人工智能的課件CH1Introduction_第1頁
人工智能的課件CH1Introduction_第2頁
人工智能的課件CH1Introduction_第3頁
人工智能的課件CH1Introduction_第4頁
人工智能的課件CH1Introduction_第5頁
已閱讀5頁，還剩71頁未讀， 繼續(xù)免費(fèi)閱讀

下載本文檔

版權(quán)說明：本文檔由用戶提供并上傳，收益歸屬內(nèi)容提供方，若內(nèi)容存在侵權(quán)，請(qǐng)進(jìn)行舉報(bào)或認(rèn)領(lǐng)

文檔簡介

1.IntroductionTo

ArtificialintelligenceAutumn2012Instructor:WangXiaolongHarbinInstituteofTechnology,ShenzhenGraduateSchoolIntelligentComputationResearchCenter(ICRC,HITSGS)

1.IntroductionTo

ArtificialCourseOutlineWhyStudyAI?WhatisAI?WhatcanAIdo?WhattasksrequireAI?HowtoachieveAI?BranchesofAIExamplesofAIAIinChinaAIinHITCourseOutlineWhyStudyAI?WhyStudyAI?SearchenginesLaborScienceMedicine/DiagnosisAppliancesWhatelse?WhyStudyAI?SearchenginesLabWhyStudyAI?(Cont.)SonyAIBOWhyStudyAI?(Cont.)SonyAIBOhWhyStudyAI?(Cont.)USmilitaryWhyStudyAI?(Cont.)USmilitarWhyStudyAI?(Cont.)NaturalLanguageQuestionAnswering/projects/infolab/WhyStudyAI?(Cont.)NaturalLaWhyStudyAI?(Cont.)RobotTeamsUSCroboticsLabWhyStudyAI?(Cont.)RobotTeamWhatisAI?Theexcitingnewefforttomakecomputersthinks…machinewithminds,inthefullandliteralsense”(Haugeland1985)“Theartofcreatingmachinesthatperformfunctionsthatrequireintelligencewhenperformedbypeople”(Kurzweil,1990)“Thestudyofmentalfacultiesthroughtheuseofcomputationalmodels”(Charniaketal.1985)Afieldofstudythatseekstoexplainandemulateintelligentbehaviorintermsofcomputationalprocesses”(Schalkol,1990)SystemsthatthinklikehumansSystemsthatthinkrationallySystemsthatactlikehumansSystemsthatactrationallyViewsofAIfallintofourcategories:Thetextbookadvocates"actingrationally"WhatisAI?TheexcitingnewefWhatisAI?(Cont.)ActingHumanly:TheTuringTestAlanTuring's1950articleComputingMachineryandIntelligencediscussedconditionsforconsideringamachinetobeintelligent.“Canmachinesthink?”“Canmachinesbehaveintelligently?”TheTuringtest(TheImitationGame):Operationaldefinitionofintelligence.Computerneedstopossess:Naturallanguageprocessing,Knowledgerepresentation,Automatedreasoning,andMachinelearningPredictedthatby2000,amachinemighthavea30%chanceoffoolingalaypersonfor5minutesAnticipatedallmajorargumentsagainstAIinfollowing50yearsArethereanyproblems/limitationstotheTuringTest?problemswiththeTuringTest:Turingtestisnotreproducible,constructive,andamenabletomathematicanalysis.Whataboutphysicalinteractionwithinterrogatorandenvironment?WhatisAI?(Cont.)ActingHumanWhatisAI?(Cont.)ActingHumanly:TheFullTuringTestThecomputerpassesthetestifahumaninterrogator,afterposingsomewrittenquestions,cannottellwhetherthewrittenresponsescomefromapersonornot.Requiresphysicalinteractionandneedsperceptionandactuation.WhatwouldacomputerneedtopasstheTuringtest?Naturallanguageprocessing:tocommunicatewithexaminer.Knowledgerepresentation:tostoreandretrieveinformationprovidedbeforeorduringinterrogation.Automatedreasoning:tousethestoredinformationtoanswerquestionsandtodrawnewconclusions.Machinelearning:toadapttonewcircumstancesandtodetectandextrapolatepatterns.Vision(forTotalTuringtest):torecognizetheexaminer’sactionsandvariousobjectspresentedbytheexaminer.Motorcontrol(totaltest):toactuponobjectsasrequested.Othersenses(totaltest):suchasaudition,smell,touch,etc.WhatisAI?(Cont.)ActingHumanWhatisAI?(Cont.)ThinkingHumanly:CognitiveScience1960“CognitiveRevolution”:information-processingpsychologyreplacedbehaviorism.Cognitivesciencebringstogethertheoriesandexperimentalevidencetomodelinternalactivitiesofthebrain.Whatlevelofabstraction?“Knowledge”or“Circuits”?Howtovalidatemodels?Predictingandtestingbehaviorofhumansubjects(top-down)Directidentificationfromneurologicaldata(bottom-up)Buildingcomputer/machinesimulatedmodelsandreproduceresults(simulation)ThinkingRationally:LawsofThoughtAristotle(~450B.C.)attemptedtocodify“rightthinking”

