Dividing Up AI

Theories and Techniques

Representational techniques: semantic nets, frames, predicate calculus, distributed representations
Heuristic search
Production systems
Grammars, transition networks, and parsers
Resolution theorem proving
Bayesian reasoning
Case-based reasoning
Explanation-based learning
Constraint propagation
Connectionism (neural networks)
Genetic algorithms and classifier systems

Basic methodology
- Programming AI: if you want an intelligent machine, program in the intelligence.
- Adaptive AI: if you want an intelligent machine, make it a good learner, and send it out into the world.
What intelligence looks like
- Neat AI
  - Theories should be elegant and parsimonious.
  - We should understand precisely what our theories can do and how they behave.
  - Most (or all) of intelligence is governed by general principles.
- Scruffy AI
  - The mind is a kludge. To make things efficient, inelegant shortcuts are often appropriate.
  - It may be impossible to come to a precise understanding of our theories.
  - There are only a few general principles that apply across domains. Intelligence comes from domain knowledge.
Why do it
- Engineering-oriented AI: Build intelligent machines to solve practical problems.
- Cognitive-science-oriented AI: Build intelligent machines to help us understand biological intelligence.

Last updated: 8 January 1996
URL: http://www.indiana.edu/~gasser/Q351/divisions.html
Comments: gasser@salsa.indiana.edu
Copyright 1996, The Trustees of Indiana University