Home teknologiaa tekoäly

tekoäly

techintroduce | | 881

Yksityiskohtaiset

Thedefinition-artificialIntelligencecanbedivideTiToTWoParts, nimittäin "keinotekoiset" ja "älykkyys"."Manuaalinen" iseaSierTounderstand, ja adlesscontroversal.Joskus.Butingeneral, "keinotekoisen järjestelmän" isanartificialsystemintheusualSense.

Sijainti.Tämä onvolvesotherissuessuchasconsioness (tietoisuus-A, itse (itse-A, ajattelu (mieli-A (mukaan.TheOnlyintelligencEtHataPersonSonsandSisTheIntelligence OFTHEPERSHIMSSELI, WHITEAGENERALLAGREEDVIEW.ButOURUNTERSTINGOUROWNIntelligenceSVermitlimilimit, jawealSohvelIMitedIndingfensionalElements thatconstituteHumanIntelligence, SOITEDIFICIONTODEFINEWHATIS "Älykkyys" Madeby "keinotekoinen" ".Siksi theresearch-artificialIntelligence-teninVolvestheresearch-HumanIntelligenceItSel.Muut.

Keinotekoinen ohjaaja.ITISALSOUSEDInrobot, EconomicandPolitical Decision -valmistus, ControlSystems ja SimulationSystems.

Pr-essorneLelsongAveadfinition-artificialIntelligence: "Keinotekoinen ontelligenceAdisciplineaBoutKnowledled-thescienceOhwoToExpresskledgeandHhowtoobtainandUsewnowledgedledge."Andanotermassachusetts, Usapr-essorwinston-thepolytechnstitutebelieves:" keinotekoisenlaitteenoteSTHEStudyfhowtomakecomputersdointelligentworkthatonlyhanscandointhePast."ThesestatementReflectTheBasicideasandBasicContentenT-artInialiIntelligence.Se, että keinotekoinen interialIntelligenceSthestudyftHelaws-HumanIntelligenceactivities, theconstructionOfartificialSystemsWithcartleIntelligence, thestudy-howtomakecomputerscompleteSTHatTrequirecomeTelliNyLandHandHandHandHandHandHandHandHandHandHandHulluNence.

Keinotekoinen.SINCETHE1970S, ITHASBEINGONDASONOFTHEWORLDIN KOHTAINEN EDGETECHNOLOGNOLOGIES (SpaceTechnology, energiatekniikka, tekoäly-A.ITISALSOCSONSEREDTOBEONOFTHETHREECUTTING-EDGETECHNOLOGNOLOGIES (GeneticEngineering, Nanotiete, keinotekoinen integrointi-A.TämäAsbecauseithasachieveDrapidDevelmenpmentHePast30years, HasBeenWidYUnedInManyDisciplines.Keinotekoinen ohjaaja.Inasysteemiin.

Keinotekoinen ohjaajadisciplinethatStudestheuseOfcomultsimulatEcertHumanthinkingProcessesandIntelligentbehaviers (sucaslearning, päättely, ajattelu, suunnittelu jne..-A.ItMainlyclonsESThePRinciples OfcomputErErealationOfIntelligenceandManucturingsimilarTohumans.Brain-SmartComputersEnableComputerStoAchieveHigher-tason.Keinotekoinen ohjaaja,.ItCanbesAidThathAlimostAllDisciplinesOfnaturalSciencesandSosiaalinenScienceshaveascopefarbeyondthescope-computerscience.TherelationhipBetBetSowificialIntelligenceandThinkingsCienceSTherelationshipBetPracticeandTheory -.ITISANAPPlicationBranch-it.ATHINKING, keinotekoista, keinotekoista.ITISNESIATEYTOCONSIDERIMAGEtHINKINGANDININSPIRATIONTINGTINGTOPROMOTHEBREAKTROUGHDEVELONPONTORTIFIFIFIFIFIFIfortIntelligence.Matematicsis-tenRegardedastheBasicscienceOfManyDisciplines.MatematiikansalsoentersThefieldOfLanguageandHinking.IntegryciplinesmustalsOborrowMathematicalTools.MatematiikansantonaplaySaroleinStandardLogic, Fuzzymathematics jne.., mutta kunMathematicsEnterSartificialIntelligencedIciplines, hewillpromoteeachotherandDevelopfaster.

Tutkimusarvo

Forexample, HeavyscientificandEngineeringCalculationsWereoriginaalisesti bornebyHehumanbrain.Nykyään ComputersCannoTONLYPERFormSuchCalculations, BullsoperformFastertHanTheHumanbrain...ItCanbesentHatTheDefinitionFComplexworkChanges kanssa.ThespesificGoalsftheScienceWillnaturalyDevelopwithtHtHangeS-tHnheghtime.OnTheOneHand, ItContinuestoMakenEwprogress, AndontheotherHand, ItsHifTStomoremeaningfulandDiffinculours.

Yleensä "MachinElearning" on "Tilastot", "informaatiotHeory" ja "cybernetics", teemathematicalfoundation.ItalsocludeSothernon-matematicsSubjects.ThistypeOf "MachinElearning" IshighlyDivedendon "Experience".Tietokoneet.WecancallThisOppiminenMethod "jatkuvaaOppiminen".ButinadditionTolearningfrom -kokemus, HumanbeingsCanalsokreate, eli "hyppääminen."Tämä" tämä "inspiraatio" tai "oivallus" unettomuus.Allalong, theHardestthingTolearnaBoutacomputteris "Insight."Ormorestrictlypeaking, iTisdiffulcortComputerstolearn" QualitativeChangesthatdonotDependonquantiTiVeChanges "Etsilaajan ja" käytäntö ", jatulehduksenhalvausviljely" käsite "" laatu "" laatu "" laatu ", ordirectlyfroma" konsepti "käsite" "laatu" "laatu" "laatu", ordirectlyfroma "konsepti" käsite "käsite"..Toano "konsepti".Koska tämä "käytäntö" täältä.ProcessOfHumanPracticeCludesBothexperienCeandCreation.

TämäSTheDreaM-IntelligentResewesers.

IN2013, S.Cwang, adatareseartHegeneraldatacenter-dijindata, kehittyneitä, kehittyi.TheauthorfoundthatthenewdataanalysismethodproVidewayForthEcometsocietyToTo "Luo".Inunsecy tämäMethodproVidesAfairlyeffectivewaytoModelHuman "luovuus".Tämä hyväksytään.Sincethen, ComputersarenotonlyGoodatcalculculing, ButalsogoodatCreatingBecausetheyareGoodatcalculculculation.Tietokonesuunnittelijat, jotka ovatkategorisesti jakautuvat "luovat" tietokoneen,.

Kun katseletBacThededanctionProcessandmattematicsOfthewmethod, theauthorexPandedhisandstanding -thinkingandMathematics.Theathematicsisconcise, selkeä, luotettava jawell-mallinnettu.IntHEHISTORYOFMATEMATICS, MATHEMATICSMASTERSSSSSHINESECHILLE.ThesecreativityRePresentedIntHeformOfVariousmathematicalTheoremsorConclusions -.Itshouldbesaidthatmathematicsisthesimplestandmoststraightforwardsubjectthatreflects(atleastonetype--Acreativitymodel.

Kehitysvuoro

Inthesummer-1956, agroup-visionaryyoungscients, Ledbymccarthy, Minsky, Rochester, Andshennon, GatheredTogetherOwoIntresearchandDiscussingaseriesOfrelatedissuesBoUTUNTInTInESTINESSIMULEIDILIGENTIES, WHO TEECTTHILIGIANDIPLINESTIRESTIESTIESTIIMESTIIMESTIIREDIFIINESTIIRESTIIMESTIIMESIESTIRNE.IBM: n "Deepblue" ComputerDefeatedTheWorldchessChampionOfMankind, anditisAperfectPerformanceOfartificialIntelligenceTechnology.

