An algebraic foundation for truth maintenance

Authors:
Allen L. Brown;Dale E. Gaucas;Dan Benanav
Affiliations:
GE Corporate Research and Development Center, Schenectady, New York;GE Corporate Research and Development Center, Schenectady, New York;GE Corporate Research and Development Center, Schenectady, New York
Venue:
IJCAI'87 Proceedings of the 10th international joint conference on Artificial intelligence - Volume 2
Year:
1987

Citing 9
Cited 1

An assumption-based TMS

Artificial Intelligence
Extending the ATMS

Artificial Intelligence
Problem solving with the ATMS

Artificial Intelligence
Flow Analysis of Computer Programs

Flow Analysis of Computer Programs
Compiler Construction

Compiler Construction
The Design and Analysis of Computer Algorithms

The Design and Analysis of Computer Algorithms
Truth Maintenance Systems for Problem Solving

Truth Maintenance Systems for Problem Solving
Principles of Compiler Design (Addison-Wesley series in computer science and information processing)

Principles of Compiler Design (Addison-Wesley series in computer science and information processing)
A mathematical theory of global program optimization (Prentice-Hall series in automatic computation)

A mathematical theory of global program optimization (Prentice-Hall series in automatic computation)

A correct non-monotonic ATMS

IJCAI'89 Proceedings of the 11th international joint conference on Artificial intelligence - Volume 2

Quantified Score

Hi-index	0.00

Visualization

Abstract

We have recast the problem of truth maintenance in a setting of algebraic equations over Boolean lattices. If a method of labeling propositions to justify them according to some reasoning agent's constraints of belief happens to conform to the postulates of Boolean lattices, the labeling system can be reformulated as an algebraic equation solving system. All truth maintenance systems known to us can be so reformulated. This note summarizes our investigations into the existence and structure of solutions of these algebraic systems. Our central result is a unique factorization theorem for lattice equational systems and their solutions. Our theoretical results are interpreted to compare various styles of truth maintenance and to reveal certain computational difficulties implicit in the algebraic structure of truth maintenance.