Whatarecorrectarguments/thoughtprocesses?E.g.,“Socratesisaman,allmenaremortal;thereforeSocratesismortal”SeveralGreekschoolsdevelopedvariousformsoflogic:

notationplusrulesofderivationforthoughts.Problems:Uncertainty:Notallfactsarecertain(e.g.,theflightmightbedelayed).Resourcelimitations:-Notenoughtimetocompute/process-Insufficientmemory/disk/etc-Etc.WhatisAI?(Cont.)ThinkingHumWhatisAI?(Cont.)ActingRationally:TheRationalAgentRationalbehavior:Doingtherightthing!Therightthing:ThatwhichisexpectedtomaximizetheexpectedreturnProvidesthemostgeneralviewofAIbecauseitincludes:Correctinference(“Lawsofthought”)UncertaintyhandlingResourcelimitationconsiderations(e.g.,reflexvs.deliberation)Cognitiveskills(NLP,AR,knowledgerepresentation,ML,etc.)Advantages:MoregeneralItsgoalofrationalityiswelldefinedRationalagentsWhatisAI?(Cont.)ActingRatioWhatCanAIDo?

Quiz:Whichofthefollowingcanbedoneatpresent?Playadecentgameoftabletennis?Drivesafelyalongacurvingmountainroad?DrivesafelyalongTelegraphAvenue?Buyaweek'sworthofgroceriesontheweb?Buyaweek'sworthofgroceriesatBerkeleyBowl?Discoverandproveanewmathematicaltheorem?Conversesuccessfullywithanotherpersonforanhour?Performacomplexsurgicaloperation?Unloadadishwasherandputeverythingaway?TranslatespokenEnglishintospokenSwedishinrealtime?Writeanintentionallyfunnystory?WhatCanAIDo?

Quiz:WhichofWhattasksrequireAI?“AIisthescienceandengineeringofmakingintelligentmachineswhichcanperformtasksthatrequireintelligencewhenperformedbyhumans…”TasksthatrequireAI:SolvingadifferentialequationBrainsurgeryInventingstuffPlayingJeopardyPlayingWheelofFortuneWhataboutwalking?Whataboutgrabbingstuff?Whataboutpullingyourhandawayfromfire?WhataboutwatchingTV?Whataboutdaydreaming?WhattasksrequireAI?“AIistHowtoachieveAI?HowisAIresearchdone?AIresearchhasboththeoreticalandexperimental

sides.Theexperimentalsidehasbothbasicandappliedaspects.Therearetwomainlinesofresearch:Oneisbiological,basedontheideathatsincehumansareintelligent,AIshouldstudyhumansandimitatetheirpsychologyorphysiology.Theotherisphenomenal,basedonstudyingandformalizingcommonsensefactsabouttheworldandtheproblemsthattheworldpresentstotheachievementofgoals.Thetwoapproachesinteracttosomeextent,andbothshouldeventuallysucceed.Itisarace,butbothracersseemtobewalking.[JohnMcCarthy]HowtoachieveAI?HowisAIreBranchesofAILogicalAISearchNaturallanguageprocessingPatternrecognitionKnowledgerepresentationInferenceFromsomefacts,otherscanbeinferred.

AutomatedreasoningLearningfromexperiencePlanningTogenerateastrategyforachievingsomegoal.EpistemologyStudyofthekindsofknowledgethatarerequiredforsolvingproblemsintheworld.

