关于广义同余神经网络结构与算法的研究

讨论了一种新的神经网络－广义同余神经网络（ＧＣＮＮ）网络中的所有神经细胞以同余的形式被激活，并阐述了它的结构，学习规则的一种算法，模拟结果表明，这种网络在取样点上具有满意的通用特性，由于该神经网络易于操作和实行，有必要就相关理论及其应用作进一步的研究。

The Research on the Construction and Algorithms of Generalized Congruence Neural Network

In thispaper, a novel class of neural networks called generalized congruence neural network(GCNN)is proposed. All neurons in the neural networks are activated in the form of congruence.The architectures, learning rules and one algorithms are presented. Simulation resultsindicate that such network has satisfactory generalization properties near the samplepoints. Since this kind of neural nets can be easily operated and implemented, it isappropriated to make further research concerning the theory and applications of GCNN.