Quotient normalized maximum likelihood criterion for learning Bayesian network structures

Loading...
Thumbnail Image

Access rights

openAccess

URL

Journal Title

Journal ISSN

Volume Title

A4 Artikkeli konferenssijulkaisussa

Date

2018-01-01

Major/Subject

Mcode

Degree programme

Language

en

Pages

10
948-957

Series

International Conference on Artificial Intelligence and Statistics, 9-11 April 2018, Playa Blanca, Lanzarote, Canary Islands, Proceedings of Machine Learning Research, Volume 84

Abstract

We introduce an information theoretic criterion for Bayesian network structure learning which we call quotient normalized maximum likelihood (qNML). In contrast to the closely related factorized normalized maximum likelihood criterion, qNML satisfies the property of score equivalence. It is also decomposable and completely free of adjustable hyperparameters. For practical computations, we identify a remarkably accurate approximation proposed earlier by Szpankowski and Weinberger. Experiments on both simulated and real data demonstrate that the new criterion leads to parsimonious models with good predictive accuracy.

Description

Keywords

Other note

Citation

Silander, T, Leppä-Aho, J, Jääsaari, E & Roos, T 2018, Quotient normalized maximum likelihood criterion for learning Bayesian network structures . in A Storkey & F Perez-Cruz (eds), International Conference on Artificial Intelligence and Statistics, 9-11 April 2018, Playa Blanca, Lanzarote, Canary Islands . Proceedings of Machine Learning Research, vol. 84, JMLR, pp. 948-957, International Conference on Artificial Intelligence and Statistics, Playa Blanca, Spain, 09/04/2018 . < http://proceedings.mlr.press/v84/silander18a.html >