Matches in SemOpenAlex for { <https://semopenalex.org/work/W4313028293> ?p ?o ?g. }
Showing items 1 to 96 of
96
with 100 items per page.
- W4313028293 endingPage "414" @default.
- W4313028293 startingPage "398" @default.
- W4313028293 abstract "The ability to learn new concepts continually is necessary in this ever-changing world. However, deep neural networks suffer from catastrophic forgetting when learning new categories. Many works have been proposed to alleviate this phenomenon, whereas most of them either fall into the stability-plasticity dilemma or take too much computation or storage overhead. Inspired by the gradient boosting algorithm to gradually fit the residuals between the target model and the previous ensemble model, we propose a novel two-stage learning paradigm FOSTER, empowering the model to learn new categories adaptively. Specifically, we first dynamically expand new modules to fit the residuals between the target and the output of the original model. Next, we remove redundant parameters and feature dimensions through an effective distillation strategy to maintain the single backbone model. We validate our method FOSTER on CIFAR-100 and ImageNet-100/1000 under different settings. Experimental results show that our method achieves state-of-the-art performance. Code is available at https://github.com/G-U-N/ECCV22-FOSTER ." @default.
- W4313028293 created "2023-01-06" @default.
- W4313028293 creator A5014747290 @default.
- W4313028293 creator A5027946213 @default.
- W4313028293 creator A5065180062 @default.
- W4313028293 creator A5077697772 @default.
- W4313028293 date "2022-01-01" @default.
- W4313028293 modified "2023-10-11" @default.
- W4313028293 title "FOSTER: Feature Boosting and Compression for Class-Incremental Learning" @default.
- W4313028293 cites W1678356000 @default.
- W4313028293 cites W2024046085 @default.
- W4313028293 cites W2060277733 @default.
- W4313028293 cites W2108598243 @default.
- W4313028293 cites W2131140847 @default.
- W4313028293 cites W2132851185 @default.
- W4313028293 cites W2177950380 @default.
- W4313028293 cites W2194775991 @default.
- W4313028293 cites W2473930607 @default.
- W4313028293 cites W2884282566 @default.
- W4313028293 cites W2949736877 @default.
- W4313028293 cites W2954929116 @default.
- W4313028293 cites W2963691377 @default.
- W4313028293 cites W2964189064 @default.
- W4313028293 cites W2978098801 @default.
- W4313028293 cites W3030364939 @default.
- W4313028293 cites W3034856281 @default.
- W4313028293 cites W3094953068 @default.
- W4313028293 cites W3102476541 @default.
- W4313028293 cites W3107810305 @default.
- W4313028293 cites W3173493671 @default.
- W4313028293 cites W3177034761 @default.
- W4313028293 cites W3180392831 @default.
- W4313028293 cites W3181161034 @default.
- W4313028293 cites W3198636227 @default.
- W4313028293 cites W3207854039 @default.
- W4313028293 cites W4225484930 @default.
- W4313028293 cites W4226134030 @default.
- W4313028293 cites W4232478844 @default.
- W4313028293 doi "https://doi.org/10.1007/978-3-031-19806-9_23" @default.
- W4313028293 hasPublicationYear "2022" @default.
- W4313028293 type Work @default.
- W4313028293 citedByCount "13" @default.
- W4313028293 countsByYear W43130282932023 @default.
- W4313028293 crossrefType "book-chapter" @default.
- W4313028293 hasAuthorship W4313028293A5014747290 @default.
- W4313028293 hasAuthorship W4313028293A5027946213 @default.
- W4313028293 hasAuthorship W4313028293A5065180062 @default.
- W4313028293 hasAuthorship W4313028293A5077697772 @default.
- W4313028293 hasBestOaLocation W43130282932 @default.
- W4313028293 hasConcept C111919701 @default.
- W4313028293 hasConcept C11413529 @default.
- W4313028293 hasConcept C119857082 @default.
- W4313028293 hasConcept C138885662 @default.
- W4313028293 hasConcept C154945302 @default.
- W4313028293 hasConcept C2776401178 @default.
- W4313028293 hasConcept C2777212361 @default.
- W4313028293 hasConcept C2779960059 @default.
- W4313028293 hasConcept C41008148 @default.
- W4313028293 hasConcept C41895202 @default.
- W4313028293 hasConcept C45374587 @default.
- W4313028293 hasConcept C46686674 @default.
- W4313028293 hasConcept C50644808 @default.
- W4313028293 hasConcept C7149132 @default.
- W4313028293 hasConceptScore W4313028293C111919701 @default.
- W4313028293 hasConceptScore W4313028293C11413529 @default.
- W4313028293 hasConceptScore W4313028293C119857082 @default.
- W4313028293 hasConceptScore W4313028293C138885662 @default.
- W4313028293 hasConceptScore W4313028293C154945302 @default.
- W4313028293 hasConceptScore W4313028293C2776401178 @default.
- W4313028293 hasConceptScore W4313028293C2777212361 @default.
- W4313028293 hasConceptScore W4313028293C2779960059 @default.
- W4313028293 hasConceptScore W4313028293C41008148 @default.
- W4313028293 hasConceptScore W4313028293C41895202 @default.
- W4313028293 hasConceptScore W4313028293C45374587 @default.
- W4313028293 hasConceptScore W4313028293C46686674 @default.
- W4313028293 hasConceptScore W4313028293C50644808 @default.
- W4313028293 hasConceptScore W4313028293C7149132 @default.
- W4313028293 hasLocation W43130282931 @default.
- W4313028293 hasLocation W43130282932 @default.
- W4313028293 hasOpenAccess W4313028293 @default.
- W4313028293 hasPrimaryLocation W43130282931 @default.
- W4313028293 hasRelatedWork W1987859285 @default.
- W4313028293 hasRelatedWork W1996541855 @default.
- W4313028293 hasRelatedWork W3151529617 @default.
- W4313028293 hasRelatedWork W3159988495 @default.
- W4313028293 hasRelatedWork W3195168932 @default.
- W4313028293 hasRelatedWork W3200719183 @default.
- W4313028293 hasRelatedWork W4225307033 @default.
- W4313028293 hasRelatedWork W4292969247 @default.
- W4313028293 hasRelatedWork W4313488044 @default.
- W4313028293 hasRelatedWork W1629725936 @default.
- W4313028293 isParatext "false" @default.
- W4313028293 isRetracted "false" @default.
- W4313028293 workType "book-chapter" @default.