A fuzzy based decisionbased neural network(FDBNN) based on DBNN(decisionbased neural networks) is proposed in this paper, which introduces conceptions of vigilance and tolerance.
在网络训练中引入置信度和容噪度的概念，提高了网络分类的稳定性，同时克服了（ＤＢＮＮ）在训练样本混有噪声时学习困难和泛化能力不高的缺点。
