Show simple item record

hal.structure.identifierUniversity of Tokyo, Graduate School of Frontier Sciences
dc.contributor.authorYamane, Ikko
hal.structure.identifierLaboratoire d'analyse et modélisation de systèmes pour l'aide à la décision [LAMSADE]
dc.contributor.authorYger, Florian
HAL ID: 17768
ORCID: 0000-0002-7182-8062
hal.structure.identifierLaboratoire d'Informatique, de Traitement de l'Information et des Systèmes [LITIS]
dc.contributor.authorBerar, Maxime
HAL ID: 178769
hal.structure.identifierUniversity of Tokyo, Graduate School of Frontier Sciences
dc.contributor.authorSugiyama, Masashi
dc.date.accessioned2017-01-27T12:43:37Z
dc.date.available2017-01-27T12:43:37Z
dc.date.issued2016
dc.identifier.urihttps://basepub.dauphine.fr/handle/123456789/16212
dc.language.isoenen
dc.subjectdimensionality reductionen
dc.subjectmultitask learningen
dc.subjectRiemannian geometryen
dc.subjectGrass- mann manifolden
dc.subject.ddc006.3en
dc.titleMultitask Principal Component Analysisen
dc.typeCommunication / Conférence
dc.description.abstractenPrincipal Component Analysis (PCA) is a canonical and well-studied tool for dimension- ality reduction. However, when few data are available, the poor quality of the covariance estimator at its core may compromise its performance. We leverage this issue by casting the PCA into a multitask framework, and doing so, we show how to solve simultaneously several related PCA problems. Hence, we propose a novel formulation of the PCA prob- lem relying on a novel regularization. This regularization is based on a distance between subspaces, and the whole problem is solved as an optimization problem over a Riemannian manifold. We experimentally demonstrate the usefulness of our approach as pre-processing for EEG signals.en
dc.identifier.citationpages302-317en
dc.relation.ispartoftitleProceedings of The 8th Asian Conference on Machine Learningen
dc.relation.ispartofeditorDurrant, Robert J.
dc.relation.ispartofeditorKim, Kee-Eung
dc.relation.ispartofpublnameJMLR: Workshop and Conference Proceedingsen
dc.relation.ispartofdate2016
dc.relation.ispartofpages460en
dc.identifier.urlsitehttp://jmlr.csail.mit.edu/proceedings/papers/v63/yamane65.htmlen
dc.contributor.countryeditoruniversityotherJAPAN
dc.subject.ddclabelIntelligence artificielleen
dc.relation.conftitle8th Asian Conference on Machine Learning (ACML 2016)en
dc.relation.confdate2016-11
dc.relation.confcityHamiltonen
dc.relation.confcountryNew Zealanden
dc.relation.forthcomingnonen
dc.description.ssrncandidatenonen
dc.description.halcandidateouien
dc.description.readershiprechercheen
dc.description.audienceInternationalen
dc.relation.Isversionofjnlpeerreviewednonen
dc.relation.Isversionofjnlpeerreviewednonen
dc.date.updated2017-01-27T10:52:52Z
hal.identifierhal-01447945*
hal.version1*
hal.author.functionaut
hal.author.functionaut
hal.author.functionaut
hal.author.functionaut


Files in this item

FilesSizeFormatView

There are no files associated with this item.

This item appears in the following Collection(s)

Show simple item record