Approximate Probability Propagation with Mixtures of Truncated Exponentials*
Mixtures of truncated exponentials (MTEs) are a powerful alternative to discretisation when working with hybrid Bayesian networks. One of the features of the MTE model is that standard propagation algorithms can be used. However, the complexity of the process is too high and therefore approximate me...
| Main Authors: | , |
|---|---|
| Format: | info:eu-repo/semantics/article |
| Language: | English |
| Published: |
2017
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/10835/4890 https://doi.org/10.1016/j.ijar.2006.06.007 |
| _version_ | 1789406540211421184 |
|---|---|
| author | Rumí, Rafael Salmerón Cerdán, Antonio |
| author_facet | Rumí, Rafael Salmerón Cerdán, Antonio |
| author_sort | Rumí, Rafael |
| collection | DSpace |
| description | Mixtures of truncated exponentials (MTEs) are a powerful alternative to discretisation when working with hybrid Bayesian networks. One of the features of the MTE model is that standard propagation algorithms can be used. However, the complexity of the process is too high and therefore approximate methods, which tradeoff complexity for accuracy, become necessary. In this paper we propose an approximate propagation algorithm for MTE networks which is based on the Penniless propagation method already known for discrete variables. We also consider how to use Markov Chain Monte Carlo to carry out the probability propagation. The performance of the proposed methods is analysed in a series of experiments with random networks. |
| format | info:eu-repo/semantics/article |
| id | oai:repositorio.ual.es:10835-4890 |
| institution | Universidad de Cuenca |
| language | English |
| publishDate | 2017 |
| record_format | dspace |
| spelling | oai:repositorio.ual.es:10835-48902023-04-12T19:38:19Z Approximate Probability Propagation with Mixtures of Truncated Exponentials* Rumí, Rafael Salmerón Cerdán, Antonio Hybrid Bayesian networks Mixtures of truncated exponentials Continuous variables Probability propagation Penniless propagation MCMC Mixtures of truncated exponentials (MTEs) are a powerful alternative to discretisation when working with hybrid Bayesian networks. One of the features of the MTE model is that standard propagation algorithms can be used. However, the complexity of the process is too high and therefore approximate methods, which tradeoff complexity for accuracy, become necessary. In this paper we propose an approximate propagation algorithm for MTE networks which is based on the Penniless propagation method already known for discrete variables. We also consider how to use Markov Chain Monte Carlo to carry out the probability propagation. The performance of the proposed methods is analysed in a series of experiments with random networks. 2017-07-07T07:16:29Z 2017-07-07T07:16:29Z 2007 info:eu-repo/semantics/article http://hdl.handle.net/10835/4890 https://doi.org/10.1016/j.ijar.2006.06.007 en Attribution-NonCommercial-NoDerivatives 4.0 Internacional http://creativecommons.org/licenses/by-nc-nd/4.0/ info:eu-repo/semantics/openAccess |
| spellingShingle | Hybrid Bayesian networks Mixtures of truncated exponentials Continuous variables Probability propagation Penniless propagation MCMC Rumí, Rafael Salmerón Cerdán, Antonio Approximate Probability Propagation with Mixtures of Truncated Exponentials* |
| title | Approximate Probability Propagation with Mixtures of Truncated Exponentials* |
| title_full | Approximate Probability Propagation with Mixtures of Truncated Exponentials* |
| title_fullStr | Approximate Probability Propagation with Mixtures of Truncated Exponentials* |
| title_full_unstemmed | Approximate Probability Propagation with Mixtures of Truncated Exponentials* |
| title_short | Approximate Probability Propagation with Mixtures of Truncated Exponentials* |
| title_sort | approximate probability propagation with mixtures of truncated exponentials* |
| topic | Hybrid Bayesian networks Mixtures of truncated exponentials Continuous variables Probability propagation Penniless propagation MCMC |
| url | http://hdl.handle.net/10835/4890 https://doi.org/10.1016/j.ijar.2006.06.007 |
| work_keys_str_mv | AT rumirafael approximateprobabilitypropagationwithmixturesoftruncatedexponentials AT salmeroncerdanantonio approximateprobabilitypropagationwithmixturesoftruncatedexponentials |