OntologyStudyofthekindsofthingsthatexist.InAI,theprogramsandsentencesdealwithvariouskindsofobjects,andwestudywhatthesekindsareandwhattheirbasicpropertiesare.

GeneticprogrammingEmotions???…BranchesofAILogicalAIExamplesofAIQ&AWastonNamedafterthefounderofIBM--ThomasJ.WastonWon$1,000,000inGameJeopardyin2011,defeated2humancompetetersSource:encyclopaedia,wikipedia,webpages,structuraldata100algorithms,90servers,3sresponsetimeJeopardyGameRules:-Getpointsforcorrectanswers-LosepointsforincorrectonesExamplesofAIQ&AWastonJeopaExamplesofAIQ&AALICEArtificialLinguisticInternetComputerEntityChatrobot,computerprogramWon2000,2001,2004LoebnerPrizeExamplesofAIQ&AALICEExamplesofAIQ&AWolframAlphaAbundantknowledgebaseAbilitytodoscientificcalculationExamplesofAIQ&AWolframAlpExamplesofAIQ&ASiriStartedoniPhone4S,voicecontrolContextanalysis(eg.weather),Googlesearchwikipedia,YelpWolframAlphaExamplesofAIQ&ASiriAiinChinaSpeechinput863NaturallanguageprocessingPatternrecognitionMachinelearningMachineTranslationMayIhelpyou?Howareyou?Howoldareyou?…AiinChinaSpeechinputAiinHITRobotGameMachineTranslationSentenceinputSiguoGameQ&ASearchEngine…AiinHITRobotGameAiinHITRobotGameAiinHITRobotGameAiinHITQ&AAiinHITQ&AAIPrehistoryPhilosophyLogic,methodsofreasoning,mindasphysicalsystemfoundationsoflearning,language,rationalityMathematicsFormalrepresentationandproofalgorithms,computation,(un)decidability,(in)tractability,probabilityEconomicsutility,decisiontheoryNeurosciencephysicalsubstrateformentalactivityPsychologyphenomenaofperceptionandmotorcontrol,experimentaltechniquesComputerEngineeringbuildingfastcomputersControltheorydesignsystemsthatmaximizeanobjectivefunctionovertimeLinguisticsknowledgerepresentation,grammarAIPrehistoryPhilosophyAIHistoryAIHistoryAIStateoftheartHavethefollowingbeenachievedbyAI?World-classchessplayingPlayingtabletennisCross-countrydrivingSolvingmathematicalproblemsDiscoverandprovemathematicaltheoriesEngageinameaningfulconversationUnderstandspokenlanguageObserveandunderstandhumanemotionsExpressemotions…DeepBluedefeatedthereigningworldchesschampionGarryKasparovin1997Provedamathematicalconjecture(Robbinsconjecture)unsolvedfordecadesNohandsacrossAmerica(drivingautonomously98%ofthetimefromPittsburghtoSanDiego)Duringthe1991GulfWar,USforcesdeployedanAIlogisticsplanningandschedulingprogramthatinvolvedupto50,000vehicles,cargo,andpeopleNASA'son-boardautonomousplanningprogramcontrolledtheschedulingofoperationsforaspacecraftProverbsolvescrosswordpuzzlesbetterthanmosthumans…AIStateoftheartHavethefoCourseOverviewGeneralIntroduction01-Introduction.

[AIMACh1]CourseSchedule.Homeworks,examsandgrading.Coursematerial,TAs.WhystudyAI?WhatisAI?TheTuringtest.Rationality.BranchesofAI.Researchdisciplinesconn-ectedtoandatthefoundationofAI.BriefhistoryofAI.Challengesforthefuture.Overviewofclasssyllabus.02-IntelligentAgents.

[AIMACh2]Whatis anintelligentagent?Examples.Doingtheright thing(rationalaction).Performancemeasure. Autonomy.Environmentandagentdesign. Structureofagents.Agenttypes.Reflexagents. Reactiveagents.Reflexagentswithstate. Goal-basedagents.Utility-basedagents.Mobile agents.Informationagents.sensorseffectorsAgentCourseOverviewGeneralIntroduCourseOverview(Cont.)Howcanwesolvecomplexproblems?03-Problemsolvingandsearch.[AIMACh3]Example:measuringproblem.Typesofproblems.Moreexampleproblems.Basicideabehindsearchalgorithms.Complexity.CombinatorialexplosionandNPcompleteness.Polynomialhierarchy.

