Matches in SemOpenAlex for { <https://semopenalex.org/work/W2807461489> ?p ?o ?g. }
- W2807461489 abstract "The need for consistent treatment of uncertainty has recently triggered increased interest in probabilistic deep learning methods. However, most current approaches have severe limitations when it comes to inference, since many of these models do not even permit to evaluate exact data likelihoods. Sum-product networks (SPNs), on the other hand, are an excellent architecture in that regard, as they allow to efficiently evaluate likelihoods, as well as arbitrary marginalization and conditioning tasks. Nevertheless, SPNs have not been fully explored as serious deep learning models, likely due to their special structural requirements, which complicate learning. In this paper, we make a drastic simplification and use random SPN structures which are trained in a classical deep learning manner, i.e. employing automatic differentiation, SGD, and GPU support. The resulting models, called RAT-SPNs, yield prediction results comparable to deep neural networks, while still being interpretable as generative model and maintaining well-calibrated uncertainties. This property makes them highly robust under missing input features and enables them to naturally detect outliers and peculiar samples." @default.
- W2807461489 created "2018-06-13" @default.
- W2807461489 creator A5010820865 @default.
- W2807461489 creator A5037636074 @default.
- W2807461489 creator A5057485787 @default.
- W2807461489 creator A5060407332 @default.
- W2807461489 creator A5069110696 @default.
- W2807461489 creator A5085989495 @default.
- W2807461489 creator A5087861569 @default.
- W2807461489 date "2018-09-05" @default.
- W2807461489 modified "2023-10-01" @default.
- W2807461489 title "Probabilistic Deep Learning using Random Sum-Product Networks" @default.
- W2807461489 cites W1533861849 @default.
- W2807461489 cites W1835282959 @default.
- W2807461489 cites W1862639047 @default.
- W2807461489 cites W1880262756 @default.
- W2807461489 cites W1909320841 @default.
- W2807461489 cites W1959608418 @default.
- W2807461489 cites W2044758663 @default.
- W2807461489 cites W2070450685 @default.
- W2807461489 cites W2095705004 @default.
- W2807461489 cites W2099471712 @default.
- W2807461489 cites W2119900738 @default.
- W2807461489 cites W2128744540 @default.
- W2807461489 cites W2135181320 @default.
- W2807461489 cites W2139776446 @default.
- W2807461489 cites W2153074847 @default.
- W2807461489 cites W2158396040 @default.
- W2807461489 cites W2159687189 @default.
- W2807461489 cites W2168906135 @default.
- W2807461489 cites W2170111110 @default.
- W2807461489 cites W2196348858 @default.
- W2807461489 cites W2335728318 @default.
- W2807461489 cites W2399803916 @default.
- W2807461489 cites W2416891345 @default.
- W2807461489 cites W2420777250 @default.
- W2807461489 cites W2735471653 @default.
- W2807461489 cites W2788951710 @default.
- W2807461489 cites W2788983691 @default.
- W2807461489 cites W2949117887 @default.
- W2807461489 cites W2953318193 @default.
- W2807461489 cites W2963276753 @default.
- W2807461489 cites W2964121744 @default.
- W2807461489 doi "https://doi.org/10.17863/cam.26973" @default.
- W2807461489 hasPublicationYear "2018" @default.
- W2807461489 type Work @default.
- W2807461489 sameAs 2807461489 @default.
- W2807461489 citedByCount "11" @default.
- W2807461489 countsByYear W28074614892019 @default.
- W2807461489 countsByYear W28074614892020 @default.
- W2807461489 crossrefType "posted-content" @default.
- W2807461489 hasAuthorship W2807461489A5010820865 @default.
- W2807461489 hasAuthorship W2807461489A5037636074 @default.
- W2807461489 hasAuthorship W2807461489A5057485787 @default.
- W2807461489 hasAuthorship W2807461489A5060407332 @default.
- W2807461489 hasAuthorship W2807461489A5069110696 @default.
- W2807461489 hasAuthorship W2807461489A5085989495 @default.
- W2807461489 hasAuthorship W2807461489A5087861569 @default.
- W2807461489 hasConcept C108583219 @default.
- W2807461489 hasConcept C111472728 @default.
- W2807461489 hasConcept C119857082 @default.
- W2807461489 hasConcept C138885662 @default.
- W2807461489 hasConcept C154945302 @default.
- W2807461489 hasConcept C167966045 @default.
- W2807461489 hasConcept C189950617 @default.
- W2807461489 hasConcept C2524010 @default.
- W2807461489 hasConcept C2776214188 @default.
- W2807461489 hasConcept C2984842247 @default.
- W2807461489 hasConcept C33923547 @default.
- W2807461489 hasConcept C39890363 @default.
- W2807461489 hasConcept C41008148 @default.
- W2807461489 hasConcept C49937458 @default.
- W2807461489 hasConcept C50644808 @default.
- W2807461489 hasConcept C79337645 @default.
- W2807461489 hasConcept C90673727 @default.
- W2807461489 hasConceptScore W2807461489C108583219 @default.
- W2807461489 hasConceptScore W2807461489C111472728 @default.
- W2807461489 hasConceptScore W2807461489C119857082 @default.
- W2807461489 hasConceptScore W2807461489C138885662 @default.
- W2807461489 hasConceptScore W2807461489C154945302 @default.
- W2807461489 hasConceptScore W2807461489C167966045 @default.
- W2807461489 hasConceptScore W2807461489C189950617 @default.
- W2807461489 hasConceptScore W2807461489C2524010 @default.
- W2807461489 hasConceptScore W2807461489C2776214188 @default.
- W2807461489 hasConceptScore W2807461489C2984842247 @default.
- W2807461489 hasConceptScore W2807461489C33923547 @default.
- W2807461489 hasConceptScore W2807461489C39890363 @default.
- W2807461489 hasConceptScore W2807461489C41008148 @default.
- W2807461489 hasConceptScore W2807461489C49937458 @default.
- W2807461489 hasConceptScore W2807461489C50644808 @default.
- W2807461489 hasConceptScore W2807461489C79337645 @default.
- W2807461489 hasConceptScore W2807461489C90673727 @default.
- W2807461489 hasLocation W28074614891 @default.
- W2807461489 hasOpenAccess W2807461489 @default.
- W2807461489 hasPrimaryLocation W28074614891 @default.
- W2807461489 hasRelatedWork W1511986666 @default.
- W2807461489 hasRelatedWork W1831449718 @default.
- W2807461489 hasRelatedWork W2113078405 @default.
- W2807461489 hasRelatedWork W2182706877 @default.
- W2807461489 hasRelatedWork W2196348858 @default.