Matches in SemOpenAlex for { <https://semopenalex.org/work/W2905144382> ?p ?o ?g. }
Showing items 1 to 59 of
59
with 100 items per page.
- W2905144382 abstract "We consider the problem of metric learning for multi-view data and present a general method for learning within-view as well as between-view metrics in vector-valued kernel spaces, as a way to capture multimodal structure of the data. We formulate a general convex optimization problem in this context to jointly learn the metric and the classifier or regressor in kernel feature spaces. The formulated multi-view metric learning (MVML) can be applied to data with any number of views, not just two, while as a kernel-based method it allows for various data types. Indeed, it is not required for the views to have the same data type, as long as all of them are individually kernelizable. We give concrete realizations of our iterative algorithm in both classification and regression settings, where the metric operating between views is also learned, either a full metric or a view-sparse one. In order to scale the computation to large training sets, a block-wise Nystrom approximation of the multi-view kernel matrix is introduced. We justify our approach theoretically and experimentally, and show its performance on real-world datasets against relevant state-of-the-art methods." @default.
- W2905144382 created "2018-12-22" @default.
- W2905144382 creator A5003793755 @default.
- W2905144382 creator A5048948839 @default.
- W2905144382 creator A5079506428 @default.
- W2905144382 date "2018-11-27" @default.
- W2905144382 modified "2023-09-25" @default.
- W2905144382 title "General Framework for Multi-View Metric Learning" @default.
- W2905144382 cites W1541700983 @default.
- W2905144382 cites W1725492276 @default.
- W2905144382 cites W1984073246 @default.
- W2905144382 cites W2020114117 @default.
- W2905144382 cites W2048679005 @default.
- W2905144382 cites W2053559248 @default.
- W2905144382 cites W2062649377 @default.
- W2905144382 cites W2103024562 @default.
- W2905144382 cites W2115003579 @default.
- W2905144382 cites W2963973173 @default.
- W2905144382 cites W3101749733 @default.
- W2905144382 cites W4206212643 @default.
- W2905144382 cites W4211246278 @default.
- W2905144382 cites W4250589301 @default.
- W2905144382 doi "https://doi.org/10.1007/978-3-030-01872-6_11" @default.
- W2905144382 hasPublicationYear "2018" @default.
- W2905144382 type Work @default.
- W2905144382 sameAs 2905144382 @default.
- W2905144382 citedByCount "0" @default.
- W2905144382 crossrefType "book-chapter" @default.
- W2905144382 hasAuthorship W2905144382A5003793755 @default.
- W2905144382 hasAuthorship W2905144382A5048948839 @default.
- W2905144382 hasAuthorship W2905144382A5079506428 @default.
- W2905144382 hasConcept C127413603 @default.
- W2905144382 hasConcept C154945302 @default.
- W2905144382 hasConcept C176217482 @default.
- W2905144382 hasConcept C21547014 @default.
- W2905144382 hasConcept C41008148 @default.
- W2905144382 hasConceptScore W2905144382C127413603 @default.
- W2905144382 hasConceptScore W2905144382C154945302 @default.
- W2905144382 hasConceptScore W2905144382C176217482 @default.
- W2905144382 hasConceptScore W2905144382C21547014 @default.
- W2905144382 hasConceptScore W2905144382C41008148 @default.
- W2905144382 hasLocation W29051443821 @default.
- W2905144382 hasLocation W29051443822 @default.
- W2905144382 hasOpenAccess W2905144382 @default.
- W2905144382 hasPrimaryLocation W29051443821 @default.
- W2905144382 hasRelatedWork W2358668433 @default.
- W2905144382 hasRelatedWork W2390279801 @default.
- W2905144382 hasRelatedWork W2748952813 @default.
- W2905144382 hasRelatedWork W2899084033 @default.
- W2905144382 hasRelatedWork W2992707672 @default.
- W2905144382 hasRelatedWork W3015810174 @default.
- W2905144382 hasRelatedWork W3090091644 @default.
- W2905144382 hasRelatedWork W4286783870 @default.
- W2905144382 hasRelatedWork W4295088746 @default.
- W2905144382 hasRelatedWork W2004336531 @default.
- W2905144382 isParatext "false" @default.
- W2905144382 isRetracted "false" @default.
- W2905144382 magId "2905144382" @default.
- W2905144382 workType "book-chapter" @default.