Matches in SemOpenAlex for { <https://semopenalex.org/work/W3045501090> ?p ?o ?g. }
- W3045501090 abstract "Recent advances in Generative Adversarial Networks (GANs) have led to their widespread adoption for the purposes of generating high quality synthetic imagery. While capable of generating photo-realistic images, these models often produce unrealistic samples which fall outside of the data manifold. Several recently proposed techniques attempt to avoid spurious samples, either by rejecting them after generation, or by truncating the model's latent space. While effective, these methods are inefficient, as a large fraction of training time and model capacity are dedicated towards samples that will ultimately go unused. In this work we propose a novel approach to improve sample quality: altering the training dataset via instance selection before model training has taken place. By refining the empirical data distribution before training, we redirect model capacity towards high-density regions, which ultimately improves sample fidelity, lowers model capacity requirements, and significantly reduces training time. Code is available at this https URL." @default.
- W3045501090 created "2020-08-03" @default.
- W3045501090 creator A5024536150 @default.
- W3045501090 creator A5034466911 @default.
- W3045501090 creator A5064658629 @default.
- W3045501090 date "2020-07-30" @default.
- W3045501090 modified "2023-10-01" @default.
- W3045501090 title "Instance Selection for GANs" @default.
- W3045501090 cites W2011762057 @default.
- W3045501090 cites W2101234009 @default.
- W3045501090 cites W2108598243 @default.
- W3045501090 cites W2125027820 @default.
- W3045501090 cites W2183341477 @default.
- W3045501090 cites W2194775991 @default.
- W3045501090 cites W2243049275 @default.
- W3045501090 cites W2549139847 @default.
- W3045501090 cites W2621175638 @default.
- W3045501090 cites W2732026016 @default.
- W3045501090 cites W2786672974 @default.
- W3045501090 cites W2804078698 @default.
- W3045501090 cites W2884805522 @default.
- W3045501090 cites W2891158090 @default.
- W3045501090 cites W2893749619 @default.
- W3045501090 cites W2901368259 @default.
- W3045501090 cites W2953030256 @default.
- W3045501090 cites W2960404386 @default.
- W3045501090 cites W2962770929 @default.
- W3045501090 cites W2962771722 @default.
- W3045501090 cites W2962785568 @default.
- W3045501090 cites W2963373786 @default.
- W3045501090 cites W2963703197 @default.
- W3045501090 cites W2963981733 @default.
- W3045501090 cites W2971128425 @default.
- W3045501090 cites W2971337561 @default.
- W3045501090 cites W2982511758 @default.
- W3045501090 cites W2990706204 @default.
- W3045501090 cites W2995327724 @default.
- W3045501090 cites W3005601278 @default.
- W3045501090 cites W3011675874 @default.
- W3045501090 cites W3034723908 @default.
- W3045501090 cites W3034918919 @default.
- W3045501090 cites W3036224891 @default.
- W3045501090 cites W3038930935 @default.
- W3045501090 cites W3043547428 @default.
- W3045501090 cites W3104876213 @default.
- W3045501090 cites W3106068426 @default.
- W3045501090 hasPublicationYear "2020" @default.
- W3045501090 type Work @default.
- W3045501090 sameAs 3045501090 @default.
- W3045501090 citedByCount "1" @default.
- W3045501090 countsByYear W30455010902021 @default.
- W3045501090 crossrefType "posted-content" @default.
- W3045501090 hasAuthorship W3045501090A5024536150 @default.
- W3045501090 hasAuthorship W3045501090A5034466911 @default.
- W3045501090 hasAuthorship W3045501090A5064658629 @default.
- W3045501090 hasConcept C111472728 @default.
- W3045501090 hasConcept C119857082 @default.
- W3045501090 hasConcept C124101348 @default.
- W3045501090 hasConcept C138885662 @default.
- W3045501090 hasConcept C149629883 @default.
- W3045501090 hasConcept C154945302 @default.
- W3045501090 hasConcept C167966045 @default.
- W3045501090 hasConcept C177264268 @default.
- W3045501090 hasConcept C178790620 @default.
- W3045501090 hasConcept C185592680 @default.
- W3045501090 hasConcept C198531522 @default.
- W3045501090 hasConcept C199360897 @default.
- W3045501090 hasConcept C2776459999 @default.
- W3045501090 hasConcept C2776760102 @default.
- W3045501090 hasConcept C2779530757 @default.
- W3045501090 hasConcept C39890363 @default.
- W3045501090 hasConcept C41008148 @default.
- W3045501090 hasConcept C43617362 @default.
- W3045501090 hasConcept C76155785 @default.
- W3045501090 hasConcept C81917197 @default.
- W3045501090 hasConcept C93959086 @default.
- W3045501090 hasConcept C97256817 @default.
- W3045501090 hasConceptScore W3045501090C111472728 @default.
- W3045501090 hasConceptScore W3045501090C119857082 @default.
- W3045501090 hasConceptScore W3045501090C124101348 @default.
- W3045501090 hasConceptScore W3045501090C138885662 @default.
- W3045501090 hasConceptScore W3045501090C149629883 @default.
- W3045501090 hasConceptScore W3045501090C154945302 @default.
- W3045501090 hasConceptScore W3045501090C167966045 @default.
- W3045501090 hasConceptScore W3045501090C177264268 @default.
- W3045501090 hasConceptScore W3045501090C178790620 @default.
- W3045501090 hasConceptScore W3045501090C185592680 @default.
- W3045501090 hasConceptScore W3045501090C198531522 @default.
- W3045501090 hasConceptScore W3045501090C199360897 @default.
- W3045501090 hasConceptScore W3045501090C2776459999 @default.
- W3045501090 hasConceptScore W3045501090C2776760102 @default.
- W3045501090 hasConceptScore W3045501090C2779530757 @default.
- W3045501090 hasConceptScore W3045501090C39890363 @default.
- W3045501090 hasConceptScore W3045501090C41008148 @default.
- W3045501090 hasConceptScore W3045501090C43617362 @default.
- W3045501090 hasConceptScore W3045501090C76155785 @default.
- W3045501090 hasConceptScore W3045501090C81917197 @default.
- W3045501090 hasConceptScore W3045501090C93959086 @default.
- W3045501090 hasConceptScore W3045501090C97256817 @default.
- W3045501090 hasLocation W30455010901 @default.