FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2898043227> ?p ?o ?g. }

• W2898043227 endingPage "55" @default.
• W2898043227 startingPage "43" @default.
• W2898043227 abstract "The variational auto-encoder (VAE) is a powerful and scalable deep generative model. Under the architecture of VAE, the choice of the approximate posterior distribution is one of the crucial issues, and it has a significant impact on tractability and flexibility of the VAE. Generally, latent variables are assumed to be normally distributed with a diagonal covariance matrix, however, it is not flexible enough to match the true complex posterior distribution. We introduce a novel approach to design a flexible and arbitrarily complex approximate posterior distribution. Unlike VAE, firstly, an initial density is constructed by a Gaussian mixture model, and each component has a diagonal covariance matrix. Then this relatively simple distribution is transformed into a more flexible one by applying a sequence of invertible Householder transformations until the desired complexity has been achieved. Additionally, we also give a detailed theoretical and geometric interpretation of Householder transformations. Lastly, due to this change of approximate posterior distribution, the Kullback–Leibler distance between two mixture densities is required to be calculated, but it has no closed form solution. Therefore, we redefine a new variational lower bound by virtue of its upper bound. Compared with other generative models based on similar VAE architecture, our method achieves new state-of-the-art results on benchmark datasets including MNIST, Fashion-MNIST, Omniglot and Histopathology data a more challenging medical images dataset, the experimental results show that our method can improve the flexibility of posterior distribution more effectively." @default.
• W2898043227 created "2018-10-26" @default.
• W2898043227 creator A5012009678 @default.
• W2898043227 creator A5062109513 @default.
• W2898043227 creator A5071458554 @default.
• W2898043227 creator A5078221510 @default.
• W2898043227 creator A5079800756 @default.
• W2898043227 date "2019-01-01" @default.
• W2898043227 modified "2023-10-15" @default.
• W2898043227 title "Variational inference with Gaussian mixture model and householder flow" @default.
• W2898043227 cites W1903029394 @default.
• W2898043227 cites W2033178790 @default.
• W2898043227 cites W2055735912 @default.
• W2898043227 cites W2070792261 @default.
• W2898043227 cites W2076063813 @default.
• W2898043227 cites W2092590148 @default.
• W2898043227 cites W2096192494 @default.
• W2898043227 cites W2104067967 @default.
• W2898043227 cites W2112796928 @default.
• W2898043227 cites W2136922672 @default.
• W2898043227 cites W2152321560 @default.
• W2898043227 cites W2153125595 @default.
• W2898043227 cites W2300687442 @default.
• W2898043227 cites W2542768043 @default.
• W2898043227 cites W2730106296 @default.
• W2898043227 cites W2963223306 @default.
• W2898043227 cites W2963881378 @default.
• W2898043227 doi "https://doi.org/10.1016/j.neunet.2018.10.002" @default.
• W2898043227 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/30408693" @default.
• W2898043227 hasPublicationYear "2019" @default.
• W2898043227 type Work @default.
• W2898043227 sameAs 2898043227 @default.
• W2898043227 citedByCount "14" @default.
• W2898043227 countsByYear W28980432272019 @default.
• W2898043227 countsByYear W28980432272020 @default.
• W2898043227 countsByYear W28980432272021 @default.
• W2898043227 countsByYear W28980432272022 @default.
• W2898043227 countsByYear W28980432272023 @default.
• W2898043227 crossrefType "journal-article" @default.
• W2898043227 hasAuthorship W2898043227A5012009678 @default.
• W2898043227 hasAuthorship W2898043227A5062109513 @default.
• W2898043227 hasAuthorship W2898043227A5071458554 @default.
• W2898043227 hasAuthorship W2898043227A5078221510 @default.
• W2898043227 hasAuthorship W2898043227A5079800756 @default.
• W2898043227 hasConcept C107673813 @default.
• W2898043227 hasConcept C11413529 @default.
• W2898043227 hasConcept C121332964 @default.
• W2898043227 hasConcept C130367717 @default.
• W2898043227 hasConcept C154945302 @default.
• W2898043227 hasConcept C163716315 @default.
• W2898043227 hasConcept C167966045 @default.
• W2898043227 hasConcept C177769412 @default.
• W2898043227 hasConcept C190502265 @default.
• W2898043227 hasConcept C202444582 @default.
• W2898043227 hasConcept C2524010 @default.
• W2898043227 hasConcept C33923547 @default.
• W2898043227 hasConcept C39890363 @default.
• W2898043227 hasConcept C41008148 @default.
• W2898043227 hasConcept C50644808 @default.
• W2898043227 hasConcept C57830394 @default.
• W2898043227 hasConcept C61224824 @default.
• W2898043227 hasConcept C62520636 @default.
• W2898043227 hasConcept C96442724 @default.
• W2898043227 hasConceptScore W2898043227C107673813 @default.
• W2898043227 hasConceptScore W2898043227C11413529 @default.
• W2898043227 hasConceptScore W2898043227C121332964 @default.
• W2898043227 hasConceptScore W2898043227C130367717 @default.
• W2898043227 hasConceptScore W2898043227C154945302 @default.
• W2898043227 hasConceptScore W2898043227C163716315 @default.
• W2898043227 hasConceptScore W2898043227C167966045 @default.
• W2898043227 hasConceptScore W2898043227C177769412 @default.
• W2898043227 hasConceptScore W2898043227C190502265 @default.
• W2898043227 hasConceptScore W2898043227C202444582 @default.
• W2898043227 hasConceptScore W2898043227C2524010 @default.
• W2898043227 hasConceptScore W2898043227C33923547 @default.
• W2898043227 hasConceptScore W2898043227C39890363 @default.
• W2898043227 hasConceptScore W2898043227C41008148 @default.
• W2898043227 hasConceptScore W2898043227C50644808 @default.
• W2898043227 hasConceptScore W2898043227C57830394 @default.
• W2898043227 hasConceptScore W2898043227C61224824 @default.
• W2898043227 hasConceptScore W2898043227C62520636 @default.
• W2898043227 hasConceptScore W2898043227C96442724 @default.
• W2898043227 hasFunder F4320321001 @default.
• W2898043227 hasFunder F4320321793 @default.
• W2898043227 hasFunder F4320325174 @default.
• W2898043227 hasLocation W28980432271 @default.
• W2898043227 hasLocation W28980432272 @default.
• W2898043227 hasOpenAccess W2898043227 @default.
• W2898043227 hasPrimaryLocation W28980432271 @default.
• W2898043227 hasRelatedWork W2885835503 @default.
• W2898043227 hasRelatedWork W2906272760 @default.
• W2898043227 hasRelatedWork W2950475743 @default.
• W2898043227 hasRelatedWork W2961264179 @default.
• W2898043227 hasRelatedWork W2978098801 @default.
• W2898043227 hasRelatedWork W3104832546 @default.
• W2898043227 hasRelatedWork W3160820590 @default.
• W2898043227 hasRelatedWork W4244110343 @default.
• W2898043227 hasRelatedWork W4289105138 @default.

• next