Matches in SemOpenAlex for { <https://semopenalex.org/work/W3215977089> ?p ?o ?g. }
Showing items 1 to 92 of
92
with 100 items per page.
- W3215977089 abstract "An important propertyfor deep neural networks to possess is the ability to perform robust out of distribution detection (OOD) on previously unseen data. This property is essential for safety purposes when deploying models for real world applications. Recent studies show that probabilistic generative models can perform poorly on this task, which is surprising given that they seek to estimate the likelihood of training data. To alleviate this issue, we propose the exponentially tilted Gaussian prior distribution for the Variational Autoencoder (VAE). With this prior, we are able to achieve state-of-the art results using just the negative log likelihood that the VAE naturally assigns, while being orders of magnitude faster than some competitive methods. We also show that our model produces high quality image samples which are more crisp than that of a standard Gaussian VAE. The new prior distribution has a very simple implementation which uses a Kullback Leibler divergence that compares the difference between a latent vector's length, and the radius of a sphere." @default.
- W3215977089 created "2021-12-06" @default.
- W3215977089 creator A5049658956 @default.
- W3215977089 creator A5053492737 @default.
- W3215977089 creator A5063069144 @default.
- W3215977089 date "2021-11-30" @default.
- W3215977089 modified "2023-09-27" @default.
- W3215977089 title "Exponentially Tilted Gaussian Prior for Variational Autoencoder" @default.
- W3215977089 cites W1959608418 @default.
- W3215977089 cites W2462906003 @default.
- W3215977089 cites W2559626456 @default.
- W3215977089 cites W2810676804 @default.
- W3215977089 cites W2904981516 @default.
- W3215977089 cites W2913300775 @default.
- W3215977089 cites W2963043971 @default.
- W3215977089 cites W2963090522 @default.
- W3215977089 cites W2963135265 @default.
- W3215977089 cites W2963139417 @default.
- W3215977089 cites W2963384319 @default.
- W3215977089 cites W2963522047 @default.
- W3215977089 cites W2963546708 @default.
- W3215977089 cites W2963636093 @default.
- W3215977089 cites W2963799213 @default.
- W3215977089 cites W2970014727 @default.
- W3215977089 cites W2970946347 @default.
- W3215977089 cites W2973674981 @default.
- W3215977089 cites W2980402966 @default.
- W3215977089 cites W2981187293 @default.
- W3215977089 cites W2995981555 @default.
- W3215977089 cites W2996156127 @default.
- W3215977089 cites W3042418402 @default.
- W3215977089 cites W3158108632 @default.
- W3215977089 cites W3172266654 @default.
- W3215977089 hasPublicationYear "2021" @default.
- W3215977089 type Work @default.
- W3215977089 sameAs 3215977089 @default.
- W3215977089 citedByCount "0" @default.
- W3215977089 crossrefType "posted-content" @default.
- W3215977089 hasAuthorship W3215977089A5049658956 @default.
- W3215977089 hasAuthorship W3215977089A5053492737 @default.
- W3215977089 hasAuthorship W3215977089A5063069144 @default.
- W3215977089 hasConcept C101738243 @default.
- W3215977089 hasConcept C107673813 @default.
- W3215977089 hasConcept C111472728 @default.
- W3215977089 hasConcept C11413529 @default.
- W3215977089 hasConcept C121332964 @default.
- W3215977089 hasConcept C138885662 @default.
- W3215977089 hasConcept C153180895 @default.
- W3215977089 hasConcept C154945302 @default.
- W3215977089 hasConcept C163716315 @default.
- W3215977089 hasConcept C167966045 @default.
- W3215977089 hasConcept C171752962 @default.
- W3215977089 hasConcept C177769412 @default.
- W3215977089 hasConcept C189950617 @default.
- W3215977089 hasConcept C207390915 @default.
- W3215977089 hasConcept C28826006 @default.
- W3215977089 hasConcept C33923547 @default.
- W3215977089 hasConcept C39890363 @default.
- W3215977089 hasConcept C41008148 @default.
- W3215977089 hasConcept C41895202 @default.
- W3215977089 hasConcept C49937458 @default.
- W3215977089 hasConcept C50644808 @default.
- W3215977089 hasConcept C62520636 @default.
- W3215977089 hasConceptScore W3215977089C101738243 @default.
- W3215977089 hasConceptScore W3215977089C107673813 @default.
- W3215977089 hasConceptScore W3215977089C111472728 @default.
- W3215977089 hasConceptScore W3215977089C11413529 @default.
- W3215977089 hasConceptScore W3215977089C121332964 @default.
- W3215977089 hasConceptScore W3215977089C138885662 @default.
- W3215977089 hasConceptScore W3215977089C153180895 @default.
- W3215977089 hasConceptScore W3215977089C154945302 @default.
- W3215977089 hasConceptScore W3215977089C163716315 @default.
- W3215977089 hasConceptScore W3215977089C167966045 @default.
- W3215977089 hasConceptScore W3215977089C171752962 @default.
- W3215977089 hasConceptScore W3215977089C177769412 @default.
- W3215977089 hasConceptScore W3215977089C189950617 @default.
- W3215977089 hasConceptScore W3215977089C207390915 @default.
- W3215977089 hasConceptScore W3215977089C28826006 @default.
- W3215977089 hasConceptScore W3215977089C33923547 @default.
- W3215977089 hasConceptScore W3215977089C39890363 @default.
- W3215977089 hasConceptScore W3215977089C41008148 @default.
- W3215977089 hasConceptScore W3215977089C41895202 @default.
- W3215977089 hasConceptScore W3215977089C49937458 @default.
- W3215977089 hasConceptScore W3215977089C50644808 @default.
- W3215977089 hasConceptScore W3215977089C62520636 @default.
- W3215977089 hasLocation W32159770891 @default.
- W3215977089 hasOpenAccess W3215977089 @default.
- W3215977089 hasPrimaryLocation W32159770891 @default.
- W3215977089 isParatext "false" @default.
- W3215977089 isRetracted "false" @default.
- W3215977089 magId "3215977089" @default.
- W3215977089 workType "article" @default.