Matches in SemOpenAlex for { <https://semopenalex.org/work/W3089818419> ?p ?o ?g. }
- W3089818419 abstract "Deep kernel map networks have shown excellent performances in various classification problems including image annotation. Their general recipe consists in aggregating several layers of singular value decompositions (SVDs) - that map data from input spaces into high dimensional spaces - while preserving the similarity of the underlying kernels. However, the potential of these deep map networks has not been fully explored as the original setting of these networks focuses mainly on the approximation quality of their kernels and ignores their discrimination power.In this paper, we introduce a novel “end-to-end” design for deep kernel map learning that balances the approximation quality of kernels and their discrimination power. Our method proceeds in two steps; first, layerwise SVD is applied in order to build initial deep kernel map approximations and then an “end-to-end” supervised learning is employed to further enhance their discrimination power while maintaining their efficiency. Extensive experiments, conducted on the challenging ImageCLEF annotation benchmark, show the high efficiency and the out-performance of this two-step process with respect to different related methods." @default.
- W3089818419 created "2020-10-08" @default.
- W3089818419 creator A5017257056 @default.
- W3089818419 creator A5029896607 @default.
- W3089818419 date "2020-10-01" @default.
- W3089818419 modified "2023-09-25" @default.
- W3089818419 title "End-to-End Deep Kernel Map Design for Image Annotation" @default.
- W3089818419 cites W119301944 @default.
- W3089818419 cites W1510073064 @default.
- W3089818419 cites W1546411676 @default.
- W3089818419 cites W1601437336 @default.
- W3089818419 cites W1751437809 @default.
- W3089818419 cites W2022137768 @default.
- W3089818419 cites W2022508996 @default.
- W3089818419 cites W2039182213 @default.
- W3089818419 cites W2067022618 @default.
- W3089818419 cites W2069797086 @default.
- W3089818419 cites W2097117768 @default.
- W3089818419 cites W2102116870 @default.
- W3089818419 cites W2109235804 @default.
- W3089818419 cites W2112545207 @default.
- W3089818419 cites W2112796928 @default.
- W3089818419 cites W2118585731 @default.
- W3089818419 cites W2123872146 @default.
- W3089818419 cites W2127069950 @default.
- W3089818419 cites W2132285904 @default.
- W3089818419 cites W2136922672 @default.
- W3089818419 cites W2137055149 @default.
- W3089818419 cites W2144902422 @default.
- W3089818419 cites W2148603752 @default.
- W3089818419 cites W2163605009 @default.
- W3089818419 cites W2163922914 @default.
- W3089818419 cites W2167608136 @default.
- W3089818419 cites W2194775991 @default.
- W3089818419 cites W2222792052 @default.
- W3089818419 cites W2294422256 @default.
- W3089818419 cites W2399164823 @default.
- W3089818419 cites W2585201405 @default.
- W3089818419 cites W2587063199 @default.
- W3089818419 cites W2595315035 @default.
- W3089818419 cites W2660819698 @default.
- W3089818419 cites W2903909270 @default.
- W3089818419 cites W2963173190 @default.
- W3089818419 cites W2963446712 @default.
- W3089818419 cites W2963766931 @default.
- W3089818419 doi "https://doi.org/10.1109/icip40778.2020.9191156" @default.
- W3089818419 hasPublicationYear "2020" @default.
- W3089818419 type Work @default.
- W3089818419 sameAs 3089818419 @default.
- W3089818419 citedByCount "1" @default.
- W3089818419 countsByYear W30898184192021 @default.
- W3089818419 crossrefType "proceedings-article" @default.
- W3089818419 hasAuthorship W3089818419A5017257056 @default.
- W3089818419 hasAuthorship W3089818419A5029896607 @default.
- W3089818419 hasBestOaLocation W30898184192 @default.
- W3089818419 hasConcept C103278499 @default.
- W3089818419 hasConcept C108583219 @default.
- W3089818419 hasConcept C111919701 @default.
- W3089818419 hasConcept C11413529 @default.
- W3089818419 hasConcept C114614502 @default.
- W3089818419 hasConcept C115961682 @default.
- W3089818419 hasConcept C122280245 @default.
- W3089818419 hasConcept C12267149 @default.
- W3089818419 hasConcept C13280743 @default.
- W3089818419 hasConcept C153180895 @default.
- W3089818419 hasConcept C154945302 @default.
- W3089818419 hasConcept C185798385 @default.
- W3089818419 hasConcept C205649164 @default.
- W3089818419 hasConcept C22789450 @default.
- W3089818419 hasConcept C2776321320 @default.
- W3089818419 hasConcept C33923547 @default.
- W3089818419 hasConcept C41008148 @default.
- W3089818419 hasConcept C74193536 @default.
- W3089818419 hasConcept C74296488 @default.
- W3089818419 hasConcept C98045186 @default.
- W3089818419 hasConceptScore W3089818419C103278499 @default.
- W3089818419 hasConceptScore W3089818419C108583219 @default.
- W3089818419 hasConceptScore W3089818419C111919701 @default.
- W3089818419 hasConceptScore W3089818419C11413529 @default.
- W3089818419 hasConceptScore W3089818419C114614502 @default.
- W3089818419 hasConceptScore W3089818419C115961682 @default.
- W3089818419 hasConceptScore W3089818419C122280245 @default.
- W3089818419 hasConceptScore W3089818419C12267149 @default.
- W3089818419 hasConceptScore W3089818419C13280743 @default.
- W3089818419 hasConceptScore W3089818419C153180895 @default.
- W3089818419 hasConceptScore W3089818419C154945302 @default.
- W3089818419 hasConceptScore W3089818419C185798385 @default.
- W3089818419 hasConceptScore W3089818419C205649164 @default.
- W3089818419 hasConceptScore W3089818419C22789450 @default.
- W3089818419 hasConceptScore W3089818419C2776321320 @default.
- W3089818419 hasConceptScore W3089818419C33923547 @default.
- W3089818419 hasConceptScore W3089818419C41008148 @default.
- W3089818419 hasConceptScore W3089818419C74193536 @default.
- W3089818419 hasConceptScore W3089818419C74296488 @default.
- W3089818419 hasConceptScore W3089818419C98045186 @default.
- W3089818419 hasLocation W30898184191 @default.
- W3089818419 hasLocation W30898184192 @default.
- W3089818419 hasLocation W30898184193 @default.
- W3089818419 hasLocation W30898184194 @default.
- W3089818419 hasLocation W30898184195 @default.