Fast factorisation of probabilistic potentials and its application to approximate inference in Bayesian networks
We present an efficient procedure for factorising probabilistic potentials represented as probability trees. This new procedure is able to detect some regularities that cannot be captured by existing methods. In cases where an exact decomposition is not achievable, we propose a heuristic way to c...
| Main Authors: | , , , |
|---|---|
| Format: | info:eu-repo/semantics/article |
| Language: | English |
| Published: |
2017
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/10835/4885 https://doi.org/10.1142/S0218488512500110 |
| _version_ | 1789406387299680256 |
|---|---|
| author | Cano, Andrés Gómez Olmedo, Manuel Pérez-Ariza, Cora B. Salmerón Cerdán, Antonio |
| author_facet | Cano, Andrés Gómez Olmedo, Manuel Pérez-Ariza, Cora B. Salmerón Cerdán, Antonio |
| author_sort | Cano, Andrés |
| collection | DSpace |
| description | We present an efficient procedure for factorising probabilistic potentials represented as
probability trees. This new procedure is able to detect some regularities that cannot be
captured by existing methods. In cases where an exact decomposition is not achievable,
we propose a heuristic way to carry out approximate factorisations guided by a parameter
called factorisation degree, which is fast to compute. We show how this parameter can be
used to control the tradeoff between complexity and accuracy in approximate inference
algorithms for Bayesian networks. |
| format | info:eu-repo/semantics/article |
| id | oai:repositorio.ual.es:10835-4885 |
| institution | Universidad de Cuenca |
| language | English |
| publishDate | 2017 |
| record_format | dspace |
| spelling | oai:repositorio.ual.es:10835-48852023-04-12T19:37:05Z Fast factorisation of probabilistic potentials and its application to approximate inference in Bayesian networks Cano, Andrés Gómez Olmedo, Manuel Pérez-Ariza, Cora B. Salmerón Cerdán, Antonio Bayesian netwoorks Probability trees Factorisation Probabilistic inference We present an efficient procedure for factorising probabilistic potentials represented as probability trees. This new procedure is able to detect some regularities that cannot be captured by existing methods. In cases where an exact decomposition is not achievable, we propose a heuristic way to carry out approximate factorisations guided by a parameter called factorisation degree, which is fast to compute. We show how this parameter can be used to control the tradeoff between complexity and accuracy in approximate inference algorithms for Bayesian networks. 2017-07-05T08:37:24Z 2017-07-05T08:37:24Z 2012 info:eu-repo/semantics/article http://hdl.handle.net/10835/4885 https://doi.org/10.1142/S0218488512500110 en http://www.worldscientific.com/doi/pdf/10.1142/S0218488512500110 Attribution-NonCommercial-NoDerivatives 4.0 Internacional http://creativecommons.org/licenses/by-nc-nd/4.0/ info:eu-repo/semantics/openAccess Preprint of an article submitted for consideration in International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems © 2012 [copyright World Scientific Publishing Company] http://www.worldscientific.com/worldscinet/ijufks |
| spellingShingle | Bayesian netwoorks Probability trees Factorisation Probabilistic inference Cano, Andrés Gómez Olmedo, Manuel Pérez-Ariza, Cora B. Salmerón Cerdán, Antonio Fast factorisation of probabilistic potentials and its application to approximate inference in Bayesian networks |
| title | Fast factorisation of probabilistic potentials and its application to approximate inference in Bayesian networks |
| title_full | Fast factorisation of probabilistic potentials and its application to approximate inference in Bayesian networks |
| title_fullStr | Fast factorisation of probabilistic potentials and its application to approximate inference in Bayesian networks |
| title_full_unstemmed | Fast factorisation of probabilistic potentials and its application to approximate inference in Bayesian networks |
| title_short | Fast factorisation of probabilistic potentials and its application to approximate inference in Bayesian networks |
| title_sort | fast factorisation of probabilistic potentials and its application to approximate inference in bayesian networks |
| topic | Bayesian netwoorks Probability trees Factorisation Probabilistic inference |
| url | http://hdl.handle.net/10835/4885 https://doi.org/10.1142/S0218488512500110 |
| work_keys_str_mv | AT canoandres fastfactorisationofprobabilisticpotentialsanditsapplicationtoapproximateinferenceinbayesiannetworks AT gomezolmedomanuel fastfactorisationofprobabilisticpotentialsanditsapplicationtoapproximateinferenceinbayesiannetworks AT perezarizacorab fastfactorisationofprobabilisticpotentialsanditsapplicationtoapproximateinferenceinbayesiannetworks AT salmeroncerdanantonio fastfactorisationofprobabilisticpotentialsanditsapplicationtoapproximateinferenceinbayesiannetworks |