COSC 6368 and

1
COSC 6368 and What is AI?

• Introduction to AI (today, and TH)
• What is AI?
• Sub-fields of AI
• Problems investigated by AI research
• Course Information

2
Part1a Definitions of AI

• AI centers on the simulation of intelligence
  using computers
• AI develops programming paradigms, languages,
  tools, and environments for application areas for
  which conventional programming fails
• Symbolic programming (LISP)
• Functional programming
• Heuristic Programming
• Logical Programming (PROLOG)
• Rule-based Programming (Expert system shells)
• Soft Computing (Belief network tools, fuzzy logic
  tool boxes,)
• Object-oriented programming (Smalltalk)

3
More Definitions of AI

• Rich/Knight AI is the study of of how to make
  computers do things which, at the moment, people
  do better
• Winston AI is the study of computations that
  make it possible to perceive, reason, and act.
• Turing Test If an artificial intelligent system
  is not distinguishable from a human being, it is
  definitely intelligent.

4
Physical Symbol System Hypothesis

• What the brain does can be thought of at some
  level as a kind of computation
• Physical Symbol System Hypothesis (PSSH) A
  physical symbol system has the sufficient and
  necessary means for general, intelligent actions.
• Remarks PSSH
• Subjected to empirical validation
• If false ? AI is quite limited
• Important for psychology and philosophy

5
Questions/Thoughts about AI

• What are the limitations of AI? Can computers
  only do what they are told? Can computers be
  creative? Can computers think? What problems
  cannot be solved by computers today?
• Computers show promise to control the current
  waste of energy and other natural resources.
• Computer can work in environment that are
  unsuitable for human beings.
• If computers control everything --- who controls
  the computers?
• If computers are intelligent what civil rights
  should be given to computers?
• If computers can perform most of our work what
  should the human beings do?
• Only those things that can be represented in
  computers are important.
• It is fun to play with computers.

6
Topics Covered in COSC 6368

• More general topics
• heuristic search and search algorithm in general
• logical reasoning (FOPL as a language)
• making sense out of data
• AI-specific Topics
• resolution / theorem proving
• reasoning in uncertain environments and belief
  networks
• machine learning and data mining
• brief coverage of planning, evolutionary
  computing, knowledge-based systems and
  philosophical aspects of AI
• Exposure to AI tools (belief networks, decision
  trees,)

7
2009 Organization COSC 6368

• Introduction AI and Course Information (1-2
  classes)
• Heuristic Search (4-5 classes)
• Evolutionary Computing (2 classes)
• FOPL, Logical Reasoning, Resolution, and PROLOG
  (3-4 classes)
• Inductive Learning, Reinforcement Learning, Brief
  Introduction to Data Mining (4 classes)
• Knowledge-based Systems and Expert Systems (1
  class)
• Planning (1-2 classes)
• Ontologies and Philosophical Aspects of AI (1-2
  classes)
• Belief Networks and Reasoning in Uncertain
  Environments (3-4 classes)
• Other Activities midterm exam (1 class), review
  (2 classes), homework/project-related
  discussions(1 class), possibly paper walk-through
  (1 class).

8
AI in General and What Is not Covered in COSC 6368

• Robotics is a quite important sub-field of AI,
  but very few teach it in the graduate AI class.
• Natural language understanding probably will not
  be covered.
• Intelligent Agents and AI for the Internet
  could/should possibly be covered in a little more
  depth.
• Artificial intelligence programming is not
  covered.
• Techniques employed in systems that automate
  decision making in uncertain environments
  deserves more attention (e.g. fuzzy logic,
  rule-based programming languages and expert
  system shells, fuzzy controllers).

9
Positive Forces for AI

• Knowledge Discovery in Data and Data Mining (KDD)
• Intelligent Agents for WWW
• Robotics (Robot Soccer, Intelligent Driving,
  Robot Waiters, industrial robots, rovers, toy
  robots)
• Creating of Knowledge Bases and Sharing of
  Knowledge (especially for Science and
  Engineering)
• Computer Chess and Computer Games in General ---
  AI for Entertainment

10
6368 Homepage

• http//www2.cs.uh.edu/ceick/6368.html

IJCAI 2009 Homepage http//ijcai-09.org/
11
Course Elements

• ?21 Lectures
• 3 Exams (two midterms, one final exam)
• 4 Graded Assignments (review questions, exam
  style paper and pencil problems, a few more
  challenging problems that might require
  programming problems that require using AI
  tools searching for something and reporting)
• Un-graded Homeworks (solutions will usually
  discussed in class)
• 1 Paper Walk-Throughs (group activity) if class
  size lt20
• Discussion of assignments and home works
• We will try to use more demos and animations ---
  we have to see if this turns out to be useful

12
AI
AI Programming
Knowledge Representation
Knowledge-based and Expert Systems
Part1b
Planning
Coping with Vague, Incomplete and Uncertain
Knowledge
Searching Intelligently
Logical Reasoning Theorem Proving
Communicating, Perceiving and Acting
Intelligent Agents Distributed AI
Learning Knowledge Discovery
13
Part1b Examples of Problems Investigated by
Different Subfields of AI
14
Knowledge Representation
Problem Can the above chess board be cover by 31
domino pieces that cover 2 fields?
AIs contribution object-oriented and
frame-based systems, ontology languages, logical
knowledge representation frameworks, belief
networks
15
Natural Language Understanding

• I saw the Golden Gate Bridge flying to San
  Francisco.
• I ate dinner with a friend. I ate dinner with a
  fork.
• John went to a restaurant. He ordered a steak.
  After an hour John left happily.
• I went to three dentists this morning.

