Matches in SemOpenAlex for { <https://semopenalex.org/work/W2785751518> ?p ?o ?g. }
Showing items 1 to 82 of
82
with 100 items per page.
- W2785751518 abstract "In order to improve the generalization ability of deep convolutional neural networks, an improved training strategy based on anomalous sample penalty term is introduced in this paper. We first establish an anomalous sample detection mechanism on the basis of preliminary neural network model, then we use the new loss function based on sample distribution to search an optimized feature boundary. A better boundary can improve the performance of the network over the test set. Then we discuss the effect of the convolutional neural network from the point of view of the sample distribution through a clean database created by ourselves. In the experiments, we compare the classification results of seven typical deep convolutional neural networks on different image databases. At the cost of reducing a little accuracy of training set, the classification accuracy of test set is improved significantly. We get the classification accuracy of 95.4%, 95.7% and 85.4% on Caltech-101, Cifar-10, Cifar-100 respectively. Finally, we analyze the influence of samples with anomalous distribution on network generalization capability through dimensionality reduction visualization." @default.
- W2785751518 created "2018-02-23" @default.
- W2785751518 creator A5035441171 @default.
- W2785751518 creator A5041971632 @default.
- W2785751518 creator A5064825247 @default.
- W2785751518 creator A5087120199 @default.
- W2785751518 creator A5091644203 @default.
- W2785751518 date "2017-12-01" @default.
- W2785751518 modified "2023-10-16" @default.
- W2785751518 title "Understanding and boosting of deep convolutional neural network based on sample distribution" @default.
- W2785751518 cites W2097117768 @default.
- W2785751518 cites W2165698076 @default.
- W2785751518 cites W2194775991 @default.
- W2785751518 doi "https://doi.org/10.1109/itnec.2017.8284849" @default.
- W2785751518 hasPublicationYear "2017" @default.
- W2785751518 type Work @default.
- W2785751518 sameAs 2785751518 @default.
- W2785751518 citedByCount "5" @default.
- W2785751518 countsByYear W27857515182018 @default.
- W2785751518 countsByYear W27857515182019 @default.
- W2785751518 countsByYear W27857515182020 @default.
- W2785751518 countsByYear W27857515182022 @default.
- W2785751518 countsByYear W27857515182023 @default.
- W2785751518 crossrefType "proceedings-article" @default.
- W2785751518 hasAuthorship W2785751518A5035441171 @default.
- W2785751518 hasAuthorship W2785751518A5041971632 @default.
- W2785751518 hasAuthorship W2785751518A5064825247 @default.
- W2785751518 hasAuthorship W2785751518A5087120199 @default.
- W2785751518 hasAuthorship W2785751518A5091644203 @default.
- W2785751518 hasConcept C108583219 @default.
- W2785751518 hasConcept C115961682 @default.
- W2785751518 hasConcept C119857082 @default.
- W2785751518 hasConcept C134306372 @default.
- W2785751518 hasConcept C153180895 @default.
- W2785751518 hasConcept C154945302 @default.
- W2785751518 hasConcept C169903167 @default.
- W2785751518 hasConcept C177148314 @default.
- W2785751518 hasConcept C185592680 @default.
- W2785751518 hasConcept C198531522 @default.
- W2785751518 hasConcept C33923547 @default.
- W2785751518 hasConcept C41008148 @default.
- W2785751518 hasConcept C43617362 @default.
- W2785751518 hasConcept C46686674 @default.
- W2785751518 hasConcept C50644808 @default.
- W2785751518 hasConcept C52622490 @default.
- W2785751518 hasConcept C75294576 @default.
- W2785751518 hasConcept C81363708 @default.
- W2785751518 hasConceptScore W2785751518C108583219 @default.
- W2785751518 hasConceptScore W2785751518C115961682 @default.
- W2785751518 hasConceptScore W2785751518C119857082 @default.
- W2785751518 hasConceptScore W2785751518C134306372 @default.
- W2785751518 hasConceptScore W2785751518C153180895 @default.
- W2785751518 hasConceptScore W2785751518C154945302 @default.
- W2785751518 hasConceptScore W2785751518C169903167 @default.
- W2785751518 hasConceptScore W2785751518C177148314 @default.
- W2785751518 hasConceptScore W2785751518C185592680 @default.
- W2785751518 hasConceptScore W2785751518C198531522 @default.
- W2785751518 hasConceptScore W2785751518C33923547 @default.
- W2785751518 hasConceptScore W2785751518C41008148 @default.
- W2785751518 hasConceptScore W2785751518C43617362 @default.
- W2785751518 hasConceptScore W2785751518C46686674 @default.
- W2785751518 hasConceptScore W2785751518C50644808 @default.
- W2785751518 hasConceptScore W2785751518C52622490 @default.
- W2785751518 hasConceptScore W2785751518C75294576 @default.
- W2785751518 hasConceptScore W2785751518C81363708 @default.
- W2785751518 hasLocation W27857515181 @default.
- W2785751518 hasOpenAccess W2785751518 @default.
- W2785751518 hasPrimaryLocation W27857515181 @default.
- W2785751518 hasRelatedWork W2279398222 @default.
- W2785751518 hasRelatedWork W2766604260 @default.
- W2785751518 hasRelatedWork W2946016983 @default.
- W2785751518 hasRelatedWork W2986507176 @default.
- W2785751518 hasRelatedWork W3011074480 @default.
- W2785751518 hasRelatedWork W3099765033 @default.
- W2785751518 hasRelatedWork W3156786002 @default.
- W2785751518 hasRelatedWork W3160711233 @default.
- W2785751518 hasRelatedWork W4299822940 @default.
- W2785751518 hasRelatedWork W4366492315 @default.
- W2785751518 isParatext "false" @default.
- W2785751518 isRetracted "false" @default.
- W2785751518 magId "2785751518" @default.
- W2785751518 workType "article" @default.