Graphical model selection for Gaussian conditional random fields in the presence of latent variables
We consider the problem of learning a conditional Gaussian graphical model in the presence of latent variables. Building on recent advances in this field, we suggest a method that decomposes the parameters of a conditional Markov random field into the sum of a sparse and a low-rank matrix. We derive...
| Main Authors: | , , |
|---|---|
| Format: | Journal article |
| Published: |
Taylor and Francis
2018
|
| _version_ | 1826272699940339712 |
|---|---|
| author | Frot, B Jostins, L McVean, G |
| author_facet | Frot, B Jostins, L McVean, G |
| author_sort | Frot, B |
| collection | OXFORD |
| description | We consider the problem of learning a conditional Gaussian graphical model in the presence of latent variables. Building on recent advances in this field, we suggest a method that decomposes the parameters of a conditional Markov random field into the sum of a sparse and a low-rank matrix. We derive convergence bounds for this estimator and show that it is well-behaved in the high-dimensional regime as well as “sparsistent” (i.e., capable of recovering the graph structure). We then show how proximal gradient algorithms and semi-definite programming techniques can be employed to fit the model to thousands of variables. Through extensive simulations, we illustrate the conditions required for identifiability and show that there is a wide range of situations in which this model performs significantly better than its counterparts, for example, by accommodating more latent variables. Finally, the suggested method is applied to two datasets comprising individual level data on genetic variants and metabolites levels. We show our results replicate better than alternative approaches and show enriched biological signal. Supplementary materials for this article are available online. |
| first_indexed | 2024-03-06T22:16:43Z |
| format | Journal article |
| id | oxford-uuid:539e3785-8309-43a5-a360-cc73b04b9907 |
| institution | University of Oxford |
| last_indexed | 2024-03-06T22:16:43Z |
| publishDate | 2018 |
| publisher | Taylor and Francis |
| record_format | dspace |
| spelling | oxford-uuid:539e3785-8309-43a5-a360-cc73b04b99072022-03-26T16:32:52ZGraphical model selection for Gaussian conditional random fields in the presence of latent variablesJournal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:539e3785-8309-43a5-a360-cc73b04b9907Symplectic Elements at OxfordTaylor and Francis2018Frot, BJostins, LMcVean, GWe consider the problem of learning a conditional Gaussian graphical model in the presence of latent variables. Building on recent advances in this field, we suggest a method that decomposes the parameters of a conditional Markov random field into the sum of a sparse and a low-rank matrix. We derive convergence bounds for this estimator and show that it is well-behaved in the high-dimensional regime as well as “sparsistent” (i.e., capable of recovering the graph structure). We then show how proximal gradient algorithms and semi-definite programming techniques can be employed to fit the model to thousands of variables. Through extensive simulations, we illustrate the conditions required for identifiability and show that there is a wide range of situations in which this model performs significantly better than its counterparts, for example, by accommodating more latent variables. Finally, the suggested method is applied to two datasets comprising individual level data on genetic variants and metabolites levels. We show our results replicate better than alternative approaches and show enriched biological signal. Supplementary materials for this article are available online. |
| spellingShingle | Frot, B Jostins, L McVean, G Graphical model selection for Gaussian conditional random fields in the presence of latent variables |
| title | Graphical model selection for Gaussian conditional random fields in the presence of latent variables |
| title_full | Graphical model selection for Gaussian conditional random fields in the presence of latent variables |
| title_fullStr | Graphical model selection for Gaussian conditional random fields in the presence of latent variables |
| title_full_unstemmed | Graphical model selection for Gaussian conditional random fields in the presence of latent variables |
| title_short | Graphical model selection for Gaussian conditional random fields in the presence of latent variables |
| title_sort | graphical model selection for gaussian conditional random fields in the presence of latent variables |
| work_keys_str_mv | AT frotb graphicalmodelselectionforgaussianconditionalrandomfieldsinthepresenceoflatentvariables AT jostinsl graphicalmodelselectionforgaussianconditionalrandomfieldsinthepresenceoflatentvariables AT mcveang graphicalmodelselectionforgaussianconditionalrandomfieldsinthepresenceoflatentvariables |