16
Planning

• Objective Construct a sequence of actions that
  will achieve a goal.
• Example John want to buy a house

17
Heuristic Search

• Heuristo (greek) I find
• Copes with problems for which it is not feasible
  to look at all solutions
• Heuristics rules a thumb (help you to explore
  the more promising solutions first), based on
  experience, frequently fuzzy
• Main ideas of heuristics search space reduction,
  ordering solutions intelligently, simplifications
  of computations

Example problems puzzles, traveling salesman
problem,
18
Figure
19
Evolutionary Computing

• Evolutionary algorithms are global search
  techniques.
• They are built on Darwins theory of evolution by
  natural selection.
• Numerous potential solutions are encoded in
  structures, called chromosomes.
• During each iteration, the EA evaluates solutions
  adn generates offspring based on the fitness of
  each solution in the task.
• Substructures, or genes, of the solutions are
  then modified through genetic operators such as
  mutation or recombination.
• The idea structures that led to good solutions
  in previous evaluations can be mutated or
  combined to form even better solutions.

20
Logical Reasoning

• Learn how to represents natural language
  statements in logic (AI as language)
• Automated theorem proving
• Foundation for PROLOG

21
Soft Computing

• Conventional Programming
• Relies on two-valued logic
• Mostly uses a symbolic (non-numerical knowledge
  representation framework)
• Soft Computing (e.g. Fuzzy Logic, Belief
  Networks,..)
• Tolerance for uncertainty and imprecision
• Uses weights, probabilities, possibilities
• Strongly relies on numeric approximation and
  interpolation
• Remark There seem to be two worlds in computer
  science one views the world as consisting of
  numbers the other views the world as consisting
  of symbols.

22
Different Forms of Learning

• Learning agent receives feedback with respect to
  its actions (e.g. using a teacher)
• Supervised Learning/Learning from
  Examples/Inductive Learning feedback is received
  with respect to all possible actions of the agent
• Reinforcement Learning feedback is only
  received with respect to the taken action of the
  agent
• Unsupervised Learning Learning without feedback

23
Machine Learning Classification- Model
Construction (1)
Classification Algorithms
IF rank professor OR years gt 6 THEN tenured
yes
24
Classification Process (2) Use the Model in
Prediction
(Jeff, Professor, 4)
Tenured?
25
Knowledge Discovery in Data and Data Mining
(KDD)
Let us find something interesting!

• Definition KDD is the non-trivial process of
  identifying valid, novel, potentially useful, and
  ultimately understandable patterns in data
  (Fayyad)

26
2. General Course Information
Course Id COSC 6368 Machine Learning Time
TU/TH 1-230 Instructor Christoph F. Eick
Classroom 232 PGH E-mail
ceick_at_aol.com Homepage
http//www2.cs.uh.edu/ceick/

27
Prerequisites

• Background
• Algorithms
• basic data structures, complexity
• Sound programming skills (no knowledge of LISP or
  PROLOG is requred)
• Ability to deal with abstract mathematical
  concepts
• Basic knowledge of logic would be helpful

28
Textbook
http//aima.cs.berkeley.edu/
29
Grading
2 Exams 60 4 Assignment 40  
Remark Weights are subject to change
NOTE PLAGIARISM IS NOT TOLERATED.
30
Tentative 2009 Teaching Plan (Subject To Change)
Week Topic
Jan 20 Introduction / Search
Jan 27 Search
Feb. 3 Search/Evolutionary Computing (EC)
Feb. 10 EC, Logical Reasoning (LR)
Feb. 17 LR
Feb. 24 LR/Learning from Examples(LFE)
March 3 LFE/Reinforcement Learning
March 10 Review,/Midterm Exam
March 24 Leftovers/Knowledge-based Systems
March 31 Ontologies/ Philosophical Foundations of AI
April 7 Planning
April 14 Reasoning in Uncertain Environments (RIE)
April 21 RIE
April 28 RIE/Review for Final Exam

Remark Topics in brown color may be skipped or
replaced by something else
31
Dates to Remember
Dates to remember Events
Last day before Spring Break May 12 Exams
March 17 /19 No class (Spring Break)

32
Exams

• Will be open notes/textbook
• Will get a review list before the exam
• Exams will center (80 or more) on material that
  was covered in the lecture
• Exam scores will be immediately converted into
  number grades
• A few sample exams are available

33
Other UH-CS Courses with Overlapping Contents

• COSC 6342 Machine Learning
• Strong Overlap Decision Trees, Bayesian Belief
  Networks, Learning from Examples in general
• Medium Overlap Reinforcement Learning
• COSC 6335 Data Mining
• Overlap Decision trees, Learning from Examples
  in general
• Preprocessing/Exploratory DA, AdaBoost
• COSC 6367 Evolutionary Computing
• Overlap Search
• We also will have 2 lectures on Evolutionary
  Computing