Matches in SemOpenAlex for { <https://semopenalex.org/work/W2963129212> ?p ?o ?g. }
Showing items 1 to 81 of
81
with 100 items per page.
- W2963129212 endingPage "3301" @default.
- W2963129212 startingPage "3292" @default.
- W2963129212 abstract "In variational autoencoders, the prior on the latent codes $z$ is often treated as an afterthought, but the prior shapes the kind of latent representation that the model learns. If the goal is to learn a representation that is interpretable and useful, then the prior should reflect the ways in which the high-level factors that describe the data vary. The ``default'' prior is a standard normal, but if the natural factors of variation in the dataset exhibit discrete structure or are not independent, then the isotropic-normal prior will actually encourage learning representations that emph{mask} this structure. To alleviate this problem, we propose using a flexible Bayesian nonparametric hierarchical clustering prior based on the time-marginalized coalescent (TMC). To scale learning to large datasets, we develop a new inducing-point approximation and inference algorithm. We then apply the method without supervision to several datasets and examine the interpretability and practical performance of the inferred hierarchies and learned latent space." @default.
- W2963129212 created "2019-07-30" @default.
- W2963129212 creator A5038046410 @default.
- W2963129212 creator A5045680105 @default.
- W2963129212 creator A5069219684 @default.
- W2963129212 date "2019-04-11" @default.
- W2963129212 modified "2023-10-03" @default.
- W2963129212 title "The LORACs Prior for VAEs: Letting the Trees Speak for the Data" @default.
- W2963129212 hasPublicationYear "2019" @default.
- W2963129212 type Work @default.
- W2963129212 sameAs 2963129212 @default.
- W2963129212 citedByCount "1" @default.
- W2963129212 countsByYear W29631292122019 @default.
- W2963129212 crossrefType "proceedings-article" @default.
- W2963129212 hasAuthorship W2963129212A5038046410 @default.
- W2963129212 hasAuthorship W2963129212A5045680105 @default.
- W2963129212 hasAuthorship W2963129212A5069219684 @default.
- W2963129212 hasConcept C107673813 @default.
- W2963129212 hasConcept C111919701 @default.
- W2963129212 hasConcept C119857082 @default.
- W2963129212 hasConcept C153180895 @default.
- W2963129212 hasConcept C154945302 @default.
- W2963129212 hasConcept C160234255 @default.
- W2963129212 hasConcept C171686336 @default.
- W2963129212 hasConcept C17744445 @default.
- W2963129212 hasConcept C199539241 @default.
- W2963129212 hasConcept C2524010 @default.
- W2963129212 hasConcept C2776214188 @default.
- W2963129212 hasConcept C2776359362 @default.
- W2963129212 hasConcept C2778572836 @default.
- W2963129212 hasConcept C2781067378 @default.
- W2963129212 hasConcept C28719098 @default.
- W2963129212 hasConcept C33923547 @default.
- W2963129212 hasConcept C41008148 @default.
- W2963129212 hasConcept C73555534 @default.
- W2963129212 hasConcept C94625758 @default.
- W2963129212 hasConceptScore W2963129212C107673813 @default.
- W2963129212 hasConceptScore W2963129212C111919701 @default.
- W2963129212 hasConceptScore W2963129212C119857082 @default.
- W2963129212 hasConceptScore W2963129212C153180895 @default.
- W2963129212 hasConceptScore W2963129212C154945302 @default.
- W2963129212 hasConceptScore W2963129212C160234255 @default.
- W2963129212 hasConceptScore W2963129212C171686336 @default.
- W2963129212 hasConceptScore W2963129212C17744445 @default.
- W2963129212 hasConceptScore W2963129212C199539241 @default.
- W2963129212 hasConceptScore W2963129212C2524010 @default.
- W2963129212 hasConceptScore W2963129212C2776214188 @default.
- W2963129212 hasConceptScore W2963129212C2776359362 @default.
- W2963129212 hasConceptScore W2963129212C2778572836 @default.
- W2963129212 hasConceptScore W2963129212C2781067378 @default.
- W2963129212 hasConceptScore W2963129212C28719098 @default.
- W2963129212 hasConceptScore W2963129212C33923547 @default.
- W2963129212 hasConceptScore W2963129212C41008148 @default.
- W2963129212 hasConceptScore W2963129212C73555534 @default.
- W2963129212 hasConceptScore W2963129212C94625758 @default.
- W2963129212 hasOpenAccess W2963129212 @default.
- W2963129212 hasRelatedWork W2762409054 @default.
- W2963129212 hasRelatedWork W2785902771 @default.
- W2963129212 hasRelatedWork W2899928530 @default.
- W2963129212 hasRelatedWork W2900528039 @default.
- W2963129212 hasRelatedWork W2952444811 @default.
- W2963129212 hasRelatedWork W2956843879 @default.
- W2963129212 hasRelatedWork W2969856101 @default.
- W2963129212 hasRelatedWork W2972504940 @default.
- W2963129212 hasRelatedWork W2981471950 @default.
- W2963129212 hasRelatedWork W3081016631 @default.
- W2963129212 hasRelatedWork W3087564542 @default.
- W2963129212 hasRelatedWork W3093195268 @default.
- W2963129212 hasRelatedWork W3126858177 @default.
- W2963129212 hasRelatedWork W3127390859 @default.
- W2963129212 hasRelatedWork W3184837972 @default.
- W2963129212 hasRelatedWork W3198923159 @default.
- W2963129212 hasRelatedWork W3206531591 @default.
- W2963129212 hasRelatedWork W3214350629 @default.
- W2963129212 hasRelatedWork W97390373 @default.
- W2963129212 isParatext "false" @default.
- W2963129212 isRetracted "false" @default.
- W2963129212 magId "2963129212" @default.
- W2963129212 workType "article" @default.