Artificial Intelligence - PowerPoint 9

Document Sample
Artificial Intelligence - PowerPoint 9 Powered By Docstoc
					                          1




Artificial Intelligence
                                                                 2




     The goal of artificial intelligence
    • psychology: model human intelligence
    • engineering: automate tasks that currently
      require human intelligence
How can we determine if an AI program has satisfied its goal?

                           machine     Turing Test: can the
                                      interrogator distinguish
      interrogator                    between the responses
                           man           of the man and the
                                              machine?
                                                                           3


Faking intelligence: Eliza (aka Doctor)
Person: Men are all alike.
Eliza: In what way?
Person: They’re always bugging us about something or other.
Eliza: Can you think of a specific example?
Person: My boyfriend made me come here.
Eliza: You say your boyfriend made you come here?
Person: He says I’m depressed much of the time.
Eliza: I am sorry to hear that you are depressed.
Person: It’s true. I am unhappy.
Eliza: Do you think coming here will help you not to be unhappy?
Person: I need some help, that much seems certain.
Eliza: What would it mean to you if you got some help?
Person: Perhaps I could learn to get along with my mother.
Eliza: Tell me more about your family.
…
                                          How does the program work? ...
                                                   4

  Pattern matching of strings:
some examples of Eliza-like rules
• (X me Y)  (X you Y ?)
  e.g. “Tell me a secret”  “Tell you a secret?”

• (I remember X) 
    (Why do you recall X just now?)
• (My {family relation} is Y) 
    (Who else in your family is Y ?)
   …
• (X)  (“I understand. Please go on.”)
                                                                      5



  Search: solve problems by trying
        (almost) everything
   A problem is specified with a start state, goal
       state, and set of operators. Examples:
problem            start              goal            operators

chess            initial board      checkmate        move a piece
route planning initial location destination location follow a link
airline travel      Austin       city of destination take a flight
route phone    initiating phone   phone called       follow a cable
  call
                                                               6


Solving 8-tile puzzle: a simple example
                   2 8 3                               1 2 3
   start state:    1 6 4              goal state:      8 4
                   7 5                                 7 6 5


    operators: move the blank tile left, up, right, or down
                              2 8 3
            for example ...   1 6 4
                              7 5

                  left                 right
                                up

          2 8 3               2 8 3            2 8 3
          1 6 4               1 4              1 6 4
            7 5               7 6 5            7 5
                                                                                                                            7
                                                     2 8 3
                                                     1 6 4     start state
                                                     7 5



                             2 8 3                   2 8 3                       2 8 3
                             1 6 4                   1   4                       1 6 4
                               7 5                   7 6 5                       7 5



                                           2 8 3     2   3      2 8 3
                    2 8 3                                                                    2 8 3
                                             1 4     1 8 4      1 4
                      6 4                                                                    1 6
                                           7 6 5     7 6 5      7 6 5
                    1 7 5                                                                    7 5 4




  8 3       2 8 3             8 3       2 8 3        2 3     2 3        2 8        2 8 3             2 8 3      2 8
2 6 4       6 4             2 1 4       7 1 4      1 8 4     1 8 4      1 4 3      1 4 5             1 6        1 6 3
1 7 5       1 7 5           7 6 5         6 5      7 6 5     7 6 5      7 6 5      7 6               7 5 4      7 5 4




2 3     2 8 3     2 8 3       8 3         2 8 3     1 2 3     2 3 4      2 8         2 8 3      2 8 3        2 3        2 8 3
6 8 4   6 4       6 7 4       2 1 4       7 1 4       8 4     1 8        1 4 3       1 4 5      1 6          1 8 6      1 5 6
1 7 5   1 7 5     1 5         7 6 5       6 5       7 6 5     7 6 5      7 6 5       7 6        7 5 4        7 5 4      7 4




          8 3        8 1 3      2 8 3      2 8 3    1 2 3                        The number of nodes is
                                                             goal
          2 1 4      2 4        7 4        7 1 4    8   4                        growing exponentially!
          7 6 5      7 6 5      6 1 5      6 5      7 6 5    state
                                                 8




  Heuristics can Improve Search
• chess: prefer those states with the most
  material advantage and control of the center
  of the board.
• airline travel: prefer those states with the
  smallest linear distance to the destination,
  and the greatest number of successor states
                                                                    9




Automated reasoning with knowledge
 • Many problems can not be solved by enumerating
   alternatives.
 • Examples: language translation, expert systems, e.g.
   for medical diagnosis or legal reasoning.
 • These problems require:
       • knowledge representation: a formal language for encoding
         information
       • automated reasoning: methods for deriving new facts from
         “givens”
                                              10



         State of the Art:
        Knowledge Systems
• Programs written in computational logic
• Able to answer hard questions with
  coherent explanations
• Project Halo – a 4-month experiment in
  Chemistry. See http://www.projecthalo.com
                                                                11


  Example Questions from Project Halo

• The spectator ions in the reaction of barium nitrate with
  sodium sulfate are what? (choices)
• Although nitric acid and phosphoric acid have very
  different properties as pure substances, their aqueous
  solutions possess many common properties. List some
  general properties of these solutions and explain their
  common behavior in terms of the species present.
• Explain why a solution of HClO4 and NaClO4 cannot act
  as a buffer solution.
• Sodium azide is used in air bags to rapidly produce gas to
  inflate the bag. The products of the decomposition reaction
  are what? (choices)
                                                 12



            State of the Art:
               Perception
• Speech: from AT&T “yes-no” system to
  “full language” dictation system

• Vision: recognition of canonical objects and
  pre-marked trails
                                                        13



            State of the Art:
           Machine Translation
• Try one of the many on-line translators, e.g.
  www.google.com/translate
• Try “The rain in Spain falls mainly on the plains”.
• Try:
   – “The spirit is willing but the flesh is weak.”
   – “The economy stumbled through April bolstered by
     bullish arm sales, but threatened by the SARS
     outbreak.”
                                                14



             State of the Art:
                 Robotics
•   Mars Rover: semi-autonomous explorer
•   RoboCup dogs: vision, planning, team play
•   Bi-pedal robots: 16+ degrees of freedom
•   Intelligent cars: off road
                                                       15




                    Predictions
• Dialogue systems for travel arrangements: 1 year
• Simple robots doing fixed tasks: 5 years
• Computer as World champion:
   –   Checkers, chess, backgammon: done that
   –   Poker, Bridge: 1-3 years
   –   Soccer robots: 50 years
   –   Go: never?
• Significant scientific breakthrough by a computer:
  5-10 years
• “Digital Aristotle”: 30 years

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:4
posted:9/17/2012
language:English
pages:15