Matches in SemOpenAlex for { <https://semopenalex.org/work/W2898078303> ?p ?o ?g. }
Showing items 1 to 90 of
90
with 100 items per page.
- W2898078303 abstract "One fundamental problem of the convolutional neural network (CNN) is catastrophic forgetting, which occurs when new object classes and data are added while the original dataset is not available any more. Training the network only using the new dataset deteriorates the performance with respect to the old dataset. To overcome this problem, we propose an expanded network architecture, called the ExpandNet, to enhance the CNN incremental learning capability. Our solution keeps filters of the original networks on one hand, yet adds additional filters to the convolutional layers as well as the fully connected layers on the other hand. The proposed new architecture does not need any information of the original dataset, and it is trained using the new dataset only. Extensive evaluations based on the CIFAR −10 and the CIFAR −100 datasets show that the proposed method has a slower forgetting rate as compared to several existing incremental learning networks." @default.
- W2898078303 created "2018-10-26" @default.
- W2898078303 creator A5002394477 @default.
- W2898078303 creator A5009039461 @default.
- W2898078303 creator A5021946323 @default.
- W2898078303 creator A5027798315 @default.
- W2898078303 creator A5075063667 @default.
- W2898078303 date "2018-07-01" @default.
- W2898078303 modified "2023-10-16" @default.
- W2898078303 title "Enhancing CNN Incremental Learning Capability with an Expanded Network" @default.
- W2898078303 cites W1467353089 @default.
- W2898078303 cites W1686810756 @default.
- W2898078303 cites W1821462560 @default.
- W2898078303 cites W2117539524 @default.
- W2898078303 cites W2160684493 @default.
- W2898078303 cites W2271840356 @default.
- W2898078303 cites W2417429787 @default.
- W2898078303 cites W2473930607 @default.
- W2898078303 cites W2519796145 @default.
- W2898078303 cites W2560647685 @default.
- W2898078303 cites W2608422103 @default.
- W2898078303 cites W2949117887 @default.
- W2898078303 cites W2962690605 @default.
- W2898078303 cites W2962707369 @default.
- W2898078303 cites W2964189064 @default.
- W2898078303 cites W2964299589 @default.
- W2898078303 cites W3118608800 @default.
- W2898078303 doi "https://doi.org/10.1109/icme.2018.8486457" @default.
- W2898078303 hasPublicationYear "2018" @default.
- W2898078303 type Work @default.
- W2898078303 sameAs 2898078303 @default.
- W2898078303 citedByCount "1" @default.
- W2898078303 countsByYear W28980783032023 @default.
- W2898078303 crossrefType "proceedings-article" @default.
- W2898078303 hasAuthorship W2898078303A5002394477 @default.
- W2898078303 hasAuthorship W2898078303A5009039461 @default.
- W2898078303 hasAuthorship W2898078303A5021946323 @default.
- W2898078303 hasAuthorship W2898078303A5027798315 @default.
- W2898078303 hasAuthorship W2898078303A5075063667 @default.
- W2898078303 hasConcept C108583219 @default.
- W2898078303 hasConcept C119857082 @default.
- W2898078303 hasConcept C123657996 @default.
- W2898078303 hasConcept C124101348 @default.
- W2898078303 hasConcept C138885662 @default.
- W2898078303 hasConcept C142362112 @default.
- W2898078303 hasConcept C153180895 @default.
- W2898078303 hasConcept C153349607 @default.
- W2898078303 hasConcept C154945302 @default.
- W2898078303 hasConcept C193415008 @default.
- W2898078303 hasConcept C2780735816 @default.
- W2898078303 hasConcept C2781238097 @default.
- W2898078303 hasConcept C38652104 @default.
- W2898078303 hasConcept C41008148 @default.
- W2898078303 hasConcept C41895202 @default.
- W2898078303 hasConcept C7149132 @default.
- W2898078303 hasConcept C81363708 @default.
- W2898078303 hasConceptScore W2898078303C108583219 @default.
- W2898078303 hasConceptScore W2898078303C119857082 @default.
- W2898078303 hasConceptScore W2898078303C123657996 @default.
- W2898078303 hasConceptScore W2898078303C124101348 @default.
- W2898078303 hasConceptScore W2898078303C138885662 @default.
- W2898078303 hasConceptScore W2898078303C142362112 @default.
- W2898078303 hasConceptScore W2898078303C153180895 @default.
- W2898078303 hasConceptScore W2898078303C153349607 @default.
- W2898078303 hasConceptScore W2898078303C154945302 @default.
- W2898078303 hasConceptScore W2898078303C193415008 @default.
- W2898078303 hasConceptScore W2898078303C2780735816 @default.
- W2898078303 hasConceptScore W2898078303C2781238097 @default.
- W2898078303 hasConceptScore W2898078303C38652104 @default.
- W2898078303 hasConceptScore W2898078303C41008148 @default.
- W2898078303 hasConceptScore W2898078303C41895202 @default.
- W2898078303 hasConceptScore W2898078303C7149132 @default.
- W2898078303 hasConceptScore W2898078303C81363708 @default.
- W2898078303 hasLocation W28980783031 @default.
- W2898078303 hasOpenAccess W2898078303 @default.
- W2898078303 hasPrimaryLocation W28980783031 @default.
- W2898078303 hasRelatedWork W2731899572 @default.
- W2898078303 hasRelatedWork W2773120646 @default.
- W2898078303 hasRelatedWork W2920938200 @default.
- W2898078303 hasRelatedWork W2970686063 @default.
- W2898078303 hasRelatedWork W3110818700 @default.
- W2898078303 hasRelatedWork W3159901390 @default.
- W2898078303 hasRelatedWork W3215138031 @default.
- W2898078303 hasRelatedWork W4206751735 @default.
- W2898078303 hasRelatedWork W4225308791 @default.
- W2898078303 hasRelatedWork W4289884281 @default.
- W2898078303 isParatext "false" @default.
- W2898078303 isRetracted "false" @default.
- W2898078303 magId "2898078303" @default.
- W2898078303 workType "article" @default.