CS B351: Intro to Artificial Intelligence and Computer Simulation<br />Instructor: Kris Hauser<br />http://cs.indiana.edu/...
Basics<br />Class web site<br />http://cs.indiana.edu/courses/b351<br />Textbook<br />S. Russell and P. Norvig<br />Artifi...
Basics<br />Instructor<br />Kris Hauser (hauserk@indiana.edu)<br />AIs<br />Mark Wilson (mw54@indiana.edu)<br />3<br />
Office Hours<br />Kris Hauser<br />M,Th 11-12 in Info W 300<br />TAs<br />W 1-3 in Lindley 406<br />4<br />
Agenda<br />Intro to AI<br />Overview of class policies<br />5<br />
Intro to AI<br />6<br />
What is AI?<br />AI is the reproduction of human reasoning and intelligent behavior by computational methods <br />7<br />
What is AI?<br />AI is an attempt of reproduction of human reasoning and intelligent behavior by computational methods <br...
What is AI?<br />Discipline that systematizes and automates reasoning processes to create machines that:<br />9<br />
The goal of AI is: to build machines that operate in the same way that humans think<br />How do humans think?<br />Build m...
The goal of AI is: to build machines that perform tasks that seem to require intelligence when performed by humans<br />Ta...
The goal of AI is: to build machines that make the “best” decisions given current knowledge and resources<br />“Best” depe...
What is Intelligence?<br />“If there were machines which bore a resemblance to our bodies and imitated our actions as clos...
What is Intelligence?<br />Turing Test (c. 1950)<br />14<br />
An Application of the Turing Test<br />CAPTCHA: Completely Automatic Public Turing tests to tell Computers and Humans Apar...
Chinese Room (John Searle)<br />16<br />
Can Machines Act/Think Intelligently?<br /><ul><li>Yes, if intelligence is narrowly defined as information processingAI ha...
Some Achievements<br /><ul><li>Computers have won over world champions in several games, including Checkers, Othello, and ...
AI techniques are used in many systems: formal calculus, video games, route planning, logistics planning, pharmaceutical d...
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 ro...
Can Machines Act/Think Intelligently?<br /><ul><li>Yes, if intelligence is narrowly defined as information processingAI ha...
Maybe yes, maybe not, if intelligence cannot be separated from consciousness
Is the machine experiencing thought?
Strong vs. Weak AI</li></ul>19<br />
20<br />
Big Open Questions<br /><ul><li>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 intellig...
In a human being, where is the interface between “intelligence” and the rest of “human nature”</li></ul>Self-consciousness...
22<br /><ul><li>AI contributes to building an information processing model of human beings, just as Biochemistry contribut...
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 A...
Both try to produce new useful technologies
Neither explains (yet?) the true meaning of being human</li></li></ul><li>Main Areas of AI<br /><ul><li>Knowledge represen...
Search, especially heuristic search (puzzles, games)
Upcoming SlideShare
Loading in …5
×

Class overview, intro to AI - Belief-optimal Reasoning for Cyber ...

272 views
189 views

Published on

0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total views
272
On SlideShare
0
From Embeds
0
Number of Embeds
1
Actions
Shares
0
Downloads
3
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide

Class overview, intro to AI - Belief-optimal Reasoning for Cyber ...

