Inference with aggregation parfactors: Lifted elimination with first-order d-separation

Felipe Iwao Takiyama, Fabio Gagliardi Cozman

Research output: Contribution to conferenceConference Paper

1 Scopus citations

Abstract

© 2014 IEEE. In this paper we focus on lifted inference for statistical relational models, that is, inference that avoids complete grounding, in models that combine logical and probabilistic assertions. We focus on relational Bayesian networks that can be represented through par factors and aggregation par factors. We present a new elimination rule for lifted variable elimination, and show how to use first-order d-separation to extend the reach of existing elimination rules.
Original languageAmerican English
Pages384-389
Number of pages6
DOIs
StatePublished - 1 Jan 2014
Externally publishedYes
EventProceedings - 2014 Brazilian Conference on Intelligent Systems, BRACIS 2014 -
Duration: 1 Jan 2014 → …

Conference

ConferenceProceedings - 2014 Brazilian Conference on Intelligent Systems, BRACIS 2014
Period1/01/14 → …

Fingerprint

Dive into the research topics of 'Inference with aggregation parfactors: Lifted elimination with first-order d-separation'. Together they form a unique fingerprint.

Cite this