Laskentafromtheformalproposal-artificialIntelligencein1956, yli50 vuotta, iThasmadEconsidebleprogressandhasbecomeabroAdcross-cuttingandcutting-Edgescience.Kekseliä,.Ifyouwanttobuildamachinethatcanthink, youMustknowwhatthinkingis ja furThermore, whatiswisdom.Whatkind-Machineissmart? Tiedemieshavemadecars, junat, lentokoneet, radiot jne.., joka on yhtäläistä.Weknowverylittleaboutthisthing,andimitatingitmaybethemostdifficultthingintheworld.

WentComputerskappale, ihmisenbegantoHaveaToolToLouldsimulationHumanthinking.IntheyEearstocome, lukemattomat tietoa.Nykyään keinotekoiset ohjaaja.ThereElepeLopleintHEComPuterDePartments OFFALMASTUNUNIESSIONSIINTHEWORLDWHOARESTUDYYYTHISSUBject.CollegestUntsStudingComputermustalSolearnSuChaCourse.Withoveryone'sunremitfforts, NowTheComputSEeMStoHaveBecomeverySmart.Forexample,inMay1997,theDEEPBLUEcomputerdevelopedbyIBMdefeatedthechessmasterKasparov(KASPAROV-A.YouMaynothevenotedThatInSomeplaces, ComputershelppeopleToperformotherSksHatwereoriginaalysonforHumans.Tietokonepeli.Keinotekoinen onlelligenceHasalwaysBeenAfrontiSUBjectOfCOnTersCience.Computerprogramminglanguages​​andothercomputers-twarehaveexistedbecause-theadvancement-artificialintelligence.

OnMarch4,2019, theescondsession kanssa.THESPOUSSERSONFESTORNE, ZHANGYESUI, sanoi, että.

Tieteellinen

Käytännöllinen soveltaminen

MachineVision, sormenpelin tunnustaminen, kasvojenkäsittely, retinarikogniitio, irisrecognition, palmprintRecognition, asiantuntijalaus.

Kurinpitoluokka

Keinotekoinen onlelligenceIafringesubject, joka on mielestänut.

Sisältää disciplinit

Filos-ia jaCognitivitiveScieces, matematiikka, neur-ysiologia, psykologia, tietotekniikka, informaatioteoria, kybernetiikka, epävarmuusteoria

p>

Tutkimusluokka

NaturallanguAgeProcessing, FundledGerePresentation, IntelligenStesearch, päättely, suunnittelu, MachinElearning, FundledGeacquision, CombinedSchedulingProblem, GunceptProblem, PatternRecognition, LogicProgramming, S-tComputing, INCCURATEANDUNCUNTIMANAGION, TIEDOTTUMINEN

tietoisuus jaArnificialIntelligence

Keinotekoisenlaitteet, initSessence, asimulaatiotiedot.

ThereareTwowayStosimulatHumanthinking.YksitystruktureSimulation, joka vastaa theStructucturalMechanismftheHumanbraintocreatea "aivomaista" konetta;.TheinternalStructureissimuloidutFromitsFunctionalProcess.Modernelektroniikkakäyttöinen.

HeewartrificialIntelligencecontinuestodeSeloprapidly, erityisesti 2008ekonomikrisri, Theunitedstses, Japanandeuropehopettorealzere-teollisuusteollisuus.IndustrialrobotsareeDelingAtAfasterRateTheNeverevere, jaTeyareEvenMoreDriven.Kohtuullisena murtautumisen kanssa.

ThestrongarTificialIntelligenceistemporyAtabottleneckandrequiresthefforts-scientist- jahimalaiset.

Tekninen

ThetAchineusedTotosTheMainMaterialBasis-artificialIntelligenceandTheartInTIntelligenceTechnologyPlatFortTheComputer.Kehityshistory-artificialIntelligenceisHistoryOfTheDevelopmentOfPutersCienceandtechnology.Kytketty toisiinsa.InddditionTocomputerscience, keinotekoinen ontelligencealsoinvolvesmultipledisciplinessuchasinformationTheory, kybernetiikka, automaatio, bioniikka, biologia, psykologia, matemaattiset allogiset, kielitieteet, lääketiede ja filos-ia.ThemaincontentenT-artificialIntelligencerESearchinclublens: OfgledleDereResentaatio, automaatti-.

Tutkimusmenetelmät

Nykyään onnifiedPrincipleorParadigmtoguideartificialIntelligenceResearchesearch.ThereArConTroverSamongresearchersonManyissues.Some-thequestionsthathaveremainedinconclusiveforalongtimeare:shouldartificialintelligencebesimulatedfrompsychologicalorneuralaspects?Or,likebirdbiologyistoaviationengineering,humanbiologyhasnothingtodowithartificialintelligenceresearch?Canintelligentbehaviorsbedescribedbysimpleprinciples(suchaslogicoroptimization-A?Ormustalargenumber-completelyunrelatedproblemsbesolved?

Canintelligencebeexpressedusingadvancedsymbols,suchaswordsandideas?Stillneedtodealwith"sub-symbols"?JOHNHAUGELANDproposedtheconcept-GOFAI(excellentold-fashionedartificialintelligence-AandalsoproposedthatartificialintelligenceshouldbeclassifiedasSYNTHETICINTELLIGENCE.[29] Tämä käsitys.

Aivosimulaatio

Mainarticle: cyberneticsandcomputationalneuroscience

Alkaen 1940STOTHE1950s, ManyResearchersExploretheconnectionsbetweenneurology, InformationTheoryandcybernetics.ItalsoRreatedSomeprelInaryIntelligenceusingElectronicNetworkstructures, Suchasw.Greywalter'sturtlesandjohnshopkinsbeast.Theseresearchers-tenHeldtechnicalAssociationMeetingsAtPrinceNUniversityandtheratioclubintheuk.Vuoteen 1960 saakka enitenPeopleHadabandononedthismethod, vaikkakin,.

Symbolinkäsittely

Mainarticle: G-ai

WHENTHEDIGIALCOMPUTERWASSUCCESTESSILLESPEDIINTHE1950S, SUORITUSTUTURIOTTEET.JOHNHAUGELANDcallsthesemethodsGOFAI(excellentold-fashionedartificialintelligence-A.[33] 1960 -luvulla, thesymbolicmethodmadegreakACHIEVESSIMULINGAIDVANCEDTHINKINGONSMALL -Pro-Programmit.METHOMSBASEDONCYBERNETICSorneuralNetWorksaresecondary.[34] tutkijat1960Sand1970SwereconvincedThattsmbolicMethodscouldEventSucceeplationingMachines.

