Matches in SemOpenAlex for { <https://semopenalex.org/work/W2912119436> ?p ?o ?g. }
Showing items 1 to 80 of
80
with 100 items per page.
- W2912119436 abstract "Standard neural networks are often overconfident when presented with data outside the training distribution. We introduce HyperGAN, a new generative model for learning a distribution of neural network parameters. HyperGAN does not require restrictive assumptions on priors, and networks sampled from it can be used to quickly create very large and diverse ensembles. HyperGAN employs a novel mixer to project prior samples to a latent space with correlated dimensions, and samples from the latent space are then used to generate weights for each layer of a deep neural network. We show that HyperGAN can learn to generate parameters which label the MNIST and CIFAR-10 datasets with competitive performance to fully supervised learning, while learning a rich distribution of effective parameters. We also show that HyperGAN can also provide better uncertainty estimates than standard ensembles by evaluating on out of distribution data as well as adversarial examples." @default.
- W2912119436 created "2019-02-21" @default.
- W2912119436 creator A5065084526 @default.
- W2912119436 creator A5076851527 @default.
- W2912119436 date "2019-01-30" @default.
- W2912119436 modified "2023-10-04" @default.
- W2912119436 title "HyperGAN: A Generative Model for Diverse, Performant Neural Networks" @default.
- W2912119436 cites W1525859397 @default.
- W2912119436 cites W1719489212 @default.
- W2912119436 cites W1899249567 @default.
- W2912119436 cites W2099471712 @default.
- W2912119436 cites W2414711238 @default.
- W2912119436 cites W2535697732 @default.
- W2912119436 cites W2765725061 @default.
- W2912119436 cites W2767449908 @default.
- W2912119436 cites W2786022758 @default.
- W2912119436 cites W2951934643 @default.
- W2912119436 cites W2963207607 @default.
- W2912119436 cites W2963238274 @default.
- W2912119436 cites W2963711523 @default.
- W2912119436 cites W2964059111 @default.
- W2912119436 cites W2964155212 @default.
- W2912119436 cites W2964253222 @default.
- W2912119436 cites W2995099985 @default.
- W2912119436 cites W603908379 @default.
- W2912119436 doi "https://doi.org/10.48550/arxiv.1901.11058" @default.
- W2912119436 hasPublicationYear "2019" @default.
- W2912119436 type Work @default.
- W2912119436 sameAs 2912119436 @default.
- W2912119436 citedByCount "6" @default.
- W2912119436 countsByYear W29121194362019 @default.
- W2912119436 countsByYear W29121194362020 @default.
- W2912119436 countsByYear W29121194362021 @default.
- W2912119436 crossrefType "posted-content" @default.
- W2912119436 hasAuthorship W2912119436A5065084526 @default.
- W2912119436 hasAuthorship W2912119436A5076851527 @default.
- W2912119436 hasBestOaLocation W29121194361 @default.
- W2912119436 hasConcept C107673813 @default.
- W2912119436 hasConcept C110121322 @default.
- W2912119436 hasConcept C111919701 @default.
- W2912119436 hasConcept C119857082 @default.
- W2912119436 hasConcept C134306372 @default.
- W2912119436 hasConcept C154945302 @default.
- W2912119436 hasConcept C177769412 @default.
- W2912119436 hasConcept C190502265 @default.
- W2912119436 hasConcept C2778572836 @default.
- W2912119436 hasConcept C33923547 @default.
- W2912119436 hasConcept C39890363 @default.
- W2912119436 hasConcept C41008148 @default.
- W2912119436 hasConcept C50644808 @default.
- W2912119436 hasConceptScore W2912119436C107673813 @default.
- W2912119436 hasConceptScore W2912119436C110121322 @default.
- W2912119436 hasConceptScore W2912119436C111919701 @default.
- W2912119436 hasConceptScore W2912119436C119857082 @default.
- W2912119436 hasConceptScore W2912119436C134306372 @default.
- W2912119436 hasConceptScore W2912119436C154945302 @default.
- W2912119436 hasConceptScore W2912119436C177769412 @default.
- W2912119436 hasConceptScore W2912119436C190502265 @default.
- W2912119436 hasConceptScore W2912119436C2778572836 @default.
- W2912119436 hasConceptScore W2912119436C33923547 @default.
- W2912119436 hasConceptScore W2912119436C39890363 @default.
- W2912119436 hasConceptScore W2912119436C41008148 @default.
- W2912119436 hasConceptScore W2912119436C50644808 @default.
- W2912119436 hasLocation W29121194361 @default.
- W2912119436 hasOpenAccess W2912119436 @default.
- W2912119436 hasPrimaryLocation W29121194361 @default.
- W2912119436 hasRelatedWork W2597787948 @default.
- W2912119436 hasRelatedWork W2906272760 @default.
- W2912119436 hasRelatedWork W2936783136 @default.
- W2912119436 hasRelatedWork W2947175736 @default.
- W2912119436 hasRelatedWork W2961085424 @default.
- W2912119436 hasRelatedWork W2980541498 @default.
- W2912119436 hasRelatedWork W2996316059 @default.
- W2912119436 hasRelatedWork W3008818413 @default.
- W2912119436 hasRelatedWork W4306674287 @default.
- W2912119436 hasRelatedWork W4310699748 @default.
- W2912119436 isParatext "false" @default.
- W2912119436 isRetracted "false" @default.
- W2912119436 magId "2912119436" @default.
- W2912119436 workType "article" @default.