Matches in SemOpenAlex for { <https://semopenalex.org/work/W2936342412> ?p ?o ?g. }
- W2936342412 abstract "Generative adversarial nets (GANs) and variational auto-encoders have significantly improved our distribution modeling capabilities, showing promise for dataset augmentation, image-to-image translation and feature learning. However, to model high-dimensional distributions, sequential training and stacked architectures are common, increasing the number of tunable hyper-parameters as well as the training time. Nonetheless, the sample complexity of the distance metrics remains one of the factors affecting GAN training. We first show that the recently proposed sliced Wasserstein distance has compelling sample complexity properties when compared to the Wasserstein distance. To further improve the sliced Wasserstein distance we then analyze its `projection complexity' and develop the max-sliced Wasserstein distance which enjoys compelling sample complexity while reducing projection complexity, albeit necessitating a max estimation. We finally illustrate that the proposed distance trains GANs on high-dimensional images up to a resolution of 256x256 easily." @default.
- W2936342412 created "2019-04-25" @default.
- W2936342412 creator A5010851725 @default.
- W2936342412 creator A5015863645 @default.
- W2936342412 creator A5019600898 @default.
- W2936342412 creator A5028963874 @default.
- W2936342412 creator A5048820768 @default.
- W2936342412 creator A5049638480 @default.
- W2936342412 creator A5055113943 @default.
- W2936342412 creator A5081970325 @default.
- W2936342412 creator A5091266570 @default.
- W2936342412 date "2019-04-11" @default.
- W2936342412 modified "2023-09-23" @default.
- W2936342412 title "Max-Sliced Wasserstein Distance and its use for GANs" @default.
- W2936342412 cites W2019106840 @default.
- W2936342412 cites W2099471712 @default.
- W2936342412 cites W2240450848 @default.
- W2936342412 cites W2593729559 @default.
- W2936342412 cites W2594154005 @default.
- W2936342412 cites W2605195953 @default.
- W2936342412 cites W2605627527 @default.
- W2936342412 cites W2727413762 @default.
- W2936342412 cites W2750675958 @default.
- W2936342412 cites W2766210095 @default.
- W2936342412 cites W2766527293 @default.
- W2936342412 cites W2768368415 @default.
- W2936342412 cites W2795671189 @default.
- W2936342412 cites W2951004968 @default.
- W2936342412 cites W2953030256 @default.
- W2936342412 cites W2962793481 @default.
- W2936342412 cites W2962879692 @default.
- W2936342412 cites W2962892300 @default.
- W2936342412 cites W2962947361 @default.
- W2936342412 cites W2963006832 @default.
- W2936342412 cites W2963073614 @default.
- W2936342412 cites W2963118869 @default.
- W2936342412 cites W2963265008 @default.
- W2936342412 cites W2963398989 @default.
- W2936342412 cites W2963444790 @default.
- W2936342412 cites W2963563295 @default.
- W2936342412 cites W2963684088 @default.
- W2936342412 cites W2963709863 @default.
- W2936342412 cites W2963809789 @default.
- W2936342412 cites W2963890275 @default.
- W2936342412 cites W2963981733 @default.
- W2936342412 cites W2963989027 @default.
- W2936342412 cites W2964201867 @default.
- W2936342412 cites W3037567775 @default.
- W2936342412 cites W3159890710 @default.
- W2936342412 cites W385466589 @default.
- W2936342412 cites W967544008 @default.
- W2936342412 hasPublicationYear "2019" @default.
- W2936342412 type Work @default.
- W2936342412 sameAs 2936342412 @default.
- W2936342412 citedByCount "6" @default.
- W2936342412 countsByYear W29363424122020 @default.
- W2936342412 countsByYear W29363424122021 @default.
- W2936342412 crossrefType "posted-content" @default.
- W2936342412 hasAuthorship W2936342412A5010851725 @default.
- W2936342412 hasAuthorship W2936342412A5015863645 @default.
- W2936342412 hasAuthorship W2936342412A5019600898 @default.
- W2936342412 hasAuthorship W2936342412A5028963874 @default.
- W2936342412 hasAuthorship W2936342412A5048820768 @default.
- W2936342412 hasAuthorship W2936342412A5049638480 @default.
- W2936342412 hasAuthorship W2936342412A5055113943 @default.
- W2936342412 hasAuthorship W2936342412A5081970325 @default.
- W2936342412 hasAuthorship W2936342412A5091266570 @default.
- W2936342412 hasConcept C11413529 @default.
- W2936342412 hasConcept C115961682 @default.
- W2936342412 hasConcept C138885662 @default.
- W2936342412 hasConcept C153180895 @default.
- W2936342412 hasConcept C154945302 @default.
- W2936342412 hasConcept C185592680 @default.
- W2936342412 hasConcept C198531522 @default.
- W2936342412 hasConcept C2776401178 @default.
- W2936342412 hasConcept C2778445095 @default.
- W2936342412 hasConcept C33923547 @default.
- W2936342412 hasConcept C41008148 @default.
- W2936342412 hasConcept C41895202 @default.
- W2936342412 hasConcept C43617362 @default.
- W2936342412 hasConcept C57493831 @default.
- W2936342412 hasConceptScore W2936342412C11413529 @default.
- W2936342412 hasConceptScore W2936342412C115961682 @default.
- W2936342412 hasConceptScore W2936342412C138885662 @default.
- W2936342412 hasConceptScore W2936342412C153180895 @default.
- W2936342412 hasConceptScore W2936342412C154945302 @default.
- W2936342412 hasConceptScore W2936342412C185592680 @default.
- W2936342412 hasConceptScore W2936342412C198531522 @default.
- W2936342412 hasConceptScore W2936342412C2776401178 @default.
- W2936342412 hasConceptScore W2936342412C2778445095 @default.
- W2936342412 hasConceptScore W2936342412C33923547 @default.
- W2936342412 hasConceptScore W2936342412C41008148 @default.
- W2936342412 hasConceptScore W2936342412C41895202 @default.
- W2936342412 hasConceptScore W2936342412C43617362 @default.
- W2936342412 hasConceptScore W2936342412C57493831 @default.
- W2936342412 hasOpenAccess W2936342412 @default.
- W2936342412 hasRelatedWork W2182984760 @default.
- W2936342412 hasRelatedWork W2417433140 @default.
- W2936342412 hasRelatedWork W2560148750 @default.
- W2936342412 hasRelatedWork W2795607530 @default.