Matches in SemOpenAlex for { <https://semopenalex.org/work/W2979661719> ?p ?o ?g. }
Showing items 1 to 83 of
83
with 100 items per page.
- W2979661719 abstract "Bayesian inference is known to provide a general framework for incorporating prior knowledge or specific properties into machine learning models via carefully choosing a prior distribution. In this work, we propose a new type of prior distributions for convolutional neural networks, deep weight prior (DWP), that exploit generative models to encourage a specific structure of trained convolutional filters e.g., spatial correlations of weights. We define DWP in the form of an implicit distribution and propose a method for variational inference with such type of implicit priors. In experiments, we show that DWP improves the performance of Bayesian neural networks when training data are limited, and initialization of weights with samples from DWP accelerates training of conventional convolutional neural networks." @default.
- W2979661719 created "2019-10-18" @default.
- W2979661719 creator A5008914065 @default.
- W2979661719 creator A5028940329 @default.
- W2979661719 creator A5047487458 @default.
- W2979661719 creator A5050892798 @default.
- W2979661719 creator A5087368991 @default.
- W2979661719 date "2018-10-16" @default.
- W2979661719 modified "2023-09-26" @default.
- W2979661719 title "The Deep Weight Prior - a Prior Distribution for CNNs via Generative Modeling of Parameters of the Model" @default.
- W2979661719 hasPublicationYear "2018" @default.
- W2979661719 type Work @default.
- W2979661719 sameAs 2979661719 @default.
- W2979661719 citedByCount "0" @default.
- W2979661719 crossrefType "posted-content" @default.
- W2979661719 hasAuthorship W2979661719A5008914065 @default.
- W2979661719 hasAuthorship W2979661719A5028940329 @default.
- W2979661719 hasAuthorship W2979661719A5047487458 @default.
- W2979661719 hasAuthorship W2979661719A5050892798 @default.
- W2979661719 hasAuthorship W2979661719A5087368991 @default.
- W2979661719 hasConcept C107673813 @default.
- W2979661719 hasConcept C110121322 @default.
- W2979661719 hasConcept C11413529 @default.
- W2979661719 hasConcept C114466953 @default.
- W2979661719 hasConcept C119857082 @default.
- W2979661719 hasConcept C134306372 @default.
- W2979661719 hasConcept C153180895 @default.
- W2979661719 hasConcept C154945302 @default.
- W2979661719 hasConcept C160234255 @default.
- W2979661719 hasConcept C167966045 @default.
- W2979661719 hasConcept C177769412 @default.
- W2979661719 hasConcept C199360897 @default.
- W2979661719 hasConcept C2776214188 @default.
- W2979661719 hasConcept C33923547 @default.
- W2979661719 hasConcept C39890363 @default.
- W2979661719 hasConcept C41008148 @default.
- W2979661719 hasConcept C81363708 @default.
- W2979661719 hasConcept C8642999 @default.
- W2979661719 hasConceptScore W2979661719C107673813 @default.
- W2979661719 hasConceptScore W2979661719C110121322 @default.
- W2979661719 hasConceptScore W2979661719C11413529 @default.
- W2979661719 hasConceptScore W2979661719C114466953 @default.
- W2979661719 hasConceptScore W2979661719C119857082 @default.
- W2979661719 hasConceptScore W2979661719C134306372 @default.
- W2979661719 hasConceptScore W2979661719C153180895 @default.
- W2979661719 hasConceptScore W2979661719C154945302 @default.
- W2979661719 hasConceptScore W2979661719C160234255 @default.
- W2979661719 hasConceptScore W2979661719C167966045 @default.
- W2979661719 hasConceptScore W2979661719C177769412 @default.
- W2979661719 hasConceptScore W2979661719C199360897 @default.
- W2979661719 hasConceptScore W2979661719C2776214188 @default.
- W2979661719 hasConceptScore W2979661719C33923547 @default.
- W2979661719 hasConceptScore W2979661719C39890363 @default.
- W2979661719 hasConceptScore W2979661719C41008148 @default.
- W2979661719 hasConceptScore W2979661719C81363708 @default.
- W2979661719 hasConceptScore W2979661719C8642999 @default.
- W2979661719 hasLocation W29796617191 @default.
- W2979661719 hasOpenAccess W2979661719 @default.
- W2979661719 hasPrimaryLocation W29796617191 @default.
- W2979661719 hasRelatedWork W2419501139 @default.
- W2979661719 hasRelatedWork W2803164693 @default.
- W2979661719 hasRelatedWork W2808181858 @default.
- W2979661719 hasRelatedWork W2910171846 @default.
- W2979661719 hasRelatedWork W2950263823 @default.
- W2979661719 hasRelatedWork W2962897886 @default.
- W2979661719 hasRelatedWork W2963859455 @default.
- W2979661719 hasRelatedWork W2964073649 @default.
- W2979661719 hasRelatedWork W2981411377 @default.
- W2979661719 hasRelatedWork W2981967650 @default.
- W2979661719 hasRelatedWork W2995393114 @default.
- W2979661719 hasRelatedWork W3005727684 @default.
- W2979661719 hasRelatedWork W3045583460 @default.
- W2979661719 hasRelatedWork W3093463114 @default.
- W2979661719 hasRelatedWork W3093515944 @default.
- W2979661719 hasRelatedWork W3163692982 @default.
- W2979661719 hasRelatedWork W3179857804 @default.
- W2979661719 hasRelatedWork W3192802934 @default.
- W2979661719 hasRelatedWork W3202239698 @default.
- W2979661719 hasRelatedWork W601603264 @default.
- W2979661719 isParatext "false" @default.
- W2979661719 isRetracted "false" @default.
- W2979661719 magId "2979661719" @default.
- W2979661719 workType "article" @default.