04-Uninformedsearch.[AIMACh3]

Depth-first.Breadth-first.Uniform-cost.Depth-limited.Iterativedeepening.Examples.Properties.05/06-Informedsearch.[AIMACh4]

Best-first.A*search.Heuristics.Hillclimbing.Problemoflocalextrema.Simulatedannealing.3l5l9lUsingthese3buckets,measure7litersofwater.TravelingsalespersonproblemCourseOverview(Cont.)HowcanCourseOverview(Cont.)Practicalapplicationsofsearch.07/08-Gameplaying.

[AIMACh5]Theminimaxalgorithm.Resourcelimitations.Aplha-betapruning.Elementsof chanceandnon- deterministicgames.tic-tac-toeCourseOverview(Cont.)PracticaCourseOverview(Cont.)Buildingknowledge-basedagents:1stOrderLogic09-LogicalAgents.

[AIMACh7]Syntax.Semantics.Atomicsentences.Complexsentences.Quantifiers.Examples.FOLknowledgebase.Situationcalculus.10-First-orderlogic.

[AIMACh7]Describingactions. Planning.Actionsequences.CourseOverview(Cont.)BuildingCourseOverview(Cont.)RepresentingandOrganizingKnowledge11-Buildingaknowledgebase.

[AIMACh8]Knowledgebases.Vocabularyandrules.Ontologies.Organizingknowledge.Kahn&Mcleod,2000AnontologyforthesportsdomainCourseOverview(Cont.)RepresenCourseOverview(Cont.)ReasoningLogically12-Inferenceinfirst-orderlogic.

[AIMACh9]Proofs.Unification.Generalizedmodusponens.Forwardandbackwardchaining.ExampleofbackwardchainingCourseOverview(Cont.)ReasoninCourseOverview(Cont.)ExamplesofLogicalReasoningSystems13-Logicalreasoningsystems.

[AIMACh10]Indexing,retrieval andunification.TheProloglanguage. Theoremprovers.Framesystems andsemanticnetworks.Semanticnetworkusedinaninsightgenerator(Dukeuniversity)CourseOverview(Cont.)ExamplesCourseOverview(Cont.)SystemsthatcanPlanFutureBehavior14/15-Planning.

[AIMACh11]Definitionandgoals.Basicrepresentationsforplanning.Situationspaceandplanspace.Examples.CourseOverview(Cont.)SystemsCourseOverview(Cont.)LogicalReasoninginthePresenceofUncertainty16/17-Uncertainty.

[AIMACh13/14]IntroductiontoUncertaintyandProbabilisticReasoning

CenteroflargestareaCenterofgravityCourseOverview(Cont.)LogicalCourseOverview(Cont.)Whatchallengesremain?18-Overviewandsummary.

[alloftheabove]Whathavewelearned.Wheredowegofromhere?CourseOverview(Cont.)WhatchaHOMEWORKLearnPythonprogramminglanguagebyyourselves.CollectrelatedconferencesandjournalsonAI,eachstudentatleastgather4kindofconferencesorjournals,andlisttheirscope.Select3from13exercisesofchapter1.HOMEWORKLearnPythonprogrammi1.IntroductionTo

ArtificialintelligenceAutumn2012Instructor:WangXiaolongHarbinInstituteofTechnology,ShenzhenGraduateSchoolIntelligentComputationResearchCenter(ICRC,HITSGS)

