Matches in SemOpenAlex for { <https://semopenalex.org/work/W2368520609> ?p ?o ?g. }
Showing items 1 to 57 of
57
with 100 items per page.
- W2368520609 abstract "A common method for combating over-fitting problem is to apply pruning algorithm to reduce the number of unnecessary weights. The pseudo-entropy of hidden layer node output is defined on the principle of the shannon entropy,and these two different definitions of entropy have approximately the same effectiveness on the description of uncertainty,but the new definition of entropy avoids the undefinition on zero values of information. The cross-entropy and pseudo-entropy are used as objective function,and the entropy cycle strategy is adopted to optimize the network parameters,thus to simplify neural networks by deleting and merging the hidden layer neurons. The simulation result shows that this proposed algorithm is simple,easy in implementation,and could improve the generalization of BP neural networks." @default.
- W2368520609 created "2016-06-24" @default.
- W2368520609 creator A5084778750 @default.
- W2368520609 date "2009-01-01" @default.
- W2368520609 modified "2023-09-25" @default.
- W2368520609 title "Pruning Algorithm for Feed-forward Neural Network Based on Entropy" @default.
- W2368520609 hasPublicationYear "2009" @default.
- W2368520609 type Work @default.
- W2368520609 sameAs 2368520609 @default.
- W2368520609 citedByCount "0" @default.
- W2368520609 crossrefType "journal-article" @default.
- W2368520609 hasAuthorship W2368520609A5084778750 @default.
- W2368520609 hasConcept C106301342 @default.
- W2368520609 hasConcept C11413529 @default.
- W2368520609 hasConcept C121332964 @default.
- W2368520609 hasConcept C154945302 @default.
- W2368520609 hasConcept C41008148 @default.
- W2368520609 hasConcept C44415725 @default.
- W2368520609 hasConcept C50644808 @default.
- W2368520609 hasConcept C62520636 @default.
- W2368520609 hasConcept C9679016 @default.
- W2368520609 hasConceptScore W2368520609C106301342 @default.
- W2368520609 hasConceptScore W2368520609C11413529 @default.
- W2368520609 hasConceptScore W2368520609C121332964 @default.
- W2368520609 hasConceptScore W2368520609C154945302 @default.
- W2368520609 hasConceptScore W2368520609C41008148 @default.
- W2368520609 hasConceptScore W2368520609C44415725 @default.
- W2368520609 hasConceptScore W2368520609C50644808 @default.
- W2368520609 hasConceptScore W2368520609C62520636 @default.
- W2368520609 hasConceptScore W2368520609C9679016 @default.
- W2368520609 hasLocation W23685206091 @default.
- W2368520609 hasOpenAccess W2368520609 @default.
- W2368520609 hasPrimaryLocation W23685206091 @default.
- W2368520609 hasRelatedWork W1561165259 @default.
- W2368520609 hasRelatedWork W2156581516 @default.
- W2368520609 hasRelatedWork W2233980831 @default.
- W2368520609 hasRelatedWork W2889194929 @default.
- W2368520609 hasRelatedWork W2889287351 @default.
- W2368520609 hasRelatedWork W2892388295 @default.
- W2368520609 hasRelatedWork W2901268451 @default.
- W2368520609 hasRelatedWork W2936927208 @default.
- W2368520609 hasRelatedWork W2945452905 @default.
- W2368520609 hasRelatedWork W2982627284 @default.
- W2368520609 hasRelatedWork W2990761096 @default.
- W2368520609 hasRelatedWork W3031009752 @default.
- W2368520609 hasRelatedWork W3121557814 @default.
- W2368520609 hasRelatedWork W3140837271 @default.
- W2368520609 hasRelatedWork W3165844382 @default.
- W2368520609 hasRelatedWork W2964433619 @default.
- W2368520609 hasRelatedWork W2965956390 @default.
- W2368520609 hasRelatedWork W3076628476 @default.
- W2368520609 hasRelatedWork W3097629360 @default.
- W2368520609 hasRelatedWork W3125243024 @default.
- W2368520609 isParatext "false" @default.
- W2368520609 isRetracted "false" @default.
- W2368520609 magId "2368520609" @default.
- W2368520609 workType "article" @default.