What is artificial intelligence Definition, top 10 types and examples.pdf

What is artificial intelligence?
Although many definitions of artificial intelligence (AI) have emerged over the past few decades, John McCarthy provided the following definition in this 2004 paper (link is located outside ibm.com): MASU. Especially intelligent computer programs. It deals with the same task of using computers to understand human intelligence, but AI does not need to be limited to biologically observable methods.
Definition of artificial intelligence
Artificial intelligence is the imitation of human intelligence processes by machines, especially computer systems. Typical applications of AI include expert systems, natural language processing, speech recognition, and machine vision.
How does artificial intelligence (AI) work?
As the hype around AI grows, vendors are making efforts to promote how AI is used in their products and services. Often, what they call AI is just a component of technologies like machine learning. AI requires specialized hardware and software infrastructure to write and train machine learning algorithms. Although no single programming language is synonymous with AI, Python, R, Java, C++, and Julia have features that are popular among AI developers.
Generally, AI systems work by ingesting large amounts of labeled training data, analyzing correlations and patterns in the data, and using these patterns to predict future situations. This way, given examples of text, chatbots can learn to generate authentic-like conversations with people. Image recognition tools can also learn to recognize and describe objects in images by considering millions of examples. New and rapidly advancing generic AI technology allows you to create realistic text, images, music, and other media.
Artificial intelligence programming focuses on cognitive skills such as:
• Learn: This aspect of AI programming focuses on taking data and creating rules to turn it into actionable information. Rules, called algorithms, provide step-by-step instructions for computing devices to accomplish a particular task.
• Logic. This aspect of AI programming focuses on selecting the appropriate algorithm to achieve the desired result.
• Self-correction: This aspect of AI programming is designed to continuously improve the algorithms and provide the most accurate results possible.
• Creativity. This aspect of AI uses neural networks, rule-based systems, statistical methods, and other AI techniques to generate new images, new text, new music, and new ideas.
Differences between AI, machine learning and deep learning
AI, machine learning, and deep learning are common terms in enterprise IT, especially when companies use them interchangeably in marketing materials. But there are differences too. The term AI was coined in the 1950s and refers to the emulation of human intelligence by machines. A constantly changing set of capabilities is incorporated as new technologies are developed. Technologies falling under the umbrella of AI include machine learning and deep lea

1. Whatisartificialintelligence?
Definition,top10typesand
examples

2. Whatisartificialintelligence?
Althoughmanydefinitionsofartificialintelligence(AI)haveemergedoverthe
pastfewdecades,JohnMcCarthyprovidedthefollowingdefinitioninthis2004
paper(linkislocatedoutsideibm.com):MASU.Especiallyintelligentcomputer
programs.Itdealswiththesametaskofusingcomputerstounderstand
humanintelligence,butAIdoesnotneedtobelimitedtobiologically
observablemethods.
Definitionofartificialintelligence
Artificialintelligenceistheimitationofhumanintelligenceprocessesby
machines,especiallycomputersystems.TypicalapplicationsofAIinclude
expertsystems,naturallanguageprocessing,speechrecognition,and
machinevision.

3. Howdoesartificialintelligence(AI)
work?
AsthehypearoundAIgrows,vendorsaremakingeffortstopromotehowAIis
usedintheirproductsandservices.Often,whattheycallAIisjusta
componentoftechnologieslikemachinelearning.AIrequiresspecialized
hardwareandsoftwareinfrastructuretowriteandtrainmachinelearning
algorithms.AlthoughnosingleprogramminglanguageissynonymouswithAI,
Python,R,Java,C++,andJuliahavefeaturesthatarepopularamongAI
developers.
Generally,AIsystemsworkbyingestinglargeamountsoflabeledtraining
data,analyzingcorrelationsandpatternsinthedata,andusingthesepatterns
topredictfuturesituations.Thisway,givenexamplesoftext,chatbotscan
learntogenerateauthentic-likeconversationswithpeople.Imagerecognition
toolscanalsolearntorecognizeanddescribeobjectsinimagesby
consideringmillionsofexamples.NewandrapidlyadvancinggenericAI
technologyallowsyoutocreaterealistictext,images,music,andothermedia.
Artificialintelligenceprogrammingfocuseson
cognitiveskillssuchas:
•Learn:ThisaspectofAIprogrammingfocusesontakingdataandcreating
rulestoturnitintoactionableinformation.Rules,calledalgorithms,provide
step-by-stepinstructionsforcomputingdevicestoaccomplishaparticular
task.
•Logic.ThisaspectofAIprogrammingfocusesonselectingtheappropriate
algorithmtoachievethedesiredresult.

