FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2587063199> ?p ?o ?g. }

• W2587063199 endingPage "1832" @default.
• W2587063199 startingPage "1820" @default.
• W2587063199 abstract "Multiple kernel learning (MKL) is a widely used technique for kernel design. Its principle consists in learning, for a given support vector classifier, the most suitable convex (or sparse) linear combination of standard elementary kernels. However, these combinations are shallow and often powerless to capture the actual similarity between highly semantic data, especially for challenging classification tasks such as image annotation. In this paper, we redefine multiple kernels using deep multi-layer networks. In this new contribution, a deep multiple kernel is recursively defined as a multi-layered combination of nonlinear activation functions, each one involves a combination of several elementary or intermediate kernels, and results into a positive semi-definite deep kernel. We propose four different frameworks in order to learn the weights of these networks: supervised, unsupervised, kernel-based semisupervised and Laplacian-based semi-supervised. When plugged into support vector machines (SVMs), the resulting deep kernel networks show clear gain, compared to several shallow kernels for the task of image annotation. Extensive experiments and analysis on the challenging ImageCLEF photo annotation benchmark, the COREL5k database and the Banana dataset validate the effectiveness of the proposed method." @default.
• W2587063199 created "2017-02-17" @default.
• W2587063199 creator A5017257056 @default.
• W2587063199 creator A5029896607 @default.
• W2587063199 date "2017-04-01" @default.
• W2587063199 modified "2023-09-30" @default.
• W2587063199 title "Nonlinear Deep Kernel Learning for Image Annotation" @default.
• W2587063199 cites W1479807131 @default.
• W2587063199 cites W1510073064 @default.
• W2587063199 cites W1601437336 @default.
• W2587063199 cites W167790647 @default.
• W2587063199 cites W1955055330 @default.
• W2587063199 cites W1967170228 @default.
• W2587063199 cites W2011589513 @default.
• W2587063199 cites W2022508996 @default.
• W2587063199 cites W2031823405 @default.
• W2587063199 cites W2036681051 @default.
• W2587063199 cites W2037407504 @default.
• W2587063199 cites W2039182213 @default.
• W2587063199 cites W2046589280 @default.
• W2587063199 cites W2053306114 @default.
• W2587063199 cites W2087501015 @default.
• W2587063199 cites W2087544865 @default.
• W2587063199 cites W2094569395 @default.
• W2587063199 cites W2104529636 @default.
• W2587063199 cites W2112796928 @default.
• W2587063199 cites W2122829955 @default.
• W2587063199 cites W2125238156 @default.
• W2587063199 cites W2127069950 @default.
• W2587063199 cites W2127586137 @default.
• W2587063199 cites W2134380836 @default.
• W2587063199 cites W2134692386 @default.
• W2587063199 cites W2136922672 @default.
• W2587063199 cites W2136951832 @default.
• W2587063199 cites W2138826255 @default.
• W2587063199 cites W2140095548 @default.
• W2587063199 cites W2151103935 @default.
• W2587063199 cites W2153635508 @default.
• W2587063199 cites W2155867588 @default.
• W2587063199 cites W2156336347 @default.
• W2587063199 cites W2161606025 @default.
• W2587063199 cites W2162867699 @default.
• W2587063199 cites W2172231696 @default.
• W2587063199 cites W2183182206 @default.
• W2587063199 cites W2536305071 @default.
• W2587063199 cites W3099206234 @default.
• W2587063199 doi "https://doi.org/10.1109/tip.2017.2666038" @default.
• W2587063199 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/28186895" @default.
• W2587063199 hasPublicationYear "2017" @default.
• W2587063199 type Work @default.
• W2587063199 sameAs 2587063199 @default.
• W2587063199 citedByCount "65" @default.
• W2587063199 countsByYear W25870631992017 @default.
• W2587063199 countsByYear W25870631992018 @default.
• W2587063199 countsByYear W25870631992019 @default.
• W2587063199 countsByYear W25870631992020 @default.
• W2587063199 countsByYear W25870631992021 @default.
• W2587063199 countsByYear W25870631992022 @default.
• W2587063199 countsByYear W25870631992023 @default.
• W2587063199 crossrefType "journal-article" @default.
• W2587063199 hasAuthorship W2587063199A5017257056 @default.
• W2587063199 hasAuthorship W2587063199A5029896607 @default.
• W2587063199 hasConcept C108583219 @default.
• W2587063199 hasConcept C114614502 @default.
• W2587063199 hasConcept C115961682 @default.
• W2587063199 hasConcept C119857082 @default.
• W2587063199 hasConcept C122280245 @default.
• W2587063199 hasConcept C12267149 @default.
• W2587063199 hasConcept C13280743 @default.
• W2587063199 hasConcept C134517425 @default.
• W2587063199 hasConcept C140417398 @default.
• W2587063199 hasConcept C153180895 @default.
• W2587063199 hasConcept C154945302 @default.
• W2587063199 hasConcept C1667742 @default.
• W2587063199 hasConcept C182335926 @default.
• W2587063199 hasConcept C185798385 @default.
• W2587063199 hasConcept C205649164 @default.
• W2587063199 hasConcept C2776879701 @default.
• W2587063199 hasConcept C33923547 @default.
• W2587063199 hasConcept C41008148 @default.
• W2587063199 hasConcept C74193536 @default.
• W2587063199 hasConceptScore W2587063199C108583219 @default.
• W2587063199 hasConceptScore W2587063199C114614502 @default.
• W2587063199 hasConceptScore W2587063199C115961682 @default.
• W2587063199 hasConceptScore W2587063199C119857082 @default.
• W2587063199 hasConceptScore W2587063199C122280245 @default.
• W2587063199 hasConceptScore W2587063199C12267149 @default.
• W2587063199 hasConceptScore W2587063199C13280743 @default.
• W2587063199 hasConceptScore W2587063199C134517425 @default.
• W2587063199 hasConceptScore W2587063199C140417398 @default.
• W2587063199 hasConceptScore W2587063199C153180895 @default.
• W2587063199 hasConceptScore W2587063199C154945302 @default.
• W2587063199 hasConceptScore W2587063199C1667742 @default.
• W2587063199 hasConceptScore W2587063199C182335926 @default.
• W2587063199 hasConceptScore W2587063199C185798385 @default.
• W2587063199 hasConceptScore W2587063199C205649164 @default.
• W2587063199 hasConceptScore W2587063199C2776879701 @default.
• W2587063199 hasConceptScore W2587063199C33923547 @default.

• next