SOME IMPROVEMENTS TO THE SHENOY-SHAFER AND HUGIN ARCHITECTURES FOR COMPUTING MARGINALS

Citation
T. Schmidt et Pp. Shenoy, SOME IMPROVEMENTS TO THE SHENOY-SHAFER AND HUGIN ARCHITECTURES FOR COMPUTING MARGINALS, Artificial intelligence, 102(2), 1998, pp. 323-333
Citations number
9
Categorie Soggetti
Computer Science Artificial Intelligence","Computer Science Artificial Intelligence
Journal title
ISSN journal
00043702
Volume
102
Issue
2
Year of publication
1998
Pages
323 - 333
Database
ISI
SICI code
0004-3702(1998)102:2<323:SITTSA>2.0.ZU;2-9
Abstract
The main aim of this paper is to describe two modifications to the She noy-Shafer architecture with the goal of making it computationally mor e efficient in computing marginals of the joint valuation. We also des cribe a modification to the Hugin architecture. Finally, we briefly co mpare the traditional and modified architectures by solving a couple o f small Bayesian networks, and conclude with a statement of further re search. (C) 1998 Elsevier Science B.V. All rights reserved.