FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2789156501> ?p ?o ?g. }

• W2789156501 abstract "Deep Neural Networks (DNNs) provide state-of-the-art solutions in several difficult machine perceptual tasks. However, their performance relies on the availability of a large set of labeled training data, which limits the breadth of their applicability. Hence, there is a need for new {em semi-supervised learning} methods for DNNs that can leverage both (a small amount of) labeled and unlabeled training data. In this paper, we develop a general loss function enabling DNNs of any topology to be trained in a semi-supervised manner without extra hyper-parameters. As opposed to current semi-supervised techniques based on topology-specific or unstable approaches, ours is both robust and general. We demonstrate that our approach reaches state-of-the-art performance on the SVHN ($9.82%$ test error, with $500$ labels and wide Resnet) and CIFAR10 (16.38% test error, with 8000 labels and sigmoid convolutional neural network) data sets." @default.
• W2789156501 created "2018-03-06" @default.
• W2789156501 creator A5019838252 @default.
• W2789156501 creator A5047293370 @default.
• W2789156501 creator A5072713767 @default.
• W2789156501 date "2018-02-27" @default.
• W2789156501 modified "2023-09-27" @default.
• W2789156501 title "Semi-Supervised Learning Enabled by Multiscale Deep Neural Network Inversion" @default.
• W2789156501 cites W1522301498 @default.
• W2789156501 cites W1810943226 @default.
• W2789156501 cites W1839408879 @default.
• W2789156501 cites W1849277567 @default.
• W2789156501 cites W1921523184 @default.
• W2789156501 cites W1924770834 @default.
• W2789156501 cites W2039708501 @default.
• W2789156501 cites W2064675550 @default.
• W2789156501 cites W2066551529 @default.
• W2789156501 cites W2068375500 @default.
• W2789156501 cites W2095705004 @default.
• W2789156501 cites W2136655611 @default.
• W2789156501 cites W2140833774 @default.
• W2789156501 cites W2160208155 @default.
• W2789156501 cites W2171355207 @default.
• W2789156501 cites W2178768799 @default.
• W2789156501 cites W2280377497 @default.
• W2789156501 cites W2293078015 @default.
• W2789156501 cites W2401231614 @default.
• W2789156501 cites W2432004435 @default.
• W2789156501 cites W2523060838 @default.
• W2789156501 cites W2596763562 @default.
• W2789156501 cites W2619371851 @default.
• W2789156501 cites W2801466482 @default.
• W2789156501 cites W2949117887 @default.
• W2789156501 cites W2949416428 @default.
• W2789156501 cites W2951533962 @default.
• W2789156501 cites W2951970475 @default.
• W2789156501 cites W2952229419 @default.
• W2789156501 cites W2964040467 @default.
• W2789156501 cites W581956982 @default.
• W2789156501 cites W6908809 @default.
• W2789156501 cites W2803403308 @default.
• W2789156501 hasPublicationYear "2018" @default.
• W2789156501 type Work @default.
• W2789156501 sameAs 2789156501 @default.
• W2789156501 citedByCount "2" @default.
• W2789156501 countsByYear W27891565012019 @default.
• W2789156501 countsByYear W27891565012021 @default.
• W2789156501 crossrefType "posted-content" @default.
• W2789156501 hasAuthorship W2789156501A5019838252 @default.
• W2789156501 hasAuthorship W2789156501A5047293370 @default.
• W2789156501 hasAuthorship W2789156501A5072713767 @default.
• W2789156501 hasConcept C119857082 @default.
• W2789156501 hasConcept C136389625 @default.
• W2789156501 hasConcept C153083717 @default.
• W2789156501 hasConcept C153180895 @default.
• W2789156501 hasConcept C154945302 @default.
• W2789156501 hasConcept C16910744 @default.
• W2789156501 hasConcept C169903167 @default.
• W2789156501 hasConcept C199360897 @default.
• W2789156501 hasConcept C2776145971 @default.
• W2789156501 hasConcept C2944601119 @default.
• W2789156501 hasConcept C2984842247 @default.
• W2789156501 hasConcept C41008148 @default.
• W2789156501 hasConcept C50644808 @default.
• W2789156501 hasConcept C51632099 @default.
• W2789156501 hasConcept C81363708 @default.
• W2789156501 hasConcept C81388566 @default.
• W2789156501 hasConceptScore W2789156501C119857082 @default.
• W2789156501 hasConceptScore W2789156501C136389625 @default.
• W2789156501 hasConceptScore W2789156501C153083717 @default.
• W2789156501 hasConceptScore W2789156501C153180895 @default.
• W2789156501 hasConceptScore W2789156501C154945302 @default.
• W2789156501 hasConceptScore W2789156501C16910744 @default.
• W2789156501 hasConceptScore W2789156501C169903167 @default.
• W2789156501 hasConceptScore W2789156501C199360897 @default.
• W2789156501 hasConceptScore W2789156501C2776145971 @default.
• W2789156501 hasConceptScore W2789156501C2944601119 @default.
• W2789156501 hasConceptScore W2789156501C2984842247 @default.
• W2789156501 hasConceptScore W2789156501C41008148 @default.
• W2789156501 hasConceptScore W2789156501C50644808 @default.
• W2789156501 hasConceptScore W2789156501C51632099 @default.
• W2789156501 hasConceptScore W2789156501C81363708 @default.
• W2789156501 hasConceptScore W2789156501C81388566 @default.
• W2789156501 hasLocation W27891565011 @default.
• W2789156501 hasOpenAccess W2789156501 @default.
• W2789156501 hasPrimaryLocation W27891565011 @default.
• W2789156501 hasRelatedWork W2171111089 @default.
• W2789156501 hasRelatedWork W2209990155 @default.
• W2789156501 hasRelatedWork W2285131997 @default.
• W2789156501 hasRelatedWork W2430148947 @default.
• W2789156501 hasRelatedWork W2462901929 @default.
• W2789156501 hasRelatedWork W2525526117 @default.
• W2789156501 hasRelatedWork W2794759802 @default.
• W2789156501 hasRelatedWork W2801525523 @default.
• W2789156501 hasRelatedWork W2903047678 @default.
• W2789156501 hasRelatedWork W2912093838 @default.
• W2789156501 hasRelatedWork W2945906332 @default.
• W2789156501 hasRelatedWork W2952719721 @default.
• W2789156501 hasRelatedWork W2958392838 @default.
• W2789156501 hasRelatedWork W2962817490 @default.

• next