Abstract
In this paper we study the problem of exact inference in hybrid Bayesian networks using mixtures of truncated basis functions (MoTBFs). We propose a structure for handling probability potentials called Sum-Product factorized potentials, and show how these potentials facilitate efficient inference based on i) properties of the MoTBFs and ii) ideas similar to the ones underlying Lazy propagation (postponing operations and keeping factorized representations of the potentials). We report on preliminary experiments demon- strating the efficiency of the proposed method in comparison with existing algorithms.
Originalsprog | Engelsk |
---|---|
Titel | Proceedings of the Sixth European Workshop on Probabilistic Graphical Models |
Redaktører | Andrés Cano, Manuel Gómez-Olmedo, Thomas Dyhre Nielsen |
Antal sider | 8 |
Forlag | DECSAI, University of Granada |
Publikationsdato | 2012 |
Sider | 171-178 |
ISBN (Trykt) | 978-84-15536-57-4 |
Status | Udgivet - 2012 |
Begivenhed | The 6th European Workshop on Probabilistic Graphical Models - Granada, Spanien Varighed: 19 sep. 2012 → 21 sep. 2012 |
Konference
Konference | The 6th European Workshop on Probabilistic Graphical Models |
---|---|
Land/Område | Spanien |
By | Granada |
Periode | 19/09/2012 → 21/09/2012 |