摘要
本文改进了Ｋｏｓｋｏ．Ｂ提出的最大最小联想记忆网络的模糊Ｈｅｂｂ关系编码规则，给出了一种新的学习算法，新算法克服了Ｋｏｓｋｏ．Ｂ算法的缺陷，在一定条件下，本文的学习算法能将模式对完整地联想出来．另外，本文在分析网络的容错性及稳定性的基础上，提出了一种五层混合模糊联想记忆网络，五层混合网络具有良好的联想容错能力．实验结果表明，本文的学习算法及混合网络是有效的．
This paper gives a new kind of weight learning algorithm for the fuzzy associative memory proposed in paper. The new learning algorithm overcomes some disadvantages of the fuzzy Hebb learning rule described in paper , and it can recall perfectly all multiple pattern pairs with some conditions. Based on the analysis of the learning algorithm about the stability and error_correction capability,a kind of compound five_layer fuzzy neural network model is designed in this paper, which can enlarge efficiently the attraction basins of memory patterns. The performance of the learning algorithm and the compound model is reported and compared with that of several associative memories through numerous examples.
出处
《华南理工大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
1998年第6期1-7,共7页
Journal of South China University of Technology(Natural Science Edition)
基金
国家自然科学基金
攀登计划资助项目
关键词
神经网络
模糊联想记忆
学习算法
吸引域
neural network
fuzzy associative memory
learning algorithm
basins of attraction
