Matches in SemOpenAlex for { <https://semopenalex.org/work/W3171859092> ?p ?o ?g. }
- W3171859092 abstract "Deep generative models are becoming widely used across science and industry for a variety of purposes. A common challenge is achieving a precise implicit or explicit representation of the data probability density. Recent proposals have suggested using classifier weights to refine the learned density of deep generative models. We extend this idea to all types of generative models and show how latent space refinement via iterated generative modeling can circumvent topological obstructions and improve precision. This methodology also applies to cases were the target model is non-differentiable and has many internal latent dimensions which must be marginalized over before refinement. We demonstrate our Latent Space Refinement (LaSeR) protocol on a variety of examples, focusing on the combinations of Normalizing Flows and Generative Adversarial Networks." @default.
- W3171859092 created "2021-06-22" @default.
- W3171859092 creator A5022362253 @default.
- W3171859092 creator A5045754297 @default.
- W3171859092 creator A5054491740 @default.
- W3171859092 date "2021-06-01" @default.
- W3171859092 modified "2023-09-23" @default.
- W3171859092 title "Latent Space Refinement for Deep Generative Models." @default.
- W3171859092 cites W1480376833 @default.
- W3171859092 cites W1487641199 @default.
- W3171859092 cites W1840847274 @default.
- W3171859092 cites W2059448777 @default.
- W3171859092 cites W2614083378 @default.
- W3171859092 cites W2765811365 @default.
- W3171859092 cites W2805508261 @default.
- W3171859092 cites W2807595580 @default.
- W3171859092 cites W2852068008 @default.
- W3171859092 cites W2889983866 @default.
- W3171859092 cites W2896116896 @default.
- W3171859092 cites W2897864643 @default.
- W3171859092 cites W2899303573 @default.
- W3171859092 cites W2899415024 @default.
- W3171859092 cites W2903107813 @default.
- W3171859092 cites W2926003515 @default.
- W3171859092 cites W2941443959 @default.
- W3171859092 cites W2944847993 @default.
- W3171859092 cites W2950048782 @default.
- W3171859092 cites W2951004968 @default.
- W3171859092 cites W2962695743 @default.
- W3171859092 cites W2962813633 @default.
- W3171859092 cites W2962820504 @default.
- W3171859092 cites W2963090522 @default.
- W3171859092 cites W2963116946 @default.
- W3171859092 cites W2964121744 @default.
- W3171859092 cites W2964307931 @default.
- W3171859092 cites W2970971581 @default.
- W3171859092 cites W2974766464 @default.
- W3171859092 cites W2990857261 @default.
- W3171859092 cites W2990924278 @default.
- W3171859092 cites W2992005611 @default.
- W3171859092 cites W2998758436 @default.
- W3171859092 cites W3000095629 @default.
- W3171859092 cites W3004387786 @default.
- W3171859092 cites W3014279707 @default.
- W3171859092 cites W3015134397 @default.
- W3171859092 cites W3040586665 @default.
- W3171859092 cites W3048134165 @default.
- W3171859092 cites W3048339329 @default.
- W3171859092 cites W3084375081 @default.
- W3171859092 cites W3096012968 @default.
- W3171859092 cites W3096335005 @default.
- W3171859092 cites W3098353287 @default.
- W3171859092 cites W3099140308 @default.
- W3171859092 cites W3099529633 @default.
- W3171859092 cites W3103543904 @default.
- W3171859092 cites W3103963833 @default.
- W3171859092 cites W3105020119 @default.
- W3171859092 cites W3106068426 @default.
- W3171859092 cites W3112908560 @default.
- W3171859092 cites W3122181482 @default.
- W3171859092 cites W3123025874 @default.
- W3171859092 cites W3130974113 @default.
- W3171859092 cites W3132704026 @default.
- W3171859092 cites W3137850141 @default.
- W3171859092 cites W3138128220 @default.
- W3171859092 cites W3139288995 @default.
- W3171859092 cites W3154384838 @default.
- W3171859092 cites W3157901108 @default.
- W3171859092 cites W3170550558 @default.
- W3171859092 cites W586270867 @default.
- W3171859092 cites W91088564 @default.
- W3171859092 hasPublicationYear "2021" @default.
- W3171859092 type Work @default.
- W3171859092 sameAs 3171859092 @default.
- W3171859092 citedByCount "0" @default.
- W3171859092 crossrefType "posted-content" @default.
- W3171859092 hasAuthorship W3171859092A5022362253 @default.
- W3171859092 hasAuthorship W3171859092A5045754297 @default.
- W3171859092 hasAuthorship W3171859092A5054491740 @default.
- W3171859092 hasConcept C111919701 @default.
- W3171859092 hasConcept C119857082 @default.
- W3171859092 hasConcept C134306372 @default.
- W3171859092 hasConcept C136197465 @default.
- W3171859092 hasConcept C140479938 @default.
- W3171859092 hasConcept C154945302 @default.
- W3171859092 hasConcept C167966045 @default.
- W3171859092 hasConcept C17744445 @default.
- W3171859092 hasConcept C199539241 @default.
- W3171859092 hasConcept C202615002 @default.
- W3171859092 hasConcept C2776359362 @default.
- W3171859092 hasConcept C2778572836 @default.
- W3171859092 hasConcept C33923547 @default.
- W3171859092 hasConcept C39890363 @default.
- W3171859092 hasConcept C41008148 @default.
- W3171859092 hasConcept C94625758 @default.
- W3171859092 hasConcept C95623464 @default.
- W3171859092 hasConceptScore W3171859092C111919701 @default.
- W3171859092 hasConceptScore W3171859092C119857082 @default.
- W3171859092 hasConceptScore W3171859092C134306372 @default.
- W3171859092 hasConceptScore W3171859092C136197465 @default.