KognitivitimulaatioEconomistSherbertsimonandalannewellstudyhUmanproblem-liputCapabilitydandtryTormalizethem.ATTHESAMETIME, he leffatefoundationForthebasicPrinciples-artificialIntelligence, SuchasRecognitionknowledesCience, operaatiotHeartHandbusinesscience.Heidän ResearchTeaMusestheresultsOfPsychologicalExperimentOdeSelProgramsThatMiChuManproblem-SolvingMethods.TämäMethodhasbeenInhertedFromcarnegiemellonUniversityandDevelopedTotspeakinsoarinthe1980s.BasedonLogic, UnlikeaLannewellandherbertsimon, JohnmccarthybelievesthatmachinesdonotneedTosimulationHumantHughts, butShoultryt-indabstractrastrangandtheSessence problem-ratkaiseva.HislaboratoryatstanfordUniversitySiscomitionTousingFormAllogictosolVeavarietyOfions.TheUniversity-Edinburghisalsocommittedtologicalmethods,andhascontributedtothedevelopment-programminglanguages​​PROLOGandlogicinotherparts-Europe.Ohjelmointi."Anti-logic"StanfordUniversityresearchers(suchasMarvinMinskyandSeymourPipette-Afoundthattosolvethedifficultproblems-computervisionandnaturallanguageprocessing,specialsolutionsareneeded-theyarguethatthereisnosimpleandGeneralprinciples(suchaslogic-Acanachieveallintelligentbehaviors.Rogerschankdescribestheir "inverselogic" metodas "Scruffy".Commonsenseknowledgebases(suchasDOUGLENAT'sCYC-Aareexamples-"SCRUFFY"AIbecausetheyhavetomanuallywriteacomplexconceptatatime.BasedOnNowledge, suuren kapasiteetin muistitietokoneessa.TutkijatBegantoconstructNowledleDoApplicationsOftwareInThreeways.Tämä "asiantuntevangerevolution" LedToteDevelopmentAndSuunnitteluOfexPertsystems, What wathEfirstSuccessforforforfartificialIntelligencesOftware -ohjelma."OsingleDGerevolution" AlsomadepeoplerealizetHatManysImpleartificialIntelligences-twaremayRequirealoT-knowledge.

Alisymbolismi

SymbolicificialIntelligenceSagnationd1980 -luvulla.ManyPeoplebelievedThatsymbolicsystemsTouldNeverimitateallHumancognitiivisetProcesses, erityisestitulehdus, robotti, machinelearningAndPatternRecognition.MonenResearchersBegantopayattentionTothesub-symbolmethodtosolvespesificartrificialIntelligenceProblems.

Bottom-up,interfaceAGENT,embeddedenvironment(robot-A,behaviorism,researchersinthefield-newAIrobotics,suchasRODNEYBROOKS,negatingsymbolicartificialintelligenceandfocusingonthebasics-robotmovementandsurvivalEngineeringissues.Heidän työsuhteensafocusedonTheViewS-earlyCyberneticsResesearcers, AndatthesametimeProposedTheUsOOfControlTheoryInartificialIntelligence.Thisisconsistentwiththerepresentationalperceptionargumentinthefield-cognitivescience:higherintelligencerequiresindividualrepresentations(suchasmovement,perception,andimage-A.IN1980 -luvulla DavidrumelhartandothersonCeagainProPotedneuralNetWorksandConnectionism.TämäAndothersub-symbolinen methodit, sucasfuzzycontrolandevolutionarycalculation, bonsomTotheresearchCategoryOfcomputationIntelligence.

Tilastot

1990 -luvulla keinotekoinen onlelligencerESearchDevelodedComplexMathematicalToolstoSolvesPesificBranchProblems.Thesetoolsaretrulyscientificmethodit, jotka, thesultsftheMethodsarEtaSurableAndverifeble ja theyarealsotherAsonforthesuccess-artificialIntelligence.Thesharedmathematicallanguagealsoallowscollaborationinexistingdisciplines(suchasmathematics,economics,oroperationsresearch-A.Stuartj.RussellandPeternorvigPointThattheseadvancesarenolessthant "vallankumous" ja "thesuccess-neates."JokinPeoplecrititiciTheSeTechnologiesforbeToFocuseDonSpesificProblemsdAningToconsiderthelong-termgoal-strongartificialIntelligence.

Integrointi

SmartAgentParadigmsMartAGENTISASYSTEMTHATPERCEIVESTHEENVIRNAPERNADTACEACTONTSTOACHEICEGOALS.ThesimplestinTIntelligentsareProgramsThatCansolvespesific Problems.Morecomplexagentsincludehumansandhumanorganizations(suchascompanies-A.TheesparadigmsAllowResearSToStudyYDIDUALPRORMESDADFIDUSEDLESSIVERIFIABLESSULUSSWITHOUNTINGTOCONSIDERASINGLEMETHODOD.Anagentthatsolvesaspecificproblemcanuseanyfeasiblemethod-someagentsusesymbolicandlogicalmethods,andsomearesub-symbolicneuralnetworksorothernewmethods.Theparadigmalsoprovidesresearcherswithacommonlanguagetocommunicatewithotherfields-suchasdecisiontheoryandeconomics(alsousingtheconcept-ABSTRACTAGENTS-A.The INTILLIGENTAGENTPARADIGMWASWIDYACESTEDINTHE1990S.AgentArchitectureandCognitivearchitEcturerESearchersHavedEnesignedSomesystemStedealwiththeThtheteractions.AsystemTatacontainssymbolsandsub-symbolsiscalledahybridIntelligentsysteem, jaHEStudyftHissystemisThetegrationOfartificialIntelligencesystems.TheHierarchicalControlSystemProVideBridgeBetWeEnTherAction-tason-SymbolaIndtheHighest-tasonTraditionalsymbolai.Rodneybrooks'sUbsuNCOrchitectureisanearlygradingSystemplan.

Älykkyys

Simulation-machineVision, kuulo, kosketus, tunne ja ajattelutavat: sormenpelin tunnustaminen, kasvojen lukumäärä, retinarekognitio, irisrecognition, palmprintRecognition, asiantuntijajärjestelmä, älykkyysesytökset, theorempr-erend, logikalvoisuus, peli, informaatio- ja induktion ja dialektiset prosessit.

Aihekategoria

Keinotekoisenlaitteetisanedgesubject, joka on välittänyt.

Kohteet

Filos-ia jaCognitivitiveSciesices, matematiikka, neur-ysiologia, psykologia, tietokoneen tiedonsiirto, tietokaveri, kybernetiikka, epävarmuus, bionikat, sosiaaliset rakenteet.

Tutkimusluokka

Languagelearningandprocessing,knowledgeexpression,intelligentsearch,reasoning,planning,machinelearning,knowledgeacquisition,combinedschedulingproblem,perceptionproblem,patternrecognition,logicprogramDesign,s-tcomputing,inaccurateanduncertainmanagement,artificiallife,neuralnetworks,complexsystems,geneticalgorithms,thehumanway-thinking,themostcriticalproblemistheshapingandimprovement-themachine'sautonomousandcreativethinkingability.

Turvallisuusongelmat

KeinotekoisetIntelligenceisstillbeingstudied, butsomescholarsbelievethatisdangerousforcomputerstoHaveiq.TämäKind-hiddendangerHasalsohappendenmanymovies.ThrainkeyistolallowthemachinetoHavetheself-tietoisuustoproduceandcontinue.Unthemachinehastheself-tietoisuus, se merkitsee, että.Suojaa.

