%0 Journal Article
%T 双向联想记忆神经网络抗噪声的能力
%A 杜升之
%A 陈增强
%A 袁著祉
%A 张兴会
%J 自动化学报
%P 668-674
%D 2005
%X ？Thispaperanalyzesnoisesensitivityofbidirectionalassociationmemory(BAM)andshowsthattheanti-noisecapabilityofBAMrelatesnotonlytotheminimumabsolutevalueofnetinputs(MAV),assomeresearchersfound,butalsotothevarianceofweightsassociatedwithsynapseconnections.Infact,itisdeterminedbythequotientofthesetwofactors.Onthisbase,anovellearningalgorithm—smallvarianceleaningforBAM(SVBAM)isproposed,whichistodecreasethevarianceoftheweightsofsynapsematrix.Simulationexperimentsshowthatthealgorithmcandecreasethevarianceofweightsefficiently,therefore,noiseimmunityofBAMisimproved.Atthesametime,perfectrecallofalltrainingpatternpairsstillcanbeguaranteedbythealgorithm.
%K Associativememory
%K neuralnetwork
%K noisesensitivity
%K digitalpattern
%U http://www.aas.net.cn/CN/abstract/abstract16015.shtml