4. •Self-correction:ThisaspectofAIprogrammingisdesignedtocontinuously
improvethealgorithmsandprovidethemostaccurateresultspossible.
•Creativity.ThisaspectofAIusesneuralnetworks,rule-basedsystems,
statisticalmethods,andotherAItechniquestogeneratenewimages,new
text,newmusic,andnewideas.
DifferencesbetweenAI,machinelearningand
deeplearning
AI,machinelearning,anddeeplearningarecommontermsinenterpriseIT,
especiallywhencompaniesusetheminterchangeablyinmarketingmaterials.
Buttherearedifferencestoo.ThetermAIwascoinedinthe1950sandrefers
totheemulationofhumanintelligencebymachines.Aconstantlychangingset
ofcapabilitiesisincorporatedasnewtechnologiesaredeveloped.
TechnologiesfallingundertheumbrellaofAIincludemachinelearningand
deeplearning.
Machinelearningallowssoftwareapplicationstomoreaccuratelypredict
outcomeswithouthavingtobeexplicitlyprogrammed.Machinelearning
algorithmsusehistoricaldataasinputtopredictnewoutputvalues.This
approachhasbecomeveryeffectiveasthenumberoflargedatasetsonwhich
itcanbetrainedincreases.Deeplearning,asubsetofmachinelearning,is
basedonanunderstandingofhowthebrainisstructured.Theuseofartificial
neuralnetworkstructuresthroughdeeplearningpowersrecentadvancesin
AI,suchasself-drivingcarsandchatGPTs.
Whyisartificialintelligenceso
important?

5. AIisimportantbecauseithasthepotentialtochangethewaywelive,work,
andplay.Itisusedeffectivelybybusinessestoautomatetasksperformedby
humans,suchascustomerservicetasks,leadgeneration,frauddetection,
andqualitycontrol.Inmanyareas,AIcanworkmuchbetterthanhumans.
Especiallywhenitcomestorepetitive,detail-orientedtasks,suchasanalyzing
largeamountsoflegaldocumentstoensurethatrelevantfieldsarefilledout
properly,AItoolsareoftenmorethanenoughtogetthejobdone.arefaster
andwithfewererrors.BecauseAIcanprocesslargedatasets,itcanalsogive
businessesinsightsintotheiroperationsthattheywereunawareof.The
rapidlyexpandingaudienceforgenericAItoolswillbecomeimportantinareas
rangingfromeducationandmarketingtoproductdesign.
Infact,advancesinAItechnologyhavenotonlycontributedtoanexplosionin
efficiencybuthavealsoopenedthedoortoentirelynewbusiness
opportunitiesforsomelargecompanies.BeforethecurrentwaveofAI,itwas
hardtoimagineusingcomputersoftwaretoconnectpassengerswithtaxis.
ButUberbecameaFortune500companybydoingso.
AIisattheheartoftoday’slargestandmostsuccessfulcompanies,including
Alphabet,Apple,Microsoft,andMeta,whichuseAItechnologytoimprove
theiroperationsandoutperformtheircompetitors.AtAlphabetInc.’sGoogle,
forexample,AIisattheheartofGoogleBrain,whichinventedthecompany’s
searchengine,Waymoself-drivingcars,andelectricaltransformers.
Whataretheadvantagesand
disadvantagesofartificial
intelligence?
ArtificialneuralnetworksanddeeplearningAItechnologiesaredeveloping
rapidly.ThemainreasonforthisisthatAIcanprocesslargeamountsofdata
muchfasterandmakemoreaccuratepredictionsthanhumans.

