法收敛速度较慢的缺点，提出了两点改进：一是选用奇函数作为激励函数；二是改进了学习过程中不随误差变化的误差反向传播因子，并给出了该算法在LabWindows／CVI环境下的实现方法。程序运行表明，本算法使学习速度有了较显著的提高。

Implement of Artificial Neural Networks BP Algorithm Based on LabWindows/CVI

In this paper, BP algorithm of artificial neural network is improved As compared to the conventional BP algorithm, against the lowspeed of BP,two imprevements are proposed The first improvement is the using of an odd function as the active function, the other is the using of a new modified error backpropagation factor in the learning process At the same time, an implementing method of this algorithm in LabWindows/CVI is given in this paper As the program is running, we can see that the learning speed of this algorithm exceeds that of the conventional BP algorithm evidently