Computer Engineering and Applications ›› 2011, Vol. 47 ›› Issue (19): 211-213.

• 图形、图像、模式识别 • Previous Articles     Next Articles

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

LIAO Zhou1,XU Weihong1,2,ZHOU Hui1   

  1. 1.College of Computer and Communication Engineering,Changsha University of Science and Technology,Changsha 410114,China
    2.College of Computer Science and Technology,Nanjing University of Science and Technology,Nanjing 210094,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2011-07-01 Published:2011-07-01


廖 洲1,徐蔚鸿1,2,周 辉1   

  1. 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

Abstract: 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.

Key words: fuzzy associative memories, training pattern pairs, T-Norms, perturbation, robustness

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-T FAM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。

关键词: 模糊联想记忆网络, 训练模式对, T-模, 摄动, 鲁棒性