Artificial Intelligence : An Introduction for CS570 Artificial Intelligence

1
Artificial Intelligence An Introduction for
CS570 Artificial Intelligence

• Jin Hyung Kim
• KAIST Computer Science Dept.

2
Definition of AI

• Automation of activities that we associate with
  human thinking, activities such as
  decision-making, problem solving, learning,
  (Bellman, 1978)
• Study of how to make computers do things at
  which, at the moment, people are better (Rich
  Knight, 1991)
• A Branch of Computer Science that is concerned
  with the automation of intelligent behavior
  (Luger Stubblefield, 1993)
• The study of mental faculties through the use of
  computational models (Charniak McDormott, 1995)

3
Computer Science Body of Knowledge
31
43
16
10
38
10
Total 280 Core Hours
3
8
21
31
15
18
36
Source IEEE/ACM Computing Curricula 2001
Computer Science
4
Computing Disciplines, before and after 1990s
5
AI Engineering Definition

• Study of how to make machine do things which
  require intelligence when human do
• things requiring intelligence ?
• Making computer MORE smart
• Making thinking computer
• Can machine think ?
• Focus on how good it performs

6
AI Cognitive Scientific Definition

• Studying intelligence by computational means
• Programmed Human intelligence
• Artificial Mind
• Focus on how similar it works as human

7
Intelligent System
Perception, Recognition, Understanding Making
decisions, Acting
Flexibility Automation Optimization
Aims
via
8
Examples of AI systems

• Language Translation systems
• Natural Language Question answering systems
• Diagnosis Expert systems
• Avionic Expert systems vs. fly-by-wire
• Space shuttle mission planning
• Robots in factory, Auto-navigation robots
• Intelligent Traffic control system
• OCR, Handwriting Recognition System
• Speech Recognition System

9
Categorization of AI definitions
10
Go Playing Programs

• Selecting next move
• By analysis of all alternative moves
• By Analysis of Board Pattern (rule-based)
• Which one is better ?
• Which one can be better ?
• Engineering (mathematical)
• How well does it perform ?
• Performance is the key concern. Dont care of
  what method used
• Cognitive Scientific
• How similarly does it do as human ?
• Simulation of Behavior

11
Acting Humanly Turing Test

• Turing (1950) Computing Machinery and
  Intelligence
• Can machine think ? ? Can machine behave
  intelligently ?
• Operational test of intelligent behavior
  imitation game
• Predicted that by 2000, a machine might have 30
  chance of fooling a lay person in 5 minutes

12
Imitation Game
13
Issues on Turing Test

• Intelligent as much as Human
• Is dog intelligent ?
• Searles Chinese Room argument
• Strong AI and Weak AI
• ELIZA - a friend you could never have before
• http//www-ai.ijs.si/eliza-cgi-bin/eliza_script
• Imitation of Client-centered Rogerian Therapy
• Suggested major component of AI knowledge,
  reasoning, language, understanding, learning
• Any man-made system passed Turing Test ?

14
Thinking Rationally Laws of Thought

• Normative (or prescriptive ) rather than
  descriptive
• Several school of Greek schools developed various
  forms of logic, notation and rules of derivation
  of thoughts
• Mathematics and Philosophies of modern AI
• Problems
• Not all intelligent behavior is mediated by
  logical deliberation
• What is purpose of thinking ? What thought
  should I have ?
• Rational Behavior doing the right thing
• right expected to maximize goal achievement
  given available information

15
Hype Cycle (Boom-Bust-Build)
Science Fiction
Hangover
Productivity
Curiosity
16
The Hype Cycle of Emerging Technologies
? ?? Gartner, 2002
? ?? Gartner, 2002
17
Approaches to Intelligent system development

• u
• Knowledge-based Approach
• u
• Data Driven Approach
• u

18
Knowledge-base Systems

• u
• Represent Human knowledge as symbol combination
• u
• Knowledge Acquisition and Representation
• u
• Logic, Expert System, Fuzzy Logic
• u

19
Data Driven Approach

• u
• Extract common characteristics from collected
  examples
• u
• Training
• u
• Statistical Methods, Artificial Neural Network

