AI Chapter I for Computer Science Students

This is Artificial Inteligence Coures chapter one for Computer science Student

1. ARTIFICIAL INTELLIGENCE(AI)
CHAPTER I
INTRODUCTION TO AI

2. Chapter Outline
What is AI
Approaches to AI -making computer:
oThink like a human (Thinking humanly)
oAct like a human (Acting humanly)
oThink rationally (Thinking rationally)
oAct rationally (Acting rationally)
The Foundations of AI
Bits of History and the State of the Art
Types of intelligence
Why do we need to study AI?
The Foundations of AI
AI applications
Sub Areas of AI

3. INTRODUCTION TO AI
Artificial intelligence (AI) is the intelligence of a machine or computer
that enables it to imitate or mimic human capabilities.
Artificial Intelligence is concerned with the design of intelligence in an
artificial device.
The term was coined by John McCarthy in 1956.
Intelligence is the ability to acquire, understand and apply the knowledge
to achieve goals in the world.

4. INTRODUCTION TO AI (1)
AIusesmultipletechnologiesthatequipmachinestosense,comprehend,
plan,act,andlearnwithhuman-likelevelsofintelligence.
Fundamentally,AIsystemsperceiveenvironments,recognizeobjects,
contributetodecisionmaking,solvecomplexproblems,learnfrompast
experiences,andimitatepatterns.Theseabilitiesarecombinedto
accomplishtaskslikedrivingacarorrecognizingfacestounlockdevice
screens.

5. INTRODUCTION TO AI (2)
Artificial Intelligence is the development of computer
systems that are able to perform tasks that would require
human intelligence.
Which is created by humans by loading suitable intelligent
algorithms and a large volume of data to solve real world
complex problems.

6. INTRODUCTION TO AI (3)
AIcannowtakeonmanyofthemostrepetitivejobs
performedbyhumanswithmoreaccuracyandmuch
faster.
Sothathumanswillbeabletoworklesshours,having
morefreetimetoenjoytheirlife,family,hobbies,and
friends.

7. INTRODUCTION TO AI (4)
Artificial Intelligence made-up of two words:
Artificial
Produced by human art or effort, rather than
originating naturally.
Intelligence
is the ability to acquire knowledge and use it"
[Pigfordand Baur]

8. INTRODUCTION TO AI (5)

9. INTRODUCTION TO AI (6)
AI is a way to make machines think and behave intelligently.
These machines are controlled by software inside them.
It concerns with the way how machines to sense, reason,
think, act.
It is the branch of computer science by which we can create
intelligent machines which can behave like human, thing like
human and able to make decisions.

10. INTRODUCTION TO AI (7)
With AIwe don't need to preprogram the machine to do
some work,
We have create machine with programmed algorithm which
can work with own intelligent which is awesomeness of AI.
Machine Learning An application of AIthat gives
machines the ability to learn and improve without the
help of humans or new programming.

11. How AI Works
anAIsystemacceptsdatainputintheformofspeech,text,image,etc.
Thesystemthenprocessesdatabyapplyingvariousrulesandalgorithms,
interpreting,predicting,andactingontheinputdata.Uponprocessing,
thesystemprovidesanoutcome,i.e.,successorfailure,ondatainput.
Theresultisthenassessedthroughanalysis,discovery,andfeedback.
Lastly,thesystemusesitsassessmentstoadjustinputdata,rulesand
algorithms,andtargetoutcomes.Thisloopcontinuesuntilthedesired
resultisachieved.

12. How AI Works (1)

13. Key Components of AI

14. Key Components of AI (1)
1.Machinelearning:MachinelearningisanAIapplicationthat
automaticallylearnsandimprovesfromprevioussetsof
experienceswithouttherequirementforexplicitprogramming.
2.Deeplearning:DeeplearningisasubsetofMLthatlearnsby
processingdatawiththehelpofartificialneuralnetworks.
3.Neuralnetwork:arecomputersystemsthatarelooselymodeled
onneuralconnectionsinthehumanbrainandenabledeep
learning.

