Matches in SemOpenAlex for { <https://semopenalex.org/work/W1968524744> ?p ?o ?g. }
Showing items 1 to 94 of
94
with 100 items per page.
- W1968524744 endingPage "11" @default.
- W1968524744 startingPage "1" @default.
- W1968524744 abstract "Minimisation methods for training feedforward networks with back propagation are compared. Feedforward neural network training is a special case of function minimisation, where no explicit model of the data is assumed. Therefore, and due to the high dimensionality of the data, linearisation of the training problem through use of orthogonal basis functions is not desirable. The focus is on function minimisation on any basis. Quasi-Newton and conjugate gradient methods are reviewed, and the latter are shown to be a special case of error back propagation with momentum term. Three feedforward learning problems are tested with five methods. It is shown that, due to the fixed stepsize, standard error back propagation performs well in avoiding local minima. However, by using not only the local gradient but also the second derivative of the error function, a much shorter training time is required. Conjugate gradient with Powell restarts shows to be the superior method." @default.
- W1968524744 created "2016-06-24" @default.
- W1968524744 creator A5022626150 @default.
- W1968524744 date "1994-01-01" @default.
- W1968524744 modified "2023-10-01" @default.
- W1968524744 title "Minimisation methods for training feedforward neural networks" @default.
- W1968524744 cites W1498436455 @default.
- W1968524744 cites W1900101064 @default.
- W1968524744 cites W2006544565 @default.
- W1968524744 cites W2010431242 @default.
- W1968524744 cites W2043382734 @default.
- W1968524744 cites W2137983211 @default.
- W1968524744 cites W2150499794 @default.
- W1968524744 cites W2155530592 @default.
- W1968524744 cites W4248591302 @default.
- W1968524744 doi "https://doi.org/10.1016/0893-6080(94)90052-3" @default.
- W1968524744 hasPublicationYear "1994" @default.
- W1968524744 type Work @default.
- W1968524744 sameAs 1968524744 @default.
- W1968524744 citedByCount "203" @default.
- W1968524744 countsByYear W19685247442012 @default.
- W1968524744 countsByYear W19685247442013 @default.
- W1968524744 countsByYear W19685247442014 @default.
- W1968524744 countsByYear W19685247442015 @default.
- W1968524744 countsByYear W19685247442016 @default.
- W1968524744 countsByYear W19685247442017 @default.
- W1968524744 countsByYear W19685247442018 @default.
- W1968524744 countsByYear W19685247442019 @default.
- W1968524744 countsByYear W19685247442021 @default.
- W1968524744 countsByYear W19685247442022 @default.
- W1968524744 countsByYear W19685247442023 @default.
- W1968524744 crossrefType "journal-article" @default.
- W1968524744 hasAuthorship W1968524744A5022626150 @default.
- W1968524744 hasConcept C105795698 @default.
- W1968524744 hasConcept C111030470 @default.
- W1968524744 hasConcept C11413529 @default.
- W1968524744 hasConcept C127413603 @default.
- W1968524744 hasConcept C133731056 @default.
- W1968524744 hasConcept C134306372 @default.
- W1968524744 hasConcept C153258448 @default.
- W1968524744 hasConcept C154945302 @default.
- W1968524744 hasConcept C155032097 @default.
- W1968524744 hasConcept C186633575 @default.
- W1968524744 hasConcept C202286095 @default.
- W1968524744 hasConcept C2775924081 @default.
- W1968524744 hasConcept C33923547 @default.
- W1968524744 hasConcept C38858127 @default.
- W1968524744 hasConcept C41008148 @default.
- W1968524744 hasConcept C47446073 @default.
- W1968524744 hasConcept C47702885 @default.
- W1968524744 hasConcept C50644808 @default.
- W1968524744 hasConcept C81184566 @default.
- W1968524744 hasConcept C86941820 @default.
- W1968524744 hasConceptScore W1968524744C105795698 @default.
- W1968524744 hasConceptScore W1968524744C111030470 @default.
- W1968524744 hasConceptScore W1968524744C11413529 @default.
- W1968524744 hasConceptScore W1968524744C127413603 @default.
- W1968524744 hasConceptScore W1968524744C133731056 @default.
- W1968524744 hasConceptScore W1968524744C134306372 @default.
- W1968524744 hasConceptScore W1968524744C153258448 @default.
- W1968524744 hasConceptScore W1968524744C154945302 @default.
- W1968524744 hasConceptScore W1968524744C155032097 @default.
- W1968524744 hasConceptScore W1968524744C186633575 @default.
- W1968524744 hasConceptScore W1968524744C202286095 @default.
- W1968524744 hasConceptScore W1968524744C2775924081 @default.
- W1968524744 hasConceptScore W1968524744C33923547 @default.
- W1968524744 hasConceptScore W1968524744C38858127 @default.
- W1968524744 hasConceptScore W1968524744C41008148 @default.
- W1968524744 hasConceptScore W1968524744C47446073 @default.
- W1968524744 hasConceptScore W1968524744C47702885 @default.
- W1968524744 hasConceptScore W1968524744C50644808 @default.
- W1968524744 hasConceptScore W1968524744C81184566 @default.
- W1968524744 hasConceptScore W1968524744C86941820 @default.
- W1968524744 hasIssue "1" @default.
- W1968524744 hasLocation W19685247441 @default.
- W1968524744 hasOpenAccess W1968524744 @default.
- W1968524744 hasPrimaryLocation W19685247441 @default.
- W1968524744 hasRelatedWork W1604847762 @default.
- W1968524744 hasRelatedWork W1968524744 @default.
- W1968524744 hasRelatedWork W2086999410 @default.
- W1968524744 hasRelatedWork W2110398104 @default.
- W1968524744 hasRelatedWork W2144183224 @default.
- W1968524744 hasRelatedWork W2405196115 @default.
- W1968524744 hasRelatedWork W2554138451 @default.
- W1968524744 hasRelatedWork W2788727425 @default.
- W1968524744 hasRelatedWork W2131501411 @default.
- W1968524744 hasRelatedWork W2171561041 @default.
- W1968524744 hasVolume "7" @default.
- W1968524744 isParatext "false" @default.
- W1968524744 isRetracted "false" @default.
- W1968524744 magId "1968524744" @default.
- W1968524744 workType "article" @default.