On theoretical pProperties of sum-product networks

  • Robert Peharz
  • , Sebastian Tschiatschek
  • , Franz Pernkopf
  • , Pedro Domingos

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

Abstract

Sum-product networks (SPNs) are a promising avenue for probabilistic modeling and have been successfully applied to various tasks. However, some theoretic properties about SPNs are not yet well understood. In this paper we fill some gaps in the theoretic foundation of SPNs. First, we show that the weights of any complete and consistent SPN can be transformed into locally normalized weights without changing the SPN distribution. Second, we show that consistent SPNs cannot model distributions significantly (exponentially) more compactly than decomposable SPNs. As a third contribution, we extend the inference mechanisms known for SPNs with finite states to generalized SPNs with arbitrary input distributions.
Original languageEnglish
Title of host publicationProceedings of the Conference on Artificial Intelligence and Statistics (AISTATS)
Pages744-752
Publication statusPublished - 2015
Externally publishedYes
Event18th International Conference on Artificial Intelligence and Statistics (AISTATS 2015) - San Diego, United States
Duration: 9 May 201512 May 2015

Publication series

NameProceedings of Machine Learning Research
Volume18

Conference

Conference18th International Conference on Artificial Intelligence and Statistics (AISTATS 2015)
Abbreviated titleAISTATS2015
Country/TerritoryUnited States
CitySan Diego
Period9/05/1512/05/15

Fingerprint

Dive into the research topics of 'On theoretical pProperties of sum-product networks'. Together they form a unique fingerprint.

Cite this