模糊联想记忆对训练模式对摄动的鲁棒性研究  

Robustness research of fuzzy associative memories with perturbation of training pattern pairs

作　　者：廖洲[1] 徐蔚鸿[1,2] 周辉[1] 

机构地区：[1]长沙理工大学计算机与通信工程学院,长沙410114 [2]南京理工大学计算机科学与技术学院,南京210094

出　　处：《计算机工程与应用》2011年第19期211-213,共3页Computer Engineering and Applications

基　　金：国家教育部重点科研基金No.208098;湖南省教育厅重点科研基金(No.07A056)~~

摘　　要：基于模糊取大算子(V)和T-模的模糊合成,构建了一类模糊联想记忆网络(V-T FAM)。利用T-模的模糊蕴涵算子,给出了这类V-T FAM的学习算法。针对训练模式对小幅摄动可能对模糊神经网络的性能产生副作用,提出V-T FAM对训练模式对摄动的鲁棒性概念。理论研究表明,当T-模满足Lipschitz条件时,采用上述学习算法的V-TFAM对训练模式对摄动幅度,在系数为β的条件下全局拥有好的鲁棒性。最后用V-T FAM在图像联想方面的实验验证了理论结果。This paper sets up a class of fuzzy associative memories based on the fuzzy composition of max operation（V） and T-Norms,so be called V-T FAM（Fuzzy Associative Memory）.With the fuzzy implication operator of T-Norms,a general learning algorithm is proposed for a class of such V-T FAMs.Since small perturbations of training pattern pairs may cause some disadvantages to performance of a fuzzy neural network,a new concept is established for the robustness of V-T FAMs to perturbations of training pattern pairs.The theoretical researches show that when T-Norms satisfy Lipschitz condition,V-T FAMs have good robustness under the condition of the perturbation factor of β of training pattern pairs by the proposed learning algorithm.Finally,the experiment with which the V-T FAM associated an image with another image is given to testify the theoretical results.

关 键 词：模糊联想记忆网络 训练模式对 T-模 摄动 鲁棒性 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]