6. Whilehumanresearchersareoverwhelmedbythevastamountsofdata
createdeveryday,AIapplicationsusingmachinelearningcantakethatdata
andinstantlytransformitintoactionableinformation.Atthetimeofthiswriting,
themaindrawbackofAIisthehighcostofprocessingthelargeamountsof
datarequiredforAIprogramming.AsAItechnologyisincorporatedintomore
productsandservices,organizationsmustalsopayattentiontothepotential
ofAItocreatebiasedanddiscriminatorysystems,intentionallyor
unintentionally.
Advantagesofartificialintelligence
Someofthebenefitsofartificialintelligenceinclude:
•Goodatdetailedwork.AIhasproventobeasgoodorbetterthandoctors
atdiagnosingsomecancers,suchasbreastcancerandmelanoma.
•Spendlesstimeondata-intensivetasks.AIiswidelyusedin
data-intensiveindustriessuchasbanking,securities,pharmaceuticals,and
insurancetoreducethetimetakentoanalyzelargedatasets.Forexample,
financialservicesroutinelyuseAItoprocessloanapplicationsanddetect
fraud.
•Laborsavingsandincreasedproductivity.Anexamplehereistheuseof
warehouseautomation.Thishasincreasedduringthepandemicandis
expectedtoincreasewiththeintegrationofAIandmachinelearning.
•Providesconsistentresults.ThebestAItranslationtoolsprovideahigh
levelofconsistencyandgivesmallbusinessestheabilitytoreachcustomers
intheirnativelanguage.
•Increasecustomersatisfactionthroughpersonalization.AIcan
personalizecontent,messages,advertisements,recommendations,and
websitesforindividualcustomers.

7. AI-poweredvirtualagentsarealwaysavailable.AIprogramsdonotneedto
sleeportakebreaks,andtheyprovide24/7service.
Disadvantagesofartificialintelligence
Thedisadvantagesofartificialintelligence
include:
•expensive.
requiresdeeptechnicalexpertise.
•ThereisalimitedsupplyofqualifiedworkerstobuildAItools.
•Reflecttrainingdatabiasatscale.
•Lackofabilitytogeneralizefromonetasktoanother.
•Endhumanemploymentandincreaseunemployment.
StrongAIvs.weakAI
AIcanbeclassifiedasweakorstrong.

8. •WeakAI(alsoknownasnarrowAI)isdesignedandtrainedtoperform
specifictasks.Industrialrobotsandvirtualpersonalassistants,likeApple’s
Siri,useweakAI.
•StrongAI,alsoknownasartificialgeneralintelligence(AGI),refersto
programmingthatcanreplicatethecognitiveabilitiesofthehumanbrain.
Givenanunfamiliartask,powerfulAIsystemscanusefuzzylogictoapply
knowledgefromonedomaintoanotherandautonomouslyfindasolution.In
theory,apowerfulAIprogramshouldbeabletopassboththeTuringTestand
theChineseRoomArgument.
Whatarethefourtypesofartificial
intelligence?
ErlendHintz,assistantprofessorofintegrativebiologyandcomputerscience
andengineeringatMichiganStateUniversity,saidAIcanbeclassifiedinto
fourtypes.AIbeginswithtask-specificintelligentsystemswidelyusedtoday
andevolvesintoperceptualsystems.Stillexists.Thecategoriesareas
follows:.
•Type1:ReactiveMachine.TheseAIsystemshavenomemoryandare
task-specific.AnexampleisDeepBlue,IBM’schessprogramthatdefeated
GarryKasparovinthe1990s.DeepBluecanidentifyandpredictpiecesona
chessboard,butbecausehehasnomemory,hecannottransferpast
experiencestofutureexperiences.
•Type2:Limitedmemory.TheseAIsystemshavememories,sotheycanuse
pastexperiencetomakefuturedecisions.Someofthedecision-making
capabilitiesofself-drivingcarsaredesignedthisway.
•Type3:TheoryofMind.Thetheoryofmindisapsychologicalterm.When
appliedtoAI,itgivessystemsthesocialintelligencetounderstandemotions.

9. ThistypeofAIwouldbeabletoinferhumanintentionsandpredictbehavior.
ThisisaskillthatAIsystemsneedtobecomeessentialmembersofhuman
teams.
•Type4:self-aware.Inthiscategory,AIsystemsareself-aware,whichgives
themconsciousness.Aself-awaremachineunderstandsitscurrentstate.This
typeofAIdoesnotexistyet.
Whatareexamplesofartificial
intelligencetechnology,andhowisit
usedtoday?
AIisembeddedinmanydifferenttypesof
technology.Herearesevenexamples.
Automation.AutomationtoolscanbecombinedwithAItechnologytomeasure
thenumberandtypeoftasksperformed.Oneexampleisroboticprocess
automation(RPA),atypeofsoftwarethatautomatesrepetitive,rule-based
dataprocessingtaskstraditionallyperformedbyhumans.CombiningRPAwith
machinelearningandnewAItoolscanautomatelargepartsofacompany’s
operations,allowingRPA’sstrategicbotstodrawintelligencefromAIand
respondtoprocesschanges.
machinelearning.Itisthescienceofrunningacomputerwithout
programming.Deeplearningisasubsetofmachinelearningand,invery
simpleterms,canbethoughtofasautomatedpredictiveanalysis.Thereare
threetypesofmachinelearningalgorithms.
•supervisedlearning.Datasetsarelabeledsothatpatternscanbedetected
andusedtolabelnewdatasets.

