Matches in SemOpenAlex for { <https://semopenalex.org/work/W3211144631> ?p ?o ?g. }
- W3211144631 abstract "Few-shot learning describes the challenging problem of recognizing samples from unseen classes given very few labeled examples. In many cases, few-shot learning is cast as learning an embedding space that assigns test samples to their corresponding class prototypes. Previous methods assume that data of all few-shot learning tasks comply with a fixed geometrical structure, mostly a Euclidean structure. Questioning this assumption that is clearly difficult to hold in real-world scenarios and incurs distortions to data, we propose to learn a task-aware curved embedding space by making use of the hyperbolic geometry. As a result, task-specific embedding spaces where suitable curvatures are generated to match the characteristics of data are constructed, leading to more generic embedding spaces. We then leverage on intra-class and inter-class context information in the embedding space to generate class prototypes for discriminative classification. We conduct a comprehensive set of experiments on inductive and transductive few-shot learning, demonstrating the benefits of our proposed method over existing embedding methods." @default.
- W3211144631 created "2021-11-08" @default.
- W3211144631 creator A5021790939 @default.
- W3211144631 creator A5064291165 @default.
- W3211144631 creator A5071656520 @default.
- W3211144631 creator A5075244507 @default.
- W3211144631 date "2021-10-01" @default.
- W3211144631 modified "2023-10-18" @default.
- W3211144631 title "Curvature Generation in Curved Spaces for Few-Shot Learning" @default.
- W3211144631 cites W2018050723 @default.
- W3211144631 cites W2104657103 @default.
- W3211144631 cites W2152825437 @default.
- W3211144631 cites W2170653751 @default.
- W3211144631 cites W2400540722 @default.
- W3211144631 cites W2551829638 @default.
- W3211144631 cites W2796346823 @default.
- W3211144631 cites W2798836702 @default.
- W3211144631 cites W2962987395 @default.
- W3211144631 cites W2963070905 @default.
- W3211144631 cites W2963680240 @default.
- W3211144631 cites W2963841265 @default.
- W3211144631 cites W2963845150 @default.
- W3211144631 cites W2964105864 @default.
- W3211144631 cites W2982049331 @default.
- W3211144631 cites W2988501586 @default.
- W3211144631 cites W2989701968 @default.
- W3211144631 cites W2997294353 @default.
- W3211144631 cites W3012255272 @default.
- W3211144631 cites W3030509278 @default.
- W3211144631 cites W3034312118 @default.
- W3211144631 cites W3034354010 @default.
- W3211144631 cites W3034587791 @default.
- W3211144631 cites W3034694537 @default.
- W3211144631 cites W3035102141 @default.
- W3211144631 cites W3035143213 @default.
- W3211144631 cites W3035700349 @default.
- W3211144631 cites W3177377177 @default.
- W3211144631 cites W4226044115 @default.
- W3211144631 cites W4244709323 @default.
- W3211144631 doi "https://doi.org/10.1109/iccv48922.2021.00857" @default.
- W3211144631 hasPublicationYear "2021" @default.
- W3211144631 type Work @default.
- W3211144631 sameAs 3211144631 @default.
- W3211144631 citedByCount "10" @default.
- W3211144631 countsByYear W32111446312022 @default.
- W3211144631 countsByYear W32111446312023 @default.
- W3211144631 crossrefType "proceedings-article" @default.
- W3211144631 hasAuthorship W3211144631A5021790939 @default.
- W3211144631 hasAuthorship W3211144631A5064291165 @default.
- W3211144631 hasAuthorship W3211144631A5071656520 @default.
- W3211144631 hasAuthorship W3211144631A5075244507 @default.
- W3211144631 hasBestOaLocation W32111446312 @default.
- W3211144631 hasConcept C111919701 @default.
- W3211144631 hasConcept C129782007 @default.
- W3211144631 hasConcept C151730666 @default.
- W3211144631 hasConcept C153083717 @default.
- W3211144631 hasConcept C154945302 @default.
- W3211144631 hasConcept C186450821 @default.
- W3211144631 hasConcept C202444582 @default.
- W3211144631 hasConcept C2524010 @default.
- W3211144631 hasConcept C2777212361 @default.
- W3211144631 hasConcept C2778572836 @default.
- W3211144631 hasConcept C2779343474 @default.
- W3211144631 hasConcept C33923547 @default.
- W3211144631 hasConcept C41008148 @default.
- W3211144631 hasConcept C41608201 @default.
- W3211144631 hasConcept C86803240 @default.
- W3211144631 hasConcept C97931131 @default.
- W3211144631 hasConceptScore W3211144631C111919701 @default.
- W3211144631 hasConceptScore W3211144631C129782007 @default.
- W3211144631 hasConceptScore W3211144631C151730666 @default.
- W3211144631 hasConceptScore W3211144631C153083717 @default.
- W3211144631 hasConceptScore W3211144631C154945302 @default.
- W3211144631 hasConceptScore W3211144631C186450821 @default.
- W3211144631 hasConceptScore W3211144631C202444582 @default.
- W3211144631 hasConceptScore W3211144631C2524010 @default.
- W3211144631 hasConceptScore W3211144631C2777212361 @default.
- W3211144631 hasConceptScore W3211144631C2778572836 @default.
- W3211144631 hasConceptScore W3211144631C2779343474 @default.
- W3211144631 hasConceptScore W3211144631C33923547 @default.
- W3211144631 hasConceptScore W3211144631C41008148 @default.
- W3211144631 hasConceptScore W3211144631C41608201 @default.
- W3211144631 hasConceptScore W3211144631C86803240 @default.
- W3211144631 hasConceptScore W3211144631C97931131 @default.
- W3211144631 hasLocation W32111446311 @default.
- W3211144631 hasLocation W32111446312 @default.
- W3211144631 hasOpenAccess W3211144631 @default.
- W3211144631 hasPrimaryLocation W32111446311 @default.
- W3211144631 hasRelatedWork W1488249307 @default.
- W3211144631 hasRelatedWork W2907464407 @default.
- W3211144631 hasRelatedWork W2921232085 @default.
- W3211144631 hasRelatedWork W2951415044 @default.
- W3211144631 hasRelatedWork W2959752353 @default.
- W3211144631 hasRelatedWork W3146864044 @default.
- W3211144631 hasRelatedWork W3210523300 @default.
- W3211144631 hasRelatedWork W4287252481 @default.
- W3211144631 hasRelatedWork W4288416546 @default.
- W3211144631 hasRelatedWork W4385434001 @default.
- W3211144631 isParatext "false" @default.
- W3211144631 isRetracted "false" @default.