Matches in SemOpenAlex for { <https://semopenalex.org/work/W3093528280> ?p ?o ?g. }
Showing items 1 to 92 of
92
with 100 items per page.
- W3093528280 abstract "Variational auto-encoders (VAEs) have proven to be a well suited tool for performing dimensionality reduction by extracting latent variables lying in a potentially much smaller dimensional space than the data. Their ability to capture meaningful information from the data can be easily apprehended when considering their capability to generate new realistic samples or perform potentially meaningful interpolations in a much smaller space. However, such generative models may perform poorly when trained on small data sets which are abundant in many real-life fields such as medicine. This may, among others, come from the lack of structure of the latent space, the geometry of which is often under-considered. We thus propose in this paper to see the latent space as a Riemannian manifold endowed with a parametrized metric learned at the same time as the encoder and decoder networks. This metric is then used in what we called the Riemannian Hamiltonian VAE which extends the Hamiltonian VAE introduced by Caterini et al. (2018) to better exploit the underlying geometry of the latent space. We argue that such latent space modelling provides useful information about its underlying structure leading to far more meaningful interpolations, more realistic data-generation and more reliable clustering." @default.
- W3093528280 created "2020-10-29" @default.
- W3093528280 creator A5031106777 @default.
- W3093528280 creator A5036395269 @default.
- W3093528280 creator A5058162067 @default.
- W3093528280 date "2020-10-20" @default.
- W3093528280 modified "2023-10-10" @default.
- W3093528280 title "Geometry-Aware Hamiltonian Variational Auto-Encoder" @default.
- W3093528280 cites W1513873506 @default.
- W3093528280 cites W1544770838 @default.
- W3093528280 cites W1545319692 @default.
- W3093528280 cites W1655728402 @default.
- W3093528280 cites W1906598733 @default.
- W3093528280 cites W1909320841 @default.
- W3093528280 cites W1959608418 @default.
- W3093528280 cites W2059448777 @default.
- W3093528280 cites W2069739265 @default.
- W3093528280 cites W2087353192 @default.
- W3093528280 cites W2101234009 @default.
- W3093528280 cites W2105824687 @default.
- W3093528280 cites W2150534249 @default.
- W3093528280 cites W2169528473 @default.
- W3093528280 cites W2478027467 @default.
- W3093528280 cites W2556467266 @default.
- W3093528280 cites W2611204545 @default.
- W3093528280 cites W2753738274 @default.
- W3093528280 cites W2784067649 @default.
- W3093528280 cites W2796335507 @default.
- W3093528280 cites W2805234167 @default.
- W3093528280 cites W2808058004 @default.
- W3093528280 cites W2890571755 @default.
- W3093528280 cites W2911859283 @default.
- W3093528280 cites W2950873402 @default.
- W3093528280 cites W2962703949 @default.
- W3093528280 cites W2962851448 @default.
- W3093528280 cites W2963090522 @default.
- W3093528280 cites W2963145887 @default.
- W3093528280 cites W2963275229 @default.
- W3093528280 cites W2963289229 @default.
- W3093528280 cites W2964011399 @default.
- W3093528280 cites W2964231450 @default.
- W3093528280 cites W2971246770 @default.
- W3093528280 cites W2972588190 @default.
- W3093528280 cites W2999284236 @default.
- W3093528280 hasPublicationYear "2020" @default.
- W3093528280 type Work @default.
- W3093528280 sameAs 3093528280 @default.
- W3093528280 citedByCount "2" @default.
- W3093528280 countsByYear W30935282802021 @default.
- W3093528280 crossrefType "posted-content" @default.
- W3093528280 hasAuthorship W3093528280A5031106777 @default.
- W3093528280 hasAuthorship W3093528280A5036395269 @default.
- W3093528280 hasAuthorship W3093528280A5058162067 @default.
- W3093528280 hasBestOaLocation W30935282801 @default.
- W3093528280 hasConcept C111919701 @default.
- W3093528280 hasConcept C118505674 @default.
- W3093528280 hasConcept C121332964 @default.
- W3093528280 hasConcept C126255220 @default.
- W3093528280 hasConcept C130787639 @default.
- W3093528280 hasConcept C2524010 @default.
- W3093528280 hasConcept C33923547 @default.
- W3093528280 hasConcept C41008148 @default.
- W3093528280 hasConcept C74650414 @default.
- W3093528280 hasConceptScore W3093528280C111919701 @default.
- W3093528280 hasConceptScore W3093528280C118505674 @default.
- W3093528280 hasConceptScore W3093528280C121332964 @default.
- W3093528280 hasConceptScore W3093528280C126255220 @default.
- W3093528280 hasConceptScore W3093528280C130787639 @default.
- W3093528280 hasConceptScore W3093528280C2524010 @default.
- W3093528280 hasConceptScore W3093528280C33923547 @default.
- W3093528280 hasConceptScore W3093528280C41008148 @default.
- W3093528280 hasConceptScore W3093528280C74650414 @default.
- W3093528280 hasLocation W30935282801 @default.
- W3093528280 hasLocation W30935282802 @default.
- W3093528280 hasLocation W30935282803 @default.
- W3093528280 hasLocation W30935282804 @default.
- W3093528280 hasOpenAccess W3093528280 @default.
- W3093528280 hasPrimaryLocation W30935282801 @default.
- W3093528280 hasRelatedWork W1587224694 @default.
- W3093528280 hasRelatedWork W1979597421 @default.
- W3093528280 hasRelatedWork W2007980826 @default.
- W3093528280 hasRelatedWork W2061531152 @default.
- W3093528280 hasRelatedWork W2077600819 @default.
- W3093528280 hasRelatedWork W2142036596 @default.
- W3093528280 hasRelatedWork W2911598644 @default.
- W3093528280 hasRelatedWork W3002753104 @default.
- W3093528280 hasRelatedWork W4225152035 @default.
- W3093528280 hasRelatedWork W4245490552 @default.
- W3093528280 isParatext "false" @default.
- W3093528280 isRetracted "false" @default.
- W3093528280 magId "3093528280" @default.
- W3093528280 workType "article" @default.