FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2981660215> ?p ?o ?g. }

• W2981660215 abstract "Generative Adversarial Networks (GANs) are known to be difficult to train, despite considerable research effort. Several regularization techniques for stabilizing training have been proposed, but they introduce non-trivial computational overheads and interact poorly with existing techniques like spectral normalization. In this work, we propose a simple, effective training stabilizer based on the notion of consistency regularization---a popular technique in the semi-supervised learning literature. In particular, we augment data passing into the GAN discriminator and penalize the sensitivity of the discriminator to these augmentations. We conduct a series of experiments to demonstrate that consistency regularization works effectively with spectral normalization and various GAN architectures, loss functions and optimizer settings. Our method achieves the best FID scores for unconditional image generation compared to other regularization methods on CIFAR-10 and CelebA. Moreover, Our consistency regularized GAN (CR-GAN) improves state-of-the-art FID scores for conditional generation from 14.73 to 11.48 on CIFAR-10 and from 8.73 to 6.66 on ImageNet-2012." @default.
• W2981660215 created "2019-11-01" @default.
• W2981660215 creator A5006249840 @default.
• W2981660215 creator A5024743253 @default.
• W2981660215 creator A5050852420 @default.
• W2981660215 creator A5064419280 @default.
• W2981660215 date "2019-10-26" @default.
• W2981660215 modified "2023-09-27" @default.
• W2981660215 title "Consistency Regularization for Generative Adversarial Networks." @default.
• W2981660215 cites W2099471712 @default.
• W2981660215 cites W2117539524 @default.
• W2981660215 cites W2129068307 @default.
• W2981660215 cites W2194775991 @default.
• W2981660215 cites W2548275288 @default.
• W2981660215 cites W2564591810 @default.
• W2981660215 cites W2592796789 @default.
• W2981660215 cites W2617573776 @default.
• W2981660215 cites W2746314669 @default.
• W2981660215 cites W2752620087 @default.
• W2981660215 cites W2775288145 @default.
• W2981660215 cites W2788191623 @default.
• W2981660215 cites W2804078698 @default.
• W2981660215 cites W2893749619 @default.
• W2981660215 cites W2938075888 @default.
• W2981660215 cites W2943865428 @default.
• W2981660215 cites W2946856970 @default.
• W2981660215 cites W2950365520 @default.
• W2981660215 cites W2951970475 @default.
• W2981660215 cites W2953327099 @default.
• W2981660215 cites W2962369866 @default.
• W2981660215 cites W2962754210 @default.
• W2981660215 cites W2962760235 @default.
• W2981660215 cites W2962770929 @default.
• W2981660215 cites W2962879692 @default.
• W2981660215 cites W2963373786 @default.
• W2981660215 cites W2963435192 @default.
• W2981660215 cites W2963684088 @default.
• W2981660215 cites W2963836885 @default.
• W2981660215 cites W2963873275 @default.
• W2981660215 cites W2963956526 @default.
• W2981660215 cites W2963981733 @default.
• W2981660215 cites W2964057425 @default.
• W2981660215 cites W2964137095 @default.
• W2981660215 cites W2964201867 @default.
• W2981660215 cites W2964259506 @default.
• W2981660215 cites W3118608800 @default.
• W2981660215 hasPublicationYear "2019" @default.
• W2981660215 type Work @default.
• W2981660215 sameAs 2981660215 @default.
• W2981660215 citedByCount "11" @default.
• W2981660215 countsByYear W29816602152019 @default.
• W2981660215 countsByYear W29816602152020 @default.
• W2981660215 countsByYear W29816602152021 @default.
• W2981660215 crossrefType "posted-content" @default.
• W2981660215 hasAuthorship W2981660215A5006249840 @default.
• W2981660215 hasAuthorship W2981660215A5024743253 @default.
• W2981660215 hasAuthorship W2981660215A5050852420 @default.
• W2981660215 hasAuthorship W2981660215A5064419280 @default.
• W2981660215 hasConcept C11413529 @default.
• W2981660215 hasConcept C119857082 @default.
• W2981660215 hasConcept C136886441 @default.
• W2981660215 hasConcept C144024400 @default.
• W2981660215 hasConcept C154945302 @default.
• W2981660215 hasConcept C19165224 @default.
• W2981660215 hasConcept C2776135515 @default.
• W2981660215 hasConcept C2776436953 @default.
• W2981660215 hasConcept C2779803651 @default.
• W2981660215 hasConcept C37736160 @default.
• W2981660215 hasConcept C39890363 @default.
• W2981660215 hasConcept C41008148 @default.
• W2981660215 hasConcept C76155785 @default.
• W2981660215 hasConcept C94915269 @default.
• W2981660215 hasConceptScore W2981660215C11413529 @default.
• W2981660215 hasConceptScore W2981660215C119857082 @default.
• W2981660215 hasConceptScore W2981660215C136886441 @default.
• W2981660215 hasConceptScore W2981660215C144024400 @default.
• W2981660215 hasConceptScore W2981660215C154945302 @default.
• W2981660215 hasConceptScore W2981660215C19165224 @default.
• W2981660215 hasConceptScore W2981660215C2776135515 @default.
• W2981660215 hasConceptScore W2981660215C2776436953 @default.
• W2981660215 hasConceptScore W2981660215C2779803651 @default.
• W2981660215 hasConceptScore W2981660215C37736160 @default.
• W2981660215 hasConceptScore W2981660215C39890363 @default.
• W2981660215 hasConceptScore W2981660215C41008148 @default.
• W2981660215 hasConceptScore W2981660215C76155785 @default.
• W2981660215 hasConceptScore W2981660215C94915269 @default.
• W2981660215 hasLocation W29816602151 @default.
• W2981660215 hasOpenAccess W2981660215 @default.
• W2981660215 hasPrimaryLocation W29816602151 @default.
• W2981660215 hasRelatedWork W1834627138 @default.
• W2981660215 hasRelatedWork W2099471712 @default.
• W2981660215 hasRelatedWork W2108598243 @default.
• W2981660215 hasRelatedWork W2173520492 @default.
• W2981660215 hasRelatedWork W2194775991 @default.
• W2981660215 hasRelatedWork W2593414223 @default.
• W2981660215 hasRelatedWork W2766527293 @default.
• W2981660215 hasRelatedWork W2785678896 @default.
• W2981660215 hasRelatedWork W2794455536 @default.
• W2981660215 hasRelatedWork W2893749619 @default.
• W2981660215 hasRelatedWork W2962793481 @default.

• next