Matches in SemOpenAlex for { <https://semopenalex.org/work/W3048772196> ?p ?o ?g. }
Showing items 1 to 86 of
86
with 100 items per page.
- W3048772196 endingPage "888" @default.
- W3048772196 startingPage "888" @default.
- W3048772196 abstract "One of the major shortcomings of variational autoencoders is the inability to produce generations from the individual modalities of data originating from mixture distributions. This is primarily due to the use of a simple isotropic Gaussian as the prior for the latent code in the ancestral sampling procedure for data generations. In this paper, we propose a novel formulation of variational autoencoders, conditional prior VAE (CP-VAE), with a two-level generative process for the observed data where continuous z and a discrete c variables are introduced in addition to the observed variables x. By learning data-dependent conditional priors, the new variational objective naturally encourages a better match between the posterior and prior conditionals, and the learning of the latent categories encoding the major source of variation of the original data in an unsupervised manner. Through sampling continuous latent code from the data-dependent conditional priors, we are able to generate new samples from the individual mixture components corresponding, to the multimodal structure over the original data. Moreover, we unify and analyse our objective under different independence assumptions for the joint distribution of the continuous and discrete latent variables. We provide an empirical evaluation on one synthetic dataset and three image datasets, FashionMNIST, MNIST, and Omniglot, illustrating the generative performance of our new model comparing to multiple baselines." @default.
- W3048772196 created "2020-08-18" @default.
- W3048772196 creator A5044131928 @default.
- W3048772196 creator A5054750463 @default.
- W3048772196 creator A5084831490 @default.
- W3048772196 date "2020-08-13" @default.
- W3048772196 modified "2023-10-17" @default.
- W3048772196 title "Data-Dependent Conditional Priors for Unsupervised Learning of Multimodal Data" @default.
- W3048772196 cites W2112796928 @default.
- W3048772196 cites W2194321275 @default.
- W3048772196 cites W2617118670 @default.
- W3048772196 cites W2963223306 @default.
- W3048772196 doi "https://doi.org/10.3390/e22080888" @default.
- W3048772196 hasPubMedCentralId "https://www.ncbi.nlm.nih.gov/pmc/articles/7517502" @default.
- W3048772196 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/33286658" @default.
- W3048772196 hasPublicationYear "2020" @default.
- W3048772196 type Work @default.
- W3048772196 sameAs 3048772196 @default.
- W3048772196 citedByCount "2" @default.
- W3048772196 countsByYear W30487721962021 @default.
- W3048772196 crossrefType "journal-article" @default.
- W3048772196 hasAuthorship W3048772196A5044131928 @default.
- W3048772196 hasAuthorship W3048772196A5054750463 @default.
- W3048772196 hasAuthorship W3048772196A5084831490 @default.
- W3048772196 hasBestOaLocation W30487721961 @default.
- W3048772196 hasConcept C105795698 @default.
- W3048772196 hasConcept C107673813 @default.
- W3048772196 hasConcept C108583219 @default.
- W3048772196 hasConcept C119857082 @default.
- W3048772196 hasConcept C153180895 @default.
- W3048772196 hasConcept C154945302 @default.
- W3048772196 hasConcept C160920958 @default.
- W3048772196 hasConcept C167966045 @default.
- W3048772196 hasConcept C177769412 @default.
- W3048772196 hasConcept C18653775 @default.
- W3048772196 hasConcept C190502265 @default.
- W3048772196 hasConcept C33923547 @default.
- W3048772196 hasConcept C39890363 @default.
- W3048772196 hasConcept C41008148 @default.
- W3048772196 hasConcept C43555835 @default.
- W3048772196 hasConcept C51167844 @default.
- W3048772196 hasConcept C79772020 @default.
- W3048772196 hasConcept C8038995 @default.
- W3048772196 hasConceptScore W3048772196C105795698 @default.
- W3048772196 hasConceptScore W3048772196C107673813 @default.
- W3048772196 hasConceptScore W3048772196C108583219 @default.
- W3048772196 hasConceptScore W3048772196C119857082 @default.
- W3048772196 hasConceptScore W3048772196C153180895 @default.
- W3048772196 hasConceptScore W3048772196C154945302 @default.
- W3048772196 hasConceptScore W3048772196C160920958 @default.
- W3048772196 hasConceptScore W3048772196C167966045 @default.
- W3048772196 hasConceptScore W3048772196C177769412 @default.
- W3048772196 hasConceptScore W3048772196C18653775 @default.
- W3048772196 hasConceptScore W3048772196C190502265 @default.
- W3048772196 hasConceptScore W3048772196C33923547 @default.
- W3048772196 hasConceptScore W3048772196C39890363 @default.
- W3048772196 hasConceptScore W3048772196C41008148 @default.
- W3048772196 hasConceptScore W3048772196C43555835 @default.
- W3048772196 hasConceptScore W3048772196C51167844 @default.
- W3048772196 hasConceptScore W3048772196C79772020 @default.
- W3048772196 hasConceptScore W3048772196C8038995 @default.
- W3048772196 hasIssue "8" @default.
- W3048772196 hasLocation W30487721961 @default.
- W3048772196 hasLocation W30487721962 @default.
- W3048772196 hasLocation W30487721963 @default.
- W3048772196 hasLocation W30487721964 @default.
- W3048772196 hasLocation W30487721965 @default.
- W3048772196 hasOpenAccess W3048772196 @default.
- W3048772196 hasPrimaryLocation W30487721961 @default.
- W3048772196 hasRelatedWork W2786010873 @default.
- W3048772196 hasRelatedWork W2884772108 @default.
- W3048772196 hasRelatedWork W2963529210 @default.
- W3048772196 hasRelatedWork W2963987720 @default.
- W3048772196 hasRelatedWork W3048772196 @default.
- W3048772196 hasRelatedWork W3093172222 @default.
- W3048772196 hasRelatedWork W3137809915 @default.
- W3048772196 hasRelatedWork W3205589360 @default.
- W3048772196 hasRelatedWork W4286907481 @default.
- W3048772196 hasRelatedWork W4293659211 @default.
- W3048772196 hasVolume "22" @default.
- W3048772196 isParatext "false" @default.
- W3048772196 isRetracted "false" @default.
- W3048772196 magId "3048772196" @default.
- W3048772196 workType "article" @default.