Belief updating and learning in semi-qualitative probabilistic networks

Cassio Polpo De Campos, Fabio Gagliardi Cozman

Research output: Contribution to conferenceConference Paper

26 Scopus citations

Abstract

This paper explores semi-qualitative probabilistic networks (SQPNs) that combine numeric and qualitative information. We first show that exact inferences with SQPNs are NPPP-Complete. We then show that existing qualitative relations in SQPNs (plus probabilistic logic and imprecise assessments) can be dealt effectively through multilinear programming. We then discuss learning: we consider a maximum likelihood method that generates point estimates given a SQPN and empirical data, and we describe a Bayesian-minded method that employs the Imprecise Dirichlet Model to generate set-valued estimates.
Original languageAmerican English
Pages153-160
Number of pages8
StatePublished - 1 Dec 2005
Externally publishedYes
EventProceedings of the 21st Conference on Uncertainty in Artificial Intelligence, UAI 2005 -
Duration: 1 Dec 2005 → …

Conference

ConferenceProceedings of the 21st Conference on Uncertainty in Artificial Intelligence, UAI 2005
Period1/12/05 → …

Fingerprint

Dive into the research topics of 'Belief updating and learning in semi-qualitative probabilistic networks'. Together they form a unique fingerprint.

Cite this