10. •Unsupervisedlearning.Thedatasetisunlabeledandsortedaccordingto
similarityordifference.
•Reinforcementlearning.Althoughthedatasetisnotlabeled,itprovides
feedbacktotheAIsystemafterperformingoneormoreactions.
Machinevision.Thistechnologygivesmachinestheabilitytosee.Machine
visionusescameras,analog-to-digitalconversion,anddigitalsignal
processingtocaptureandanalyzevisualinformation.Althoughoften
comparedtohumanvision,machinevisionisnottiedtobiologyandcan,for
example,beprogrammedtoseethroughwalls.Itisusedinawidevarietyof
applications,fromsignaturerecognitiontomedicalimageanalysis.Computer
visionfocusesonmachine-basedimageprocessingandisoftenconfused
withmachinevision.
NaturalLanguageProcessing(NLP).Itistheprocessingofhuman
languagebyacomputerprogram.Theoldestandbest-knownexampleofNLP
isspamdetection,whichexaminesthesubjectandbodyofanemailto
determinewhetheritisspam.CurrentapproachestoNLParebasedon
machinelearning.NLPtasksincludetexttranslation,sentimentanalysis,and
speechrecognition.
Robotics.Thisengineeringfieldfocusesonthedesignandmanufacturingof
robots.Robotsareoftenusedtoperformtasksthataredifficultforhumansto
performorthataredifficulttoperformconsistently.Forexample,robotsare
usedoncarmanufacturingassemblylinesandbyNASAtomovelarge
objectsinspace.Researchersarealsousingmachinelearningtocreate
robotsthatcaninteractinsocialenvironments.
Self-drivingcar.Self-drivingcarsuseacombinationofcomputervision,
imagerecognition,anddeeplearningtocreateautomatedskillstosteerthe
vehiclewhilestayinginaspecificlaneandavoidingunexpectedobstacles
suchaspedestrians.Meat.

11. Creationoftext,images,andaudio.GenerativeAItechniquesthatcreate
differenttypesofmediafromtextpromptsarebeingwidelyusedacross
enterprisestocreateunlimitedtypesofcontent,fromphotorealisticarttoemail
responsesandscreenplays.Wasimplemented.
Whataretheapplicationsofartificial
intelligence?
Artificialintelligenceismakinginroadsinto
variousmarkets.Hereare11examples.
AIinhealthcareThebiggeststakesaretoimprovepatientoutcomesand
reducecosts.Companiesareusingmachinelearningtomakebetterand
fastermedicaldiagnosesthanhumans.Oneofthemostfamoushealthcare
technologiesisIBMWatson.Understandnaturallanguageandbeableto
answernaturallanguagequestions.Thesystemminespatientdataandother
availabledatasourcestogeneratehypothesesandpresentthemusinga
confidencescoringscheme.OtherAIapplicationsincludeusingonlinevirtual
medicalassistantsandchatbotstohelppatientsandhealthcarecustomers
findmedicalinformation,scheduleappointments,understandbilling
processes,andcompleteotheradministrativeprocesses.Theseinclude:A
varietyofAItechnologiesarealsobeingusedtopredict,fight,andunderstand
pandemicssuchasthecoronavirusdisease(COVID-19).
AIinbusinessMachinelearningalgorithmsareintegratedintoanalyticsand
customerrelationshipmanagement(CRM)platformstogaininsightsonhow
tobetterservecustomers.Chatbotsareintegratedintowebsitestoprovide
immediateservicetocustomers.TherapidadvancementofgenericAI
technologieslikeChatGPTisexpectedtohavefar-reachingeffects,including
jobcuts,changesinproductdesign,anddisruptingbusinessmodels.

