Matches in SemOpenAlex for { <https://semopenalex.org/work/W3100162595> ?p ?o ?g. }
Showing items 1 to 78 of
78
with 100 items per page.
- W3100162595 endingPage "379" @default.
- W3100162595 startingPage "366" @default.
- W3100162595 abstract "Many approaches to training generative models by distinct training objectives have been proposed in the past. Variational Autoencoder (VAE) is an outstanding model of them based on log-likelihood. In this paper, we propose a novel learnable prior, Pull-back Prior, for VAEs by adjusting the density of the prior through a discriminator that can assess the quality of data. It involves the discriminator from the theory of GANs to enrich the prior in VAEs. Based on it, we propose a more general framework, VAE with a Pull-back Prior (VAEPP), which uses existing techniques of VAEs and WGANs, to improve the log-likelihood, quality of sampling and stability of training. In MNIST and CIFAR-10, the log-likelihood of VAEPP outperforms models without autoregressive components and is comparable to autoregressive models. In MNIST, Fashion-MNIST, CIFAR-10 and CelebA, the FID of VAEPP is comparable to GANs and SOTA of VAEs." @default.
- W3100162595 created "2020-11-23" @default.
- W3100162595 creator A5030501342 @default.
- W3100162595 creator A5046419834 @default.
- W3100162595 creator A5075634052 @default.
- W3100162595 creator A5079058939 @default.
- W3100162595 creator A5089304305 @default.
- W3100162595 date "2020-01-01" @default.
- W3100162595 modified "2023-10-13" @default.
- W3100162595 title "VAEPP: Variational Autoencoder with a Pull-Back Prior" @default.
- W3100162595 cites W1834627138 @default.
- W3100162595 cites W2194321275 @default.
- W3100162595 cites W2802061317 @default.
- W3100162595 cites W2949649223 @default.
- W3100162595 doi "https://doi.org/10.1007/978-3-030-63836-8_31" @default.
- W3100162595 hasPublicationYear "2020" @default.
- W3100162595 type Work @default.
- W3100162595 sameAs 3100162595 @default.
- W3100162595 citedByCount "3" @default.
- W3100162595 countsByYear W31001625952021 @default.
- W3100162595 countsByYear W31001625952022 @default.
- W3100162595 crossrefType "book-chapter" @default.
- W3100162595 hasAuthorship W3100162595A5030501342 @default.
- W3100162595 hasAuthorship W3100162595A5046419834 @default.
- W3100162595 hasAuthorship W3100162595A5075634052 @default.
- W3100162595 hasAuthorship W3100162595A5079058939 @default.
- W3100162595 hasAuthorship W3100162595A5089304305 @default.
- W3100162595 hasConcept C101738243 @default.
- W3100162595 hasConcept C105795698 @default.
- W3100162595 hasConcept C108583219 @default.
- W3100162595 hasConcept C112972136 @default.
- W3100162595 hasConcept C119857082 @default.
- W3100162595 hasConcept C153180895 @default.
- W3100162595 hasConcept C154945302 @default.
- W3100162595 hasConcept C159877910 @default.
- W3100162595 hasConcept C167966045 @default.
- W3100162595 hasConcept C190502265 @default.
- W3100162595 hasConcept C2779803651 @default.
- W3100162595 hasConcept C33923547 @default.
- W3100162595 hasConcept C39890363 @default.
- W3100162595 hasConcept C41008148 @default.
- W3100162595 hasConcept C76155785 @default.
- W3100162595 hasConcept C94915269 @default.
- W3100162595 hasConceptScore W3100162595C101738243 @default.
- W3100162595 hasConceptScore W3100162595C105795698 @default.
- W3100162595 hasConceptScore W3100162595C108583219 @default.
- W3100162595 hasConceptScore W3100162595C112972136 @default.
- W3100162595 hasConceptScore W3100162595C119857082 @default.
- W3100162595 hasConceptScore W3100162595C153180895 @default.
- W3100162595 hasConceptScore W3100162595C154945302 @default.
- W3100162595 hasConceptScore W3100162595C159877910 @default.
- W3100162595 hasConceptScore W3100162595C167966045 @default.
- W3100162595 hasConceptScore W3100162595C190502265 @default.
- W3100162595 hasConceptScore W3100162595C2779803651 @default.
- W3100162595 hasConceptScore W3100162595C33923547 @default.
- W3100162595 hasConceptScore W3100162595C39890363 @default.
- W3100162595 hasConceptScore W3100162595C41008148 @default.
- W3100162595 hasConceptScore W3100162595C76155785 @default.
- W3100162595 hasConceptScore W3100162595C94915269 @default.
- W3100162595 hasLocation W31001625951 @default.
- W3100162595 hasOpenAccess W3100162595 @default.
- W3100162595 hasPrimaryLocation W31001625951 @default.
- W3100162595 hasRelatedWork W2002563186 @default.
- W3100162595 hasRelatedWork W2912987408 @default.
- W3100162595 hasRelatedWork W2937381246 @default.
- W3100162595 hasRelatedWork W2980541498 @default.
- W3100162595 hasRelatedWork W3004801820 @default.
- W3100162595 hasRelatedWork W4230582276 @default.
- W3100162595 hasRelatedWork W4281672036 @default.
- W3100162595 hasRelatedWork W4296978181 @default.
- W3100162595 hasRelatedWork W4310699748 @default.
- W3100162595 hasRelatedWork W4313444753 @default.
- W3100162595 isParatext "false" @default.
- W3100162595 isRetracted "false" @default.
- W3100162595 magId "3100162595" @default.
- W3100162595 workType "book-chapter" @default.