Gaussian Processes for Source Separation in Overdetermined Bilinear Mixtures

Abstract : In this work, we consider the nonlinear Blind Source Separation (BSS) problem in the context of overdetermined Bilinear Mixtures, in which a linear structure can be employed for performing separation. Based on the Gaussian Process (GP) framework, two approaches are proposed: the predictive distribution and the maximization of the marginal likelihood. In both cases, separation can be achieved by assuming that the sources are Gaussian and temporally correlated. The results with synthetic data are favorable to the proposal.
Document type :
Conference papers
Complete list of metadatas

Cited literature [12 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-01480992
Contributor : Christian Jutten <>
Submitted on : Thursday, March 2, 2017 - 9:39:36 AM
Last modification on : Monday, July 8, 2019 - 3:08:59 PM
Long-term archiving on : Wednesday, May 31, 2017 - 1:14:52 PM

File

Fantinato_LQ_Gaussian_Process-...
Files produced by the author(s)

Identifiers

Citation

Denis Fantinato, Leonardo Duarte, Bertrand Rivet, Bahram Ehsandoust, Romis Attux, et al.. Gaussian Processes for Source Separation in Overdetermined Bilinear Mixtures. 13th International Conference on Latent Variable Analysis and Signal Separation (LVA/ICA 2017), Olivier Michel; Nadège Thirion-Moreau, Feb 2017, Grenoble, France. pp.300 - 309, ⟨10.1007/978-3-319-53547-0_29⟩. ⟨hal-01480992⟩

Share

Metrics

Record views

965

Files downloads

271