12. AIineducationAIautomatesgrading,freeingupteacherstospendmore
timeonothertasks.Assessyourstudents,adapttotheirneeds,andhelp
themlearnattheirownpace.AItutorscanprovideextrasupporttostudents
andkeepthemontrack.Thistechnologycouldalsochangewhereandhow
studentslearnandperhapsevenreplacesometeachers.Asdemonstratedby
ChatGPIT,GoogleBard,andotherlarge-scalelanguagemodels,generative
AIcanhelpteacherscreatecurriculumandothercontentandengage
studentsinnewways.Theadventofthesetoolsalsorequiresteachersto
rethinkstudentassignmentsandtestsandmodifyplagiarismpolicies.
AIinfinance.AIinpersonalfinanceapplicationslikeIntuitMintandTurboTax
isdisruptingfinancialinstitutions.Suchapplicationscollectpersonaldataand
providefinancialadvice.Otherprograms,suchasIBMWatson,havebeen
appliedtothehome-buyingprocess.Today,artificialintelligencesoftware
drivesmostofthetradingonWallStreet.
AIinlawThelegaldiscoveryprocess,orreviewingdocuments,isoftena
difficulttaskforhumans.UsingAItoautomatelabor-intensiveprocessesinthe
legalindustrycansavetimeandimprovecustomerservice.Lawfirmsuse
machinelearningtodescribedataandpredictoutcomes,usecomputervision
toclassifyandextractinformationfromdocuments,anduseNLPtointerpret
requestsforinformation.Are.
AIinentertainmentandmediaEntertainmentbusinessesuseAI
technologyfortargetedadvertising,contentrecommendationsanddelivery,
frauddetection,screenwriting,andfilmproduction.Automatedjournalism
helpsnewsroomsstreamlinemediaworkflowswhilereducingtime,costs,and
complexity.NewsroomsuseAItoautomateroutinetaskslikedataentryand
proofreading.Researchtopicsandhelpwithtitles.Questionsremainabout
howjournalismcanreliablyuseChatGPTandothergenerativeAItogenerate
content.
AIinsoftwarecodingandITprocesses.NewgenerativeAItoolscan
generateapplicationcodebasedonnaturallanguagesignals,butthesetools
arestillintheirinfancyandareunlikelytoreplacesoftwareengineersinthe

13. nearfuture.AIisalsobeingusedtoautomatemanyITprocesses,including
dataentry,frauddetection,customerservice,predictiveness,andsecurity.
Security.SecurityvendorsrankAIandmachinelearninghighonthelistof
buzzwordstheyusetomarkettheirproducts,sobuyersshouldapproachthem
withcaution.Nevertheless,AItechniqueshavebeensuccessfullyappliedto
variousaspectsofcybersecurity,suchasdetectinganomalies,solvingfalse
positiveproblems,andanalyzingbehavioralthreats.Organizationsuse
machinelearninginsecurityinformationandeventmanagement(SIEM)
softwareandrelatedareastodetectanomaliesandidentifysuspiciousactivity
thatsignalsathreat.Byanalyzingdataandusinglogictoidentifysimilarities
toknownmaliciouscode,AIcancounternewattacksmuchfasterthanhuman
workersorpreviousiterationsofthetechnology.
AIinbanking.Bankshavesuccessfullydeployedchatbotstoinform
customersaboutservicesandoffersandprocesstransactionswithouthuman
intervention.AIvirtualassistantsareusedtoimproveandreducethecostof
bankingregulatorycompliance.BankingorganizationsareusingAItoimprove
lendingdecisions,setloanlimits,andidentifyinvestmentopportunities.
AIintransportationsectorManufacturinghasbeenattheforefrontof
incorporatingrobotsintotheworkflow.Inadditiontoitsfundamentalroleinthe
operationofautonomousvehicles,AItechnologyisalsobeingusedinthe
transportationsectortomanagetraffic,predictflightdelays,andimprovethe
safetyandefficiencyofmaritimetransportation.Inthesupplychain,AIis
replacingtraditionalmethodsofforecastingdemandandanticipating
disruption.ThistrendwasacceleratedbyCOVID-19,whenmanycompanies
becamedistressedbytheimpactoftheglobalpandemiconthesupplyand
demandofgoods.
Howdoesartificialintelligence(AI)
work?