15. Key Components of AI (2)
4.Cognitivecomputing:Cognitivecomputingaimstorecreatethe
humanthoughtprocessinacomputermodel.Itseekstoimitate
andimprovetheinteractionbetweenhumansandmachinesby
understandinghumanlanguageandthemeaningofimages.
5.Naturallanguageprocessing(NLP):NLPisatoolthatallows
computerstocomprehend,recognize,interpret,andproduce
humanlanguageandspeech.
6.Computervision:Computervisionemploysdeeplearningand
patternidentificationtointerpretimagecontent(graphs,tables,
PDFpictures,andvideos).

16. What is the intelligence?
Ability of problem solving
The ability to think, plan and schedule
The ability to Information manipulation
Ability to tackle ambiguous problems
Ability to learn and recognize
Ability to understand and perceive

17. What is the intelligence? (1)
Intelligence is made-up of following components:

18. Approaches to AI
Theexcitingnewefforttomakecomputersthinks…machinewith
minds,inthefullandliteralsense”(Haugeland1985)
Theautomationofactivitiesthatweassociatewithhumanthinking,
activitiessuch asdecision-making,problem solving,
learning,…(Bellman,1978)

19. Approaches to AI (1)
“Theartofcreatingmachinesthatperformfunctionsthatrequire
intelligencewhenperformedbypeople”(Kurzweil,1990)
“Thestudyofhowtomakecomputersdothingsatwhich,atthe
moment,peopledobetter”,(RichandKnight,1991)

20. Systems that act like humans
You enter a room which has a computer terminal. You have a fixed period of
time to type what you want into the terminal, and study the replies. At the
other end of the line is either a human being or a computer system.
If it is a computer system, and at the end of the period you cannot reliably
determine whether it is a system or a human, then the system is deemed to
be intelligent.
?

21. Systems that act like humans
The Turing Test approach
a human questioner cannot tell if
there is a computer or a human answering his question, via
teletype (remote communication)
The computer must behave intelligently
Intelligent behavior
to achieve human-level performance in all cognitive tasks

22. Systems that act like humans
These cognitive tasks include:
Natural language processing
for communication with human
Knowledge representation
to store information effectively & efficiently
Automated reasoning
to retrieve & answer questions using the stored information
Machine learning
to adapt to new circumstances

23. The total Turing Test
Includes two more issues:
Computer vision
to perceive objects (seeing)
Robotics
to move objects (acting)

24. Approaches to AI (2)
“Thestudyofmentalfacultiesthroughtheuseof
computationalmodels”,(Charniaketal.1985)
“Thestudyofthecomputationsthatmakeitpossibleto
perceive,reasonandact”,(Winston,1992)

25. Approaches to AI (3)
“Computational Intelligence is the study of the design of intelligent
agents”(Poole et al, 1998)
“AI….is concerned with intelligent behavior in artifact”, (Nilsson, 1998)

26. How to Achieve AI?
AI
Acting
humanly
Thinking
rationally
Acting
rationally
Thinking
humanly
26

27. Types of intelligence
Logical-mathematical intelligence.
Linguistic intelligence.
Spatial intelligence.
Musical intelligence.
Intrapersonal intelligence.
Interpersonal intelligence.
Naturalistic intelligence.

28. Why do we need to study AI?
One of the main reasons we want to study AI is to automate several
real world complex problems.
To
Handle large amounts of data in an efficient way.
Collect data simultaneously from multiple sources without any
lag.
Index and organize data in a way that allows us to derive
insights.
Learn from new data and update constantly using the right
learning algorithms.
Think and respond to situations based on the conditions in real
time.
Solving a problem in an easy manner.

29. Why do we need to study AI? (1)
With help of AI, we can create machines which can solve real-
world complex problems very easily and accurately such as health
issues, marketing, traffic issues.
With help of AI, we can create our own personal virtual assistance
such as google assistance, siri,etc
With help of AI , we can build intelligent robots which can work in
environment where survival of human can be at risk.

