Matches in SemOpenAlex for { <https://semopenalex.org/work/W2949649223> ?p ?o ?g. }
Showing items 1 to 98 of
98
with 100 items per page.
- W2949649223 endingPage "5073" @default.
- W2949649223 startingPage "5066" @default.
- W2949649223 abstract "The variational autoencoder (VAE) is a powerful generative model that can estimate the probability of a data point by using latent variables. In the VAE, the posterior of the latent variable given the data point is regularized by the prior of the latent variable using Kullback Leibler (KL) divergence. Although the standard Gaussian distribution is usually used for the prior, this simple prior incurs over-regularization. As a sophisticated prior, the aggregated posterior has been introduced, which is the expectation of the posterior over the data distribution. This prior is optimal for the VAE in terms of maximizing the training objective function. However, KL divergence with the aggregated posterior cannot be calculated in a closed form, which prevents us from using this optimal prior. With the proposed method, we introduce the density ratio trick to estimate this KL divergence without modeling the aggregated posterior explicitly. Since the density ratio trick does not work well in high dimensions, we rewrite this KL divergence that contains the high-dimensional density ratio into the sum of the analytically calculable term and the lowdimensional density ratio term, to which the density ratio trick is applied. Experiments on various datasets show that the VAE with this implicit optimal prior achieves high density estimation performance." @default.
- W2949649223 created "2019-06-27" @default.
- W2949649223 creator A5006786260 @default.
- W2949649223 creator A5034538103 @default.
- W2949649223 creator A5036403126 @default.
- W2949649223 creator A5037036505 @default.
- W2949649223 creator A5069398453 @default.
- W2949649223 date "2019-07-17" @default.
- W2949649223 modified "2023-10-06" @default.
- W2949649223 title "Variational Autoencoder with Implicit Optimal Priors" @default.
- W2949649223 doi "https://doi.org/10.1609/aaai.v33i01.33015066" @default.
- W2949649223 hasPublicationYear "2019" @default.
- W2949649223 type Work @default.
- W2949649223 sameAs 2949649223 @default.
- W2949649223 citedByCount "25" @default.
- W2949649223 countsByYear W29496492232020 @default.
- W2949649223 countsByYear W29496492232021 @default.
- W2949649223 countsByYear W29496492232022 @default.
- W2949649223 countsByYear W29496492232023 @default.
- W2949649223 crossrefType "journal-article" @default.
- W2949649223 hasAuthorship W2949649223A5006786260 @default.
- W2949649223 hasAuthorship W2949649223A5034538103 @default.
- W2949649223 hasAuthorship W2949649223A5036403126 @default.
- W2949649223 hasAuthorship W2949649223A5037036505 @default.
- W2949649223 hasAuthorship W2949649223A5069398453 @default.
- W2949649223 hasBestOaLocation W29496492231 @default.
- W2949649223 hasConcept C101738243 @default.
- W2949649223 hasConcept C105795698 @default.
- W2949649223 hasConcept C107673813 @default.
- W2949649223 hasConcept C11413529 @default.
- W2949649223 hasConcept C121332964 @default.
- W2949649223 hasConcept C126255220 @default.
- W2949649223 hasConcept C138885662 @default.
- W2949649223 hasConcept C154945302 @default.
- W2949649223 hasConcept C163716315 @default.
- W2949649223 hasConcept C171752962 @default.
- W2949649223 hasConcept C177769412 @default.
- W2949649223 hasConcept C185429906 @default.
- W2949649223 hasConcept C189508267 @default.
- W2949649223 hasConcept C197055811 @default.
- W2949649223 hasConcept C207390915 @default.
- W2949649223 hasConcept C2776135515 @default.
- W2949649223 hasConcept C28826006 @default.
- W2949649223 hasConcept C33923547 @default.
- W2949649223 hasConcept C41008148 @default.
- W2949649223 hasConcept C41895202 @default.
- W2949649223 hasConcept C50644808 @default.
- W2949649223 hasConcept C51167844 @default.
- W2949649223 hasConcept C57830394 @default.
- W2949649223 hasConcept C61224824 @default.
- W2949649223 hasConcept C62520636 @default.
- W2949649223 hasConceptScore W2949649223C101738243 @default.
- W2949649223 hasConceptScore W2949649223C105795698 @default.
- W2949649223 hasConceptScore W2949649223C107673813 @default.
- W2949649223 hasConceptScore W2949649223C11413529 @default.
- W2949649223 hasConceptScore W2949649223C121332964 @default.
- W2949649223 hasConceptScore W2949649223C126255220 @default.
- W2949649223 hasConceptScore W2949649223C138885662 @default.
- W2949649223 hasConceptScore W2949649223C154945302 @default.
- W2949649223 hasConceptScore W2949649223C163716315 @default.
- W2949649223 hasConceptScore W2949649223C171752962 @default.
- W2949649223 hasConceptScore W2949649223C177769412 @default.
- W2949649223 hasConceptScore W2949649223C185429906 @default.
- W2949649223 hasConceptScore W2949649223C189508267 @default.
- W2949649223 hasConceptScore W2949649223C197055811 @default.
- W2949649223 hasConceptScore W2949649223C207390915 @default.
- W2949649223 hasConceptScore W2949649223C2776135515 @default.
- W2949649223 hasConceptScore W2949649223C28826006 @default.
- W2949649223 hasConceptScore W2949649223C33923547 @default.
- W2949649223 hasConceptScore W2949649223C41008148 @default.
- W2949649223 hasConceptScore W2949649223C41895202 @default.
- W2949649223 hasConceptScore W2949649223C50644808 @default.
- W2949649223 hasConceptScore W2949649223C51167844 @default.
- W2949649223 hasConceptScore W2949649223C57830394 @default.
- W2949649223 hasConceptScore W2949649223C61224824 @default.
- W2949649223 hasConceptScore W2949649223C62520636 @default.
- W2949649223 hasIssue "01" @default.
- W2949649223 hasLocation W29496492231 @default.
- W2949649223 hasLocation W29496492232 @default.
- W2949649223 hasOpenAccess W2949649223 @default.
- W2949649223 hasPrimaryLocation W29496492231 @default.
- W2949649223 hasRelatedWork W1972722672 @default.
- W2949649223 hasRelatedWork W2033178790 @default.
- W2949649223 hasRelatedWork W2115685217 @default.
- W2949649223 hasRelatedWork W2891583400 @default.
- W2949649223 hasRelatedWork W2891850914 @default.
- W2949649223 hasRelatedWork W2949649223 @default.
- W2949649223 hasRelatedWork W3009854800 @default.
- W2949649223 hasRelatedWork W3036866888 @default.
- W2949649223 hasRelatedWork W3041775337 @default.
- W2949649223 hasRelatedWork W326566402 @default.
- W2949649223 hasVolume "33" @default.
- W2949649223 isParatext "false" @default.
- W2949649223 isRetracted "false" @default.
- W2949649223 magId "2949649223" @default.
- W2949649223 workType "article" @default.