| 多输出单元BP神经网络梯度算法的收敛性 |
| |
| 引用本文： | 周凤麒,吴微,熊焱. 多输出单元BP神经网络梯度算法的收敛性[J]. 黑龙江大学自然科学学报, 2006, 23(1): 108-113 |
| |
| 作者姓名： | 周凤麒 吴微 熊焱 |
| |
| 作者单位： | 大连理工大学,应用数学系,辽宁,大连,116024 |
| |
| 摘 要： | 梯度算法广泛应用于训练前馈神经网络.对于单输出前馈神经网络的梯度算法的收敛性已经有了详细的讨论.研究了带有多个输出单元的BP神经网络的梯度算法,证明了误差函数在梯度算法所生成的权向量序列上的单调递减性,并且证明了梯度算法的弱收敛性和强收敛性.
|
| 关 键 词： | BP神经网络 梯度算法 收敛性 多输出单元 |
| 文章编号： | 1001-7011(2006)01-0108-06 |
| 修稿时间： | 2005-09-29 |
Convergence of gradient algorithm for BP neural network with multiple output units |
| |
| ZHOU Feng-qi,WU Wei,XIONG Yan. Convergence of gradient algorithm for BP neural network with multiple output units[J]. Journal of Natural Science of Heilongjiang University, 2006, 23(1): 108-113 |
| |
| Authors: | ZHOU Feng-qi WU Wei XIONG Yan |
| |
| Abstract: | Gradient algorithm has been widely used for training the weights of feedforward neural networks. The convergence of the gradient algorithm for feedforward neural networks with one output unit has thoroughly studied. The convergence of gradient algorithm for a three-layer BP neural network with multiple output units is studied. The weak and strong convergence results of the corresponding gradient algorithm and the monotonicity of the error function in the iteration weighted vectors are proved. |
| |
| Keywords: | BP neural network gradient algorithm convergence multiple output units |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
