一种基于T-模运算的模糊联想记忆学习算法

提出了基于一类最大T-模运算的模糊联想记忆连接权矩阵神经网络学习算法,并且给出了严格的理论证明,该算法成功地解决了多模糊模式对的存储问题。对于给定的模糊模式对,若这些模式对存在连接权矩阵,则该算法很容易求出它们的最大权矩阵,如果这些模式对不能够用一个连接权矩阵来存储,应用本文的算法,可以应用尽可能少的连接权矩阵来存储,从而可以有效地减少存储空间。

A Learning Algorithm of Fuzzy Associative Memories Based on a Class of T-Norm Operations

Fuzzy associative memories(FAM)associated weight matrix neural networks learning algorithm based on a class of T-norm operations and its strict theoretic proofs are presented in this paper.This algorithm solves the problem of storage for multiple fuzzy pattern pairs successfully.For the given fuzzy pattern pairs,if there is a associated weight matrix to store those pattern pairs,then the maximum weight matrix can be solved by the algorithm,if those pattern pairs can't be stored by a associated weight matrix,then those fuzzy pattern pairs can be encoded to store in FAM associated weight matrixes as few as possible by the algorithm,so it can cut down the storage space greatly.;