Matches in SemOpenAlex for { <https://semopenalex.org/work/W4387498356> ?p ?o ?g. }
Showing items 1 to 89 of
89
with 100 items per page.
- W4387498356 endingPage "110034" @default.
- W4387498356 startingPage "110034" @default.
- W4387498356 abstract "Generative adversarial network (GAN) is a framework for generating fake data using a set of real examples. However, GAN is unstable in the training stage. In order to stabilize GANs, the noise injection has been used to enlarge the overlap of the real and fake distributions at the cost of increasing variance. The diffusion process (or data smoothing in its spatial domain) removes fine details in order to capture the structure and important patterns in data but it suppresses the capability of GANs to learn high-frequency information in the training procedure. Based on these observations, we propose a data representation for the GAN training, called noisy scale-space (NSS), that recursively applies the smoothing with a balanced noise to data in order to replace the high-frequency information by random data, leading to a coarse-to-fine training of GANs. We experiment with NSS using DCGAN and StyleGAN2 based on benchmark datasets in which the NSS-based GANs outperforms the state-of-the-arts in most cases." @default.
- W4387498356 created "2023-10-11" @default.
- W4387498356 creator A5021189167 @default.
- W4387498356 creator A5022648850 @default.
- W4387498356 creator A5065943331 @default.
- W4387498356 date "2023-10-01" @default.
- W4387498356 modified "2023-10-12" @default.
- W4387498356 title "Generative adversarial networks via a composite annealing of noise and diffusion" @default.
- W4387498356 cites W1834627138 @default.
- W4387498356 cites W2021130884 @default.
- W4387498356 cites W2111835955 @default.
- W4387498356 cites W2112796928 @default.
- W4387498356 cites W2533598788 @default.
- W4387498356 cites W2593414223 @default.
- W4387498356 cites W2890139949 @default.
- W4387498356 cites W2953327099 @default.
- W4387498356 cites W2962770929 @default.
- W4387498356 cites W2962793481 @default.
- W4387498356 cites W2963420272 @default.
- W4387498356 cites W2964024144 @default.
- W4387498356 cites W2965141076 @default.
- W4387498356 cites W2992133484 @default.
- W4387498356 cites W3014099862 @default.
- W4387498356 cites W3035231706 @default.
- W4387498356 cites W3083907077 @default.
- W4387498356 cites W3120254195 @default.
- W4387498356 cites W3145726664 @default.
- W4387498356 cites W3165906701 @default.
- W4387498356 cites W3170237120 @default.
- W4387498356 cites W4295123323 @default.
- W4387498356 cites W4320525814 @default.
- W4387498356 cites W4364355572 @default.
- W4387498356 cites W4380884624 @default.
- W4387498356 doi "https://doi.org/10.1016/j.patcog.2023.110034" @default.
- W4387498356 hasPublicationYear "2023" @default.
- W4387498356 type Work @default.
- W4387498356 citedByCount "0" @default.
- W4387498356 crossrefType "journal-article" @default.
- W4387498356 hasAuthorship W4387498356A5021189167 @default.
- W4387498356 hasAuthorship W4387498356A5022648850 @default.
- W4387498356 hasAuthorship W4387498356A5065943331 @default.
- W4387498356 hasConcept C11413529 @default.
- W4387498356 hasConcept C115961682 @default.
- W4387498356 hasConcept C119857082 @default.
- W4387498356 hasConcept C124101348 @default.
- W4387498356 hasConcept C153180895 @default.
- W4387498356 hasConcept C154945302 @default.
- W4387498356 hasConcept C17744445 @default.
- W4387498356 hasConcept C199539241 @default.
- W4387498356 hasConcept C2776359362 @default.
- W4387498356 hasConcept C31972630 @default.
- W4387498356 hasConcept C3770464 @default.
- W4387498356 hasConcept C39890363 @default.
- W4387498356 hasConcept C41008148 @default.
- W4387498356 hasConcept C94625758 @default.
- W4387498356 hasConcept C99498987 @default.
- W4387498356 hasConceptScore W4387498356C11413529 @default.
- W4387498356 hasConceptScore W4387498356C115961682 @default.
- W4387498356 hasConceptScore W4387498356C119857082 @default.
- W4387498356 hasConceptScore W4387498356C124101348 @default.
- W4387498356 hasConceptScore W4387498356C153180895 @default.
- W4387498356 hasConceptScore W4387498356C154945302 @default.
- W4387498356 hasConceptScore W4387498356C17744445 @default.
- W4387498356 hasConceptScore W4387498356C199539241 @default.
- W4387498356 hasConceptScore W4387498356C2776359362 @default.
- W4387498356 hasConceptScore W4387498356C31972630 @default.
- W4387498356 hasConceptScore W4387498356C3770464 @default.
- W4387498356 hasConceptScore W4387498356C39890363 @default.
- W4387498356 hasConceptScore W4387498356C41008148 @default.
- W4387498356 hasConceptScore W4387498356C94625758 @default.
- W4387498356 hasConceptScore W4387498356C99498987 @default.
- W4387498356 hasLocation W43874983561 @default.
- W4387498356 hasOpenAccess W4387498356 @default.
- W4387498356 hasPrimaryLocation W43874983561 @default.
- W4387498356 hasRelatedWork W1978572805 @default.
- W4387498356 hasRelatedWork W1987225439 @default.
- W4387498356 hasRelatedWork W1989941125 @default.
- W4387498356 hasRelatedWork W1997992934 @default.
- W4387498356 hasRelatedWork W2019977573 @default.
- W4387498356 hasRelatedWork W2125114371 @default.
- W4387498356 hasRelatedWork W2149980199 @default.
- W4387498356 hasRelatedWork W2383807498 @default.
- W4387498356 hasRelatedWork W2391139946 @default.
- W4387498356 hasRelatedWork W3125766170 @default.
- W4387498356 isParatext "false" @default.
- W4387498356 isRetracted "false" @default.
- W4387498356 workType "article" @default.