Matches in SemOpenAlex for { <https://semopenalex.org/work/W2955680640> ?p ?o ?g. }
Showing items 1 to 95 of
95
with 100 items per page.
- W2955680640 endingPage "32" @default.
- W2955680640 startingPage "17" @default.
- W2955680640 abstract "Generative adversarial networks (GAN) are trained through a minimax game between a generator and a discriminator to generate data that mimics observations. While being widely used, GAN training is known to be empirically unstable. This paper presents a new theory for generative adversarial methods that does not rely on the traditional minimax formulation. Our theory shows that with a strong discriminator, a good generator can be obtained by composite functional gradient learning, so that several distance measures (including the KL divergence and the JS divergence) between the probability distributions of real data and generated data are simultaneously improved after each functional gradient step until converging to zero. This new point of view leads to stable procedures for training generative models. It also gives a new theoretical insight into the original GAN. Empirical results on image generation show the effectiveness of our new method." @default.
- W2955680640 created "2019-07-12" @default.
- W2955680640 creator A5037709424 @default.
- W2955680640 creator A5052225329 @default.
- W2955680640 date "2021-01-01" @default.
- W2955680640 modified "2023-10-01" @default.
- W2955680640 title "A Framework of Composite Functional Gradient Methods for Generative Adversarial Models" @default.
- W2955680640 cites W1678356000 @default.
- W2955680640 cites W1970789124 @default.
- W2955680640 cites W2023163512 @default.
- W2955680640 cites W2078179989 @default.
- W2955680640 cites W2194775991 @default.
- W2955680640 cites W2566832195 @default.
- W2955680640 cites W2593414223 @default.
- W2955680640 cites W2780786077 @default.
- W2955680640 cites W2964024144 @default.
- W2955680640 cites W4293404332 @default.
- W2955680640 doi "https://doi.org/10.1109/tpami.2019.2924428" @default.
- W2955680640 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/31247543" @default.
- W2955680640 hasPublicationYear "2021" @default.
- W2955680640 type Work @default.
- W2955680640 sameAs 2955680640 @default.
- W2955680640 citedByCount "5" @default.
- W2955680640 countsByYear W29556806402020 @default.
- W2955680640 countsByYear W29556806402021 @default.
- W2955680640 countsByYear W29556806402023 @default.
- W2955680640 crossrefType "journal-article" @default.
- W2955680640 hasAuthorship W2955680640A5037709424 @default.
- W2955680640 hasAuthorship W2955680640A5052225329 @default.
- W2955680640 hasConcept C11413529 @default.
- W2955680640 hasConcept C119857082 @default.
- W2955680640 hasConcept C121332964 @default.
- W2955680640 hasConcept C126255220 @default.
- W2955680640 hasConcept C138885662 @default.
- W2955680640 hasConcept C149728462 @default.
- W2955680640 hasConcept C153180895 @default.
- W2955680640 hasConcept C154945302 @default.
- W2955680640 hasConcept C163258240 @default.
- W2955680640 hasConcept C207390915 @default.
- W2955680640 hasConcept C2524010 @default.
- W2955680640 hasConcept C2779803651 @default.
- W2955680640 hasConcept C2780992000 @default.
- W2955680640 hasConcept C28719098 @default.
- W2955680640 hasConcept C33923547 @default.
- W2955680640 hasConcept C37736160 @default.
- W2955680640 hasConcept C39890363 @default.
- W2955680640 hasConcept C41008148 @default.
- W2955680640 hasConcept C41895202 @default.
- W2955680640 hasConcept C62520636 @default.
- W2955680640 hasConcept C76155785 @default.
- W2955680640 hasConcept C94915269 @default.
- W2955680640 hasConceptScore W2955680640C11413529 @default.
- W2955680640 hasConceptScore W2955680640C119857082 @default.
- W2955680640 hasConceptScore W2955680640C121332964 @default.
- W2955680640 hasConceptScore W2955680640C126255220 @default.
- W2955680640 hasConceptScore W2955680640C138885662 @default.
- W2955680640 hasConceptScore W2955680640C149728462 @default.
- W2955680640 hasConceptScore W2955680640C153180895 @default.
- W2955680640 hasConceptScore W2955680640C154945302 @default.
- W2955680640 hasConceptScore W2955680640C163258240 @default.
- W2955680640 hasConceptScore W2955680640C207390915 @default.
- W2955680640 hasConceptScore W2955680640C2524010 @default.
- W2955680640 hasConceptScore W2955680640C2779803651 @default.
- W2955680640 hasConceptScore W2955680640C2780992000 @default.
- W2955680640 hasConceptScore W2955680640C28719098 @default.
- W2955680640 hasConceptScore W2955680640C33923547 @default.
- W2955680640 hasConceptScore W2955680640C37736160 @default.
- W2955680640 hasConceptScore W2955680640C39890363 @default.
- W2955680640 hasConceptScore W2955680640C41008148 @default.
- W2955680640 hasConceptScore W2955680640C41895202 @default.
- W2955680640 hasConceptScore W2955680640C62520636 @default.
- W2955680640 hasConceptScore W2955680640C76155785 @default.
- W2955680640 hasConceptScore W2955680640C94915269 @default.
- W2955680640 hasIssue "1" @default.
- W2955680640 hasLocation W29556806401 @default.
- W2955680640 hasLocation W29556806402 @default.
- W2955680640 hasOpenAccess W2955680640 @default.
- W2955680640 hasPrimaryLocation W29556806401 @default.
- W2955680640 hasRelatedWork W2784622812 @default.
- W2955680640 hasRelatedWork W2808401425 @default.
- W2955680640 hasRelatedWork W2953246223 @default.
- W2955680640 hasRelatedWork W2955680640 @default.
- W2955680640 hasRelatedWork W2963604896 @default.
- W2955680640 hasRelatedWork W2989596816 @default.
- W2955680640 hasRelatedWork W3005996785 @default.
- W2955680640 hasRelatedWork W3110333877 @default.
- W2955680640 hasRelatedWork W4287591810 @default.
- W2955680640 hasRelatedWork W4320812420 @default.
- W2955680640 hasVolume "43" @default.
- W2955680640 isParatext "false" @default.
- W2955680640 isRetracted "false" @default.
- W2955680640 magId "2955680640" @default.
- W2955680640 workType "article" @default.