Docstoc

Belief-optimal Reasoning for Cyber-physical Systems_1_

Document Sample
Belief-optimal Reasoning for Cyber-physical Systems_1_ Powered By Docstoc
					    CS B551: Elements of
    Artificial Intelligence
1   Instructor: Kris Hauser
    http://cs.indiana.edu/~hauserk
Basics
Class web site
   http://cs.indiana.edu/classes/b551
Textbook
   S. Russell and P. Norvig
   Artificial Intelligence: a Modern Approach
   3rd edition
    2nd edition can be used, but is not preferable




                                                      2
Basics
Instructor
   Kris Hauser (hauserk@indiana.edu)
AIs
   Kai Song (kaisong@indiana.edu)




                                        3
Office Hours
Kris Hauser
   Tu 10-11,W 12-1 in Info E 257 (connector building)
Kai Song
   TBA




                                                         4
Agenda
Intro to AI
Overview of class policies




                              5
What is AI?
AI is the reproduction of human reasoning and
 intelligent behavior by computational methods




                                                 6
What is AI?
AI is an attempt of reproduction of human
 reasoning and intelligent behavior by
 computational methods




                                             7
What is AI?
Discipline that systematizes and automates
 reasoning processes to create machines that:




 Think like humans          Think rationally
 Act like humans            Act rationally



                                                8
 Think like humans         Think rationally
 Act like humans           Act rationally

The goal of AI is: to build machines that
 operate in the same way that humans think
   How do humans think?
   Build machines according to theory, test how
    behavior matches mind’s behavior
   Cognitive Science
Manipulation of symbolic knowledge
How does hardware affect reasoning? Discrete
 machines, analog minds

                                                   9
 Think like humans                Think rationally
 Act like humans                  Act rationally
The goal of AI is: to build machines that perform tasks
 that seem to require intelligence when performed by
 humans
Take a task at which people are better, e.g.:
     Prove a theorem
     Play chess
     Plan a surgical operation
     Diagnose a disease
     Navigate in a building
and build a computer system that does it
 automatically
But do we want to duplicate human imperfections?
                                                      10
 Think like humans          Think rationally
 Act like humans            Act rationally

The goal of AI is: to build machines that make
 the “best” decisions given current knowledge
 and resources
“Best” depending on some utility function
   Influences from economics, control theory
How do self-consciousness, hopes, fears,
 compulsions, etc. impact intelligence?
Where do utilities come from?
                                                11
What is Intelligence?
“If there were machines which bore a resemblance to our
bodies and imitated our actions as closely as possible for
all practical purposes, we should still have two very
certain means of recognizing that they were not real men.
The first is that they could never use words, or put
together signs, as we do in order to declare our thoughts to
others… Secondly, even though some machines might do
some things as well as we do them, or perhaps even better,
they would inevitably fail in others, which would reveal
that they are acting not from understanding, …”
Discourse on the Method, by Descartes (1598-1650)

                                                        12
What is Intelligence?
Turing Test (c. 1950)




                         13
What is intelligence?
An Application of the Turing Test
CAPTCHA: Completely Automatic Public Turing
 tests to tell Computers and Humans Apart




                                               15
Chinese Room (John Searle)




                             16
Can Machines Act/Think
Intelligently?
§ Yes, if intelligence is narrowly defined as
  information processing
  AI has made impressive achievements showing
  that tasks initially assumed to require
  intelligence can be automated
  Each success of AI seems to push further the
  limits of what we consider “intelligence”




                                                 17
  Some Achievements
§ Computers have won over world
  champions in several games,
  including Checkers, Othello, and
  Chess, but still do not do well in
  Go
§ AI techniques are used in many
  systems: formal calculus, video
  games, route planning, logistics
  planning, pharmaceutical drug
  design, medical diagnosis,
  hardware and software trouble-
  shooting, speech recognition,
  traffic monitoring, facial
  recognition, medical image
  analysis, part inspection, etc...
§ DARPA Grand Challenge:
  robotic car autonomously
  traversed 132 miles of desert
§ IBM’s Watson competes with
  Jeopardy champs
§ Some industries (automobile,
  electronics) are highly robotized,
  while other robots perform brain
  and heart surgery, are rolling
  on Mars, fly autonomously, …,        18
  but home robots still remain
  a thing of the future
                                       18
Can Machines Act/Think
Intelligently?
§ Yes, if intelligence is narrowly defined as
  information processing
  AI has made impressive achievements showing
  that tasks initially assumed to require
  intelligence can be automated
§ Maybe yes, maybe not, if intelligence cannot be
  separated from consciousness
  § Is the machine experiencing thought?
  § Strong vs. Weak AI


