论文部分内容阅读
本文分析了一种新的神经元模型──二次Sigmoidal神经元对前向种经网络的分类能力的改进程度．结果表明：在隐层及输出层无论采用多阈值还是单阈值二次Sigmoidal神经元，三层前向网络的分类能力比CommittedMachine只能提高3倍．
This paper analyzes a new neuron model ─ ─ quadratic Sigmoidal neurons to improve the classification of the ability to forward the network classification. The results show that the classification ability of the three-layer forward network can only be increased by 3 times than that of Committed Machine in both the hidden layer and the output layer, no matter using multi-threshold or single-threshold quadratic Sigmoidal neurons.