14. Butdecadesbeforethisdefinition,theconversationaboutartificialintelligence
beganwithAlanTuring’sseminalbookComputingMachineryandIntelligence,
publishedin1950(linkislocatedoutsideibm.com).Itwasshown.Inthis
paper,Turingoftensays:Knownasthe“FatherofComputerScience,heasks
thequestion,“Canmachinesthink?”Fromthere,heproposedatestnow
famouslyknownasthe“TuringTest.”.Inthistest,ahumaninterrogator
attemptstodistinguishbetweencomputerandhumantextresponses.
Althoughthistesthasreceivedmuchscrutinysinceitsrelease,itisstillan
importantpartofthehistoryofAIaswellasanongoingconceptwithin
philosophy,asitleveragesideasaboutlinguistics..
StuartRussellandPeterNervingthenpublishedArtificialIntelligence:A
ModernApproach(linkoffibm.com),whichbecameoneoftheleading
textbooksinAIresearch.Init,theyhighlightfourpossiblegoalsordefinitions
ofAIthatdifferentiatecomputersystemsbasedonrationality,thinking,and
action.
Humanapproach:
●Systemsthatthinklikehumans
●Systemsthatactlikehumans
Idealapproach:
●Systemsthatthinkrationally
●Systemsthatactrationally
AlanTuring’sdefinitionwouldfallintothecategoryof“systemsthat
behavelikehumans.”.
Initssimplestform,artificialintelligenceisafieldthatcombinescomputer
sciencewithrobustdatasetstoenableproblem-solving.Italsoincludesthe
subfieldsofmachinelearninganddeeplearning,whichareoftenmentioned

15. alongsideartificialintelligence.ThesefieldsincludeAIalgorithmsthataimto
createexpertsystemsthatmakepredictionsandclassificationsbasedon
inputdata.
Overthepastfewyears,artificialintelligencehasgonethroughseveralcycles
ofhype,butevenforskeptics,OpenAI’sreleaseofChatGPTseemstobea
turningpoint.ThelasttimegenerativeAIbecamethisbig,thebreakthroughs
wereincomputervision,andnowwe’reseeingbreakthroughsinnatural
languageprocessing.Andit’snotjustlanguage.Generativemodelscanalso
learngrammarsforsoftwarecode,molecules,naturalimages,andmanyother
typesofdata.
Theapplicationsofthistechnologyaregrowingeveryday,andwearejust
beginningtoexploreitspotential.ButasthehypearoundtheuseofAIin
businessgrows,theconversationaroundethicsbecomesincreasingly
important.LearnmoreaboutIBM’spositionintheAIethicsconversation.
Typesofartificialintelligence:
WeakAIvs.strongAI
WeakAI(alsoknownasnarrowAIornarrowartificialintelligence(ANI))isAI
thatistrainedandfocusedonperformingaspecifictask.MostoftheAI
aroundustodayispoweredbyweakAI.ThistypeofAIisfarfromweak,so
“narrow”mightbeamoreaccuratedescription.Thisenablesextremely
powerfulapplicationssuchasApple’sSiri,Amazon’sAlexa,IBMWatson,and
self-drivingcars.
StrongAIincludesartificialgeneralintelligence(AGI)andartificial
superintelligence(ASI).Artificialgeneralintelligence(AGI),orgeneralAI,isa
theoreticalformofAIinwhichmachineshaveintelligencecomparabletothat
ofhumans.Itwillhaveaself-awareconsciousnesswiththeabilitytosolve
problems,learn,andplanforthefuture.Artificialsuperintelligence(ASI),also

16. knownassuperintelligence,willexceedtheintelligenceandcapabilitiesofthe
humanbrain.AlthoughpowerfulAIisstillentirelytheoreticalandcurrentlyhas
nopracticalexamplesofitsuse,thisdoesnotmeanthatAIresearchersare
notconsideringdevelopingit.Ontheotherhand,thebestexampleofanASI
maybeinsciencefictionworkssuchasHAL,theextraterrestrialandevil
computerassistantfrom2001:ASpaceOdyssey.
ComparisonofDeepLearningand
MachineLearning
Thetermsdeeplearningandmachinelearningareusedinterchangeably,soit
isworthpayingattentiontothenuancesbetweenthetwo.Asmentioned
earlier,deeplearningandmachinelearningarebothsubfieldsofartificial
intelligence,anddeeplearningisactuallyasubfieldofmachinelearning.
Deeplearningactuallyinvolvesneuralnetworks.“Depth”indeeplearning
referstoaneuralnetworkthathasthreeormorelayerswithinputsand
outputsandcanbeconsideredadeeplearningalgorithm.Thisisusually
representedusingthefollowingdiagram:.
Thedifferencebetweendeeplearningandmachinelearningliesinthe
learningmethodofeachalgorithm.Deeplearningautomatesmuchofthe
featureextractionpartoftheprocess,eliminatingsomeofthenecessary
manualinterventionandallowingtheuseoflargerdatasets.AsLexFriedman
saidaboveinhisMITlecture,deeplearningcanbethoughtofas“scalable
machinelearning.”.Classical,or“shallow,”machinelearningreliesheavilyon
humaninterventiontolearn.Humanexpertsdetermineahierarchyoffeatures
tounderstandthedifferencesbetweendatainputs.Trainingusuallyrequires
morestructureddata.
“Deep”machinelearningcantakeadvantageoflabeleddatasetstoinform
algorithms,alsoknownassupervisedlearning,butdoesnotnecessarily
requirelabeleddatasets.Itcanencapsulateunstructureddatainitsrawform

