000029586 001__ 29586
000029586 005__ 20130225122324.0
000029586 0248_ $$aoai:doc.rero.ch:20120713111649-JP$$ppostprint$$prero_explore$$zthesis_urn$$zreport$$zthesis$$zcdu004$$zbook$$zjournal$$zcdu16$$zhegge$$zpreprint$$zcdu1$$zdissertation$$zcdu34
000029586 041__ $$aeng
000029586 080__ $$a004
000029586 100__ $$aWang, Jun$$uCUI, Université de Genève
000029586 245__ $$9eng$$aMetric learning with multiple kernels
000029586 269__ $$c2011
000029586 520__ $$9eng$$aMetric learning has become a very active research field. The most popular representative–Mahalanobis metric learning–can be seen as learning a linear transformation and then computing the Euclidean metric in the transformed space. Since a linear transformation might not always be appropriate for a given learning problem, kernelized versions of various metric learning algorithms exist. However, the problem then becomes finding the appropriate kernel function. Multiple kernel learning addresses this limitation by learning a linear combination of a number of predefined kernels; this approach can be also readily used in the context of multiple-source learning to fuse different data sources. Surprisingly, and despite the extensive work on multiple kernel learning for SVMs, there has been no work in the area of metric learning with multiple kernel learning. In this paper we fill this gap and present a general approach for metric learning with multiple kernel learning. Our approach can be instantiated with different metric learning algorithms provided that they satisfy some constraints. Experimental evidence suggests that our approach outperforms metric learning with an unweighted kernel combination and metric learning with cross-validation based kernel selection.
000029586 695__ $$9eng$$ametric learning ; kernel ; Support Vector Machines
000029586 700__ $$aDo, Huyen$$uCUI, Université de Genève
000029586 700__ $$aWoznica, Adam$$uCUI, Université de Genève
000029586 700__ $$aKalousis, Alexandros$$uHaute école de gestion de Genève
000029586 773__ $$g2011/24//1-9$$tAdvances in Neural Information Processing Systems, Vol. 24 : 25th Annual Conference on Neural Information Processing Systems 2011. Proceedings of a meeting held 12-14 December 2011, Granada, Spain. 2011
000029586 775__ $$gNIPS 2011 Online papers$$ohttp://books.nips.cc/papers/files/nips24/NIPS2011_0683.pdf
000029586 8564_ $$fKalousis_2011_Metric_learning_with_multiple_kernels.pdf$$qapplication/pdf$$s174511$$uhttp://doc.rero.ch/record/29586/files/Kalousis_2011_Metric_learning_with_multiple_kernels.pdf$$yorder:1$$zTexte intégral
000029586 918__ $$aHaute école de gestion de Genève$$bCampus de Battelle, Bâtiment F, 7 route de Drize, 1227 Carouge$$cCentre de recherche appliqué en gestion (CRAG)
000029586 919__ $$aHaute école de gestion de Genève$$bGenève$$ddoc.support@rero.ch
000029586 980__ $$aPOSTPRINT$$bHEGGE$$fART_INPROC
000029586 990__ $$a20120713111649-JP