Epistemological Problems of Artificial Intelligence by John McCarthy

1
Epistemological Problems of Artificial
Intelligenceby John McCarthy

• Presented by Francis M. Kagunda
• September 24, 2002

2
Outline

• Introduction
• Epistemological problems
• Circumscription
• Concepts as objects
• Philosophical notes
• Conclusion

3
Introduction

• Parts of AI problems
• (a). Epistemological
• (b). Heuristic
• Advantage of similar problems
• Single solution variety of heuristic approaches
  to a problem
• AI must take into account what is knowable with
  available observational and computational
  facilities

4
Epistemological Problems

• Co-operation of other people or overcoming their
  opposition?
• Does the strategy involve the acquisition of
  knowledge?
• Involvement of concurrent events and actions
• Ability to express knowledge about space and the
  locations, shapes and layouts of objects in
  space

5
Epistemological problems continued

• Relation between multidimensional objects
• Modal concepts
• The frame problem
• Dove andremo?

6
Circumscription

• Humans and intelligent computer programs must
  often jump to the conclusion that the objects
  they can determine to have certain properties or
  relations are the only objects that do.
  Circumscription formalizes such conjectural
  reasoning.
• Circumscription is non-monotonic

7
examples

• Semantic ways of looking at circumscription
• Objects that can be shown to have a certain
  property P by reasoning from certain facts A are
  all the objects that satisfy P.

8
Concepts as Objects

• Concept as object has application beyond the
  discussion of knowledge
• Knowledge and modal treatment.Belief, wanting
  and logical or physical necessity
• Relation between knowledge and ability
• Semantics of reasoning about objects is more
  complicated if one refers to them via concepts

9
Philosophical notes

• Building a view of the world into the structure
  of a program does not in itself give the program
  the ability to state the view explicitly.
• Common sense requires scientific formulation
• Dilemma imprecise and inconsistent formulations
  need to be used.
• AI formalism that uses concepts and goes beyond
  them
• Different schools of thought influence program
  construction

10
Conclusion

• AI could benefit from building simple systems and
  so might philosophy
• Refer to the suggested further reading
• Questions, concerns, comments, . . .
• Thank you all.