Timezone: »
We present a method for learning Bayesian networks from data sets containingthousands of variables without the need for structure constraints. Our approachis made of two parts. The first is a novel algorithm that effectively explores thespace of possible parent sets of a node. It guides the exploration towards themost promising parent sets on the basis of an approximated score function thatis computed in constant time. The second part is an improvement of an existingordering-based algorithm for structure optimization. The new algorithm provablyachieves a higher score compared to its original formulation. On very large datasets containing up to ten thousand nodes, our novel approach consistently outper-forms the state of the art.
Author Information
Mauro Scanagatta (IDSIA)
Cassio de Campos (Queen's University Belfast)
Giorgio Corani (IDSIA)
Marco Zaffalon (IDSIA)
More from the Same Authors
-
2021 : Causal Expectation-Maximisation »
Marco Zaffalon · Alessandro Antonucci · Rafael Cabañas -
2021 : Zaffalon, Antonucci, Cabañas - Causal Expectation-Maximisation »
Marco Zaffalon · Alessandro Antonucci · Rafael Cabañas -
2016 Poster: Learning Treewidth-Bounded Bayesian Networks with Thousands of Variables »
Mauro Scanagatta · Giorgio Corani · Cassio de Campos · Marco Zaffalon -
2014 Poster: Advances in Learning Bayesian Networks of Bounded Treewidth »
Siqi Nie · Denis Maua · Cassio P de Campos · Qiang Ji -
2014 Spotlight: Advances in Learning Bayesian Networks of Bounded Treewidth »
Siqi Nie · Denis Maua · Cassio P de Campos · Qiang Ji -
2014 Poster: Global Sensitivity Analysis for MAP Inference in Graphical Models »
Jasper De Bock · Cassio P de Campos · Alessandro Antonucci -
2011 Poster: Solving Decision Problems with Limited Information »
Denis Maua · Cassio P de Campos -
2011 Spotlight: Solving Decision Problems with Limited Information »
Denis Maua · Cassio P de Campos