1.IntroductionTo

ArtificialCourseOutlineWhyStudyAI?WhatisAI?WhatcanAIdo?WhattasksrequireAI?HowtoachieveAI?BranchesofAIExamplesofAIAIinChinaAIinHITCourseOutlineWhyStudyAI?WhyStudyAI?SearchenginesLaborScienceMedicine/DiagnosisAppliancesWhatelse?WhyStudyAI?SearchenginesLabWhyStudyAI?(Cont.)SonyAIBOWhyStudyAI?(Cont.)SonyAIBOhWhyStudyAI?(Cont.)USmilitaryWhyStudyAI?(Cont.)USmilitarWhyStudyAI?(Cont.)NaturalLanguageQuestionAnswering/projects/infolab/WhyStudyAI?(Cont.)NaturalLaWhyStudyAI?(Cont.)RobotTeamsUSCroboticsLabWhyStudyAI?(Cont.)RobotTeamWhatisAI?Theexcitingnewefforttomakecomputersthinks…machinewithminds,inthefullandliteralsense”(Haugeland1985)“Theartofcreatingmachinesthatperformfunctionsthatrequireintelligencewhenperformedbypeople”(Kurzweil,1990)“Thestudyofmentalfacultiesthroughtheuseofcomputationalmodels”(Charniaketal.1985)Afieldofstudythatseekstoexplainandemulateintelligentbehaviorintermsofcomputationalprocesses”(Schalkol,1990)SystemsthatthinklikehumansSystemsthatthinkrationallySystemsthatactlikehumansSystemsthatactrationallyViewsofAIfallintofourcategories:Thetextbookadvocates"actingrationally"WhatisAI?TheexcitingnewefWhatisAI?(Cont.)ActingHumanly:TheTuringTestAlanTuring's1950articleComputingMachineryandIntelligencediscussedconditionsforconsideringamachinetobeintelligent.“Canmachinesthink?”“Canmachinesbehaveintelligently?”TheTuringtest(TheImitationGame):Operationaldefinitionofintelligence.Computerneedstopossess:Naturallanguageprocessing,Knowledgerepresentation,Automatedreasoning,andMachinelearningPredictedthatby2000,amachinemighthavea30%chanceoffoolingalaypersonfor5minutesAnticipatedallmajorargumentsagainstAIinfollowing50yearsArethereanyproblems/limitationstotheTuringTest?problemswiththeTuringTest:Turingtestisnotreproducible,constructive,andamenabletomathematicanalysis.Whataboutphysicalinteractionwithinterrogatorandenvironment?WhatisAI?(Cont.)ActingHumanWhatisAI?(Cont.)ActingHumanly:TheFullTuringTestThecomputerpassesthetestifahumaninterrogator,afterposingsomewrittenquestions,cannottellwhetherthewrittenresponsescomefromapersonornot.Requiresphysicalinteractionandneedsperceptionandactuation.WhatwouldacomputerneedtopasstheTuringtest?Naturallanguageprocessing:tocommunicatewithexaminer.Knowledgerepresentation:tostoreandretrieveinformationprovidedbeforeorduringinterrogation.Automatedreasoning:tousethestoredinformationtoanswerquestionsandtodrawnewconclusions.Machinelearning:toadapttonewcircumstancesandtodetectandextrapolatepatterns.Vision(forTotalTuringtest):torecognizetheexaminer’sactionsandvariousobjectspresentedbytheexaminer.Motorcontrol(totaltest):toactuponobjectsasrequested.Othersenses(totaltest):suchasaudition,smell,touch,etc.WhatisAI?(Cont.)ActingHumanWhatisAI?(Cont.)ThinkingHumanly:CognitiveScience1960“CognitiveRevolution”:information-processingpsychologyreplacedbehaviorism.Cognitivesciencebringstogethertheoriesandexperimentalevidencetomodelinternalactivitiesofthebrain.Whatlevelofabstraction?“Knowledge”or“Circuits”?Howtovalidatemodels?Predictingandtestingbehaviorofhumansubjects(top-down)Directidentificationfromneurologicaldata(bottom-up)Buildingcomputer/machinesimulatedmodelsandreproduceresults(simulation)ThinkingRationally:LawsofThoughtAristotle(~450B.C.)attemptedtocodify“rightthinking”

Whatarecorrectarguments/thoughtprocesses?E.g.,“Socratesisaman,allmenaremortal;thereforeSocratesismortal”SeveralGreekschoolsdevelopedvariousformsoflogic:

notationplusrulesofderivationforthoughts.Problems:Uncertainty:Notallfactsarecertain(e.g.,theflightmightbedelayed).Resourcelimitations:-Notenoughtimetocompute/process-Insufficientmemory/disk/etc-Etc.WhatisAI?(Cont.)ThinkingHumWhatisAI?(Cont.)ActingRationally:TheRationalAgentRationalbehavior:Doingtherightthing!Therightthing:ThatwhichisexpectedtomaximizetheexpectedreturnProvidesthemostgeneralviewofAIbecauseitincludes:Correctinference(“Lawsofthought”)UncertaintyhandlingResourcelimitationconsiderations(e.g.,reflexvs.deliberation)Cognitiveskills(NLP,AR,knowledgerepresentation,ML,etc.)Advantages:MoregeneralItsgoalofrationalityiswelldefinedRationalagentsWhatisAI?(Cont.)ActingRatioWhatCanAIDo?