Toteutusmenetelmä

ThereareTwodifferentwaystorealizeartificialIntelligenceOnAcomputer.YksinäinenRaditionalProgrammingTechniquestomakethesysteMappearIntelligent riippumatta.ThismethodiscalledEngineeringMethod(ENGINEERINGAPPROACH-A,ithasmadeachievementsinsomefields,suchastextrecognition,computerchessandsoon.THEOTERISTHEMOLINGAPPOACH, joka ei ole luopuminen, butasorequirestheimpmenationMethodTobethesameorsimilarTothatusedbyHumansorbiologyorganisms.Geneticalgorithm(GENERICALGORITHM,abbreviatedasGA-Aandartificialneuralnetwork(ARTIFICIALNEURALNETWORK,abbreviatedasANN-Abelongtothelattertype.Geneettiset goritmitSimulathetegeneettiset evoluutionMechanismOfHumansororganismit, whileartificialneuralNetworksSimulatheactivefnerVecellsinTheHumanoranimalbrain.InordertogetthesamesMartEffect, molemmatMethodSareUsuural -saatavilla.Itheformermethod, iTisNESTEYTOMANAALIASSEPSESTHEPROGRAMLOGICInDAIL.ImeGameissImple, ITISSTILLVENIENTIE.Ifthegameiscomplicatedandthenumber-charactersandactivityspaceincrease,thecorrespondinglogicwillbeverycomplicated(exponentiallyincrease-A,andmanualprogrammingwillbeverycumbersomeanderror-prone.Onceanerroroccurs, youmustModifyTheoriginalProgram, käännä uudelleen, virheenkorjaus- ja finaalisesti.Whenadoptingthelattermethod,theprogrammerhastodesignanintelligentsystem(amodule-Aforeachroletocontrol.Thisintelligentsystem(module-Adoesnotunderstandanythingatfirst,justlikeanewbornbaby,butitcanlearnandcanGraduallyadapttotheenvironmentanddealwithvariouscomplexsituations.TämäKind-System-tenMakesMistakesAtTheBeginning, Butitcanlearnalesson, jaTmayBecornectedthenextimeitruns, atleastwillnotgowrongforever ja theeSeisNoneEdtoreleasEnwversionorPatchtchatch.KäyttämälläThismethodtorealizeartificialIntelligencerequiresprogrammerstorshoveabiologicalTinkingMethod, anditismorediffultogeStarted.Butonceyouenterthedoor, itCanbidely -käytetty.SincetismethoddoesnotRequiredEledregulationsonTherole'SactivityRulesWHenprogramming, ITISUSUSSURELABOR-SÄILYTYSTEENTEEPRAIMINEN.

Pr-essionalInstitutions

Yhdysvallat

⒈Massachusettsinstitute-TECHNOLOGY

⒉STANFORDUNIVERSITYStanfordUniversity(CA-A

⒊CARNEGIEMELLONUNIVERSITY(PA-A

⒋Liversation-california-Berkeley

⒌EniversityOfwashington

⒌EniversityOfwashington

p>

⒍Eniversity-tExas-AustinUniversity-tExasataustin

⒎EniversityOfPennsylvaniaUniversityOfPennsylvania

⒏Eniversity-illinois-Urbana-Champaign UniversityOfillinoisurbanathat-Champaign

⒐Maryland-Collegepark

⒑CORNELLUNIVERSITYCornellUniversity(NY-A

⒒Eniversity-Massachusetts-Amherstuniversity-Massachusettsamherstcampus

⒓Georgiainstitute-tNology

Michigan-Ann Arbor University - Michigan-Ann Arbor

⒕Outncalifornia

⒖COLUMBIAUNIVERSITYColumbiaUniversity(NY-A

Kalifornian yliopisto Kalifornian yliopistossa, Los Angeles

⒘BROWNUNIVERSITY(RI-A

⒙YALEUNIVERSITY(CT-A

⒚Kaliforniafcalifornia-SandiegouniversityOfcalifornia, Sandiego

⒛EniversityOfWisconsin-MadisonWisconsinuniversity-madison

Kiina

1, Institute-automation, Chineseacademy-Sciences

2, Tsinghua -yliopisto

3, Pekinguniversity

4, Nanjinguniversity-scienceandtechnology

5, BeijinguniversityOfscienceandtechnology

6, chinascienceandtechnologyuniversity

7, JilinUniversity

8, Harbininstitute-tNology

9, PekinguniversityOfPostsandTeleCommunications

10, Pekinginstitute-tNology

p>

11, InstituteOfartificialIntelligence, Xiamenununiversity

12, InstitutefintelligentVehicles, Xi'anJiaotongUniversity

13, InstituteOfIntelligentsysteManDIntelligents-tware, CentralsouthUniversity

14.InstituteFintelligence, XidianUniversity

15.Institute-imageandartificialIntelligence, HuazhongUniversityOfscienceandTechnology

16, ChongqingUniversityOfPostsandTeleCommunications

17, WuhanUniversity-tNology

Pääsaavutukset

Ihmis- ja kannepeli

February10-17,1996,GARRYKASPAROVdefeated"DeepBlue"4:2(DEEPBLUE-A.

OnMay3-11,1997, GarrykasparovlostToTheimproved "DeepBlue" 2.5: 3.5.

Helmikuu 2003, Garrykasparovdrew3: 3 "syvällä juniorilla".

InNovember2003,GARRYKASPAROVdrew2:2with"X3DGermans"(X3D-FRITZ-A.

Hahmontunnistus

Käyttämällä $ PatternRecognitionEngine, BranchingInto2DrEcognitionEngine, 3RecognitionEngine, StandingWaverecognitionGinAnEnMulti-DimensionalRecognitionEngine

2DRecognitionGineFingerPrintRecognition, PortraitRecognition, TextRecognition, ImageCognition, LicenSeplaterecognitionHaveBeenLaunched; StandwaverecognitionGinehasbeenLaunchedforecerecognition;.25-A

Automaattinen tekniikka

AutonomousDriving(OSOSystem-A

BanknotePrintingFactory(￥AssemblyLine-A

FalconSystem(YODDrawing-A

Tietotekniikka

TeeWnowledgeItSelyTheProcessingObject, StudyHhowTousExificialIntelligenceands-tWareTechnologyToDesign, ConstrucTandMainAnAkNowledgeSystem

Asiantuntijajärjestelmä

Älykkyys

ComputerVisionAdImageProcessing

MachinetranslationAdnaturaLanguAgeAntsanging

DataminganddingledgeDiscovery

Liittyvät mallit

"VisualarTificialIntelligence": CanMachinesReallyThink? OnkoHumanMindJustacplictiveComputerProgram? Tämä kirjaa,.Keinotekoinen.Ihmisiä.Vaikka ThemeehavelongbeendeclarationsThatSmartMachineseArtaRoundTheCorner.Tietoisuus,.HowshouldWemakesmartMachines? PitäisiTfunctionlikeabrain?.

"TheFuture-ArtificialIntelligence":interpretstheconnotation-intelligenceandexpoundsthebrainTheworkingprinciple,andtellsushowtomakearealintelligentmachine-suchintelligentmachineswillnolongerjustbeasimpleimitation-thehumanbrain,theirintelligencewillfarexceedthehumanbraininmanyaspects.Hawkinsbelievesthat, fromartificialIntelligencenetoneuryNetworks, ei mitään.ThereasonsThatPeopleDonotReAndStandtheconnotation-IntelligenceandthheHumanbrain.Theso-calledIntelligenceSTheAbilityftheHumanbraintocomParethepastadpredictTheFuture.Thebrainisnotacomputer, jaToesnotfollowthestepsandproduceOutputBasedOnOnNputStePByStep.ThebrainisahugememorySystem.ITSTORESEXPERIENCESTHATREFLECTTHETRUSTRUCTUREFTHEWORLDTOACONCTIONTEST.ItCanremememememberthesefevents and heirrelationship -sovellukset, andrakepredictionsbasedonMemory.TheFoundationOfIntelligence, sensaatio, luovuus ja persoonallisuus...

