简介：

简介：

简介：

简介：

简介：ProfessorKowk-faiSo,theeditor-in-chiefofNeuralRegenerationResearch,hasbeennamedaFellowoftheNationalAcademyofInventors(NAI)ProfessorKwok-faiSo,DepartmentofOphthalmology,LiKaShing,FacultyofMedicine,TheUniversityofHongKong(HKU),hasbeennamedaFellowoftheNationalAcademyofInventors(NAI).

简介：

简介：

简介：Thispaperproposesanewneuralfuzzyinferencesystemthatmainlyconsistsoffourparts.Thefirstpartisabouthowtouseneuralnetworktoexpresstherelationwithinafuzzyrule.Thesecondpartisthesimplificationofthefirstpart,andexperimentsshowthatthesesimplificationswork.Onthecontrarytothesecondpart,thethirdpartistheenhancementofthefirstpartanditcanbeusedwhenthefirstpartcannotworkverywellinthefuzzyinferencealgorithm,whichwouldbeintroducedinthefourthpart.Finally,thefourthpart"neuralfuzzyinferencealgorithm"isbeenintroduced.Itcaninferencethenewmembershipfunctionoftheoutputbasedonpreviousfuzzyrules.Theaccuracyofthefuzzyinferencealgorithmisdependentonneuralnetworkgeneralizationability.Evenifthegeneralizationabilityoftheneuralnetworkweusedisgood,westillgetinaccurateresultssincethenewcomingrulemaynotberelatedtoanyofthepreviousrules.Experimentsshowthisalgorithmissuccessfulinsituationswhichsatisfytheseconditions.

简介：ThetypicalBDI(beliefdesireintention)modelofagentisnotefficientlycomputableandthestrictlogicexpressionisnoteasilyapplicabletotheAUV(autonomousunderwatervehicle)domainwithuncertainties.Inthispaper,anAUVfuzzyneuralBDImodelisproposed.Themodelisafuzzyneuralnetworkcomposedoffivelayers:input(beliefsanddesires),fuzzification,commitment,fuzzyintention,anddefuzzificationlayer.Inthemodel,thefuzzycommitmentrulesandneuralnetworkarecombinedtoformintentionsfrombeliefsanddesires.ThemodelisdemonstratedbysolvingPEG(pursuit-evasiongame),andthesimulationresultissatisfactory.

简介：Inthispaper,weproposetwoweightedlearningmethodsfortheconstructionofsinglehiddenlayerfeedforwardneuralnetworks.Bothmethodsincorporateweightedleastsquares.Ourideaistoallowthetraininginstancesnearertothequerytoofferbiggercontributionstotheestimatedoutput.Byminimizingtheweightedmeansquareerrorfunction,optimalnetworkscanbeobtained.Theresultsofanumberofexperimentsdemonstratetheeffectivenessofourproposedmethods.

简介：

简介：Inthispaper,weintroduceatypeofapproximationoperatorsofneuralnetworkswithsigmodalfunctionsoncompactintervals,andobtainthepointwiseanduniformestimatesoftheapproximation.Toimprovetheapproximationrate,wefurtherintroduceatypeofcombinationsofneuralnetworks.Moreover,weshowthatthederivativesoffunctionscanalsobesimultaneouslyapproximatedbythederivativesofthecombinations.Wealsoapplyourmethodtoconstructapproximationoperatorsofneuralnetworkswithsigmodalfunctionsoninfiniteintervals.

简介：

简介：ＷｅｌｄｑｕａｌｉｔｙｃｏｎｔｒｏｌｂｙｎｅｕｒａｌｎｅｔｗｏｒｋＨｕａｎｇＳｈｉｓｈｅｎｇ；ＬｉＤｉａｎｄＳｏｎｇＹｏｎｇｌｕｎ（ＳｏｕｔｈＣｈｉｎａＵｎｉｖｅｒｓｉｔｙｏｆＴｅｃｈｎｏｌｏｇｙ）Ａｂｓｔｒａｃｔ：Ａｒｔｉｆｉｃｉａｌｎｅｕｒａｌｎ...

简介：

简介：NeuralRegenerationResearch(NRR)isafirst-classinternationalacademicjournalpublishedinEnglish.ItissupervisedbytheMinistryofHealth,P.R.China.NRR(CN11-5422/R,ISSN1675-5374)isamonthlyjournal,distributedallovertheworld.

简介：

简介：

简介：

简介：Basedoncurrentresearchonapplicationsofchaoticneuronnetworkforinformationprocessing,thestabilityandconvergenceofchaoticneuronnetworkareprovedfromtheviewpointofenergyfunction.Moreover,anewauto-associativematrixisdevisedforartificialneuralnetworkcomposedofchaoticneurons,thus,animprovedchaoticneuronnetworkforassociativememoryisbuiltup.Finally,theassociativerecallingprocessofthenetworkisanalyzedindetailandexplanationsofimprovementaregiven.