FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3012026596> ?p ?o ?g. }

• W3012026596 abstract "We show that the sum of the implicit generator log-density $log p_g$ of a GAN with the logit score of the discriminator defines an energy function which yields the true data density when the generator is imperfect but the discriminator is optimal, thus making it possible to improve on the typical generator (with implicit density $p_g$). To make that practical, we show that sampling from this modified density can be achieved by sampling in latent space according to an energy-based model induced by the sum of the latent prior log-density and the discriminator output score. This can be achieved by running a Langevin MCMC in latent space and then applying the generator function, which we call Discriminator Driven Latent Sampling~(DDLS). We show that DDLS is highly efficient compared to previous methods which work in the high-dimensional pixel space and can be applied to improve on previously trained GANs of many types. We evaluate DDLS on both synthetic and real-world datasets qualitatively and quantitatively. On CIFAR-10, DDLS substantially improves the Inception Score of an off-the-shelf pre-trained SN-GAN~citep{sngan} from $8.22$ to $9.09$ which is even comparable to the class-conditional BigGAN~citep{biggan} model. This achieves a new state-of-the-art in unconditional image synthesis setting without introducing extra parameters or additional training." @default.
• W3012026596 created "2020-03-23" @default.
• W3012026596 creator A5019862529 @default.
• W3012026596 creator A5037065865 @default.
• W3012026596 creator A5057271863 @default.
• W3012026596 creator A5064986566 @default.
• W3012026596 creator A5085080733 @default.
• W3012026596 creator A5086198262 @default.
• W3012026596 creator A5086329948 @default.
• W3012026596 date "2020-03-12" @default.
• W3012026596 modified "2023-10-01" @default.
• W3012026596 title "Your GAN is Secretly an Energy-based Model and You Should use Discriminator Driven Latent Sampling" @default.
• W3012026596 cites W1496559305 @default.
• W3012026596 cites W1505878979 @default.
• W3012026596 cites W1578208063 @default.
• W3012026596 cites W189596042 @default.
• W3012026596 cites W1970391837 @default.
• W3012026596 cites W2099471712 @default.
• W3012026596 cites W2100495367 @default.
• W3012026596 cites W2116064496 @default.
• W3012026596 cites W2116825644 @default.
• W3012026596 cites W2152790380 @default.
• W3012026596 cites W2161914416 @default.
• W3012026596 cites W2416272112 @default.
• W3012026596 cites W2519536754 @default.
• W3012026596 cites W2521028896 @default.
• W3012026596 cites W2557449848 @default.
• W3012026596 cites W2566467060 @default.
• W3012026596 cites W2606176153 @default.
• W3012026596 cites W2788289427 @default.
• W3012026596 cites W2796731119 @default.
• W3012026596 cites W2883147591 @default.
• W3012026596 cites W2893749619 @default.
• W3012026596 cites W2914221386 @default.
• W3012026596 cites W2920868047 @default.
• W3012026596 cites W2950247758 @default.
• W3012026596 cites W2951875216 @default.
• W3012026596 cites W2962760235 @default.
• W3012026596 cites W2962771722 @default.
• W3012026596 cites W2962793481 @default.
• W3012026596 cites W2962879692 @default.
• W3012026596 cites W2963082441 @default.
• W3012026596 cites W2963170156 @default.
• W3012026596 cites W2963277051 @default.
• W3012026596 cites W2963373786 @default.
• W3012026596 cites W2963444790 @default.
• W3012026596 cites W2963636093 @default.
• W3012026596 cites W2963684088 @default.
• W3012026596 cites W2963836885 @default.
• W3012026596 cites W2963971375 @default.
• W3012026596 cites W2963981733 @default.
• W3012026596 cites W2964151866 @default.
• W3012026596 cites W2965850471 @default.
• W3012026596 cites W2970049491 @default.
• W3012026596 cites W2970607325 @default.
• W3012026596 cites W2970747737 @default.
• W3012026596 cites W2970803184 @default.
• W3012026596 cites W2970938715 @default.
• W3012026596 cites W2971034910 @default.
• W3012026596 cites W2971161260 @default.
• W3012026596 cites W2979473621 @default.
• W3012026596 cites W2989574682 @default.
• W3012026596 cites W2990706204 @default.
• W3012026596 cites W2995085126 @default.
• W3012026596 cites W2995327724 @default.
• W3012026596 cites W2995890624 @default.
• W3012026596 cites W3035101153 @default.
• W3012026596 cites W3035166812 @default.
• W3012026596 cites W3102021454 @default.
• W3012026596 cites W3140968660 @default.
• W3012026596 cites W632087424 @default.
• W3012026596 hasPublicationYear "2020" @default.
• W3012026596 type Work @default.
• W3012026596 sameAs 3012026596 @default.
• W3012026596 citedByCount "0" @default.
• W3012026596 crossrefType "posted-content" @default.
• W3012026596 hasAuthorship W3012026596A5019862529 @default.
• W3012026596 hasAuthorship W3012026596A5037065865 @default.
• W3012026596 hasAuthorship W3012026596A5057271863 @default.
• W3012026596 hasAuthorship W3012026596A5064986566 @default.
• W3012026596 hasAuthorship W3012026596A5085080733 @default.
• W3012026596 hasAuthorship W3012026596A5086198262 @default.
• W3012026596 hasAuthorship W3012026596A5086329948 @default.
• W3012026596 hasConcept C105795698 @default.
• W3012026596 hasConcept C111919701 @default.
• W3012026596 hasConcept C11413529 @default.
• W3012026596 hasConcept C121332964 @default.
• W3012026596 hasConcept C140779682 @default.
• W3012026596 hasConcept C163258240 @default.
• W3012026596 hasConcept C186370098 @default.
• W3012026596 hasConcept C2778572836 @default.
• W3012026596 hasConcept C2779803651 @default.
• W3012026596 hasConcept C2780992000 @default.
• W3012026596 hasConcept C28826006 @default.
• W3012026596 hasConcept C33923547 @default.
• W3012026596 hasConcept C41008148 @default.
• W3012026596 hasConcept C62520636 @default.
• W3012026596 hasConcept C73586568 @default.
• W3012026596 hasConcept C76155785 @default.
• W3012026596 hasConcept C94915269 @default.

• next