On-line neural network learning algorithm with exponential convergence rate
On-line neural network learning algorithm with exponential convergence rate
复制标题
具有指数收敛速度的在线神经网络学习算法
DOI：
10.1049/el:19960895
复制
发表时间：
1996
影响因子：
1.1
通讯作者：
Y. Zhao
中科院分区：
文献类型：
--
作者：
Y. Zhao
提出了一种新的前馈神经网络在线学习算法。该算法是通过使用一种技术来实现的最小跟踪的时间相关的目标函数。理论分析表明，该算法具有指数收敛性，仿真结果表明该算法是有效的。
A new on-line learning algorithm for feedforward neural networks is presented. This algorithm is realised by using a technique for minimum tracking of a time-dependent objective function. Theoretical analysis shows that it converges exponentially, and simulations show that it is very effective.