Matches in SemOpenAlex for { <https://semopenalex.org/work/W4384961716> ?p ?o ?g. }
Showing items 1 to 79 of
79
with 100 items per page.
- W4384961716 abstract "Generative Adversarial Networks (GANs) can produce high-quality samples, but do not provide an estimate of the probability density around the samples. However, it has been noted that maximizing the log-likelihood within an energy-based setting can lead to an adversarial framework where the discriminator provides unnormalized density (often called energy). We further develop this perspective, incorporate importance sampling, and show that 1) Wasserstein GAN performs a biased estimate of the partition function, and we propose instead to use an unbiased estimator; and 2) when optimizing for likelihood, one must maximize generator entropy. This is hypothesized to provide a better mode coverage. Different from previous works, we explicitly compute the density of the generated samples. This is the key enabler to designing an unbiased estimator of the partition function and computation of the generator entropy term. The generator density is obtained via a new type of flow network, called one-way flow network, that is less constrained in terms of architecture, as it does not require a tractable inverse function. Our experimental results show that our method converges faster, produces comparable sample quality to GANs with similar architecture, successfully avoids over-fitting to commonly used datasets and produces smooth low-dimensional latent representations of the training data." @default.
- W4384961716 created "2023-07-22" @default.
- W4384961716 creator A5020546938 @default.
- W4384961716 creator A5020938711 @default.
- W4384961716 creator A5028670907 @default.
- W4384961716 creator A5052173004 @default.
- W4384961716 creator A5065396778 @default.
- W4384961716 date "2023-07-19" @default.
- W4384961716 modified "2023-10-05" @default.
- W4384961716 title "Adversarial Likelihood Estimation With One-Way Flows" @default.
- W4384961716 doi "https://doi.org/10.48550/arxiv.2307.09882" @default.
- W4384961716 hasPublicationYear "2023" @default.
- W4384961716 type Work @default.
- W4384961716 citedByCount "0" @default.
- W4384961716 crossrefType "posted-content" @default.
- W4384961716 hasAuthorship W4384961716A5020546938 @default.
- W4384961716 hasAuthorship W4384961716A5020938711 @default.
- W4384961716 hasAuthorship W4384961716A5028670907 @default.
- W4384961716 hasAuthorship W4384961716A5052173004 @default.
- W4384961716 hasAuthorship W4384961716A5065396778 @default.
- W4384961716 hasBestOaLocation W43849617161 @default.
- W4384961716 hasConcept C105795698 @default.
- W4384961716 hasConcept C106301342 @default.
- W4384961716 hasConcept C11413529 @default.
- W4384961716 hasConcept C114614502 @default.
- W4384961716 hasConcept C121332964 @default.
- W4384961716 hasConcept C126255220 @default.
- W4384961716 hasConcept C163258240 @default.
- W4384961716 hasConcept C185429906 @default.
- W4384961716 hasConcept C189508267 @default.
- W4384961716 hasConcept C19499675 @default.
- W4384961716 hasConcept C197055811 @default.
- W4384961716 hasConcept C2779803651 @default.
- W4384961716 hasConcept C2780992000 @default.
- W4384961716 hasConcept C33923547 @default.
- W4384961716 hasConcept C41008148 @default.
- W4384961716 hasConcept C42812 @default.
- W4384961716 hasConcept C45374587 @default.
- W4384961716 hasConcept C52740198 @default.
- W4384961716 hasConcept C62520636 @default.
- W4384961716 hasConcept C76155785 @default.
- W4384961716 hasConcept C94915269 @default.
- W4384961716 hasConceptScore W4384961716C105795698 @default.
- W4384961716 hasConceptScore W4384961716C106301342 @default.
- W4384961716 hasConceptScore W4384961716C11413529 @default.
- W4384961716 hasConceptScore W4384961716C114614502 @default.
- W4384961716 hasConceptScore W4384961716C121332964 @default.
- W4384961716 hasConceptScore W4384961716C126255220 @default.
- W4384961716 hasConceptScore W4384961716C163258240 @default.
- W4384961716 hasConceptScore W4384961716C185429906 @default.
- W4384961716 hasConceptScore W4384961716C189508267 @default.
- W4384961716 hasConceptScore W4384961716C19499675 @default.
- W4384961716 hasConceptScore W4384961716C197055811 @default.
- W4384961716 hasConceptScore W4384961716C2779803651 @default.
- W4384961716 hasConceptScore W4384961716C2780992000 @default.
- W4384961716 hasConceptScore W4384961716C33923547 @default.
- W4384961716 hasConceptScore W4384961716C41008148 @default.
- W4384961716 hasConceptScore W4384961716C42812 @default.
- W4384961716 hasConceptScore W4384961716C45374587 @default.
- W4384961716 hasConceptScore W4384961716C52740198 @default.
- W4384961716 hasConceptScore W4384961716C62520636 @default.
- W4384961716 hasConceptScore W4384961716C76155785 @default.
- W4384961716 hasConceptScore W4384961716C94915269 @default.
- W4384961716 hasLocation W43849617161 @default.
- W4384961716 hasOpenAccess W4384961716 @default.
- W4384961716 hasPrimaryLocation W43849617161 @default.
- W4384961716 hasRelatedWork W1528820368 @default.
- W4384961716 hasRelatedWork W1569550976 @default.
- W4384961716 hasRelatedWork W1570464650 @default.
- W4384961716 hasRelatedWork W1937731701 @default.
- W4384961716 hasRelatedWork W2043285515 @default.
- W4384961716 hasRelatedWork W2061808167 @default.
- W4384961716 hasRelatedWork W2187148935 @default.
- W4384961716 hasRelatedWork W2543710962 @default.
- W4384961716 hasRelatedWork W2585385340 @default.
- W4384961716 hasRelatedWork W63219142 @default.
- W4384961716 isParatext "false" @default.
- W4384961716 isRetracted "false" @default.
- W4384961716 workType "article" @default.