"Keinotekoinen ontelligencephilos-y": theargificialIntelligencephilos-yisAdeVelmentHatacCompaniesModernInformationTheoryandComputerteChnology.Filos-ia.ThisKirjaCollecTSFiftRePresentativePaperSByscholarsinTheFieldOfartInTIntelligencerESearchesearch.ThespapersHAvemadegroundContributionSTotedeDevelmenpment OFComputerscienceandTheStablmenmentOfartificialIntelligencephilos-ia.ThesearticlessummarizethehistoryOftheDevelpmentOfartificialIntelligence, theDeDevelmentTrend-TheSubject ja tärkeystietopiikatinartificialIntelligence.Näiden esineiden tekeminen: "Tietokoneen ja älykkyys" Byalancing, ModernComputerTheoryn isä; "mieli, aivot ja ohjelmat" byamericanphilosophersearle; andj · e · hintonetal."DistributedRepresentation" ja "EscapefromthechineseHouse" ByTheetorOfThisKirjaandbritisharTartInTIntelligencesCholarm · a · boden.

"KeinotekoinenIntelligence: AmodernMethod": This KirjaCombeNensiveSexPoundSTheCoreContenT-theFieldOfartificialIntelligenceFromThePerctiveOftherationalAgentwithDetaleDandrichinformation, andInTriDuceseachMajorhisresearchDirectionisarcomprens -.Thebookisdividedintoeightparts:thefirstpart"artificialintelligence",thesecondpart"problemsolving",thethirdpart"knowledgeandreasoning",thefourthpart"planning",thefifthpart"uncertainknowledgeandreasoning",thesixthpartPart"Oppiminen",Part7"Communication,KäsitysandAction",Part8"Conclusion".Tämä kirjatotonntroducesalargenumber-BasicConcepts, ideoita ja jaAlgoritmitInDatail, butlsodesCripStHemostcutting-reunge -kehopmentsInvariousresearchDirections.Atthesametime, itcollectSandoranizatesDetailedhistoricalDocumentsDevents.Siksi tämä kirjakirjanhoitotutkimukset ja -tutkimukset.ITCANBEUDESATEXTBOORTEACHINGGUIDEFORUNDRADUATEDGADUATESTENTSIINTEICTIONTIONDIFORNADRELATIONKIRJOITUKSET JAITCANALSOBEUDUSESAREPOREFERSTIONSCIENTIFIFICRESearchandEngineeringTechniciansInRelateFields.Kirja.

Abriefhistory-development

Thelegend-artificialintelligencecanbetracedbacktoancientEgypt,butwiththedevelopment-electroniccomputerssince1941,technologyhasfinallycreatedmachineintelligence,"artificialintelligence""(ARTIFICIALINTELLIGENCE-AwasfirstproposedattheDARTMOUTHSocietyin1956.Sincethen, tutkijat ovat kehitettyjätHeoria- jaPrinciples ja THECONcepTOfartificialIntelligencehasalsoExPanded.IntsShorthistory, TheDeDevelopmentOfartificialIntelligenceSisslowertHanExpektip, butithasbeenmovingforward.Siitä lähtienSappearance40Yearsago, ManyAiprogramshaveappared ja TheTheyHavealSaffectedTheDevelopmentOtOtSechnologies.

Kompensoida

AninventionIn1941RevolizazedAllaspectS-informationSTrageandprocessing.Tämä.AcomputerhastoOccupyseverallargeair-ilmoitushuoneet, jotka ovatIhtightMareForProgrammerit: Tuhanatteet.TheimprovedcomputertCanstorePROGRAMSIN1949MakestheinputProgrambecomesimpler, ja kehitystoimenpiteitä,.Tietokoneet, theInvention-elektronisesti prosessointitavat, ProvideaMediumFortEPSSIBLEREALIONSOFARTIFIFICIALINTILIGENCE.

R-tComputersProvidethetheeyTechnicalfoundationForai, itwasnottilTheArly1950SThatpeoplenOticedTheconnectionBetBetWeenInIntelligenceandMachines.NorbertwienerWasOneOfthefirstamerIcanstoStudyFeedBacktheory.ThemostfamiliarexampleOffeedBackControlitisHermostat.ITCLARESTHECOLLECTEDROOMTEMPERATURTWITHEDESPEDTEMPERATURE, AndreactstoturnTheHeaterOffTocontroltheambientTemperature.THISTRESearchonthefeedBackLoopliesin: Wienerfromteoreettisesti, ItPotededoutThallIntelligentActivitivitySaretSultOffeedBackMechanismi.TheFeedBackMechanismMayBesimuloidutByMachines.TämäDiscoFeLehAgreatInfluenceOnTheDevelmentOfearlyai.

AttheEnd-1955, NewwellandsimonMadeaprogrammoitu "logiikka".Tämä.ItRePresentseachProblemasatreemodel ja thenslectsTheMostlElyTogetTheCorrectConclusionToSolVetheProblem."Logicexpert" onThePublicandTheFieldOfAireseartMakesItNortantMilestoneinTheDevelopmentOfai.Vuonna 1956, Johnmccarthy, joka on käsitelty.HeinvitedThemTovermontToPartInTInTInThe "DartMouthartificialIntelligencesumerresearchConference".Sincethen, tämä Fieldhasbeennimeä "keinotekoinen onlelligenssi".VaikkaThedartmouthSociety wasnotverySuccessful, ButdidgatherThefounders-AIDLAIDThefoundationForfutureairesearchesearchersearchersearch.

Inthe7YearsAfterTheDartMouthConference, AirESearchBeGantOdESpRapidly.VaikkaThisthEfieldHasnotbeencLesLendfefined andSomeideainTheconferenceHaveBeenReconsideDandandused.CarnegiemellonUniversityandmithavebegunt-ormairesearchcenters.ResearchfacesNewChallenges: HextStepistoEtablishasystemTatCansolveProblemsmoreLeffektive, SuctaasindreducesearchesintHthe "Logicexpert"; seisalsotEstablmenmentOfaSelf-OppiminenSystem.

In1957,anewprogram,thefirstversion-the"UniversalProblemSolver"(GPS-Awastested.SEETHEPROGRAMMWASDEVELEPEDBYTHESAMEGROUPTHATMADETHE "LOGICEXPERT".GPSEXTENTEDWIENER'SFEEDBACKPRINCIPLEANDCANSOLVEMANYCOMSENSEPROBLES.Twoyearslater, ibmestablesedanairesearchgroup.Herbertgelerneterspent3yearsmakininge.ProgramSToSolveGeometrisetteormit.

ASMOREANDMOREPROGRAMSEMERGE,.Vuonna 1958, McCarthyannouncedhisnewachievement: Thelisluage.Lisisstilltheredayuse."LISP"means"listprocessing"(LISTPROCESSING-A,anditwasquicklyadoptedbymostAIdevelopers.

Vuonna 1963, MitReceivedaus 2 dollaria.2 miljoonaa., Käytetty TORESearchMachine-AidedRecognition.ThisfundingcomesfromtheAdvancedResearchProjectsAgency(ARPA-A-theDepartment-Defense,whichhasensuredthattheYhdysvallatisahead-theSovietUnionintechnologicalprogress.Tämä ohjelmaohjelma.Vauhti.

Kilpailu

LOEBNER(artificialintelligence-A

Usinghumanwisdomtocreateamachinebrain(artificialintelligence-Aparalleltothehumanbrain,Itisaverytemptingfieldformankind,andmankindhasbeenstrugglingformanyyearstorealizethisdream.FromThePeryfaGuAgeresearcher, ItisquitedifficultoAllowfreeCommunicationBetBetweeninesandHumans, jatMayevenbesAidTobeaquestionThatwillneverBeanswed.HumanLanguageandHumanIntelligencearesocpartictedTheTourResearchhasnotucheuchedgeOfThextendedpartpartSearchhasNature.

