摘要
基于模糊取大算子和三角模T的模糊合成,构建了一类模糊双向自联想记忆网络Max-T FBAM。利用三角模T的伴随蕴涵算子,为这类Max-TFBAM提出了学习算法,并理论上证明了该学习算法确定的连接权矩阵是网络最大的连接权矩阵。对任意输入能使Max-T FBAM迭代一步内就进入稳定态,该类网络具有全局稳定性和可靠的存储能力。当三角模T满足利普希兹条件时,采用上述学习算法时自联想Max-T FBAM对训练模式的摄动全局拥有好的鲁棒性。最后用实验证实了理论研究,也为图像的可靠存储提供了参考。
Based on the fuzzy composition of max operation and any triangular norm T,a type of fuzzy bidirectional auto-associative memory （Max-T FBAM） was proposed. By means of concomitant implication operator of a triangular norm, a general learning algorithm was proposed for a class of such Max-T FBAMs. It was proved theoretically that the learning algorithm can ensure the Max-T FBAMs have maximal pair of connection weight matrices, which can be convergent to an equilibrium state in one iterative process for any input, and have reliable store capabilities. When triangular norms satisfy Lipschitz condition, Max-T FBAMs have good robustness of small perturbations of training pattern pairs by the learning algorithm. Finally the experiment is given to not only testify the theoretical research, but also provide a reference for reliable storing images.
出处
《计算机科学》
CSCD
北大核心
2009年第2期238-240,267,共4页
Computer Science
基金
国家自然科学基金项目(No.60632050)
湖南省教育厅科研基金项目(No.07C522)
湖南省自然科学基金项目(No.05JJ40004)的资助
关键词
三角模
模糊双向联想记忆网络
学习算法
稳定性
鲁棒性
Triangular norm, Fuzzy bidirectional associative memory, Learning algorithm, Stability, Robustness
