introduction to artificial intelligence and few examples

RaviKumarBN,Asst.Prof,CSE,BMSIT 1

What is Human intelligence? •  It’sacomposi?onofabili?eslike

RaviKumarBN,Asst.Prof,CSE,BMSIT3

Learning

Understanding of Language

Perceiving Reasoning

Feeling

What is intelligence?

•  Theabilitytolearnorunderstandfromexperience•  Theabilitytoacquireandretainknowledge•  Theabilitytorespondquicklyandsuccessfullytoanewsitua?on

•  TheabilitytousereasontosolveproblemsIf intelligence is learning, understanding, retaining,responding,andusingreasonthenwhatisAI?


Quick Answer from Academia

q Modelinghumancogni8onusingcomputers.q StudyofmakingcomputersdothingswhichatthemomentpeoplearebeNerat..q Makingcomputersdothingswhichrequireintelligence.



More Formal Definition of AI q AIisabranchofcomputersciencewhichisconcernedwiththestudyandcrea?onofcomputersystemsthatexhibit

Someformofintelligence.OrThosecharacteris?cswhichweassociatewithintelligenceinhumanbehavior.q Itisthescienceandengineeringofmakingintelligentmachines,especiallyintelligentcomputerprograms.

What’s Involved in Intelligence? q Abilitytointeractwiththerealworld

Ø Toperceive,understand,andact•  e.g.,speechrecogni?onandunderstanding

q Searchingthebestsolu?on-medicalq ReasoningandPlanning

Ø Modelingtheexternalworld–deliveryrobotØ Solvingnewproblems,planning,andmakingdecisionsØ Abilitytodealwithunexpectedproblems,uncertain?es

q LearningandAdapta?onØ Wearecon?nuouslylearningandadap?ngØ ourinternalmodelsarealwaysbeing“updated”

•  e.g.,ababylearningtocategorizeandrecognizeanimals


John McCarthy

•  (September4,1927–October24,2011)wasanAmericanComputerScien?stAndCogni?veScien?st.

• McCarthywasoneofthefoundersofthedisciplineofAr?ficialIntelligence.

• Hecoinedtheterm"Ar?ficialIntelligence"(AI)


Twodimensions

•  Thinking/Reasoningvs.Behavior/Ac?on.

•  Successaccordingtohumanstandardsvs.successaccordingtoanidealconceptofintelligence(ra?onality):

FourCategories.

ü  Systemsthatthinklikehumans(focusonreasoningandhumanframework).

ü  Systemsthatthinkra?onally(focusonreasoningandageneralconceptofintelligence).

ü  Systemsthatactlikehumans(focusonbehaviorandhumanframework).

ü  Systemsthatactra?onally(focusonbehaviorandageneralconceptofintelligence).

Views of AI fall into four categories in Two dimensions:


Definition of AI


Systems that Think like Humans “The exciting new effort to make computers think…. Machine with minds,….” (Haugeland,1985) “[The automation of] activities that we associated human thinking, activities such as dec i s ion –making , p rob lem so lv ing , learning…”(Bellman,1978)

Systems that Think Rationally “The study of mental faculties through the use of computational models” (Charnaik and McDermott,1985) “The study of the computations that make it possible to perceive, reason and act” (Wintson, 1992)

Systems that Act like Humans “The art of creating machines that perform functions that require intelligence when performed by people” (Kurzwell, 1990) “The study of how to make computers do things at which, at the moment, people are better” (Rich and Knight,1991)

Systems that Act Rationally “A field of study that seeks to explain and emulate intelligent behavior in terms of computational processes” (Schalkoff,1990) “The branch of computer science that is concerned with the automation of intelligent behavior” (Luger and Stubble field)

Acting Humanly : Turing Test


AlanTuringBorn:23JUN1912,LondonDied:17JUN1954computerscien?st,mathema?cian,logician,cryptanalystandtheore?calbiologist.

¨ “CanMachinethink?”->“CanMachinesbehaveintelligently”¨ Opera?onaltestforintelligentbehavior:theImita?onGame.¨ Thecomputerwouldneedtopossessthefollowingcapabili?es:

•  NaturalLanguageProcessing•  KnowledgeRepresenta<on•  AutomatedReasoning•  MachineLearning

Place both a human and a machine mimicking human responses outside the field of direct observation and use an unbiased interface to interrogate them. If the responses are distinguishable, the machine is not displaying intelligence.


Thinking Humanly q Cogni?veScienceapproach-  Trytoget“inside”ourminds-Introspec?on-tryingtocatchourownthoughtsastheygobyandthroughpsychologicalexperiments.

-  E.g..Conductexperimentswithpeopletotryto“reverse-engineer”howwereason,learning,remember,predict.

q Problems-Humansdon’tbehavera?onally.-  Thereverseengineeringisveryhardtodo.-  Thebrain’shardwareisverydifferenttoacomputerprogram.


Thinking Rationally: The “laws of thought” approach

q TheGreekphilosopherAristotlewasoneofthefirsttoaNempttocodify``rightthinking,''thatis,irrefutablereasoningprocesses.q HegaveSyllogismsthatalwaysyieldedcorrectconclusionwhencorrectpremisesaregiven.q Theselawsofthoughtweresupposedtogoverntheopera?onofthemind,andini?atedthefieldoflogic.q Thelogicisttradi?oninAIhopestocreateintelligentsystemsusinglogicprogramming.q Howevertherearetwoobstaclestothisapproach.

