CS451CS551EE565 ARTIFICIAL INTELLIGENCE

1
CS451/CS551/EE565ARTIFICIAL INTELLIGENCE

• Knowledge Representation
• 10-30-2006
• Prof. Janice T. Searleman
• jets_at_clarkson.edu, jetsza

2
Outline

• Knowledge Representation
• Ontological engineering
• Categories and objects
• Actions, situations and events
• Reading Assignment AIMA
• Chapter 10 Knowledge Representation
• Projects
• be sure to include sufficient good references
• the project has a research component a small
  programming/modelling component
• be sure to relate the research to AI in general
  and agents specifically

3
First Order Logic

• sufficiently powerful to express most
  mathematical concepts
• things that are difficult to represent in FOL
• It is very cold today.
• If theres no evidence to the contrary, assume
  that any adult you meet knows how to read.
• Blond-haired people often have blue eyes.
• I know that Bill thinks that the Giants will win,
  but I think that they are going to lose.
• cannot quantify over a function or a predicate

4
Other Logics

• Higher-order logics
• e.g. allow functions predicates to be
  quantified
• Modal logics different degrees of truth
• e.g. possibly true, true-ish, absolutely true
• also, probabilistic logics
• Multivalued logics true, false, neither
• Fuzzy logics true to a certain degree
• Temporal logics
• etc.

5
Reasoning Systems

• deduction
• abduction
• induction

from infer A
B A gt B
from infer B
A A gt B
from infer p(a)
p(X) p(b) p(c)
6
What is Knowledge?

• textbook knowledge
• e.g. facts, theories, methods,
• rules of use
• heuristics which capture practical guidelines
  for effective use of knowledge
• Can be well-informed, yet unskillful inexpert.
• An expert makes use of empirical associations,
  heuristics for dealing with uncertain or
  incomplete information, and constraints.
  Results speed, focus on relevant facts, fewer
  errors, adaptability robustness.

7
KB operations

• store facts (memory)
• want the KB to be well-formed, consistent
• retrieve knowledge (explicit knowledge)
• yes/no Is Socrates a person?
• return a list of all known people
• inference (implicit knowledge)
• Is Socrates mortal?
• Who is mortal?
• other operations remove facts, modify facts, etc.

8
Procedural vs. Declarative

• Declarative just the facts, maam
• knowledge is specified, how to use it is not
• adv. flexible, modular, easy to add facts
• Procedural how to do something (e.g. make a cup
  of tea)
• control info required to use the knowledge is
  embedded into the knowledge itself
• adv. control of the search for answers
• Tacit knowledge not expressed in language (how
  to move hand)
• Causal knowledge cause and effect
• others?

9
Knowledge Engineer

• Populates KB with facts and relations
• Must study and understand the domain to pick
  important objects and relationships
• Main steps
• Decide what to talk about
• Decide on vocabulary of predicates, functions
  constants
• Encode general knowledge about domain
• Encode description of specific problem instance
• Pose queries to inference procedure and get
  answers

10
Properties of good knowledge bases

• Expressive
• Concise
• Unambiguous
• Context-insensitive
• Effective
• Clear
• Correct
• Trade-offs e.g. sacrifice some correctness if it
  enhances brevity.

11
Efficiency

• Ideally Not the knowledge engineers problem
• The inference procedure should obtain same
  answers no matter how knowledge is implemented.
• In practice
• - use automated optimization
• - knowledge engineer should have some
• understanding of how inference is done

12
Debugging

• In principle, easier than debugging a program,
• because we can look at each logic sentence in
  isolation and tell whether it is correct.
• Example
• x, Animal(x) ? ? b, BrainOf(x) b
• means
• there is some object that is the value of the
  BrainOf function applied to an animal
• and can be corrected to mean
• every animal has a brain
• without looking at other sentences.

13
Ontology

• Collection of concepts and inter-relationships
• Widely used in the database community to
  translate queries and concepts from one
  database to another, so that multiple databases
  can be used conjointly (database federation)

14
Ontological engineering

• How to create more general and flexible
  representations.
• Concepts like actions, time, physical object and
  beliefs
• Operates on a bigger scale than K.E.
• Define general framework of concepts
• Upper ontology
• Limitations of logic representation
• Red, green and yellow tomatos exceptions and
  uncertainty

15
The upper ontology of the world
16
Ontology Example
Khan McLeod, 2000
17
Towards a general ontology

• Need to develop good representations for
• categories
• measures
• composite objects
• time, space and change
• events and processes
• physical objects
• substances
• mental objects and beliefs
• other?
• A general-purpose ontology should be applicable
  in more or less any special-purpose domain.