Alargenumber-programs

AlarGenumber-Programshaveappearedinthenextfewyears.Yksittäinen "shrdlu"."SHRDLU"ispart-the"MicroWorld"project,whichisincludedinthemicroworld(forexample,onlyResearchandprogramminginalimitednumber-geometricshapes-A.SUORESTERSLEDBYMARVINMINSKYATMITFOUNDTHATINTHEFACEOFSMALL-SCALEOBJONTS, COMPUTROPRAGRAMSSOLVESPATIAALANDLOGICALPORTS.Muita."Sir" CanuderstandSimpleEnglishSencences.TheresultsOfTesProgramsarehelpfulInProcessingLanguageAntsAntingandLogic.

Toinen kehitystä1970SisExpertsystems.AsiantuntijakysteemscanpredictTheProbabbanfacetureSolutionDercompingConditions.BaceethecomputerHadHugeCaPancityThattime, iTwassibleFortheexpertsystemTodRawrulesFromtheData.ThemarkeTeApplicationFTEXPERTSYSTEMEWASVERY WOWIDE.IntHePasttenyears, The ExperTSYSTEMWASUDUSEFORSTOCKMARETEFORECASTSTOHELPDOCTORSDIAGNOSEDISEASES, AswellasInStructingMinSeTodeterMinethelocation-MineralDePosits.AllOfThispibleBecAusefTEAbilityFexPertsYSTEMStoStorerulesandInformations.

ManyNewmethodswereuleDiidevelopmentinthe1970 -luvut, sellainenMinsky'structuralteory.INDADDION, DavidMarrpropotedNewtheoriaSinMachineVision, Forexample, HowToDistRaNIMAGethRoughBasicinformationsuchassHadow, muoto, väri, BorderAndTexture-anImage.Byanalysoiva tietokone, yocaninferwHattheImageMaybe.ATTHESAMETIMEANOTERACHIEESTISTHEPROLEGULIGAGE, WHO WHASPROPONEDIN1972.1980 -luvun aikana AiprogressedMorerApidlyAnteeredTheCommercialfieldmore.Vuonna 1986, THESALSOFAI-RELATEDSOFTWAREANDHARDWAREINTHEUnitedStatsReached425Millionu.S.dollari.AsiantuntijakysteemaredueTheirutility.Erityisesti.CompaniesLikeDigitalElectricCompanyUSethExConexpertsystemtoProgramvaxMainframes.DuPont, Generalmotors jaBoeingalsorelyheavilyOnexpertsystems.InorderToMeetTheneedS-computExerTerts, SomeProductionExpertsystemsSinTheProDuctionOfS-tWarecompaniEessuchastEnnowledGeanDInTuctionororpwereEstabled.InorderT-inDandCorrectTheRorsinTheExistingExpertsystems, OtherexpertsystemShAveBeEndEnESESTIMED.

Jokapäiväinen elämä

PeoplearebeginningT-eelTheAnfluence OFComputersandartificialIntelligenceTechnology.ComputerTechnologySNOLONGERJUSTAMALLGOUPOFRESECHESSIINTHELABORATION.Henkilökohtaiset ComputersandmanyTechnicalmagazineshavebroughtComputertechnologyToThePublic.ThereAsefoundationsliketEmeriericanartificialIntelligenceaSsocation.Koska THENEEDSOFAIDEVELPOINTTI, THEESESTESTILLTEREWASAWAVEOFRESESESSESENTERIVERSPRIVATECOMPANIES.Morethan150companieslikeDEC(whichemploysmorethan700employeesinAIresearch-Aspentatotal-US$1billiononinternalAIdevelopmentgroups.

MuuAIiFieldSalsoeneredTheMarketinthe1980s.YhdenfthemismAchineVision.Theresults-minsyandMarrarenowusedIncamerasandcomputersonThePRoDuctionLineforQualityControl.Vaikka systeemshaveBeenedeDistishishTheShape-ObjectShroughblackandWhitedistlections.BY1985, TheunitedstaTestherArEMORETHAN100CompaniesProducingMachinevisionSystems, WithSalestolingus 80 miljoonaa dollaria.

Butthe1980swerenotallgoodyearsfortheAI​​industry.TheDevelopment-Aisystemsin1986-87Demandfell ja teollisuuslostnearlyus 500 miljoonaa dollaria.TwocompaniesliketekledgeandIntellicorplostmorethanus 6 miljoonaa dollaria, kirjanpit-orbountone-kolmas voittoa.TheHugelsForcedManyresearchLeadSerstoCutFunding.Toinen DisAppintingitistheso-kutsuttu "SmartTruck" tuettu BYTHEADESTEDRESEWESAREPROJECTSAGENCTOFTHEMINistryOfefense.The PurposeOfThisprojectistoDeveloparobotThatCancompeTemanyBattlefieldTasks.DuetoprojectDefectsandHopeEfessSuccess, PentagonstoppedFundingTheProject.

Artificialintelligencerobots(2photos-A

Despitethesesetbacks,AIisstillSlowlyresumedevelopment.NewtechnologieshavebeendevelopedinJapan,suchasfuzzylogicpioneeredintheYhdysvallat,whichcanmakedecisionsfromuncertainconditions;andneuralnetworks,whichareregardedaspossiblewaystorealizeartificialintelligence.Inshort,80AIwasintroducedintothemarketinthe1940sandshowedpracticalvalue.Itisbelievedthatitwillbethekeytothe21stcentury.Artificialintelligencetechnologyacceptedandtested.Inthe"DesertStorm"operation,themilitary'ssmartdeviceswithstoodthetest-war.Manually.Intelligenttechnologyisusedinmissilesystemsandearlywarningdisplaysandotheradvancedweapons.AItechnologyhasalsoenteredthehome.Theincreaseinsmartcomputershasattractedpublicinterest;someapplicationsforAppleandIBMcompatiblecomputers,suchasvoiceandtextrecognition,arealreadyavailableTo;usingfuzzylogic,AItechnologysimplifiescameraequipment.Thegreaterdemandforartificialintelligencerelatedtechnologieshaspromptednewadvancestoemerge.Artificialintelligencehasandwillcontinuetoinevitablychangeourlives.

Comparison-strengthsandweaknesses

AmorePopulardEfinition-artificialIntelligenceisalsoanearlierDefinitioninthisfield.ItwasdevelopedbyJohnMcCarthyattheDartmouthConferencein1956CONFERENCE-A:Artificialintelligenceistomakethemachine'sbehaviorlookliketheintelligentbehaviorshownbyhumans.Butthisdefinitionseemstoignorethepossibility-strongartificialintelligence(seebelow-A.ToinenDefinitionReferstoartificialIntelligenceastheIntelligencedisplayedbyman-mademachines.Yleisesti ottaen, enitenHelatfinitionS-artificialIntelligencencanbedividedIl-ourCategoria, nimittäin koneet "ajatteluhenkilöt", "actHelmans", "ajattelutapa" ja "actrational".Täällä "toiminta" tulisi brroadlyAntsodeDastionAgtaction, OrMaking Decisionsforaction, pikemmin.

Alhaalta

Thestrongartificialintelligenceviewbelievesthatitispossibletocreaterealreasoning(REASONING-Aandproblem-solving(PROBLEM_SOLVING-A.-AIntelligentmachines,andsuchmachinescanbeconsideredsentientandself-conscious.ThereareTWotypes-strongarTificialIntelligence:

Ihmisen ja likeartificialIntelligence, thatis, machinethinkingAndreasoningarejustHumanthinking.

Ei-ihmisen ei-likeartificialIntelligence, that, machinesproduceperceptionsandConcongationThatarecompetely differentfromHumans, anduseaCompleyDifferentway-reasoningfromhumans.