17. (text,images,etc.)andautomaticallydetermineahierarchyoffeaturesthat
distinguishdifferentcategoriesofdatafromeachother.Unlikemachine
learning,processingdatarequiresnohumanintervention,whichallowsyouto
extendmachinelearninginmoreinterestingways.
Theriseofgenerativemodels
GenerativeAIreferstodeeplearningmodelsthatcantakerawdata(for
example,theentireWikipediaoracollectionofRembrandts)and“learn”to
generatestatisticallyprobableoutputsinresponsetoprompts.Broadly
speaking,generativemodelsentailsimplification.
Representtrainingdataandextractquotesfromittocreatesimilarnewtasks.
However,itisnotthesameastheoriginaldata.
Generativemodelshavebeenusedinstatisticsformanyyearstoanalyze
numericaldata.However,theriseofdeeplearninghasmadeitpossibleto
extenddeeplearningtoimages,audio,andothercomplexdatatypes.The

18. firstclassofmodelstoaccomplishthiscrossoverfeatwastheVariable
Autoencoder(VAE),introducedin2013.VAEwasthefirstdeeplearning
modelwidelyusedtogeneraterealisticimagesandsounds.
“VAEopensthedoortodeepergenerativemodelingbymakingiteasierto
createmodels.
AkashSrivastava,agenericAIexpertattheMIT-IBMWatsonAILab.
“AlotofwhatwethinkofasgenericAItodaystartedhere.”
EarlyexamplesfrommodelslikeGPT-3,BERT,andDALL-E2showwhatis
possible.Thefuturewillhavemodelstrainedonawiderangeofunlabeled
datasetsthatcanbeusedforavarietyoftaskswithminimalfine-tuning.
Systemsthatperformspecifictasksinasingledomainarebeingreplacedby
pervasiveAIthatlearnsmoregenerallyandworksacrossdomainsand
problems.Fundamentalmodelstrainedonlarge,unlabeleddatasetsand
fine-tunedfordifferentapplicationsaredrivingthischange.
WhenitcomestogenericAI,theunderlyingmodelsarepredictedtochange
dramatically.
AccelerateAIadoptioninyourenterprise.Reducinglabelingrequirements
wouldleadtosignificantimprovements
TheaccuracyandefficiencyofAI-poweredautomationenabledbyAIwill
enablefarmorecompaniestodeployAIinawiderrangeofmission-critical
situations.ForIBM,thehopeisthatthepowerofthefoundationalmodelwill
eventuallybeavailabletoallbusinessesinafrictionlesshybridcloud
environment.
Artificialintelligenceapplications

19. Manyreal-worldapplicationsofAIsystemscurrentlyexist.Belowaresomeof
themostcommonusecases.
•Speechrecognition:Alsoknownasautomaticspeechrecognition(ASR),
computerspeechrecognition,orspeech-to-text,istheabilitytoprocess
humanspeechinwrittenformusingnaturallanguageprocessing(NLP)..
Manymobiledeviceshavevoicerecognitionbuiltintotheirsystemstoperform
voicesearches.Siri—orimproveaccessibilitywhenitcomestotexting.
•CustomerService:Onlinevirtualagentsarereplacinghumanagentsinthe
customerjourney.Theseincludecustomerengagementonwebsitesand
socialmediaplatformsbyansweringfrequentlyaskedquestions(FAQs)on
topicssuchasshippingandprovidingpersonalizedadvice,cross-selling
products,andsizesuggestionstousers.ChangethewayyouthinkExamples
includemessagingbotsone-commercesitesthatusevirtualagents,
messagingappslikeSlackandFacebookMessenger,andtaskstypically
performedbyvirtualorvoiceassistants.
•Computervision:ThisAItechnologyallowscomputersandsystemstoderive
meaningfulinformationfromdigitalimages,videos,andothervisualinputs
andtakeactionsbasedonthoseinputs.Thisabilitytoprovide
recommendationsdifferentiatesitfromimagerecognitiontasks.Computer
visionpoweredbyconvolutionalneuralnetworksisusedinphototaggingin
socialmedia,radiologyimaginginmedicine,self-drivingcarsinthe
automotiveindustry,andmuchmore.
•RecommendationEngine:AIalgorithmsusehistoricalconsumerbehavior
datatohelpdiscoverdatatrendsthatcanbeusedtodevelopmoreeffective
cross-sellingstrategies.Itisusedtorecommendrelevantadd-onsto
customersduringthecheckoutprocessatanonlineretailer.
•Automatedstocktrading:AI-poweredhigh-frequencytradingplatforms
designedtooptimizestockportfoliosexecutethousandsorevenmillionsof
tradesperdaywithouthumanintervention.

