Matches in SemOpenAlex for { <https://semopenalex.org/work/W3126214278> ?p ?o ?g. }
Showing items 1 to 84 of
84
with 100 items per page.
- W3126214278 abstract "Standard formulations of GANs, where a continuous function deforms a connected latent space, have been shown to be misspecified when fitting disconnected manifolds. In particular, when covering different classes of images, the generator will necessarily sample some low quality images in between the modes. Rather than modify the learning procedure, a line of works aims at improving the sampling quality from trained generators. Thus, it is now common to introduce a rejection step within the generation procedure. Building on this, we propose to train an additional network and transform the latent space via an adversarial learning of importance weights. This idea has several advantages: 1) it provides a way to inject disconnectedness on any GAN architecture, 2) the rejection avoids going through both the generator and the discriminator saving computation time, 3) this importance weights formulation provides a principled way to estimate the Wasserstein's distance to the true distribution, enabling its minimization. We demonstrate the effectiveness of our method on different datasets, both synthetic and high dimensional, and stress its superiority on highly disconnected data." @default.
- W3126214278 created "2021-02-15" @default.
- W3126214278 creator A5002574498 @default.
- W3126214278 creator A5008216150 @default.
- W3126214278 creator A5021382990 @default.
- W3126214278 creator A5053968661 @default.
- W3126214278 date "2021-05-04" @default.
- W3126214278 modified "2023-09-26" @default.
- W3126214278 title "Learning Disconnected Manifolds: Avoiding The No Gan's Land by Latent Rejection" @default.
- W3126214278 hasPublicationYear "2021" @default.
- W3126214278 type Work @default.
- W3126214278 sameAs 3126214278 @default.
- W3126214278 citedByCount "1" @default.
- W3126214278 countsByYear W31262142782020 @default.
- W3126214278 crossrefType "journal-article" @default.
- W3126214278 hasAuthorship W3126214278A5002574498 @default.
- W3126214278 hasAuthorship W3126214278A5008216150 @default.
- W3126214278 hasAuthorship W3126214278A5021382990 @default.
- W3126214278 hasAuthorship W3126214278A5053968661 @default.
- W3126214278 hasConcept C106131492 @default.
- W3126214278 hasConcept C111919701 @default.
- W3126214278 hasConcept C11413529 @default.
- W3126214278 hasConcept C121332964 @default.
- W3126214278 hasConcept C126255220 @default.
- W3126214278 hasConcept C140779682 @default.
- W3126214278 hasConcept C147764199 @default.
- W3126214278 hasConcept C154945302 @default.
- W3126214278 hasConcept C163258240 @default.
- W3126214278 hasConcept C2778572836 @default.
- W3126214278 hasConcept C2779803651 @default.
- W3126214278 hasConcept C2780992000 @default.
- W3126214278 hasConcept C31972630 @default.
- W3126214278 hasConcept C33923547 @default.
- W3126214278 hasConcept C41008148 @default.
- W3126214278 hasConcept C45374587 @default.
- W3126214278 hasConcept C62520636 @default.
- W3126214278 hasConcept C76155785 @default.
- W3126214278 hasConcept C94915269 @default.
- W3126214278 hasConceptScore W3126214278C106131492 @default.
- W3126214278 hasConceptScore W3126214278C111919701 @default.
- W3126214278 hasConceptScore W3126214278C11413529 @default.
- W3126214278 hasConceptScore W3126214278C121332964 @default.
- W3126214278 hasConceptScore W3126214278C126255220 @default.
- W3126214278 hasConceptScore W3126214278C140779682 @default.
- W3126214278 hasConceptScore W3126214278C147764199 @default.
- W3126214278 hasConceptScore W3126214278C154945302 @default.
- W3126214278 hasConceptScore W3126214278C163258240 @default.
- W3126214278 hasConceptScore W3126214278C2778572836 @default.
- W3126214278 hasConceptScore W3126214278C2779803651 @default.
- W3126214278 hasConceptScore W3126214278C2780992000 @default.
- W3126214278 hasConceptScore W3126214278C31972630 @default.
- W3126214278 hasConceptScore W3126214278C33923547 @default.
- W3126214278 hasConceptScore W3126214278C41008148 @default.
- W3126214278 hasConceptScore W3126214278C45374587 @default.
- W3126214278 hasConceptScore W3126214278C62520636 @default.
- W3126214278 hasConceptScore W3126214278C76155785 @default.
- W3126214278 hasConceptScore W3126214278C94915269 @default.
- W3126214278 hasLocation W31262142781 @default.
- W3126214278 hasOpenAccess W3126214278 @default.
- W3126214278 hasPrimaryLocation W31262142781 @default.
- W3126214278 hasRelatedWork W1483683744 @default.
- W3126214278 hasRelatedWork W2414413599 @default.
- W3126214278 hasRelatedWork W2422281811 @default.
- W3126214278 hasRelatedWork W2517579295 @default.
- W3126214278 hasRelatedWork W2605145092 @default.
- W3126214278 hasRelatedWork W2735856532 @default.
- W3126214278 hasRelatedWork W2765404437 @default.
- W3126214278 hasRelatedWork W2770828701 @default.
- W3126214278 hasRelatedWork W2869762888 @default.
- W3126214278 hasRelatedWork W2949342585 @default.
- W3126214278 hasRelatedWork W2950123785 @default.
- W3126214278 hasRelatedWork W2954098871 @default.
- W3126214278 hasRelatedWork W2964001144 @default.
- W3126214278 hasRelatedWork W2965141076 @default.
- W3126214278 hasRelatedWork W2993087067 @default.
- W3126214278 hasRelatedWork W3034163353 @default.
- W3126214278 hasRelatedWork W3128448487 @default.
- W3126214278 hasRelatedWork W3161074134 @default.
- W3126214278 hasRelatedWork W3166635318 @default.
- W3126214278 hasRelatedWork W3200917001 @default.
- W3126214278 isParatext "false" @default.
- W3126214278 isRetracted "false" @default.
- W3126214278 magId "3126214278" @default.
- W3126214278 workType "article" @default.