PowerPoint Presentation - Artificial Intelligence by zhangyun

VIEWS: 246 PAGES: 19

									Artificial Intelligence

 Universitatea Politehnica Bucuresti
             2007-2008

     Adina Magda Florea
 http://turing.cs.pub.ro/ai_07
       L1 – Course content
                    Introduction to AI
                    Search techniques
Knowledge           Symbolic logic
representation
                    Theorem proving. Prolog
                    Rule based systems
                                 Partial exam - 1 h
                    Frame based systems
                    Uncertain reasoning
                    Automatic planning
Aplications         Machine learning
                    Natural laguage understanding
                                 Final exam - 3h
       Course materials
Books
 A. Florea, A. Boangiu. Elemente de Inteligenta Artificiala.

 A. Florea. Bazele logice ale Inteligentei Artificiale. Cap. 4, 6.

 A. Florea, S. Radu, A. Mogos. Tehnici de programare Prolog
  pentru inteligenta artificiala – in print

On-line articles
   "Computing Machinery and Intelligence", Turing, A.M. Mind, 59, 1950, 433-
    460.
   "AI's Greatest Trends and Controversies", IEEE Intelligent Systems
    January/February 2000.
   "A Framework for Representing Knowledge", Marvin Minsky, MIT-AI
    Laboratory Memo 306, June, 1974.
Requirements
   Laboratory: min 6
   Laboratory homework
   Partial exam
   Final exam
Grading
   Partial exam: 20%
   Final exam: 50%
   Laboratory: 30%
Lecture No. 1
   What is artificial intelligence
   Features of AI problems
   History
   Structure of an AI system
   AI domains
Key questions
   Is it possible to make computers intelligent?
   How can we determine if a computer
    program is intelligent?
   At what level do we model intelligent
    behavior?
   Which       representations   and       which
    techniques do we use in AI?
1. What is Artificial Intelligence
   Alan Turing - “Computing Machinery and
    Intelligence”, 1950

   Simulation of human intelligence
   Emulation of human intelligence

   Symbolic AI
   Computational Intelligence

   AI Definitions
2. Features of AI problems
   General
   Dynamics
   Difficult (complexity)
   Knowledge vs data
   Heuristic knowledge, common sense
   Uncertain knowledge
   Reasoning, inferences
   Autonomy
   Learning/adaptation
Inferences
   Inference
   Rule of inference
   Inference strategy
       Complete / incomplete
       Consistent (sound) / Inconsistent (unsound)
Examples of Inference Rules
A                     Deductive inferences
AB          Modus ponens
B

nice(john)      Nondeductive inferences
nice(tom)
x nice(x)        Inductive inference
3. History
   Conference - Dartmouth College 1956 – AI - John
    McCarthy, Marvin Minsky, Alen Newell, Herbert Simon.
   1956 - 1957 A. Newell, J. Shaw si H. Simon - "The Logic
    Theorist."
   1960 AI programs
   1965 J. A. Robinson – resolution
   1965 – DENDRAL - J. Lederberg si E. Feigenbaum. –
    expert system
   1959 - Lisp (LISt Processing) - John McCarthy
    (Dartmouth College)
   1972 - Prolog (PROgrammation et LOGique) - Alain
    Colmerauer (universitatea Marseille-Aix)
   1983 - Smalltalk - Goldberg, Robson
History
   ’70 – knowledge
   Knowledge based systems
   Knowledge engineering
   MYCIN - Buchanan, Shortliffe - Stanford University - '74-
    '75
   Expert systems
   Expert systems shells
   ’80-’90 – disappointment
   ’90-’00 – interest, lot of research
   DAI
   IA and MAS
   Intelligent components, inteligent behavior
   Semantic Web
4. Structure of an AI system

              Intelligent
                system




   Action                      Perception
            Interaction with
            the environment
 Structure of an AI system

Inference engine
                                        Knowledge base
 Control strategy



                     Interaction with
                    the environment


  User interface                          Knowledge
                                          acquisition
    5. AI domains
   Knowledge representation, ontologies
   Common sense reasoning
   Games
   Automatic theorem proving
   Planning techniques
   Knowledge acquisition
   Machine learning
   NL processing
   Perception
   Expertise in specific domains
   Intelligent agents
   Multi-agent systems
           Buttler agent

   Imagine your very own mobile butler, able to travel
    with you and organise every aspect of your life
    from the meetings you have to the restaurants you
    eat in.
   The program works through mobile phones and is
    able to determine users' preferences and use the
    web to plan business and social events
   And like a real-life butler the relationship between
    phone agent and user improves as they get to know
    each other better.
   The learning algorithms will allow the butler to
    arrange meetings without the need to consult
    constantly with the user to establish their
    requirements.

                                                           16
       Robocup agents

   The goal of the annual RoboCup competitions,
    which have been in existence since 1997, is to
    produce a team of soccer-playing robots that can
    beat the human world champion soccer team by the
    year 2050.
   http://www.robocup.org/




                                                       17
    True UAV Autonomy

   In a world first, truly autonomous, Intelligent
    Agent-controlled flight was achieved by a
    Codarra ‘Avatar’ unmanned aerial vehicle
    (UAV).
   The flight tests were conducted in restricted
    airspace at the Australian Army’s Graytown
    Range about 60 miles north of Melbourne.
   The Avatar was guided by an on-board JACK™
    intelligent software agent that directed the
    aircraft’s autopilot during the course of the
    mission.

                                                      18
   ALICE robot

   Trading Agents Competition

   The Loebner Prize in AI

   AI Portal

   MONET Network of excellence
                                  19

								
To top