Stochastic backpropagation and approximate inference in deep generative models. Rezende, D., J., Mohamed, S., & Wierstra, D. 31st International Conference on Machine Learning, ICML 2014, 4:3057-3070, 2014.
![Stochastic backpropagation and approximate inference in deep generative models [pdf]](https://bibbase.org/img/filetypes/pdf.svg "pdf")
Paper abstract bibtex
Paper abstract bibtex We marry ideas from deep neural networks and approximate Bayesian inference to derive a gen-eralised class of deep, directed generative models, endowed with a new algorithm for scalable inference and learning. Our algorithm introduces a recognition model to represent an approximate posterior distribution and uses this for optimisation of a variational lower bound. We develop stochastic backpropagation - rules for gradient backpropagation through stochastic variables - And derive an algorithm that allows for joint optimisation of the parameters of both the generative and recognition models. We demonstrate on several real-world data sets that by using stochastic backpropagation and variational inference, we obtain models that are able to generate realistic samples of data, allow for accurate imputations of missing data, and provide a useful tool for high-dimensional data visualisation.
@article{
title = {Stochastic backpropagation and approximate inference in deep generative models},
type = {article},
year = {2014},
pages = {3057-3070},
volume = {4},
id = {dc6ac099-8b9f-3f71-a5fd-6ceb0f5b34f5},
created = {2021-09-09T14:35:21.128Z},
file_attached = {true},
profile_id = {235249c2-3ed4-314a-b309-b1ea0330f5d9},
group_id = {1ff583c0-be37-34fa-9c04-73c69437d354},
last_modified = {2022-03-28T09:45:09.539Z},
read = {false},
starred = {false},
authored = {false},
confirmed = {true},
hidden = {false},
citation_key = {Rezende2014},
folder_uuids = {1853f94b-7af1-40fa-b068-4758e9a02bc4},
private_publication = {false},
abstract = {We marry ideas from deep neural networks and approximate Bayesian inference to derive a gen-eralised class of deep, directed generative models, endowed with a new algorithm for scalable inference and learning. Our algorithm introduces a recognition model to represent an approximate posterior distribution and uses this for optimisation of a variational lower bound. We develop stochastic backpropagation - rules for gradient backpropagation through stochastic variables - And derive an algorithm that allows for joint optimisation of the parameters of both the generative and recognition models. We demonstrate on several real-world data sets that by using stochastic backpropagation and variational inference, we obtain models that are able to generate realistic samples of data, allow for accurate imputations of missing data, and provide a useful tool for high-dimensional data visualisation.},
bibtype = {article},
author = {Rezende, Danilo Jimenez and Mohamed, Shakir and Wierstra, Daan},
journal = {31st International Conference on Machine Learning, ICML 2014}
}Downloads: 0
{"_id":"xS26Qsr3NKcsXZPKy","bibbaseid":"rezende-mohamed-wierstra-stochasticbackpropagationandapproximateinferenceindeepgenerativemodels-2014","downloads":0,"creationDate":"2017-03-16T18:57:17.604Z","title":"Stochastic backpropagation and approximate inference in deep generative models","author_short":["Rezende, D., J.","Mohamed, S.","Wierstra, D."],"year":2014,"bibtype":"article","biburl":"https://bibbase.org/service/mendeley/bfbbf840-4c42-3914-a463-19024f50b30c","bibdata":{"title":"Stochastic backpropagation and approximate inference in deep generative models","type":"article","year":"2014","pages":"3057-3070","volume":"4","id":"dc6ac099-8b9f-3f71-a5fd-6ceb0f5b34f5","created":"2021-09-09T14:35:21.128Z","file_attached":"true","profile_id":"235249c2-3ed4-314a-b309-b1ea0330f5d9","group_id":"1ff583c0-be37-34fa-9c04-73c69437d354","last_modified":"2022-03-28T09:45:09.539Z","read":false,"starred":false,"authored":false,"confirmed":"true","hidden":false,"citation_key":"Rezende2014","folder_uuids":"1853f94b-7af1-40fa-b068-4758e9a02bc4","private_publication":false,"abstract":"We marry ideas from deep neural networks and approximate Bayesian inference to derive a gen-eralised class of deep, directed generative models, endowed with a new algorithm for scalable inference and learning. Our algorithm introduces a recognition model to represent an approximate posterior distribution and uses this for optimisation of a variational lower bound. We develop stochastic backpropagation - rules for gradient backpropagation through stochastic variables - And derive an algorithm that allows for joint optimisation of the parameters of both the generative and recognition models. We demonstrate on several real-world data sets that by using stochastic backpropagation and variational inference, we obtain models that are able to generate realistic samples of data, allow for accurate imputations of missing data, and provide a useful tool for high-dimensional data visualisation.","bibtype":"article","author":"Rezende, Danilo Jimenez and Mohamed, Shakir and Wierstra, Daan","journal":"31st International Conference on Machine Learning, ICML 2014","bibtex":"@article{\n title = {Stochastic backpropagation and approximate inference in deep generative models},\n type = {article},\n year = {2014},\n pages = {3057-3070},\n volume = {4},\n id = {dc6ac099-8b9f-3f71-a5fd-6ceb0f5b34f5},\n created = {2021-09-09T14:35:21.128Z},\n file_attached = {true},\n profile_id = {235249c2-3ed4-314a-b309-b1ea0330f5d9},\n group_id = {1ff583c0-be37-34fa-9c04-73c69437d354},\n last_modified = {2022-03-28T09:45:09.539Z},\n read = {false},\n starred = {false},\n authored = {false},\n confirmed = {true},\n hidden = {false},\n citation_key = {Rezende2014},\n folder_uuids = {1853f94b-7af1-40fa-b068-4758e9a02bc4},\n private_publication = {false},\n abstract = {We marry ideas from deep neural networks and approximate Bayesian inference to derive a gen-eralised class of deep, directed generative models, endowed with a new algorithm for scalable inference and learning. Our algorithm introduces a recognition model to represent an approximate posterior distribution and uses this for optimisation of a variational lower bound. We develop stochastic backpropagation - rules for gradient backpropagation through stochastic variables - And derive an algorithm that allows for joint optimisation of the parameters of both the generative and recognition models. We demonstrate on several real-world data sets that by using stochastic backpropagation and variational inference, we obtain models that are able to generate realistic samples of data, allow for accurate imputations of missing data, and provide a useful tool for high-dimensional data visualisation.},\n bibtype = {article},\n author = {Rezende, Danilo Jimenez and Mohamed, Shakir and Wierstra, Daan},\n journal = {31st International Conference on Machine Learning, ICML 2014}\n}","author_short":["Rezende, D., J.","Mohamed, S.","Wierstra, D."],"urls":{"Paper":"https://bibbase.org/service/mendeley/bfbbf840-4c42-3914-a463-19024f50b30c/file/7a8002bc-1bb2-6aa5-1a4c-3d8202586396/14014082.pdf.pdf"},"biburl":"https://bibbase.org/service/mendeley/bfbbf840-4c42-3914-a463-19024f50b30c","bibbaseid":"rezende-mohamed-wierstra-stochasticbackpropagationandapproximateinferenceindeepgenerativemodels-2014","role":"author","metadata":{"authorlinks":{}},"downloads":0},"search_terms":["stochastic","backpropagation","approximate","inference","deep","generative","models","rezende","mohamed","wierstra"],"keywords":[],"authorIDs":[],"dataSources":["5NxYfDAcNJEC35W7g","ya2CyA73rpZseyrZ8","2252seNhipfTmjEBQ"]}