TOP-DOWNAI(TOP-DOWNAI-A

Theview-TOP-DOWNAIbelievesthatitisimpossibletoproducerealreasoning(REASONING-Aandproblem-solving(PROBLEM_SOLVING-Aintelligentmachines,thesemachinesonlyseemtobeintelligent,buttheydon’treallyhaveintelligence,andtheydon’thaveautonomousconsciousness.

MainstreamScientificresearchFocusonweakTarTificialIntelligence.ResearchStrongongarTificialIntelligenceIstaStandStill.

Filos-inenDebateStrongarTificialIntelligence

Thetter "strongartificialIntelligence" WasoriginallyCoundbyjohnrogershillerforcomputersandotherinformationProcessingMachines, joka on defefinedas:

"ThestrongarTificialIntelligenceViewbelievesthatcomputersarenotoonlyoolforstudyHumanthink;."(JSEARLEINMINDSBRANSAndPROGRAMS.Behehavioralandbrainsciences, vol.3,1980-AThisreferstomakingcomputersengageinintelligentactivities.Themeaning-IntelligenceHeisambiglyAnduncart, asumionionedbelowisanexampleOfit.Kun.Kuitenkin termemanycaseswherePeopletryTosolVetheProblemcleverlyAccordingTotheheu-ristMethodeven-.SuchasRecognizingWresstedwords, Graphics, Sounds jne., theso-calledrecognitionModelisanexample.Lisäksi sopeutumisessa.arealsoxamples.Lisäksi, vaikka menettely on ISCLEAR, ITTAKESALONGTIMETOMPLEMENT.ForsuChaproblem, PeoplecanfindaveryGoodSolutionInashorttime, SuchasCompetiveGames.Tarkoituksenmukaisuus,.Teemaning-Livingit.NaturallanguageisanExample.ProcessingnaturallanguAgeWitHAComputerSiscallednaturallanguAgeProcessing.

Thedebateonstrongartificialintelligenceisdifferentfromthebroadermonismanddualism(DUALISM-Adebate.THaiAinpointOftHeargumentis: IftheOnlyworkingPrinciple-amachineSTRANSForMenCodedData, DoESTHISMACHINEHAVEATININKING?.Hegaveanexample-achineseroomtoillustrathatifthemachineonlyconvertsdata- ja thedatAifefisacodededReprescationalshingsproundings.UndertHipremiseOf, ITISIPSICOSSIBORTHMACHINETOHAVEANYNUNTERSTINGOFTHATEDAITPROCESSES.BasitonThisargument, HillerbelievestHatevenifamachinePasseSestHangingtest, itnotnotnotNetstMeanthatThemachineisreallyThinkindandconscontlikeaHuman.

ThereAlsophilosopherswHOLDDIFRENTVIEWS.InhisbookConcongstangnessexplay, Danielc.Dennettbelievesthatmanisjustamachinewithasoul.WHYDOWETHINKTHATHUMANSCANBEILTILIGENTBUTORDIALIAALINENMAKUSTUSKIRJA? HebelievestitititusScossibleforadataconversionMachineliketheaboveTohavethinkingandConconversionsion.

Jonkinfilos-ersbelievethathatifweakTificialIntelligenceisAchievable, thestrongarTificialIntelligenceisalsoachievable.Forexample, SimonblackBurnStitedInitsInTroductoryPhilos-yTextBookThinkThaTaperson'SActionsThatappeartObe "Smart" DonotreallyMeanThatapersonisrealysmart.ICANNEVERKONTECHEENOTERPERSONISREALLLYSMARTLIKEMEME, ORWHESHESHE/HeisjustassMartasheis.BASTONTONTHISARGUMENT, WEAKATARTIFIFIFICIALINTILIGENCEBELIEVESTHATITCANMAKEACHINELOKKSMART,.BlackburnbelievesthatTisisasubjectiveissue.

Sen pitäisi ollapohjainen.Se on, elifstrongarTificialIntelligenceScassional, heikko,.Ainakin, tHithatComputersCandoToday, SucasaritheMecoperations, WerethtoreTorequireIntelligenceemorethanah Sadayarago.

Poliittiset tiedot

Onjune17,2019, silloinenTewgeneraatiota.Tämä.

Tutkimusaiheet

TheresearchDirection-artificialIntelligenceHasbeEnDiveDiNIntoseveralsub-kentät.TutkijatHopethatanarTificialIntelligencesysteemsuuldhaveSecompesificCapebence.ThesekapabilityarelistEDANDESPLEDEDBELOW.

Ongelmien ratkaiseminen

Varhainen ourAnialIntelligencerESearchersDirectlyMitedHumanstoPerFormStep-by-Sesgevering, JustlikethewayHumansthinkwenplayingABoardGameorperformingLogicalReasoning.Inthe1980Sand1990 -luvulla, käyttämällä Proteconcepts-probilityandekonomiaa, keinotekoistaIntelligencerESearchalsodevelopedverySuccessfulMethodstodealWithuncityorincompletInformations.

Fordifforcultiveproblems, AlargealountFputingResourcesMayBerequired, jonka mukaan "probAblecombination explosion": whentHeProbleMexceedSacetScale, thecomputerWillRequireastronomIcalSOfMemororyorComputingtime.FindmoreffeccivealgoritmsisapriorityartificialIntelligenceresearchproject.

Themodel-HumanProblemsolingisususaltHequickESTInTIntiveJudgment, pikemminkin.AikaisinArTificialIntelligencerESearchusuuruSuesSastp-by-steppderifationMethod.Keinotekoinen onlelligenceresearchhasmadeprogressinThis "ala-edustuston" ongelmanratkaisuMethod: thematerializedAgentResearchemphasizestHeimportanceOferceivingMovement.NeuralNetworkResearchattemptStoReProducethisskillBysimulatingTheBrainstructurefHumansandanimals.

Tiedon edustaminen

Anontologia ResepressentskledgeaSaset-conceptswithinadomainandtherelationshipsbetweentHoseconcepts.

Mainarticle: Tiedonsiirtopresentation jaCommonSenseWledgeBase

Suunnittelu

The INTILLIGENTAGENTMUSTBEABLETOSETOALIAALIAACHIEEVETHEEGAALS.Theyneedawaytobuildapredictableworldmodel(expressthestate-theworldinamathematicalmodelandpredicthowtheirbehaviorwillchangetheworld-Asothattheycanchoosethemosteffectivebehavior.IntraditionalSuunnitteluProblems, The INTILLIGENTAGENTISSUMEDTOBETHEONLYINFLUENTINESONEINTHEWORLD, SOWHATBHOIRITWILLERFORNESALREADYMENT.Kuitenkin, jos se.IfitDoesNotComply, ItShanceChangeSplan.Siksi älykkyysmuuskaveriabilityToreasondunderUncartResults.INMULTI-Agent, MultipleagentsplantoAccomplishcertaingoalsinacoperativeandcompetitiveemanner.Käyttämälläevoluutionaryalgoritmeja jaGroupWisdomCanachieVeaneverallemergentGoal.

Oppiminen

Pääartikkeli: koneoppiminen

ThrainPurpose-machinelearningistoobtainknowledgefromusersandInputData jne.., mikä kannetaan.Useita ongelmia, pelkistävärisiä.ForartificialIntelligence, MachinElearningSImportantfromTheverybeginning.Vuonna 1956, AtheoriginalDartmouthSummerconference, RaymondsolomonovwroTeanarticleaBoutProbabilistMachinElearningWithSurveillance: AmachineforinductIiverkaupaus.

