Matches in SemOpenAlex for { <https://semopenalex.org/work/W2951187696> ?p ?o ?g. }
Showing items 1 to 72 of
72
with 100 items per page.
- W2951187696 abstract "Inference and learning for probabilistic generative networks is often very challenging and typically prevents scalability to as large networks as used for deep discriminative approaches. To obtain efficiently trainable, large-scale and well performing generative networks for semi-supervised learning, we here combine two recent developments: a neural network reformulation of hierarchical Poisson mixtures (Neural Simpletrons), and a novel truncated variational EM approach (TV-EM). TV-EM provides theoretical guarantees for learning in generative networks, and its application to Neural Simpletrons results in particularly compact, yet approximately optimal, modifications of learning equations. If applied to standard benchmarks, we empirically find, that learning converges in fewer EM iterations, that the complexity per EM iteration is reduced, and that final likelihood values are higher on average. For the task of classification on data sets with few labels, learning improvements result in consistently lower error rates if compared to applications without truncation. Experiments on the MNIST data set herein allow for comparison to standard and state-of-the-art models in the semi-supervised setting. Further experiments on the NIST SD19 data set show the scalability of the approach when a manifold of additional unlabeled data is available." @default.
- W2951187696 created "2019-06-27" @default.
- W2951187696 creator A5022342273 @default.
- W2951187696 creator A5023612311 @default.
- W2951187696 date "2017-02-07" @default.
- W2951187696 modified "2023-09-27" @default.
- W2951187696 title "Truncated Variational EM for Semi-Supervised Neural Simpletrons" @default.
- W2951187696 hasPublicationYear "2017" @default.
- W2951187696 type Work @default.
- W2951187696 sameAs 2951187696 @default.
- W2951187696 citedByCount "1" @default.
- W2951187696 countsByYear W29511876962017 @default.
- W2951187696 crossrefType "posted-content" @default.
- W2951187696 hasAuthorship W2951187696A5022342273 @default.
- W2951187696 hasAuthorship W2951187696A5023612311 @default.
- W2951187696 hasConcept C106195933 @default.
- W2951187696 hasConcept C119857082 @default.
- W2951187696 hasConcept C154945302 @default.
- W2951187696 hasConcept C167966045 @default.
- W2951187696 hasConcept C177264268 @default.
- W2951187696 hasConcept C190502265 @default.
- W2951187696 hasConcept C199360897 @default.
- W2951187696 hasConcept C2776214188 @default.
- W2951187696 hasConcept C39890363 @default.
- W2951187696 hasConcept C41008148 @default.
- W2951187696 hasConcept C48044578 @default.
- W2951187696 hasConcept C49937458 @default.
- W2951187696 hasConcept C50644808 @default.
- W2951187696 hasConcept C77088390 @default.
- W2951187696 hasConcept C97931131 @default.
- W2951187696 hasConceptScore W2951187696C106195933 @default.
- W2951187696 hasConceptScore W2951187696C119857082 @default.
- W2951187696 hasConceptScore W2951187696C154945302 @default.
- W2951187696 hasConceptScore W2951187696C167966045 @default.
- W2951187696 hasConceptScore W2951187696C177264268 @default.
- W2951187696 hasConceptScore W2951187696C190502265 @default.
- W2951187696 hasConceptScore W2951187696C199360897 @default.
- W2951187696 hasConceptScore W2951187696C2776214188 @default.
- W2951187696 hasConceptScore W2951187696C39890363 @default.
- W2951187696 hasConceptScore W2951187696C41008148 @default.
- W2951187696 hasConceptScore W2951187696C48044578 @default.
- W2951187696 hasConceptScore W2951187696C49937458 @default.
- W2951187696 hasConceptScore W2951187696C50644808 @default.
- W2951187696 hasConceptScore W2951187696C77088390 @default.
- W2951187696 hasConceptScore W2951187696C97931131 @default.
- W2951187696 hasLocation W29511876961 @default.
- W2951187696 hasOpenAccess W2951187696 @default.
- W2951187696 hasPrimaryLocation W29511876961 @default.
- W2951187696 hasRelatedWork W2044544672 @default.
- W2951187696 hasRelatedWork W2057869818 @default.
- W2951187696 hasRelatedWork W2073614810 @default.
- W2951187696 hasRelatedWork W2099987944 @default.
- W2951187696 hasRelatedWork W2101229103 @default.
- W2951187696 hasRelatedWork W2112008717 @default.
- W2951187696 hasRelatedWork W2129761946 @default.
- W2951187696 hasRelatedWork W2405274037 @default.
- W2951187696 hasRelatedWork W2564801728 @default.
- W2951187696 hasRelatedWork W2587488103 @default.
- W2951187696 hasRelatedWork W2606711537 @default.
- W2951187696 hasRelatedWork W2909206416 @default.
- W2951187696 hasRelatedWork W2943335249 @default.
- W2951187696 hasRelatedWork W2963435192 @default.
- W2951187696 hasRelatedWork W3004391146 @default.
- W2951187696 hasRelatedWork W3141470656 @default.
- W2951187696 hasRelatedWork W3158076588 @default.
- W2951187696 hasRelatedWork W3160822690 @default.
- W2951187696 hasRelatedWork W3182947351 @default.
- W2951187696 hasRelatedWork W3213422582 @default.
- W2951187696 isParatext "false" @default.
- W2951187696 isRetracted "false" @default.
- W2951187696 magId "2951187696" @default.
- W2951187696 workType "article" @default.