20
Generality vs Power

• Aims Powerful and general solutions
• General Problem Solver
• Early attempt failed
• Complexity Toy Problems Only
• Specialized Approach to get Power
• Knowledge Based Approach
• Practical Expert Systems

21
State of the Art

• Which of the following can be done at present ?
• Play a decent game of table tennis
• Drive along a curvy mountain road
• Drive in the center of Seoul city
• Play decent game of Go
• Discover and prove a new mathematical theorem
• Write an intentionally funny story
• Give competent legal advice in a specialized area
  of law
• Translate spoken Korean into spoken Japanese in
  real time

22
Axes of AI Research
Theory
Methodology
System
Application
23
Major research areas (Methodology)

• Symbolic Programming
• Knowledge Representation
• Search Planning
• Automated Reasoning
• Machine Learning, knowledge Discovery
• Artificial Neural Net
• Genetic Algorithm
• ...

24
Major research areas (Applications)

• Natural Language Understanding
• Image, Speech and pattern recognition
• Uncertainty Modeling
• Expert systems
• Virtual Reality
• ..

25
Symbolic Programming

• Program as Representation of world
• Symbol as basic element of representation
• atom, property, relationship
• Symbolic Expression as method of combination
• LISP for Symbolic programming
• PROLOG for logic programming
• Object-Oriented Concept

26
Knowledge Representation

• What kind of Knowledge needed for Problem solving
  ?
• Structure of knowledge ?
• declarative vs procedural
• Representation techniques ?
• explicit vs (implicit inference)
• logic, frame, object-oriented, semantic net,
  script
• Knowledge acquisition and update

27
Search Theory

• An Optimization method
• Analyze alternative cases and select one
• Cope with Exponential complexity, NP classes
• Try likely one first (Heuristic Search)
• Utilize local information (Hill Climbing Method)
• Optimal solution vs good solution
• Genetic Algorithm, Simulated Annealing
• Stochastic search

28
Automated Reasoning

• Qualitative Reasoning
• Utilization of qualitative knowledge such as
• Non-monotonic Reasoning
• Ostrich flys ?
• Plausible Reasoning
• Information fusion under uncertainty
• Case-based Reasoning
• Utilization of Experience

29
Machine Learning

• Performance improvement by experience
• How much of knowledge required to start learning
  ?
• Method of acquiring new knowledge and merging it
  to existing knowledge-base
• Role of teacher
• Role of examples and experience
• Parameter Adjustment
• Inductive learning
• Computational Learning Theory
• Quality of generalization capability in terms of
  Training data
• Used in Practice such as Data Mining

30
Data Mining
Knowlegre extraction for decision making
Data
Decision Making
Information / knowledge

• ????
• Point of Sale
• ATM
• ????
• ????
• ??
• ????
• ????
• ??????

• A?? ???? 80? B??? ????
• ????? ??? ???? 6??? ??
• A??? ?? ??? B??? 2?
• ?? ??? ??? ??

• ????? ?
• ??? ??
• ??? ?? ??? ?
• ????? ????? ?
• ??? ?? ???? ?
• ??? ?

31
Neural Network

• Computational model of Neurons
• Power comes from Connection of simple processing
  element - connectionism

X1
w1
w2
X2
F(X1, X2, , Xn)
S
. . .
wn
Xn
32
Neural Network

• learning link weigh adjustment
• Error-back-propagation supervised learning
• Any Functional Mapping is learnable
• Strong at Sensory Data Processing
• Symbolic Grounding
• Old Horse on the race again
• Massive parallelism, graceful degradation

33
Neural Network Classifier
Input layer
Hidden layer
Output layer
34
Genetic Algorithm

• Computational model of life evolution
• Stochastic optimization technique
• Initial chromosome creation
• New generations are made (cross over, mutation)
• survival of the fittest
• Base of artificial life research
• study evolution of life, by simulation

35
History of AI

• 50 years of rise and fall of New technologies
  after invention of computer

• Logic
• Optimization
• Proabilistic Modeling
• Search theory
• Rule-based system
• Expert systems

• Fuzzy Theory
• Neural Netwrok
• Genetic Algorithm
• Chaos theory
• Artificial life
• .....