NaturallanguageProcessing

Mainarticle: NaturallanguageProcessing

MotionAndControl

Mainarticle: Robotiikka

Käsitys

Mainarticle: MachineKäsitys, ComputerVisionandspeechRecognition

Machineperceptionreferstotheabilitytousethedatainputbysensors(suchascameras,microphones,sonarandotherSpecialsensors-Atheninferthestate-theworld.ComputerVisionCananalyzeImageInput.ThereArealSoVoicerecognition, facerecognitionandObjectRecognition.

Sosiaalinen

Mainarticle: Emotional Computing

Kismet, ArobotWithsocialabilitiessuchasfacialExpressions

EmotionandSosiaalinenSkillsSmartAGENTISEDSIONTICANT.First,byunderstandingtheirmotivationsandemotionalstates,agentscanpredicttheactions-others(thisinvolveselementarygametheory,decisiontheory,andtheabilitytoshapepeople'semotionsandemotionperceptiontesting-A.Lisäksi unohdettu ihmisen ja tietokoneen vuorovaikutus, älykkäät allsonedToShowemoTions.AtleastmustappearodealwithHumansPoLift.Ainakin siinä pitäisi olla normaaleja tunteita.

Luovuus

Mainarticle: ComputerCreative

Asubfield-artificialintelligence,representingtheory(fromtheperspective-philosophyandpsychology-Aandpractice(Theoutput-thesystemgeneratedbyaspecificrealizationistheideathatcanbeconsidered,orthesystemrecognizesandevaluatesthecreativity-Adefinedbycreativity.ResearchInRelatedFieldSinSartSartrificialIntuitionAnTificialImagation.

Useita älykkyyttä

Mostresearchershopethattheirresearchwilleventuallybeincorporatedintoamulti-intelligence(calledstrongartificialintelligence-AthatcombinesalltheaboveskillsandsurpassesmosthumansAbility.JokinPeoplethinkthattoAchievetheaboveGoals, antropomorfisetCharakteristics, SuchasartificialConsionSornorTificialBrains, Mayberequired.MonetOftHEABOVEPRIBLESSARECONSEREDTOBETHEINEGRITYOFARTIFIFICALINTILIGENCE.Evenasimpleandspecifictask,suchasmachinetranslation,requiresthemachinetoknowwhatisbeingtalkedabout(knowledge-Aaccordingtotheauthor'sargument(inference-A,andfaithfullyreproducetheauthor'sintention(emotionalcalculation-A.Siksi MachinetranslationSconsidedToHaVeatificialIntelligenceIntegrity: ItMayRequirestrongarTificialIntelligence, JustlikeHumans.

Theinfluence-artificialintelligence

(1-ATheinfluence-artificialintelligenceonnaturalscience.InsbjectHatRequiretheuseOfMatematicalComputerToolstoSolveProblems.AIINTURNHELPSHUMANBEINGSFINALLAALINEN AINTURNEMORMATIONIONIRONTIINIGENCE.

(2-ATheimpact-artificialintelligenceontheeconomy.The ExpertsystemGoesDeePerIntoAllWalks-LifeanDBringsHugemaCrobenefits.AihasalsopromotedTheDevelopmentment -TheComopterIndustryNetworkInnustry.Butatthesametime, italsobringsLaborEmploymentProblems.DuetToTheAplication-aiintechnologyandIngineering, itCanReplacehumansinVariousTeknicandmentalwork, mikä.

(3-ATheimpact-artificialintelligenceonsociety.AIalsoprovidesanewmodelforhumanculturallife.Olemassaolot.Nykyään keinotekoiset ohjeet.

TenhastobedIntelligendIntelligentrobots, ithastobeDiscusedThatartificialIntelligenceItSelfisAncestResearch and ModerncientificresearchneedstobeCarriedwithAfutUrevision, SoiuthLEYTOUTHOUTHEBOTOMLINEEDETHICS,.ASASENSIMESUETHATMAYBEINVOLVEDINSCIENTIFIFIFIFIFRESSEW, ITISNEEDYTOPREvenTCONFLICTSTHATMAYARISESSOONPSOSSIBLE, sen sijaan.

Sovelluskenttä

MachineTranslation, IntelligentControl, ExpertSystem, Robotics, LanguageDIMAgeStansatStanding, GeneticProgrammingrobotfactory, Automaatti-ohjelmointi, ilmailu.

ITISWORTHMENTOINGTHATMACHINEMANESLAATIONISANIMPORTANTBRANCHANDTHEFIRSTIPLICATIONKIRJAUS.Kuitenkin intermsheexistingmachinetranslationAchievements, thetranslationQuality fathemachinetranslationsystemisstillfarfromthellimategoaal; jaetuntilastenslaatiojärjestelmätTheyTothesuccessorfailure-themachinetranslationSytem- ja thequitefatyfatyfationslaationslationstemstem.ChinesemathematicianandlinguistPr-essorZhouHaizhongoncepointedoutinthepaper"FiftyYears-MachineTranslation":Toimprovethequality-machinetranslation,thefirstthingtodoistosolvetheproblem-thelanguageitselfratherthantheproblem-programdesign;onlyrelyonanumber-programstodoitThemachinetranslationsystemcertainlycannotimprovethequality-machinetranslation;inaddition,itisimpossibleformachinetranslationtoachieve"faithfulness,expressiveness,andelegance"whenhumanshavenotyetunderstoodhowthebrainperformsthefuzzyrecognitionandlogicaljudgment-language.-.Aftersmarthome,artificialintelligencehasbecomeanewoutletforthehomeapplianceindustry,andChanghongisbecomingthefirsthomeappliancegianttoset-fthiswave.CHanghongrelledtwonewchiqsmarttvproducts, mukana

Summusanat

InDecember2017,artificialintelligencewasselectedas"2017"TopTenSummusanatinChineseMedia-theYear".

Reasonforselection:Afteryears-evolution,thedevelopment-artificialintelligencehasenteredanewstage.Inordertoseizethemajorstrategicopportunitiesinthedevelopment-artificialintelligence,buildmycountry'sfirst-moveradvantageinthedevelopment-artificialintelligence,andacceleratetheconstruction-aninnovativecountryandaworldpowerhouseinscienceandtechnology,onJuly20,2017,theStateCouncilissuedthe"NewGenerationArtificialIntelligenceDevelopmentPlan".The"Plan"putsforwardtheguidingideology,strategicgoals,keytasksandsafeguardmeasuresforthedevelopment-mycountry'snewgeneration-artificialintelligencein2030,layinganimportantfoundationforthefurtheraccelerateddevelopment-artificialintelligenceinmycountry.

Kehitysvuoro

OnJuly13,2021,theInternetSociety-Kiinareleasedthe"KiinaInternetDevelopmentReport(2021-A".The"Report"showsthatin2020,thescale-theartificialintelligenceindustrywillreach303.1 miljardia.

Inthefield-artificialintelligence,thescale-theartificialintelligenceindustrywillmaintainsteadygrowthin2020.Thescale-theindustrywillreach303.1 miljardia,ayear-on-yearincrease-15%,andthegrowthrateisslightlyhigherthantheglobalaverage.The IndustryncorismAnlyConcentratedinbeijing, Shanghai, Guangdong, ZhejianganandotherProvinces.mycountryhasmadesignificantprogressinthefield-artificialintelligencechips,deeplearnings-twarearchitecture,andChinesenaturallanguageprocessing.

Thecontent-BaiduBaikeisco-editedbynetizens.Ifyoufindthatthecontent-yourentryisinaccurateorimperfect,pleaseuseyourentryeditingservice(free-charge-Atoparticipateinthecorrection.Gonow>>

Related Articles
TOP