Hybrid Processing of Beliefs and Constraints
Abstract
This paper explores algorithms for processing probabilistic and deterministic information when the former is represented as a belief network and the latter as a set of boolean clauses. The motivating tasks are 1. evaluating belief networks having a large number of deterministic relationships and 2. evaluating probabilities of complex boolean queries or complex evidence information over a belief network. We present and analyze a variable elimination algorithm that exploits both types of information, and provide empirical evaluation demonstrating its computational benefits.
Cite
Text
Dechter and Larkin. "Hybrid Processing of Beliefs and Constraints." Conference on Uncertainty in Artificial Intelligence, 2001.Markdown
[Dechter and Larkin. "Hybrid Processing of Beliefs and Constraints." Conference on Uncertainty in Artificial Intelligence, 2001.](https://mlanthology.org/uai/2001/dechter2001uai-hybrid/)BibTeX
@inproceedings{dechter2001uai-hybrid,
title = {{Hybrid Processing of Beliefs and Constraints}},
author = {Dechter, Rina and Larkin, David},
booktitle = {Conference on Uncertainty in Artificial Intelligence},
year = {2001},
pages = {112-119},
url = {https://mlanthology.org/uai/2001/dechter2001uai-hybrid/}
}