Quiz:Whichofthefollowingcanbedoneatpresent?Playadecentgameoftabletennis?Drivesafelyalongacurvingmountainroad?DrivesafelyalongTelegraphAvenue?Buyaweek'sworthofgroceriesontheweb?Buyaweek'sworthofgroceriesatBerkeleyBowl?Discoverandproveanewmathematicaltheorem?Conversesuccessfullywithanotherpersonforanhour?Performacomplexsurgicaloperation?Unloadadishwasherandputeverythingaway?TranslatespokenEnglishintospokenSwedishinrealtime?Writeanintentionallyfunnystory?WhatCanAIDo?

Quiz:WhichofWhattasksrequireAI?“AIisthescienceandengineeringofmakingintelligentmachineswhichcanperformtasksthatrequireintelligencewhenperformedbyhumans…”TasksthatrequireAI:SolvingadifferentialequationBrainsurgeryInventingstuffPlayingJeopardyPlayingWheelofFortuneWhataboutwalking?Whataboutgrabbingstuff?Whataboutpullingyourhandawayfromfire?WhataboutwatchingTV?Whataboutdaydreaming?WhattasksrequireAI?“AIistHowtoachieveAI?HowisAIresearchdone?AIresearchhasboththeoreticalandexperimental

sides.Theexperimentalsidehasbothbasicandappliedaspects.Therearetwomainlinesofresearch:Oneisbiological,basedontheideathatsincehumansareintelligent,AIshouldstudyhumansandimitatetheirpsychologyorphysiology.Theotherisphenomenal,basedonstudyingandformalizingcommonsensefactsabouttheworldandtheproblemsthattheworldpresentstotheachievementofgoals.Thetwoapproachesinteracttosomeextent,andbothshouldeventuallysucceed.Itisarace,butbothracersseemtobewalking.[JohnMcCarthy]HowtoachieveAI?HowisAIreBranchesofAILogicalAISearchNaturallanguageprocessingPatternrecognitionKnowledgerepresentationInferenceFromsomefacts,otherscanbeinferred.

AutomatedreasoningLearningfromexperiencePlanningTogenerateastrategyforachievingsomegoal.EpistemologyStudyofthekindsofknowledgethatarerequiredforsolvingproblemsintheworld.

OntologyStudyofthekindsofthingsthatexist.InAI,theprogramsandsentencesdealwithvariouskindsofobjects,andwestudywhatthesekindsareandwhattheirbasicpropertiesare.

