Learning hybrid Bayesian networks using mixtures of truncated exponentials
In this paper we introduce an algorithm for learning hybrid Bayesian networks from data. The result of the algorithm is a network where the conditional distribution for each variable is a mixture of truncated exponentials (MTE), so that no restrictions on the network topology are imposed. The struct...
| Main Authors: | , , |
|---|---|
| Format: | info:eu-repo/semantics/article |
| Language: | English |
| Published: |
2017
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/10835/4898 https://doi.org/10.1016/j.ijar.2005.10.004 |
| _version_ | 1789406602806165504 |
|---|---|
| author | Romero, Vanessa Rumí, Rafael Salmerón Cerdán, Antonio |
| author_facet | Romero, Vanessa Rumí, Rafael Salmerón Cerdán, Antonio |
| author_sort | Romero, Vanessa |
| collection | DSpace |
| description | In this paper we introduce an algorithm for learning hybrid Bayesian networks from data. The result of the algorithm is a network where the conditional distribution for each variable is a mixture of truncated exponentials (MTE), so that no restrictions on the network topology are imposed. The structure of the network is obtained by searching over the space of candidate networks using optimisation methods. The conditional densities are estimated by means of Gaussian kernel densities that afterwards are approximated by MTEs, so that the resulting network is appropriate for using standard algorithms for probabilistic reasoning. The behaviour of the proposed algorithm is tested using a set of real-world and arti cially generated databases. |
| format | info:eu-repo/semantics/article |
| id | oai:repositorio.ual.es:10835-4898 |
| institution | Universidad de Cuenca |
| language | English |
| publishDate | 2017 |
| record_format | dspace |
| spelling | oai:repositorio.ual.es:10835-48982023-04-12T19:38:48Z Learning hybrid Bayesian networks using mixtures of truncated exponentials Romero, Vanessa Rumí, Rafael Salmerón Cerdán, Antonio Bayesian networks Mixtures of truncated exponentials Continuous variables Parameter learning Kernel Methods Simulated In this paper we introduce an algorithm for learning hybrid Bayesian networks from data. The result of the algorithm is a network where the conditional distribution for each variable is a mixture of truncated exponentials (MTE), so that no restrictions on the network topology are imposed. The structure of the network is obtained by searching over the space of candidate networks using optimisation methods. The conditional densities are estimated by means of Gaussian kernel densities that afterwards are approximated by MTEs, so that the resulting network is appropriate for using standard algorithms for probabilistic reasoning. The behaviour of the proposed algorithm is tested using a set of real-world and arti cially generated databases. 2017-07-07T07:18:03Z 2017-07-07T07:18:03Z 2006 info:eu-repo/semantics/article http://hdl.handle.net/10835/4898 https://doi.org/10.1016/j.ijar.2005.10.004 en Attribution-NonCommercial-NoDerivatives 4.0 Internacional http://creativecommons.org/licenses/by-nc-nd/4.0/ info:eu-repo/semantics/openAccess |
| spellingShingle | Bayesian networks Mixtures of truncated exponentials Continuous variables Parameter learning Kernel Methods Simulated Romero, Vanessa Rumí, Rafael Salmerón Cerdán, Antonio Learning hybrid Bayesian networks using mixtures of truncated exponentials |
| title | Learning hybrid Bayesian networks using mixtures of truncated exponentials |
| title_full | Learning hybrid Bayesian networks using mixtures of truncated exponentials |
| title_fullStr | Learning hybrid Bayesian networks using mixtures of truncated exponentials |
| title_full_unstemmed | Learning hybrid Bayesian networks using mixtures of truncated exponentials |
| title_short | Learning hybrid Bayesian networks using mixtures of truncated exponentials |
| title_sort | learning hybrid bayesian networks using mixtures of truncated exponentials |
| topic | Bayesian networks Mixtures of truncated exponentials Continuous variables Parameter learning Kernel Methods Simulated |
| url | http://hdl.handle.net/10835/4898 https://doi.org/10.1016/j.ijar.2005.10.004 |
| work_keys_str_mv | AT romerovanessa learninghybridbayesiannetworksusingmixturesoftruncatedexponentials AT rumirafael learninghybridbayesiannetworksusingmixturesoftruncatedexponentials AT salmeroncerdanantonio learninghybridbayesiannetworksusingmixturesoftruncatedexponentials |