20. RelatedTopic:
Whatisartificialintelligence(AI)?HowmanytypesofAIarethere,andhow
manycoursesareinAI?
ChatGPT(ChatGenerativePre-TrendTransformer).
Top10BestTopicsforResearchinArtificialIntelligence
Whatisdeeplearning?
Whatisnaturallanguageprocessing?
Historyofartificialintelligence:Key
datesandnames
Theideaofthe“thinkingmachine”datesbacktoancientGreece.However,
sincetheadventofelectroniccomputing(andrelatedtosomeofthetopics
discussedinthisarticle),therehavebeensomesignificanteventsand
milestonesinthedevelopmentofartificialintelligence,including:
•1950:AlanTuringpublishedComputingMachineryandIntelligence.Turing,
famousforbreakingtheNaziEnigmacodeduringWorldWarII,proposedin
hispapertoanswerthequestion,“Canmachinesthink?”ItpresentstheTuring
testtodeterminewhethercomputerscandemonstrateintelligencesimilarto
humans(orresultssimilartointelligence).ThesignificanceoftheTuringtest
hasbeendebatedeversince.

21. •1967:FrankRosenblattcreatedtheMark1Perceptron,thefirstcomputer
basedonneuralnetworksthat“learned”throughtrialanderror.Justayear
later,MarvinMinskyandSeymourPapertpublishedabookcalled“The
Perceptron.”.Thisbookisagroundbreakingworkonneuralnetworksand,at
leastforthetimebeing,setsthetoneforfutureneuralnetworkresearch
projects.
•1980s:Neuralnetworksthattrainthemselvesusingbackpropagation
algorithmsbegintobewidelyusedinAIapplications.
•1997:IBM’sDeepBluedefeatedreigningworldchesschampionGarry
Kasparovinachessmatch(andrematch).
•2011:IBMWatsondefeateddefendingchampionsKenJenningsandBrad
RutterinJeopardy.
•2015:Baidu’sMinwasupercomputerusesaspecialtypeofdeepneural
networkcalledaconvolutionalneuralnetworktorecognizeandclassify
imageswithgreateraccuracythantheaveragehuman.
•2016:DeepMind’sAlphaGoprogram,poweredbydeepneuralnetworks,
defeatedworldGochampionLeeSodolinfivegames.Thiswinissignificant
giventhelargenumberofmovestobemadeasthegameprogresses(over
14.5trillioninjust4moves!).Afterthis,GooglereportedlyacquiredDeepMind
forUS$400million.
•2023:Withtheriseoflarge-scalelanguagemodelslikeChatGPT,orLLM;
SignificantchangesinAIperformanceanditspotentialtodriveenterprise
value.

22. ThesenewgenerativeAIpracticesallowyoutopre-traindeeplearning
models.
Largeamountsofunlabeledrawdata.
followme:Twitter,Facebook,LinkedIn,Instagram
Categories
ArtificialIntelligence,Tech,Technology
IPL2024PlayerAuction,Teams,SoldPlayer,BuyerslistandunsoldPlayerslist
WhileAppleVisionPromaynotsuiteveryuser’sneeds,itsfinelytunedfeaturesmakeita
standoutsuccessintherealmofpersonalizedtechnology.
9thoughtson“Whatisartificialintelligence?
Definition,top10typesandexamples”
​Pingback:WhatisGoogleGemini
​Pingback:Whatisnaturallanguageprocessing
​Pingback:Top10BestTopicsforResearchinArtificialIntelligence
​Pingback:Whatisdeeplearning?
​Pingback:ChatGPT
​Pingback:Whatisachatbot?Definition,Working,Types,and
Examples-DigitalTekBlog
​Pingback:WhatisconversationalAI?Definition,typesandexample
​Pingback:Whatismachinelearning?Definition,Applications,and
Types
​Pingback:MachineLearningandArtificialIntelligenceinRobotics