中文摘要：

根据Madaline网络工作原理,针对其网络特点和现有算法中存在的缺点,包括存在权值修改公式参数较多不容易协调,经验取值缺乏理论依据不够灵活,按照置信度原则进行翻转神经元会陷入＂局部震荡＂。提出改进的MRII学习算法,通过建立神经元敏感性替代置信度作为度量隐层神经元翻转的尺度,并采用感知机学习规则,减少权值调整次数,实验结果验证了该算法的优越性。

英文摘要：

In this paper, by analyzing the Madaline network work principle, according to the network characteristic and the shortcomings of the existing algorithms including the adjust weigh formula used in MRII algorithm with many parameters, but most of which come from experience in practice without theoretical reason, it made networks fall into ＂local recycle＂ with turning over the neuron based on confidence principle. We present an improved learning algorithm based on MRII, by establishing neuron＇ s sensitivity as a tool for measuring the turn of each hidden neuron, which reduces the number of weight adjustment. Some experimental results verify the effectiveness of the proposed algorithm.