Optimal Pooling of Covariance Matrix Estimates Across Multiple Classes

 |  Login

Show simple item record

dc.contributor Aalto-yliopisto fi
dc.contributor Aalto University en
dc.contributor.author Raninen, Elias
dc.contributor.author Ollila, Esa
dc.date.accessioned 2018-12-10T10:28:22Z
dc.date.available 2018-12-10T10:28:22Z
dc.date.issued 2018-09-10
dc.identifier.citation Raninen , E & Ollila , E 2018 , Optimal Pooling of Covariance Matrix Estimates Across Multiple Classes . in 2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 - Proceedings . vol. 2018-April , 8461327 , Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing , Institute of Electrical and Electronics Engineers , pp. 4224-4228 , IEEE International Conference on Acoustics, Speech and Signal Processing , Calgary , Canada , 15/04/2018 . DOI: 10.1109/ICASSP.2018.8461327 en
dc.identifier.isbn 978-1-5386-4659-5
dc.identifier.isbn 978-1-5386-4658-8
dc.identifier.issn 2379-190X
dc.identifier.other PURE UUID: b9bd4811-60df-4c47-aac6-f66cb972a2f8
dc.identifier.other PURE ITEMURL: https://research.aalto.fi/en/publications/optimal-pooling-of-covariance-matrix-estimates-across-multiple-classes(b9bd4811-60df-4c47-aac6-f66cb972a2f8).html
dc.identifier.other PURE FILEURL: https://research.aalto.fi/files/28772321/ELEC_RaninenOllilaICASSP2018AuthorsManuscript.pdf
dc.identifier.uri https://aaltodoc.aalto.fi/handle/123456789/35238
dc.description.abstract The paper considers the problem of estimating the covariance matrices of multiple classes in a low sample support condition, where the data dimensionality is comparable to, or larger than, the sample sizes of the available data sets. In such conditions' a common approach is to shrink the class sample covariance matrices (SCMs) towards the pooled SCM. The success of this approach hinges upon the ability to choose the optimal regularization parameter. Typically, a common regularization level is shared among the classes and determined via a procedure based on cross-validation. We use class-specific regularization levels since this enables the derivation of the optimal regularization parameter for each class in terms of the minimum mean squared error (MMSE). The optimal parameters depend on the true unknown class population covariances. Consistent estimators of the parameters can, however, be easily constructed under the assumption that the class populations follow (unspecified) elliptically symmetric distributions. We demonstrate the performance of the proposed method via a simulation study as well as via an application to discriminant analysis using both synthetic and real data sets. en
dc.format.extent 5
dc.format.extent 4224-4228
dc.format.mimetype application/pdf
dc.language.iso en en
dc.relation.ispartof IEEE International Conference on Acoustics, Speech and Signal Processing en
dc.relation.ispartofseries 2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 - Proceedings en
dc.relation.ispartofseries Volume 2018-April en
dc.relation.ispartofseries Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing en
dc.rights openAccess en
dc.subject.other Software en
dc.subject.other Signal Processing en
dc.subject.other Electrical and Electronic Engineering en
dc.subject.other 113 Computer and information sciences en
dc.title Optimal Pooling of Covariance Matrix Estimates Across Multiple Classes en
dc.type A4 Artikkeli konferenssijulkaisussa fi
dc.description.version Peer reviewed en
dc.contributor.department Department of Signal Processing and Acoustics
dc.subject.keyword Classification
dc.subject.keyword Covariance matrix estimation
dc.subject.keyword Elliptical distribution
dc.subject.keyword Regularization
dc.subject.keyword Software
dc.subject.keyword Signal Processing
dc.subject.keyword Electrical and Electronic Engineering
dc.subject.keyword 113 Computer and information sciences
dc.identifier.urn URN:NBN:fi:aalto-201812106253
dc.identifier.doi 10.1109/ICASSP.2018.8461327
dc.type.version acceptedVersion


Files in this item

Files Size Format View

There are no files associated with this item.

This item appears in the following Collection(s)

Show simple item record

Search archive


Advanced Search

article-iconSubmit a publication

Browse

My Account