BP网络的PID型二阶快速学习算法
Second Order Fast Learning Algorithm of BP Neural Network Via PID Control
-
摘要: 本文利用PID控制的思想,提出了BP网络的一种二阶快速学习算法,给出了学习因子选择的必要条件与较佳区域,并结合一非线性正弦函数进行了仿真研究.结果表明,较之标标准BP学习算法,利用此法可使学习收敛速度提高22倍左右.Abstract: In this paper a second order fast learning algorithm of BP neural network, based on PID concepes, is proposed and a necessary condition and a good convergence region of learning parameter are derived. An example of nonlinear sine function is illustrated by simulation, which shows that the learning speed of the procedure described here is at least increased by 22 times more than the standard Bp algorithm.
-
Key words:
- BP neural network /
- PID control /
- fast learning algorithm
计量
- 文章访问数: 2978
- HTML全文浏览量: 104
- PDF下载量: 1353
- 被引次数: 0