FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2083291282> ?p ?o ?g. }

• W2083291282 endingPage "1252" @default.
• W2083291282 startingPage "1229" @default.
• W2083291282 abstract "Mixture of experts (ME) is a modular neural network architecture for supervised learning. A double-loop Expectation-Maximization (EM) algorithm has been introduced to the ME architecture for adjusting the parameters and the iteratively reweighted least squares (IRLS) algorithm is used to perform maximization in the inner loop [Jordan, M.I., Jacobs, R.A. (1994). Hierarchical mixture of experts and the EM algorithm, Neural Computation, 6(2), 181-214]. However, it is reported in literature that the IRLS algorithm is of instability and the ME architecture trained by the EM algorithm, where IRLS algorithm is used in the inner loop, often produces the poor performance in multiclass classification. In this paper, the reason of this instability is explored. We find out that due to an implicitly imposed incorrect assumption on parameter independence in multiclass classification, an incomplete Hessian matrix is used in that IRLS algorithm. Based on this finding, we apply the Newton-Raphson method to the inner loop of the EM algorithm in the case of multiclass classification, where the exact Hessian matrix is adopted. To tackle the expensive computation of the Hessian matrix and its inverse, we propose an approximation to the Newton-Raphson algorithm based on a so-called generalized Bernoulli density. The Newton-Raphson algorithm and its approximation have been applied to synthetic data, benchmark, and real-world multiclass classification tasks. For comparison, the IRLS algorithm and a quasi-Newton algorithm called BFGS have also been applied to the same tasks. Simulation results have shown that the use of the proposed learning algorithms avoids the instability problem and makes the ME architecture produce good performance in multiclass classification. In particular, our approximation algorithm leads to fast learning. In addition, the limitation of our approximation algorithm is also empirically investigated in this paper." @default.
• W2083291282 created "2016-06-24" @default.
• W2083291282 creator A5012078602 @default.
• W2083291282 creator A5027918340 @default.
• W2083291282 creator A5069972085 @default.
• W2083291282 date "1999-11-01" @default.
• W2083291282 modified "2023-10-12" @default.
• W2083291282 title "Improved learning algorithms for mixture of experts in multiclass classification" @default.
• W2083291282 cites W1980148047 @default.
• W2083291282 cites W1994450731 @default.
• W2083291282 cites W2001619934 @default.
• W2083291282 cites W2005136695 @default.
• W2083291282 cites W2011277999 @default.
• W2083291282 cites W2015772285 @default.
• W2083291282 cites W2025653905 @default.
• W2083291282 cites W2028069051 @default.
• W2083291282 cites W2038210983 @default.
• W2083291282 cites W2039437573 @default.
• W2083291282 cites W2078409719 @default.
• W2083291282 cites W2096077802 @default.
• W2083291282 cites W2097028668 @default.
• W2083291282 cites W2103111465 @default.
• W2083291282 cites W2129244720 @default.
• W2083291282 cites W2134199473 @default.
• W2083291282 cites W2135520967 @default.
• W2083291282 cites W2142260217 @default.
• W2083291282 cites W2142726513 @default.
• W2083291282 cites W2143403191 @default.
• W2083291282 cites W2150884987 @default.
• W2083291282 cites W2165959773 @default.
• W2083291282 doi "https://doi.org/10.1016/s0893-6080(99)00043-x" @default.
• W2083291282 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/12662629" @default.
• W2083291282 hasPublicationYear "1999" @default.
• W2083291282 type Work @default.
• W2083291282 sameAs 2083291282 @default.
• W2083291282 citedByCount "108" @default.
• W2083291282 countsByYear W20832912822012 @default.
• W2083291282 countsByYear W20832912822013 @default.
• W2083291282 countsByYear W20832912822014 @default.
• W2083291282 countsByYear W20832912822015 @default.
• W2083291282 countsByYear W20832912822016 @default.
• W2083291282 countsByYear W20832912822017 @default.
• W2083291282 countsByYear W20832912822018 @default.
• W2083291282 countsByYear W20832912822019 @default.
• W2083291282 countsByYear W20832912822020 @default.
• W2083291282 countsByYear W20832912822021 @default.
• W2083291282 countsByYear W20832912822022 @default.
• W2083291282 countsByYear W20832912822023 @default.
• W2083291282 crossrefType "journal-article" @default.
• W2083291282 hasAuthorship W2083291282A5012078602 @default.
• W2083291282 hasAuthorship W2083291282A5027918340 @default.
• W2083291282 hasAuthorship W2083291282A5069972085 @default.
• W2083291282 hasConcept C105795698 @default.
• W2083291282 hasConcept C11413529 @default.
• W2083291282 hasConcept C121332964 @default.
• W2083291282 hasConcept C12267149 @default.
• W2083291282 hasConcept C123860398 @default.
• W2083291282 hasConcept C154945302 @default.
• W2083291282 hasConcept C158622935 @default.
• W2083291282 hasConcept C182081679 @default.
• W2083291282 hasConcept C203616005 @default.
• W2083291282 hasConcept C28826006 @default.
• W2083291282 hasConcept C33923547 @default.
• W2083291282 hasConcept C41008148 @default.
• W2083291282 hasConcept C45374587 @default.
• W2083291282 hasConcept C49781872 @default.
• W2083291282 hasConcept C50644808 @default.
• W2083291282 hasConcept C62520636 @default.
• W2083291282 hasConcept C85189116 @default.
• W2083291282 hasConceptScore W2083291282C105795698 @default.
• W2083291282 hasConceptScore W2083291282C11413529 @default.
• W2083291282 hasConceptScore W2083291282C121332964 @default.
• W2083291282 hasConceptScore W2083291282C12267149 @default.
• W2083291282 hasConceptScore W2083291282C123860398 @default.
• W2083291282 hasConceptScore W2083291282C154945302 @default.
• W2083291282 hasConceptScore W2083291282C158622935 @default.
• W2083291282 hasConceptScore W2083291282C182081679 @default.
• W2083291282 hasConceptScore W2083291282C203616005 @default.
• W2083291282 hasConceptScore W2083291282C28826006 @default.
• W2083291282 hasConceptScore W2083291282C33923547 @default.
• W2083291282 hasConceptScore W2083291282C41008148 @default.
• W2083291282 hasConceptScore W2083291282C45374587 @default.
• W2083291282 hasConceptScore W2083291282C49781872 @default.
• W2083291282 hasConceptScore W2083291282C50644808 @default.
• W2083291282 hasConceptScore W2083291282C62520636 @default.
• W2083291282 hasConceptScore W2083291282C85189116 @default.
• W2083291282 hasIssue "9" @default.
• W2083291282 hasLocation W20832912821 @default.
• W2083291282 hasLocation W20832912822 @default.
• W2083291282 hasOpenAccess W2083291282 @default.
• W2083291282 hasPrimaryLocation W20832912821 @default.
• W2083291282 hasRelatedWork W114133465 @default.
• W2083291282 hasRelatedWork W1560140409 @default.
• W2083291282 hasRelatedWork W1996936972 @default.
• W2083291282 hasRelatedWork W2044296933 @default.
• W2083291282 hasRelatedWork W2050698291 @default.
• W2083291282 hasRelatedWork W2463007134 @default.
• W2083291282 hasRelatedWork W2970432723 @default.

• next