36
AI Prehistory

• Philosophy
• Logic, methods of reasoning, mind as physical
  system, foundations of learning, language,
  rationality
• Mathematics
• Formal representation of proof, algorithms,
  computation, decidability, tractability,
  probability
• Psychology
• Adaption, phenomena of perception and motor
  control, experimental techniques
• Linguistics
• Knowledge representation, grammar
• Neurosicence Physical substrate for mental
  activity
• Control Theory homeostatic systems, stability,
  optimal designs

37
Potted History of AI (I)

• 1943 McCulloch Pitts Boolean Circuit model
  of Brain
• 1950 Turings Computing Machinery and
  Intelligence
• 1950s Early AI programs Samuels checker
  program, Newell Simons Logic Theorist
• 1956 Dartmouth meeting Artificial
  Intelligence adopted
• 1965 Robinsons algorithm for logical reasoning

38
Potted History of AI (II)

• 1966-74 AI discovers computational complexity
• 1969-79 Early development of knowledge-based
  systems
• 1980-88 Expert systems industry booms, AI
  Programming Machine
• 1983 1993 Japan initiated 5th generation
  computer project
• 1988-93 Expert systems industry burst AI
  Winter
• 1985-95 Neural Network back to the race
• 1988 Resurgence of probabilistic and
  decision-theoretic methods,
• Rapid increase of technical depth of mainstrean
  AI
• Nouville AI Alife, Genetic Algorithm, Soft
  computing

39
AI Success Story

• Evans ANOLOGY
• Symbolic Algebra
• Macsyma (http//www.macsyma.com/)
• Chess Program DEEP BLUE defeat Gary Kasparov
  (1996)
• Automatic Theorem Proving contest (1999)

40
AI Success Story (Planning)

• MARVEL (Schwuttke, 1992)
• Real-time Space shuttle Mission planning
• Berth assignment (KAL, 1997)
• Unmanned Vehicle
• Ground and air
• Pathfinder Rover, 1996
• Asimo a walking robot

41
Autonomous Land Vehicle(DARPAs GrandChallenge
contest)
42
AI Success Story (Language Processing)

• PEGASUS (Zue, 1994)
• Spoken Natural language for airline reservation
• Limited context, free representation
• Japanese-Korea Hotel reservation(KT, 1995)
• Chatter Bot
• ????? ?? (typing)?? ???? ???? ?
• Many machine translation
• ?? ??? ??, ?? - ???

43
AI Success Story Medical expert systems
Programs listed by Special Field

• Gynecology
• Imaging Analysis
• Internal Medicine
• Intensive Care
• Laboratory Systems
• Orthopedics
• Pediatrics
• Pulmonology Ventilation
• Surgery Post-Operative Care
• Trauma Management

• Antibiotics InfectiousDiseases
• Cancer
• Chest pain
• Dentistry
• Dermatology
• Drugs Toxicology
• Emergency
• Epilepsy
• Family Practice
• Genetics
• Geriatrics

44
Pattern Recognition Applications

• Handwriting and document recognition
• forms, postal mail, historic documents
• PDA pen recognition
• Signature, biometrics (finger, face, iris, etc.)
• Gesture, facial expression
• As a Human computer intertraction
• EEG, EKG, X-ray
• Trafic monitoring, Remote Sensing
• Smart Weapon guided missile, target homing

45
Automatic Target Recognizer
46
Postal Address Recognition
47
Handwriting Understanding
48
??? PC e-Book, Tablet PC, PDA, M-phone
49
Ubiquitous ????
50
BioInformatics / Protein Structure Analysis
51
Contribution of AI

• Practical AI OCR, ICR, Symbolic Algebra, Machine
  Translation, Many Expert systems, Planning
  systems
• New concepts and Ideas to other fields of
  computer sciences
• Programming Language OO, functional language,
  logic-based
• DataBase
• Operating System

52
Future of AI

• Making AI Easy to use
• Easy-to-use Expert system building tools
• Web auto translation system
• Recognition-based Interface Packages
• Integrated Paradigm
• Symbolic Processing Neural Processing
• AI in everywhere, AI in nowhere
• AI embedded in all products
• Ubiquitous Computing, Pervasive Computing