The effect of combination functions on the complexity of relational Bayesian networks

Denis Deratani Mau, Fabio Gagliardi Cozman

Research output: Contribution to journalConference articlepeer-review

Abstract

We study the complexity of marginal inference with Relational Bayesian Networks as parameterized by their probability formulas. We show that without combination functions, inference is #P-equivalent, displaying the same complexity as standard Bayesian networks (this is so even when relations have unbounded arity and when the domain is succinctly specified in binary notation). By allowing increasingly more expressive probability formulas using only maximization as combination, we obtain inferential complexity that ranges from #P-equivalent to FPSPACE-complete to EXP-hard. In fact, by suitable restrictions to the number of nestings of combination functions, we obtain complexity classes in all levels of the counting hierarchy. Finally, we investigate the use of arbitrary combination functions and obtain that inference is FEXP-complete even under a seemingly strong restriction.

Original languageEnglish
Pages (from-to)333-344
Number of pages12
JournalJournal of Machine Learning Research
Volume52
Issue number2016
StatePublished - 2016
Externally publishedYes
Event8th International Conference on Probabilistic Graphical Models, PGM 2016 - Lugano, Switzerland
Duration: 6 Sep 20169 Sep 2016

Keywords

  • Complexity theory
  • Probabilistic inference
  • Relational Bayesian networks

Fingerprint

Dive into the research topics of 'The effect of combination functions on the complexity of relational Bayesian networks'. Together they form a unique fingerprint.

Cite this