CS 657 Applied Logic for Artificial Intelligence

Objectives:

Applications of logic. Topics include first order predicate calculus as a reasoning agent, deductive reasoning and resolution refutation, nonmonotonic reasoning, induction, reasoning with uncertain information, reasoning about knowledge and belief, meta level representation and reasoning and and architectures for intelligent agents.

Prerequisites: CS480/580 or consent of the instructor

Textbook:

Michael R. Genesereth and Nils J. Nilsson Logical Foundations of Artificial Intelligence Morgan Kaufmann Publishers, Inc. 1987

References:

Various references cited in the textbook Various journals and conference proceedings in CS and AI

Outline:


1. Overview
2. Declarative knowledge representation.
	Applications: blocks world, circuits, algebra, natural language.
3. Logical inference
4. Resolution principle: concept, method, strategies.
	Applications: finding answers from databases, diagnosing circuit faults,
	  proving mathematical theorems.
5. Nonmonotonic reasoning
	Applications: decision making with best, insufficient knowledge.
6. Generalization and induction
	Applications: learning, perception.
7. Knowledge vs. belief, and reasoning with uncertainty
	Applications: expert systems.
8. Metaknowledge and metareasoning
	Applications: building more efficient expert systems.
9. States, actions, and changes
	Applications: robot actions.


wahab@duke.ncsl.nist.gov
Tue Apr 23 10:50:19 EDT 1996