FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3166104071> ?p ?o ?g. }

• W3166104071 abstract "Abstract Recent work has shown that Neural Ordinary Differential Equations (ODEs) can serve as generative models of images using the perspective of Continuous Normalizing Flows (CNFs). Such models offer exact likelihood calculation, and invertible generation/density estimation. In this work we introduce a Multi-Resolution variant of such models (MRCNF), by characterizing the conditional distribution over the additional information required to generate a fine image that is consistent with the coarse image. We introduce a transformation between resolutions that allows for no change in the log likelihood. We show that this approach yields comparable likelihood values for various image datasets, with improved performance at higher resolutions, with fewer parameters, using only one GPU. Further, we examine the out-of-distribution properties of MRCNFs, and find that they are similar to those of other likelihood-based generative models." @default.
• W3166104071 created "2021-06-22" @default.
• W3166104071 creator A5004824330 @default.
• W3166104071 creator A5013102998 @default.
• W3166104071 creator A5066216363 @default.
• W3166104071 creator A5075885606 @default.
• W3166104071 date "2023-06-16" @default.
• W3166104071 modified "2023-09-30" @default.
• W3166104071 title "Multi-Resolution Continuous Normalizing Flows" @default.
• W3166104071 cites W1597739853 @default.
• W3166104071 cites W1959608418 @default.
• W3166104071 cites W1973753234 @default.
• W3166104071 cites W2074163268 @default.
• W3166104071 cites W2103504761 @default.
• W3166104071 cites W2104067967 @default.
• W3166104071 cites W2117612778 @default.
• W3166104071 cites W2132984323 @default.
• W3166104071 cites W2173520492 @default.
• W3166104071 cites W2267126114 @default.
• W3166104071 cites W228380312 @default.
• W3166104071 cites W2518435514 @default.
• W3166104071 cites W2531327146 @default.
• W3166104071 cites W2557283755 @default.
• W3166104071 cites W2902630600 @default.
• W3166104071 cites W2913300775 @default.
• W3166104071 cites W2940744433 @default.
• W3166104071 cites W2963090522 @default.
• W3166104071 cites W2963139417 @default.
• W3166104071 cites W2963636093 @default.
• W3166104071 cites W2964201867 @default.
• W3166104071 cites W2980402966 @default.
• W3166104071 cites W2992035660 @default.
• W3166104071 cites W2998108143 @default.
• W3166104071 cites W3034889960 @default.
• W3166104071 cites W3035231706 @default.
• W3166104071 cites W3098510582 @default.
• W3166104071 cites W3098989009 @default.
• W3166104071 cites W3104312791 @default.
• W3166104071 cites W3104674435 @default.
• W3166104071 cites W3105493304 @default.
• W3166104071 cites W3175859295 @default.
• W3166104071 doi "https://doi.org/10.21203/rs.3.rs-3027011/v1" @default.
• W3166104071 hasPublicationYear "2023" @default.
• W3166104071 type Work @default.
• W3166104071 sameAs 3166104071 @default.
• W3166104071 citedByCount "0" @default.
• W3166104071 crossrefType "posted-content" @default.
• W3166104071 hasAuthorship W3166104071A5004824330 @default.
• W3166104071 hasAuthorship W3166104071A5013102998 @default.
• W3166104071 hasAuthorship W3166104071A5066216363 @default.
• W3166104071 hasAuthorship W3166104071A5075885606 @default.
• W3166104071 hasBestOaLocation W31661040711 @default.
• W3166104071 hasConcept C104317684 @default.
• W3166104071 hasConcept C105795698 @default.
• W3166104071 hasConcept C110121322 @default.
• W3166104071 hasConcept C11413529 @default.
• W3166104071 hasConcept C115961682 @default.
• W3166104071 hasConcept C134306372 @default.
• W3166104071 hasConcept C153180895 @default.
• W3166104071 hasConcept C154945302 @default.
• W3166104071 hasConcept C167966045 @default.
• W3166104071 hasConcept C185592680 @default.
• W3166104071 hasConcept C202444582 @default.
• W3166104071 hasConcept C204241405 @default.
• W3166104071 hasConcept C28826006 @default.
• W3166104071 hasConcept C33923547 @default.
• W3166104071 hasConcept C34862557 @default.
• W3166104071 hasConcept C39890363 @default.
• W3166104071 hasConcept C41008148 @default.
• W3166104071 hasConcept C43555835 @default.
• W3166104071 hasConcept C51544822 @default.
• W3166104071 hasConcept C55493867 @default.
• W3166104071 hasConcept C78045399 @default.
• W3166104071 hasConcept C96442724 @default.
• W3166104071 hasConceptScore W3166104071C104317684 @default.
• W3166104071 hasConceptScore W3166104071C105795698 @default.
• W3166104071 hasConceptScore W3166104071C110121322 @default.
• W3166104071 hasConceptScore W3166104071C11413529 @default.
• W3166104071 hasConceptScore W3166104071C115961682 @default.
• W3166104071 hasConceptScore W3166104071C134306372 @default.
• W3166104071 hasConceptScore W3166104071C153180895 @default.
• W3166104071 hasConceptScore W3166104071C154945302 @default.
• W3166104071 hasConceptScore W3166104071C167966045 @default.
• W3166104071 hasConceptScore W3166104071C185592680 @default.
• W3166104071 hasConceptScore W3166104071C202444582 @default.
• W3166104071 hasConceptScore W3166104071C204241405 @default.
• W3166104071 hasConceptScore W3166104071C28826006 @default.
• W3166104071 hasConceptScore W3166104071C33923547 @default.
• W3166104071 hasConceptScore W3166104071C34862557 @default.
• W3166104071 hasConceptScore W3166104071C39890363 @default.
• W3166104071 hasConceptScore W3166104071C41008148 @default.
• W3166104071 hasConceptScore W3166104071C43555835 @default.
• W3166104071 hasConceptScore W3166104071C51544822 @default.
• W3166104071 hasConceptScore W3166104071C55493867 @default.
• W3166104071 hasConceptScore W3166104071C78045399 @default.
• W3166104071 hasConceptScore W3166104071C96442724 @default.
• W3166104071 hasLocation W31661040711 @default.
• W3166104071 hasOpenAccess W3166104071 @default.
• W3166104071 hasPrimaryLocation W31661040711 @default.
• W3166104071 hasRelatedWork W1750424384 @default.

• next