复合对向－反向传播人工神经网络模型及其应用

A New Artificial Neural Network Model:Combined Counter-Back Propagation and its Application

组合Ｋｏｈｏｎｅｎ竞争学习和反向传播学习的优点，本文首次提出了复合对向－反向传播人工神经网络模型，该模型较好地体现了生物神经网络系统信息处理时的自适应、自组织、分布式存贮及并行处理等特点。它保留了反向传播网络的优点，同时较后者更易收敛，计算时间缩短，网络参数设置也更为自由。通过在临床精液检查结果分析中的成功应用，证明了该系统的有效性和可靠性。

A new artificial neural network model,combined counter-back propagation (CCBP),was developed by combining all virtues of competitive learning and back propagation learning.The characteristics of CCBP model were studied throughly, and it was showed that CCBP represented well adaptation,self-organization and paralled distributed processing of biological neural network in information ptocessing.In addition to its pcossessing all the advantages of BP,it was faster to converge and easier to choose the network parameters for CCBP than for BP. The validity of the system was confirmed through its successful application to the analysis of the data from clinical human semen test.Hence,the CCBP may be identified as one of the useful artificial neural network models.