30. Goals of AI
To create Expert Systems:
The systems which exhibit intelligent behavior, learn, demonstrate, explain,
and advice its users.
To replicate/implementhuman intelligence in machines:
Creating systems that understand, think, learn, and behave like humans.
Building a machine which can perform knowledge intensive tasks that requires
human intelligent such as:
Proving a theorem, playing chess, plan some surgical operations, and driving a
car in traffic.

31. The Foundations of AI
AcademicdisciplinesimportanttoAI:
Philosophy:Logic,methodsofreasoning,mindasphysical
system,foundationsoflearning,language,rationality.
Canamachinethink?
Mathematics:Formalrepresentationandproof,computation,
probability.
Neuroscience:neuronsasinformationprocessingunits.
Brainarchitecture

32. The Foundations of AI (1)
Academic disciplines important to AI:
Psychology/ Cognitive Science: how do people behave,
perceive, process information, represent knowledge.
Computer engineering: to building fastest computers.
Control theory: design systems that maximize an objective
function over time.
Linguistics: knowledge representation, grammar.

33. Applications of AI?
Artificial Intelligence has various applications in today's
society
Following are some sectors which have the application of
Artificial Intelligence:

34. AI applications: (1)
Identification Technologies
ID cards:
E.G., ATM cards
Can be a security risk:
Cards can be lost, stolen, passwords forgotten, etc.
Biometric identification:
Walk up to a locked door
Fingerprint device
Iris scan
Signature
Voice pattern

35. AI application: (2)
For Security application
AI uses for Security applications
For identifying face of crime makers
For identifying some one from other
Object identification

36. AI application: (3)
Predicting the Stock Market
Very difficult problem!
Given the past, predict the future
We can use learning algorithms to learn a predictive model from
historical data
Banks and financial traders used such models
Association rule mining-to identify relations of different items

37. AI application: (4)
Machine Translation
Language problems is international business
How hard is automated translation
Very difficult!
Ex: English to Russian, to Amharic, to Afanoromoo, to Tigrigna, to
Sidamalanguage, and other Ethio-local languages.
Not only the words be translated, but also their meaning also!
However AI
Use algorithms which combine dictionaries, grammar models, etc.

38. AI application: (5)
AIinEverydayLife
Postoffice
Automaticaddressrecognitionandsortingofmail
Banks
Automaticcheckreaders,signatureverificationsystems
Automatedloanapplicationclassification
Telecomcompanies
Automaticvoicerecognitionfordirectoryinquiries
Creditcardcompanies
Automatedfrauddetection

39. AI application: (6)
AI in Everyday Life
Computer companies
Automated diagnosis for help-desk applications
Movie search
Movie recommendation
Google:
Search technology

40. Sub Areas of AI:
Game Playing
Deep Blue Chess program beat world champion Gary Kasparov
Speech Recognition
PEGASUS spoken language interface to American Airlines' EAASY
SABRE reservation system,
Computer Vision
Face recognition programs in use by banks, government, etc.

41. Sub Areas of AI: (1)
Expert Systems
Diagnostic Systems:
MYCIN system for diagnosing bacterial infections
System Configuration:
DEC's XCON system for custom hardware configuration.
Radiotherapy treatment planning.
Financial Decision Making:
Credit card companies, mortgage companies, banks.
Classification Systems:
NASA developed a system for classifying astronomical images

42. Sub Areas of AI: (2)
Mathematical Theorem Proving
Use inference methods to prove new theorems.
Natural Language Understanding
AltaVista's translation of web pages.
Translation of CatepillarTruck manuals into 20 languages.
Scheduling and Planning
Automatic scheduling for manufacturing.
Artificial Neural Networks:
Machine Learning:

43. Sub Areas of AI: (3)

44. Current research areas of AI
Current scholars of AI working on the following:
Robotics
Medical diagnosis
Security application
Language processing
Vehicle riding
Gamming
Smart city / Smart environment
Expert systems

45. Reading assignment
History of AI