Learning an L1-regularized Gaussian Bayesian network in the equivalence class space.
Learning the structure of a graphical model from data is a common task in a wide range of practical applications. In this paper, we focus on Gaussian Bayesian networks, i.e., on continuous data and directed acyclic graphs with a joint probability density of all variables given by a Gaussian. We prop...
| Main Authors: | , , |
|---|---|
| Format: | Journal article |
| Language: | English |
| Published: |
2010
|
| _version_ | 1826256815220850688 |
|---|---|
| author | Vidaurre, D Bielza, C Larrañaga, P |
| author_facet | Vidaurre, D Bielza, C Larrañaga, P |
| author_sort | Vidaurre, D |
| collection | OXFORD |
| description | Learning the structure of a graphical model from data is a common task in a wide range of practical applications. In this paper, we focus on Gaussian Bayesian networks, i.e., on continuous data and directed acyclic graphs with a joint probability density of all variables given by a Gaussian. We propose to work in an equivalence class search space, specifically using the k-greedy equivalence search algorithm. This, combined with regularization techniques to guide the structure search, can learn sparse networks close to the one that generated the data. We provide results on some synthetic networks and on modeling the gene network of the two biological pathways regulating the biosynthesis of isoprenoids for the Arabidopsis thaliana plant. |
| first_indexed | 2024-03-06T18:08:15Z |
| format | Journal article |
| id | oxford-uuid:0226d912-3bcd-4bb6-949a-d3f240b16dd8 |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-03-06T18:08:15Z |
| publishDate | 2010 |
| record_format | dspace |
| spelling | oxford-uuid:0226d912-3bcd-4bb6-949a-d3f240b16dd82022-03-26T08:38:59ZLearning an L1-regularized Gaussian Bayesian network in the equivalence class space.Journal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:0226d912-3bcd-4bb6-949a-d3f240b16dd8EnglishSymplectic Elements at Oxford2010Vidaurre, DBielza, CLarrañaga, PLearning the structure of a graphical model from data is a common task in a wide range of practical applications. In this paper, we focus on Gaussian Bayesian networks, i.e., on continuous data and directed acyclic graphs with a joint probability density of all variables given by a Gaussian. We propose to work in an equivalence class search space, specifically using the k-greedy equivalence search algorithm. This, combined with regularization techniques to guide the structure search, can learn sparse networks close to the one that generated the data. We provide results on some synthetic networks and on modeling the gene network of the two biological pathways regulating the biosynthesis of isoprenoids for the Arabidopsis thaliana plant. |
| spellingShingle | Vidaurre, D Bielza, C Larrañaga, P Learning an L1-regularized Gaussian Bayesian network in the equivalence class space. |
| title | Learning an L1-regularized Gaussian Bayesian network in the equivalence class space. |
| title_full | Learning an L1-regularized Gaussian Bayesian network in the equivalence class space. |
| title_fullStr | Learning an L1-regularized Gaussian Bayesian network in the equivalence class space. |
| title_full_unstemmed | Learning an L1-regularized Gaussian Bayesian network in the equivalence class space. |
| title_short | Learning an L1-regularized Gaussian Bayesian network in the equivalence class space. |
| title_sort | learning an l1 regularized gaussian bayesian network in the equivalence class space |
| work_keys_str_mv | AT vidaurred learninganl1regularizedgaussianbayesiannetworkintheequivalenceclassspace AT bielzac learninganl1regularizedgaussianbayesiannetworkintheequivalenceclassspace AT larranagap learninganl1regularizedgaussianbayesiannetworkintheequivalenceclassspace |