Matches in SemOpenAlex for { <https://semopenalex.org/work/W2069291384> ?p ?o ?g. }
Showing items 1 to 95 of
95
with 100 items per page.
- W2069291384 endingPage "535" @default.
- W2069291384 startingPage "525" @default.
- W2069291384 abstract "A neural network learning paradigm based on information theory is proposed as a way to perform, in an unsupervised fashion, redundancy reduction among the elements of the output layer without loss of information from the sensory input. The model developed performs nonlinear decorrelation up to higher orders of the cumulant tensors and results in probabilistically independent components of the output layer. This means that we don't need to assume Gaussian distribution at either the input or the output. The theory presented is related to the unsupervised learning theory of Barlow, which proposes redundancy reduction as the goal of cognition. When nonlinear units are used (sigmoid or higher-order pi-neurons), nonlinear principal component analysis is obtained. In this case, nonlinear manifolds can be reduced to minimum dimension manifolds. If such units are used the network performs a generalized principal component analysis in the sense that non-Gaussian distributions can be linearly decorrelated and higher orders of the correlation tensors are also taken into account. The basic structure of the architecture involves a general transformation that is volume conserving and therefore the entropy, yielding a map without loss of information. Minimization of the mutual information among the output neurons eliminates the redundancy between the outputs and results in statistical decorrelation of the extracted features. This is known as factorial learning. To sum up, this paper presents a model of factorial learning for general nonlinear transformations of an arbitrary non-Gaussian (or Gaussian) environment with statistically nonlinearly correlated input. Simulations demonstrate the effectiveness of this method." @default.
- W2069291384 created "2016-06-24" @default.
- W2069291384 creator A5047963275 @default.
- W2069291384 creator A5087996237 @default.
- W2069291384 date "1995-01-01" @default.
- W2069291384 modified "2023-09-25" @default.
- W2069291384 title "Nonlinear higher-order statistical decorrelation by volume-conserving neural architectures" @default.
- W2069291384 cites W1971074050 @default.
- W2069291384 cites W1979191599 @default.
- W2069291384 cites W2011723901 @default.
- W2069291384 cites W2026295629 @default.
- W2069291384 cites W2042925217 @default.
- W2069291384 cites W2066426992 @default.
- W2069291384 cites W2081070226 @default.
- W2069291384 cites W2122925692 @default.
- W2069291384 cites W2123806929 @default.
- W2069291384 cites W2144914634 @default.
- W2069291384 cites W2151351982 @default.
- W2069291384 cites W4240035482 @default.
- W2069291384 doi "https://doi.org/10.1016/0893-6080(94)00108-x" @default.
- W2069291384 hasPublicationYear "1995" @default.
- W2069291384 type Work @default.
- W2069291384 sameAs 2069291384 @default.
- W2069291384 citedByCount "115" @default.
- W2069291384 countsByYear W20692913842013 @default.
- W2069291384 countsByYear W20692913842016 @default.
- W2069291384 countsByYear W20692913842017 @default.
- W2069291384 countsByYear W20692913842018 @default.
- W2069291384 countsByYear W20692913842019 @default.
- W2069291384 countsByYear W20692913842020 @default.
- W2069291384 countsByYear W20692913842021 @default.
- W2069291384 countsByYear W20692913842022 @default.
- W2069291384 countsByYear W20692913842023 @default.
- W2069291384 crossrefType "journal-article" @default.
- W2069291384 hasAuthorship W2069291384A5047963275 @default.
- W2069291384 hasAuthorship W2069291384A5087996237 @default.
- W2069291384 hasConcept C105795698 @default.
- W2069291384 hasConcept C106301342 @default.
- W2069291384 hasConcept C111919701 @default.
- W2069291384 hasConcept C11413529 @default.
- W2069291384 hasConcept C121332964 @default.
- W2069291384 hasConcept C152124472 @default.
- W2069291384 hasConcept C152139883 @default.
- W2069291384 hasConcept C153180895 @default.
- W2069291384 hasConcept C154945302 @default.
- W2069291384 hasConcept C158622935 @default.
- W2069291384 hasConcept C163716315 @default.
- W2069291384 hasConcept C177860922 @default.
- W2069291384 hasConcept C27438332 @default.
- W2069291384 hasConcept C28826006 @default.
- W2069291384 hasConcept C33923547 @default.
- W2069291384 hasConcept C41008148 @default.
- W2069291384 hasConcept C50644808 @default.
- W2069291384 hasConcept C52622258 @default.
- W2069291384 hasConcept C62520636 @default.
- W2069291384 hasConceptScore W2069291384C105795698 @default.
- W2069291384 hasConceptScore W2069291384C106301342 @default.
- W2069291384 hasConceptScore W2069291384C111919701 @default.
- W2069291384 hasConceptScore W2069291384C11413529 @default.
- W2069291384 hasConceptScore W2069291384C121332964 @default.
- W2069291384 hasConceptScore W2069291384C152124472 @default.
- W2069291384 hasConceptScore W2069291384C152139883 @default.
- W2069291384 hasConceptScore W2069291384C153180895 @default.
- W2069291384 hasConceptScore W2069291384C154945302 @default.
- W2069291384 hasConceptScore W2069291384C158622935 @default.
- W2069291384 hasConceptScore W2069291384C163716315 @default.
- W2069291384 hasConceptScore W2069291384C177860922 @default.
- W2069291384 hasConceptScore W2069291384C27438332 @default.
- W2069291384 hasConceptScore W2069291384C28826006 @default.
- W2069291384 hasConceptScore W2069291384C33923547 @default.
- W2069291384 hasConceptScore W2069291384C41008148 @default.
- W2069291384 hasConceptScore W2069291384C50644808 @default.
- W2069291384 hasConceptScore W2069291384C52622258 @default.
- W2069291384 hasConceptScore W2069291384C62520636 @default.
- W2069291384 hasIssue "4" @default.
- W2069291384 hasLocation W20692913841 @default.
- W2069291384 hasOpenAccess W2069291384 @default.
- W2069291384 hasPrimaryLocation W20692913841 @default.
- W2069291384 hasRelatedWork W1976025425 @default.
- W2069291384 hasRelatedWork W2103378973 @default.
- W2069291384 hasRelatedWork W2112100270 @default.
- W2069291384 hasRelatedWork W2151857959 @default.
- W2069291384 hasRelatedWork W2380927352 @default.
- W2069291384 hasRelatedWork W2747609880 @default.
- W2069291384 hasRelatedWork W2941406960 @default.
- W2069291384 hasRelatedWork W2993993911 @default.
- W2069291384 hasRelatedWork W3032136380 @default.
- W2069291384 hasRelatedWork W4300011720 @default.
- W2069291384 hasVolume "8" @default.
- W2069291384 isParatext "false" @default.
- W2069291384 isRetracted "false" @default.
- W2069291384 magId "2069291384" @default.
- W2069291384 workType "article" @default.