GeneticprogrammingEmotions???…BranchesofAILogicalAIExamplesofAIQ&AWastonNamedafterthefounderofIBM--ThomasJ.WastonWon$1,000,000inGameJeopardyin2011,defeated2humancompetetersSource:encyclopaedia,wikipedia,webpages,structuraldata100algorithms,90servers,3sresponsetimeJeopardyGameRules:-Getpointsforcorrectanswers-LosepointsforincorrectonesExamplesofAIQ&AWastonJeopaExamplesofAIQ&AALICEArtificialLinguisticInternetComputerEntityChatrobot,computerprogramWon2000,2001,2004LoebnerPrizeExamplesofAIQ&AALICEExamplesofAIQ&AWolframAlphaAbundantknowledgebaseAbilitytodoscientificcalculationExamplesofAIQ&AWolframAlpExamplesofAIQ&ASiriStartedoniPhone4S,voicecontrolContextanalysis(eg.weather),Googlesearchwikipedia,YelpWolframAlphaExamplesofAIQ&ASiriAiinChinaSpeechinput863NaturallanguageprocessingPatternrecognitionMachinelearningMachineTranslationMayIhelpyou?Howareyou?Howoldareyou?…AiinChinaSpeechinputAiinHITRobotGameMachineTranslationSentenceinputSiguoGameQ&ASearchEngine…AiinHITRobotGameAiinHITRobotGameAiinHITRobotGameAiinHITQ&AAiinHITQ&AAIPrehistoryPhilosophyLogic,methodsofreasoning,mindasphysicalsystemfoundationsoflearning,language,rationalityMathematicsFormalrepresentationandproofalgorithms,computation,(un)decidability,(in)tractability,probabilityEconomicsutility,decisiontheoryNeurosciencephysicalsubstrateformentalactivityPsychologyphenomenaofperceptionandmotorcontrol,experimentaltechniquesComputerEngineeringbuildingfastcomputersControltheorydesignsystemsthatmaximizeanobjectivefunctionovertimeLinguisticsknowledgerepresentation,grammarAIPrehistoryPhilosophyAIHistoryAIHistoryAIStateoftheartHavethefollowingbeenachievedbyAI?World-classchessplayingPlayingtabletennisCross-countrydrivingSolvingmathematicalproblemsDiscoverandprovemathematicaltheoriesEngageinameaningfulconversationUnderstandspokenlanguageObserveandunderstandhumanemotionsExpressemotions…DeepBluedefeatedthereigningworldchesschampionGarryKasparovin1997Provedamathematicalconjecture(Robbinsconjecture)unsolvedfordecadesNohandsacrossAmerica(drivingautonomously98%ofthetimefromPittsburghtoSanDiego)Duringthe1991GulfWar,USforcesdeployedanAIlogisticsplanningandschedulingprogramthatinvolvedupto50,000vehicles,cargo,andpeopleNASA'son-boardautonomousplanningprogramcontrolledtheschedulingofoperationsforaspacecraftProverbsolvescrosswordpuzzlesbetterthanmosthumans…AIStateoftheartHavethefoCourseOverviewGeneralIntroduction01-Introduction.

[AIMACh1]CourseSchedule.Homeworks,examsandgrading.Coursematerial,TAs.WhystudyAI?WhatisAI?TheTuringtest.Rationality.BranchesofAI.Researchdisciplinesconn-ectedtoandatthefoundationofAI.BriefhistoryofAI.Challengesforthefuture.Overviewofclasssyllabus.02-IntelligentAgents.

[AIMACh2]Whatis anintelligentagent?Examples.Doingtheright thing(rationalaction).Performancemeasure. Autonomy.Environmentandagentdesign. Structureofagents.Agenttypes.Reflexagents. Reactiveagents.Reflexagentswithstate. Goal-basedagents.Utility-basedagents.Mobile agents.Informationagents.sensorseffectorsAgentCourseOverviewGeneralIntroduCourseOverview(Cont.)Howcanwesolvecomplexproblems?03-Problemsolvingandsearch.[AIMACh3]Example:measuringproblem.Typesofproblems.Moreexampleproblems.Basicideabehindsearchalgorithms.Complexity.Combi

溫馨提示

  • 1. 本站所有資源如無特殊說明，都需要本地電腦安裝OFFICE2007和PDF閱讀器。圖紙軟件為CAD,CAXA,PROE,UG,SolidWorks等.壓縮文件請(qǐng)下載最新的WinRAR軟件解壓。
  • 2. 本站的文檔不包含任何第三方提供的附件圖紙等，如果需要附件，請(qǐng)聯(lián)系上傳者。文件的所有權(quán)益歸上傳用戶所有。
  • 3. 本站RAR壓縮包中若帶圖紙，網(wǎng)頁內(nèi)容里面會(huì)有圖紙預(yù)覽，若沒有圖紙預(yù)覽就沒有圖紙。
  • 4. 未經(jīng)權(quán)益所有人同意不得將文件中的內(nèi)容挪作商業(yè)或盈利用途。
  • 5. 人人文庫網(wǎng)僅提供信息存儲(chǔ)空間，僅對(duì)用戶上傳內(nèi)容的表現(xiàn)方式做保護(hù)處理，對(duì)用戶上傳分享的文檔內(nèi)容本身不做任何修改或編輯，并不能對(duì)任何下載內(nèi)容負(fù)責(zé)。
  • 6. 下載文件中如有侵權(quán)或不適當(dāng)內(nèi)容，請(qǐng)與我們聯(lián)系，我們立即糾正。
  • 7. 本站不保證下載資源的準(zhǔn)確性、安全性和完整性, 同時(shí)也不承擔(dān)用戶因使用這些下載資源對(duì)自己和他人造成任何形式的傷害或損失。

最新文檔

評(píng)論

0/150

提交評(píng)論