Generalized SURE for optimal shrinkage of singular values in low-rank matrix denoising
| hal.structure.identifier | Institut de Mathématiques de Bordeaux [IMB] | |
| dc.contributor.author | BIGOT, Jérémie | |
| hal.structure.identifier | Institut de Mathématiques de Bordeaux [IMB] | |
| dc.contributor.author | DELEDALLE, Charles-Alban | |
| hal.structure.identifier | Institut de Mathématiques de Bordeaux [IMB] | |
| dc.contributor.author | FÉRAL, Delphine | |
| dc.date.accessioned | 2024-04-04T03:10:17Z | |
| dc.date.available | 2024-04-04T03:10:17Z | |
| dc.date.created | 2017-04-22 | |
| dc.date.issued | 2017-11 | |
| dc.identifier.issn | 1532-4435 | |
| dc.identifier.uri | https://oskar-bordeaux.fr/handle/20.500.12278/193678 | |
| dc.description.abstractEn | We consider the problem of estimating a low-rank signal matrix from noisy measurements under the assumption that the distribution of the data matrix belongs to an exponential family. In this setting, we derive generalized Stein's unbiased risk estimation (SURE) formulas that hold for any spectral estimators which shrink or threshold the singular values of the data matrix. This leads to new data-driven spectral estimators, whose optimality is discussed using tools from random matrix theory and through numerical experiments. Under the spiked population model and in the asymptotic setting where the dimensions of the data matrix are let going to infinity, some theoretical properties of our approach are compared to recent results on asymptotically optimal shrinking rules for Gaussian noise. It also leads to new procedures for singular values shrinkage in finite-dimensional matrix denoising for Gamma-distributed and Poisson-distributed measurements. | |
| dc.language.iso | en | |
| dc.publisher | Microtome Publishing | |
| dc.subject.en | matrix denoising | |
| dc.subject.en | singular value decomposition | |
| dc.subject.en | low-rank model | |
| dc.subject.en | spectral estimator | |
| dc.subject.en | Stein's unbiased risk estimate | |
| dc.subject.en | random matrix theory | |
| dc.subject.en | exponential family | |
| dc.subject.en | optimal shrinkage rule | |
| dc.subject.en | degrees of freedom | |
| dc.title.en | Generalized SURE for optimal shrinkage of singular values in low-rank matrix denoising | |
| dc.type | Article de revue | |
| dc.subject.hal | Statistiques [stat] | |
| dc.subject.hal | Mathématiques [math] | |
| dc.identifier.arxiv | 1605.07412 | |
| bordeaux.journal | Journal of Machine Learning Research | |
| bordeaux.hal.laboratories | Institut de Mathématiques de Bordeaux (IMB) - UMR 5251 | * |
| bordeaux.institution | Université de Bordeaux | |
| bordeaux.institution | Bordeaux INP | |
| bordeaux.institution | CNRS | |
| bordeaux.peerReviewed | oui | |
| hal.identifier | hal-01323285 | |
| hal.version | 1 | |
| hal.popular | non | |
| hal.audience | Internationale | |
| hal.origin.link | https://hal.archives-ouvertes.fr//hal-01323285v1 | |
| bordeaux.COinS | ctx_ver=Z39.88-2004&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.jtitle=Journal%20of%20Machine%20Learning%20Research&rft.date=2017-11&rft.eissn=1532-4435&rft.issn=1532-4435&rft.au=BIGOT,%20J%C3%A9r%C3%A9mie&DELEDALLE,%20Charles-Alban&F%C3%89RAL,%20Delphine&rft.genre=article |
Fichier(s) constituant ce document
| Fichiers | Taille | Format | Vue |
|---|---|---|---|
|
Il n'y a pas de fichiers associés à ce document. |
|||