Matches in SemOpenAlex for { <https://semopenalex.org/work/W2949457404> ?p ?o ?g. }
- W2949457404 abstract "We show that a generative random field model, which we call generative ConvNet, can be derived from the commonly used discriminative ConvNet, by assuming a ConvNet for multi-category classification and assuming one of the categories is a base category generated by a reference distribution. If we further assume that the non-linearity in the ConvNet is Rectified Linear Unit (ReLU) and the reference distribution is Gaussian white noise, then we obtain a generative ConvNet model that is unique among energy-based models: The model is piecewise Gaussian, and the means of the Gaussian pieces are defined by an auto-encoder, where the filters in the bottom-up encoding become the basis functions in the top-down decoding, and the binary activation variables detected by the filters in the bottom-up convolution process become the coefficients of the basis functions in the top-down deconvolution process. The Langevin dynamics for sampling the generative ConvNet is driven by the reconstruction error of this auto-encoder. The contrastive divergence learning of the generative ConvNet reconstructs the training images by the auto-encoder. The maximum likelihood learning algorithm can synthesize realistic natural image patterns." @default.
- W2949457404 created "2019-06-27" @default.
- W2949457404 creator A5034228010 @default.
- W2949457404 creator A5041455283 @default.
- W2949457404 creator A5055935194 @default.
- W2949457404 creator A5064307909 @default.
- W2949457404 date "2016-02-10" @default.
- W2949457404 modified "2023-09-27" @default.
- W2949457404 title "A Theory of Generative ConvNet" @default.
- W2949457404 cites W1505878979 @default.
- W2949457404 cites W1802356529 @default.
- W2949457404 cites W189596042 @default.
- W2949457404 cites W1959608418 @default.
- W2949457404 cites W1984469511 @default.
- W2949457404 cites W1990838964 @default.
- W2949457404 cites W1993845689 @default.
- W2949457404 cites W2013035813 @default.
- W2949457404 cites W2023443487 @default.
- W2949457404 cites W2052961007 @default.
- W2949457404 cites W2079930597 @default.
- W2949457404 cites W2099972257 @default.
- W2949457404 cites W2105464873 @default.
- W2949457404 cites W2108581046 @default.
- W2949457404 cites W2112796928 @default.
- W2949457404 cites W2116064496 @default.
- W2949457404 cites W2124601597 @default.
- W2949457404 cites W2126174118 @default.
- W2949457404 cites W2128084896 @default.
- W2949457404 cites W2130325614 @default.
- W2949457404 cites W2131686571 @default.
- W2949457404 cites W2140622310 @default.
- W2949457404 cites W2141200610 @default.
- W2949457404 cites W2161914416 @default.
- W2949457404 cites W2185528074 @default.
- W2949457404 cites W2191540403 @default.
- W2949457404 cites W2206858481 @default.
- W2949457404 cites W2272361785 @default.
- W2949457404 cites W2614634292 @default.
- W2949457404 cites W2951603627 @default.
- W2949457404 cites W2952264928 @default.
- W2949457404 cites W2953066166 @default.
- W2949457404 cites W2963846894 @default.
- W2949457404 hasPublicationYear "2016" @default.
- W2949457404 type Work @default.
- W2949457404 sameAs 2949457404 @default.
- W2949457404 citedByCount "26" @default.
- W2949457404 countsByYear W29494574042016 @default.
- W2949457404 countsByYear W29494574042017 @default.
- W2949457404 countsByYear W29494574042018 @default.
- W2949457404 countsByYear W29494574042019 @default.
- W2949457404 countsByYear W29494574042020 @default.
- W2949457404 countsByYear W29494574042021 @default.
- W2949457404 countsByYear W29494574042022 @default.
- W2949457404 crossrefType "posted-content" @default.
- W2949457404 hasAuthorship W2949457404A5034228010 @default.
- W2949457404 hasAuthorship W2949457404A5041455283 @default.
- W2949457404 hasAuthorship W2949457404A5055935194 @default.
- W2949457404 hasAuthorship W2949457404A5064307909 @default.
- W2949457404 hasConcept C11413529 @default.
- W2949457404 hasConcept C121332964 @default.
- W2949457404 hasConcept C12426560 @default.
- W2949457404 hasConcept C153180895 @default.
- W2949457404 hasConcept C154945302 @default.
- W2949457404 hasConcept C163716315 @default.
- W2949457404 hasConcept C167966045 @default.
- W2949457404 hasConcept C174576160 @default.
- W2949457404 hasConcept C2524010 @default.
- W2949457404 hasConcept C33923547 @default.
- W2949457404 hasConcept C39890363 @default.
- W2949457404 hasConcept C41008148 @default.
- W2949457404 hasConcept C45347329 @default.
- W2949457404 hasConcept C50644808 @default.
- W2949457404 hasConcept C62520636 @default.
- W2949457404 hasConcept C97931131 @default.
- W2949457404 hasConceptScore W2949457404C11413529 @default.
- W2949457404 hasConceptScore W2949457404C121332964 @default.
- W2949457404 hasConceptScore W2949457404C12426560 @default.
- W2949457404 hasConceptScore W2949457404C153180895 @default.
- W2949457404 hasConceptScore W2949457404C154945302 @default.
- W2949457404 hasConceptScore W2949457404C163716315 @default.
- W2949457404 hasConceptScore W2949457404C167966045 @default.
- W2949457404 hasConceptScore W2949457404C174576160 @default.
- W2949457404 hasConceptScore W2949457404C2524010 @default.
- W2949457404 hasConceptScore W2949457404C33923547 @default.
- W2949457404 hasConceptScore W2949457404C39890363 @default.
- W2949457404 hasConceptScore W2949457404C41008148 @default.
- W2949457404 hasConceptScore W2949457404C45347329 @default.
- W2949457404 hasConceptScore W2949457404C50644808 @default.
- W2949457404 hasConceptScore W2949457404C62520636 @default.
- W2949457404 hasConceptScore W2949457404C97931131 @default.
- W2949457404 hasLocation W29494574041 @default.
- W2949457404 hasOpenAccess W2949457404 @default.
- W2949457404 hasPrimaryLocation W29494574041 @default.
- W2949457404 hasRelatedWork W1522301498 @default.
- W2949457404 hasRelatedWork W1834627138 @default.
- W2949457404 hasRelatedWork W1893585201 @default.
- W2949457404 hasRelatedWork W1993845689 @default.
- W2949457404 hasRelatedWork W2099471712 @default.
- W2949457404 hasRelatedWork W2116064496 @default.
- W2949457404 hasRelatedWork W2126174118 @default.