摘要
泛函网络是最近提出的一种对神经网络的一般化推广。与神经网络不同,它处理的只是一般的实值泛函模型,针对该问题,将实值泛函神经元推广到复值泛函神经元,再对复值泛函神经元的结构作了变形,提出了一种复值泛函网络新模型,给出了基于梯度下降法的复值可分离泛函网络学习算法。采用复分析的方法,利用单一泛函神经元模型,借助于正交边界和实步长函数概念求解复值XOR分类问题。通过理论分析可看出,相比复值神经网络,用复值泛函网络解决问题具有很强的计算能力。
Functional network is extension of neural network, it deals with general real-valued functional model. The structure of functional neuron is changed, and functional neuron is expanded to complex-valued neuron. A kind of separable complex-valued functional network model is proposed,a fully complex separable functional network structure that yields a simplified complex-valued back-propagation algorithm is presented. The XOR problem that cannot be solved with two-layered real-valued neural network can be solved by a single complex-valued functional neuron with the orthogonal decision boundaries, which reveals a potent computational power of complex-valued functional network.
出处
《系统工程与电子技术》
EI
CSCD
北大核心
2006年第8期1244-1248,共5页
Systems Engineering and Electronics
基金
国家自然科学基金(60461001)
广西自然科学基金资助课题(0542048)
关键词
复基函数簇
复值泛函网络
学习算法
复XOR分类
正交边界
实步长函数
complex-valued base function series
complex-valued functional network
learning algorithm
complex XOR classify
orthogonal decision boundaries
real step function