  1. 1. CS B351: Intro to Artificial Intelligence and Computer Simulation<br />Instructor: Kris Hauser<br />http://cs.indiana.edu/~hauserk<br />1<br />
  2. 2. Basics<br />Class web site<br />http://cs.indiana.edu/courses/b351<br />Textbook<br />S. Russell and P. Norvig<br />Artificial Intelligence: a Modern Approach<br />2nd edition or 3rd edition<br />2<br />
  3. 3. Basics<br />Instructor<br />Kris Hauser (hauserk@indiana.edu)<br />AIs<br />Mark Wilson (mw54@indiana.edu)<br />3<br />
  4. 4. Office Hours<br />Kris Hauser<br />M,Th 11-12 in Info W 300<br />TAs<br />W 1-3 in Lindley 406<br />4<br />
  5. 5. Agenda<br />Intro to AI<br />Overview of class policies<br />5<br />
  6. 6. Intro to AI<br />6<br />
  7. 7. What is AI?<br />AI is the reproduction of human reasoning and intelligent behavior by computational methods <br />7<br />
  8. 8. What is AI?<br />AI is an attempt of reproduction of human reasoning and intelligent behavior by computational methods <br />8<br />
  9. 9. What is AI?<br />Discipline that systematizes and automates reasoning processes to create machines that:<br />9<br />
  10. 10. The goal of AI is: to build machines that operate in the same way that humans think<br />How do humans think?<br />Build machines according to theory, test how behavior matches mind’s behavior<br />Cognitive Science<br />Manipulation of symbolic knowledge<br />How does hardware affect reasoning? Discrete machines, analog minds<br />10<br />
  11. 11. The goal of AI is: to build machines that perform tasks that seem to require intelligence when performed by humans<br />Take a task at which people are better, e.g.:<br />Prove a theorem<br />Play chess<br />Plan a surgical operation<br />Diagnose a disease<br />Navigate in a building<br />and build a computer system that does it automatically<br />But do we want to duplicate human imperfections?<br />11<br />
  12. 12. The goal of AI is: to build machines that make the “best” decisions given current knowledge and resources<br />“Best” depending on some utility function<br />Influences from economics, control theory<br />How do self-consciousness, hopes, fears, compulsions, etc. impact intelligence?<br />Where do utilities come from?<br />12<br />
  13. 13. What is Intelligence?<br />“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, …”<br />Discourse on the Method, by Descartes (1598-1650)<br />13<br />
  14. 14. What is Intelligence?<br />Turing Test (c. 1950)<br />14<br />
  15. 15. An Application of the Turing Test<br />CAPTCHA: Completely Automatic Public Turing tests to tell Computers and Humans Apart<br />15<br />
  16. 16. Chinese Room (John Searle)<br />16<br />
  17. 17. Can Machines Act/Think Intelligently?<br /><ul><li>Yes, if intelligence is narrowly defined as information processingAI has made impressive achievements showing that tasks initially assumed to require intelligence can be automated</li></ul>Each success of AI seems to push further the limits of what we consider “intelligence”<br />17<br />
  18. 18. Some Achievements<br /><ul><li>Computers have won over world champions in several games, including Checkers, Othello, and Chess, but still do not do well in Go
  19. 19. 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...
  20. 20. DARPA Grand Challenge: robotic car autonomously traversed 132 miles of desert
  21. 21. IBM’s Watson competes with Jeopardy champs
  22. 22. Some industries (automobile, electronics) are highly robotized, while other robots perform brain and heart surgery, are rolling on Mars, fly autonomously, …, but home robots still remain a thing of the future</li></ul>18<br />18<br />
  23. 23. Can Machines Act/Think Intelligently?<br /><ul><li>Yes, if intelligence is narrowly defined as information processingAI has made impressive achievements showing that tasks initially assumed to require intelligence can be automated
  24. 24. Maybe yes, maybe not, if intelligence cannot be separated from consciousness
  25. 25. Is the machine experiencing thought?
  26. 26. Strong vs. Weak AI</li></ul>19<br />
  27. 27. 20<br />
  28. 28. Big Open Questions<br /><ul><li>Is intelligent behavior just information processing?(Physical symbol system hypothesis)
  29. 29. If so, can the human brain solve problems that are inherently intractable for computers? Will a general theory of intelligence emerge from neuroscience?
