By the same authors

Bayesian network learning by compiling to weighted MAX-SAT

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Full text download(s)

Links

Author(s)

Department/unit(s)

Publication details

Title of host publicationProceedings of the 24th Conference on Uncertainty in Artificial Intelligence (UAI 2008)
DatePublished - 2008
Pages105-112
Number of pages8
PublisherAUAI Press
Place of PublicationCorvallis, Oregon
EditorsDavid McAllester, Petri Myllymaki
Original languageEnglish
ISBN (Print)0-9749039-4-9

Abstract

The problem of learning discrete Bayesian networks from data is encoded as a weighted MAX-SAT problem and the MaxWalkSat local search algorithm is used to address it. For each dataset, the per-variable summands of the (BDeu) marginal likelihood for different choices of parents (?family scores?) are computed prior to applying MaxWalkSat. Each permissible choice of parents for each variable is encoded as a distinct propositional atom and the associated family score encoded as a ?soft? weighted single-literal clause. Two approaches to enforcing acyclicity are considered: either by encoding the ancestor relation or by attaching a total order to each graph and encoding that. The latter approach gives better results. Learning experiments have been conducted on 21 synthetic datasets sampled from 7 BNs. The largest dataset has 10,000 datapoints and 60 variables producing (for the ?ancestor? encoding) a weighted CNF input file with 19,932 atoms and 269,367 clauses. For most datasets, MaxWalkSat quickly finds BNs with higher BDeu score than the ?true? BN. The effect of adding prior information is assessed. It is further shown that Bayesian model averaging can be effected by collecting BNs generated during the search.

Discover related content

Find related publications, people, projects, datasets and more using interactive charts.

View graph of relations