基于一类模运算的模糊神经网络的模糊联想记忆

Fuzzy Associative Memories of Fuzzy Neural Networks Based on a Class of Norm Operations

本文通过Ｔ－模建立的一类模糊神经网络推广了ｍａｘ－ｍｉｎ神经网络。对于给定的模糊模式对（Ｘ１，Ｙ１），…，（Ｘｐ，Ｙｐ），我们得到了这类网络能够存储这族模糊模式对的充分条件和必要条件。并将所讨论的方法应用于一类带阈值的ｍａｘ－ｍｉｎ神经网络，证明了这类网络能够存储给定的一族模糊模式对的等价条件。最后用实例验证了我们的结论。

In this paper, we generalize max-min neural networks by constructing a class of fuzzy neural networks with T-norm. For given fuzzy pattern pairs (X 1,Y 1), …,(X p,Y p),we obtain sufficient conditions and necessary conditions which the family of given fuzzy pattern pairs can be stored in the fuzzy neural networks. With the same method, we study the max-min neural networks with thresholds. The equivalent conditions which the family can be stored in the max-min neural networks are shown. Finally, we give examples to illustrate our conclusions.