Ø First,Itisnoteasytotakeinformalknowledgeandstateintheformaltermsrequiredbylogicalnota?on,par?cularlywhenknowledgeisnot100%certain.Ø Second,solvingproblemprincipallyisdifferentfromdoingitinprac?ce.


Acting Rationally: The rational agent approach

q Whatmeans“behavera<onally”foraperson/system:Ø Taketheright/bestac?ontoachievethegoals,basedonhis/itsknowledgeandbelief

q Example.AssumeIdon’tliketogetwet(mygoal),soIbringanumbrella(myac?on).DoIbehavera?onally?Ø TheanswerisdependentonmyknowledgeandbeliefØ IfI’veheardtheforecastforrainandIbelieveit,thenbringingtheumbrellaisra?onal.

Ø IfI’venotheardtheforecastforrainandIdonotbelievethatitisgoingtorain,thenbringingtheumbrellaisnotra?onal.


The rational agent approach

q Anagentisen?tythatperceivesitsenvironmentandisabletoexecuteac?onstochangeit.

q Agentshaveinherentgoalsthattheywanttoachieve.

q ARa?onalagentactsinawaytomaximizetheachievementofitsgoals

q Truemaximiza?onsofgoalsrequiresomniscienceandunlimitedcomputa?onalabili?es

q Limitedra?onalityinvolvesmaximizinggoalswithinthecomputa?onalandotherresourcesavailable.


History 1943 McCulloch&PiNs:Booleancircuitmodelofbrain1950 Turing’s“Compu?ngMachineryandIntelligence”1950s Early AI programs, including Samuel’s checkers program, Newell & Simon’sLogicTheorist,Gelernter’sGeometryEngine.1956Dartmouthmee?ng:“Ar?ficialIntelligence”adopted1965Robinson’scompletealgorithmforlogicalreasoning1966-73AIdiscoverscomputa?onalcomplexityNeuralnetworkresearchalmostdisappears1969-79Earlydevelopmentofknowledge-basedsystems1980 AIbecomesanindustry1986Neuralnetworksreturntopopularity1987AIbecomesascience1995Theemergenceofintelligentagents


Applications of AI:

q NaturalLanguageUnderstandingq ExpertSystemsq PlanningandRobo?csq MachineLearningq GamePlaying


Natural Language Processing

q Todesignandbuildsoqwarethatwillanalyzeunderstandandgeneratelanguagesthathumanusenaturally.


Modes of communication

q Textbased.

q Dialoguebased.


Speech Recognition

q Processofconver?ngsoundsignalcapturedbymicrophoneormobile/telephonetoasetofwords.

q 70-100words/minwithaccuracyof90%


Computer Vision

q Abilityofamachinetoextractinforma?onfromanimagethatisnecessarytosolveatask

�  ImageAcquisi?on�  ImageProcessing�  ImageAnalysis�  Imageunderstanding


Intelligent Robot

q Tendtomimichumansensinganddecisionmaking abili?es so that they can adoptthemselves to certain condi?ons andmodifytheirac?ons.


Expert Systems

q TheseareSoqwaresusedfordecisionmaking.q AutomatedReasoningandTheoremProving.q Troubleshoo?ngExpertSystems.q StockMarketExpertSystem.


Fields of AI

q Computerscience:�  GraphicalUserInterface�  Automa?cStoragemanagement

�  ObjectOrientedProgramming�  Datamiming�  computergaming

q Telecommunica?on:• AutomatedOnlineAssistants• Voicedialing•  SpeechRecogni?on


Fields of AI Avia?on&Automa?on:• NASA'sfightresearchcenter.• Voicerecogni?oninfighterjets.• Direc?onstoA.Ipilotsthroughairtrafficcontrollers.

• Automa?cGearingSysteminCars.


Fields of AI

Robo?cs:• AssemblingRobots• WeldingRobots• Behaviorbasedrobo?cs• DancingRobots• Robotnaviga?on


Daily life applications

• HomeSecurity• Bank• Postoffice• Websites• Digitalcameras•  Newsandpublishing•  Financialtrades• Healthandmedicine• Gamesandtoys


How AI is different????????


Ar?ficialIntelligence

NonCrea?ve

Precise

Consistency

Mul?tasking

NaturalIntelligence

Crea?ve

MayContainError

NonConsistent

Can’tHandle

Drawbacks of A.I

•  LimitedAbility•  SlowRealTimeResponse• Can’tHandleEmergencySitua?on• Difficultcode• HighCost



ManyHollywoodmoviescomewiththefuturetechnology,thattechnologywecanseeinourfuture.Havealook

someofthem….

In this movie

people to

purchase

remote

controlled

humanoid

robots

through which

they interact

with society.

These fit, attractive, remotely controlled robots

ultimately assume their life roles, enabling

people to experience life vicariously from the

comfort and safety of their own homes.

Alex (Robocop) at first

rejects his current

condition upon seeing

that his original body

now consists of only

lungs, throat, head and

right hand when the

armor and

Cybernetic components

are fully removed, but

he is convinced by

Norton to be strong for

his wife and son.

He had loose his body parts in a car bomb blast. Alex has him outfitted with the cybernetic body and software.

introduction to artificial intelligence and few examples

Engineering