                                                    19
Big Open Questions
§ Is intelligent behavior just information
  processing?
 (Physical symbol system hypothesis)
§ If so, can the human brain solve problems
  that are inherently intractable for
  computers? Will a general theory of
  intelligence emerge from neuroscience?
§ In a human being, where is the interface
  between “intelligence” and the rest of
  “human nature”
   Self-consciousness, emotions, compulsions
§ What is the role of the body?
 (Mind-body problem)                            20
§ AI contributes to building an information
  processing model of human beings, just as
  Biochemistry contributes to building a model
  of human beings based on bio-molecular
  interactions
§ Both try to explain how a human being
  operates
§ Both also explore ways to avoid human
  imperfections (in Biochemistry, by engineering new
 proteins and drug molecules; in AI, by designing
 rational reasoning methods)
§ Both try to produce new useful technologies
§ Neither explains (yet?) the true meaning of
  being human
                                                       21
 Main Areas of AI
§ Knowledge representation
  (including formal logic)       Agent                 Perception
§ Search, especially heuristic              Robotics
  search (puzzles, games)
§ Planning                     Reasoning
§ Reasoning under uncertainty,              Search
  including probabilistic                                Learning
  reasoning
                                           Knowledge Constraint
§ Learning                                    rep.
                                Planning            satisfaction
§ Robotics and perception
§ Natural language processing
                                Natural
                                               ...       Expert 22
                                language
                                                        Systems
Bits of History
§ 1956: The name “Artificial Intelligence” is coined
§ 60’s: Search and games, formal logic and theorem
  proving
§ 70’s: Robotics, perception, knowledge
  representation, expert systems
§ 80’s: More expert systems, AI becomes an
  industry
§ 90’s: Rational agents, probabilistic reasoning,
  machine learning
§ 00’s: Systems integrating many AI methods,
  machine learning, natural language processing,
  reasoning under uncertainty, robotics again          23
AI References
Conferences
   IJCAI, ECAI, AAAI, NIPS
Journals
   AI, Comp. I, IEEE Trans. Pattern Anal. Mach. Intel.,
    IEEE Int. Sys., JAIR
Societies
   AAAI, SIGART, AISB
AI Magazine (Editor: IU’s David Leake)




                                                           24
Careers in AI
‘Pure’ AI
   Academia, industry labs
Applied AI
   Almost any area of CS!
   NLP, vision, robotics
   Economics
Cognitive Science




                              25
Syllabus
Introduction to AI
   Philosophy, history, agent frameworks
Search
   Uninformed search, heuristic search, heuristics, game
    playing
Reasoning under uncertainty
   Probability, planning under uncertainty, Bayesian
    networks, probabilistic inference, temporal sequences
Machine learning
   Neural nets, decision tree learning, support vector
    machines, etc.
Applications
   Constraint satisfaction, motion planning, computer      26
    vision
  Computer Vision        Knowledge representation and
                         learning                       Biologically-inspired computing


      B657                    B552   S626      S675          B553       I486

                                 B555     B556
Algorithms for Optimization
and Learning                                                      Game theory

      B553                           B551
                                                                      E626

       Robotics

                                     Topics in AI        Natural Language Processing
 B335       I400

                                        B659                        B651
 Q360       Q570                                                                    27
     Class Policies
28
Prerequisites
C211
I recommend:
   Two semesters programming
   Basic knowledge of data structures
   Basic knowledge of algorithmic complexity




                                                29
Programming Assignments
Projects will be written in Python
Easy to learn
2 weeks for each assignment




                                      30
Grading
75% Homework
   6 assignments, lowest score will be dropped
25% Final




                                                  31
Homework Policy
Due at end of class on due date
   Typically Thursdays
   No “slip days”
Extensions only granted in rare cases
   Require advance notice except emergencies




                                                32
Final Project
Encouraged if you are intending to do
 research or coursework in AI, pursue
 higher degree
   Individual or small groups (up to 3)
   Counts as two homework assignments
Content
   Software, new research, or technical report
   Mid-semester project proposal
   End-of-year report and in-class presentation
Enrollment
Add/drop deadline w/o penalty: Aug 27
Waitlist deadline: Aug 25




                                         34
Takeaways
AI has many interpretations
   Act vs. think, human-like vs. rational
   Concept has evolved
“Intelligence” has many interpretations
   Turing test
   Chinese room
AI success stories from each perspective




                                             35
Homework
Register
Textbook
Survey
http://cs.indiana.edu/classes/b551
Readings:
   R&N Ch. 1, 26 (introduction and historical
    perspectives)
   R&N 3.1-3



                                                 36

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:0
posted:7/2/2014
language:English
pages:36