Efficiency analysis of multiple-context TMSS in scene representation

Authors:
Gregory M. Provan
Affiliations:
Mathematical Institute, Oxford University, Oxford, England
Venue:
AAAI'87 Proceedings of the sixth National conference on Artificial intelligence - Volume 1
Year:
1987

Citing 7
Cited 0

A qualitative physics based on confluences

Artificial Intelligence - Special volume on qualitative reasoning about physical systems
An assumption-based TMS

Artificial Intelligence
Problem solving with the ATMS

Artificial Intelligence
Incremental reconstruction of 3D scenes from multiple, complex images

Artificial Intelligence
Diagnosing multiple faults

Artificial Intelligence
A retrospective view of the Hearsay-II architecture

IJCAI'77 Proceedings of the 5th international joint conference on Artificial intelligence - Volume 2
Reasoning in multiple belief spaces

IJCAI'83 Proceedings of the Eighth international joint conference on Artificial intelligence - Volume 1

Quantified Score

Hi-index	0.00

Visualization

Abstract

Multiple possible solutions can arise in many domains, such as scene interpretation and speech recognition. This paper examines the eficiency of multiple-context TMSs, such as the ATMS, in solving a scene representation problem which we call the Vision Constraint Recognition problem. The ATMS has been claimed to be quite eficient for solving problems with multiple possible solutions, even for problems with large databases. However, we present evidence that for large databases with multiple possible solutions (which we argue occur frequently in practice), such multiple-context TMSs can be very inefficient. We present a class of problems for which using a multiple-context TMS is both intrinsically interesting and ideal, but which will be computationally infeasible because of the exponential size of the database which the TMS must explore. To circumvent such infeasiblity, appropriate control must be exerted by the problem solver.