  30. 30. In a human being, where is the interface between “intelligence” and the rest of “human nature”</li></ul>Self-consciousness, emotions, compulsions<br /><ul><li>What is the role of the body?(Mind-body problem)</li></ul>21<br />
  31. 31. 22<br /><ul><li>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
  32. 32. Both try to explain how a human being operates
  33. 33. Both also explore ways to avoid human imperfections (in Biochemistry, by engineering new proteins and drug molecules; in AI, by designing rational reasoning methods)
  34. 34. Both try to produce new useful technologies
  35. 35. Neither explains (yet?) the true meaning of being human</li></li></ul><li>Main Areas of AI<br /><ul><li>Knowledge representation (including formal logic)
  36. 36. Search, especially heuristic search (puzzles, games)
  37. 37. Planning
  38. 38. Reasoning under uncertainty, including probabilistic reasoning
  39. 39. Learning
  40. 40. Robotics and perception
  41. 41. Natural language processing</li></ul>23<br />Agent<br />Perception<br />Robotics<br />Reasoning<br />Search<br />Learning<br />Knowledgerep.<br />Constraintsatisfaction <br />Planning<br />Naturallanguage<br />...<br />Expert<br />Systems<br />
  42. 42. Bits of History<br /><ul><li>1956: The name “Artificial Intelligence” is coined
  43. 43. 60’s: Search and games, formal logic and theorem proving
  44. 44. 70’s: Robotics, perception, knowledge representation, expert systems
  45. 45. 80’s: More expert systems, AI becomes an industry
  46. 46. 90’s: Rational agents, probabilistic reasoning, machine learning
  47. 47. 00’s: Systems integrating many AI methods, machine learning, natural language processing, reasoning under uncertainty, robotics again</li></ul>24<br />
  48. 48. Syllabus<br />Introduction to AI<br />Philosophy, history, agent frameworks<br />Search<br />Uninformed search, heuristic search, heuristics<br />Search applications (and variants)<br />Constraint satisfaction, planning, game playing, motion planning<br />Reasoning under uncertainty<br />Probability, planning under uncertainty, Bayesian networks, probabilistic inference, dynamic modeling<br />Intro to machine learning<br />Neural nets, decision tree learning, support vector machines, etc.<br />25<br />
  49. 49. 26<br />Game theory<br />Biologically-inspired computing<br />Computer Vision<br />E626<br />B553<br />I486<br />B657<br />B351<br />Knowledge representation and learning<br />B551<br />B552<br />S626<br />S675<br />Robotics<br />B335<br />I400<br />Natural Language Processing <br />Topics in AI<br />B651<br />B659<br />Q360<br />Q570<br />
  50. 50. Careers in AI<br />‘Pure’ AI<br />Academia, industry labs<br />Applied AI<br />Almost any area of CS!<br />NLP, vision, robotics<br />Economics<br />Cognitive Science<br />27<br />
  51. 51. AI References<br />Conferences<br />IJCAI, ECAI, AAAI, NIPS<br />Journals<br />AI, Comp. I, IEEE Trans. Pattern Anal. Mach. Intel., IEEE Int. Sys., Journal of AIR<br />Societies<br />AAAI, SIGART, AISB<br />AI Magazine (Editor: IU’s David Leake)<br />28<br />
  52. 52. Class Policies<br />29<br />
  53. 53. Prerequisites<br />C211<br />I recommend:<br />Two semesters programming<br />Basic knowledge of data structures<br />Basic knowledge of algorithmic complexity<br />30<br />
  54. 54. Programming Assignments<br />Projects will be written in Python<br />Great for scripting<br />Peter Norvig, Director of Research at Google, and textbook author<br />Easy to learn<br />2 weeks for each assignment<br />31<br />
  55. 55. Grading<br />50% Homework<br />6 assignments - lowest score will be dropped<br />30% Final<br />20% Midterm<br />32<br />
  56. 56. Homework Policy<br />Due at end of class on due date<br />Typically Tuesdays<br />Extensions only granted in rare cases<br />Require advance notice except emergencies<br />33<br />
  57. 57. Enrollment<br />Add/drop deadline<br />No penalty: Sept 3<br />Late drop/add: Oct 27<br />Waitlist deadline: Sept 4<br />34<br />
  58. 58. Takeaways<br />AI has many interpretations<br />Act vs. think, human-like vs. rational<br />Concept has evolved<br />‘I’ has many interpretations<br />Turing test<br />Chinese room<br />AI success stories from each perspective<br />35<br />
  59. 59. Homework<br />Register<br />Textbook<br />Survey<br />http://cs.indiana.edu/classes/b351<br />Readings: R&N Ch. 1, 2, 26<br />36<br />

×