Bounded Conditioning: A Flexible Probabilistic Inference Algorithm

From Tetherless World Wiki

Jump to: navigation, search

Citation: Eric Horvitz and Henri Jacques Suermondt. (1990) Bounded Conditioning: A Flexible Probabilistic Inference Algorithm. In KSL-88-36, October,1990.

Publication techreport ( Edit )
type Technical Report
bibtype techreport
Bibtex basics
author Eric Horvitz and Henri Jacques Suermondt
title Bounded Conditioning: A Flexible Probabilistic Inference Algorithm
number KSL-88-36
institution Knowledge Systems, AI Laboratory
year 1990
month October
Bibtex more
note 20 pages.
Access Paper
abstract We introduce a graceful approach to probabilistic inference called bounded conditioning. Bounded conditioning monotonically refines the bounds on posterior probabilities in a belief network with computation, and converges on final probabilities of interest with the allocation of a complete resource fraction. The approach allows a reasoner to exchange arbitrary quantities of computational resource for incremental gains in inference quality. As such, bounded conditioning holds promise as a useful inference technique for reasoning under the general conditions of uncertain and varying reasoning resources. The algorithm solves a probabilistic bounding problem in complex belief networks by breaking the problem into a set of mutually exclusive, tractable subproblems and ordering their solution by the expected effect that each subproblem will have on the final answer. We introduce the algorithm discuss its characterization, and present its performance on several belief networks, including a complex model for reasoning about problems in intensive-care medicine.

KSL Technical Report ID: KSL-88-36
Facts about Bounded Conditioning: A Flexible Probabilistic Inference AlgorithmRDF feed
Abstract We introduce a graceful approach to probab … We introduce a graceful approach to probabilistic inference called bounded conditioning. Bounded conditioning monotonically refines the bounds on posterior probabilities in a belief network with computation, and converges on final probabilities of interest with the allocation of a complete resource fraction. The approach allows a reasoner to exchange arbitrary quantities of computational resource for incremental gains in inference quality. As such, bounded conditioning holds promise as a useful inference technique for reasoning under the general conditions of uncertain and varying reasoning resources. The algorithm solves a probabilistic bounding problem in complex belief networks by breaking the problem into a set of mutually exclusive, tractable subproblems and ordering their solution by the expected effect that each subproblem will have on the final answer. We introduce the algorithm discuss its characterization, and present its performance on several belief networks, including a complex model for reasoning about problems in intensive-care medicine. about problems in intensive-care medicine.
Author Eric Horvitz and Henri Jacques Suermondt  +
Bibtype techreport  +
Has author Eric Horvitz and Henri Jacques Suermondt  +
Has identifier KSL-88-36  +
Has publishing details October,1990  +
Has title Bounded Conditioning: A Flexible Probabilistic Inference Algorithm  +
Has where published KSL-88-36  +
Has year 1990  +
Institution Knowledge Systems, AI Laboratory  +
Ksl tr id KSL-88-36  +
Month October  +
Note 20 pages.
Number KSL-88-36  +
Process note YES  +
Title Bounded Conditioning: A Flexible Probabilistic Inference Algorithm  +
Year 1990  +
Retrieved from "http://tw.rpi.edu/wiki/Bounded_Conditioning:_A_Flexible_Probabilistic_Inference_Algorithm"
Personal tools