Skip to Main content Skip to Navigation
Conference papers

Mining the Bilinear Structure of Data with Approximate Joint Diagonalization

Louis Korczowski 1, * Florent Bouchard 1 Christian Jutten 1 Marco Congedo 1
* Corresponding author
1 GIPSA-VIBS [2016-2019] - GIPSA - Vision and Brain Signal Processing
GIPSA-DIS [2016-2019] - Département Images et Signal
Abstract : Approximate Joint Diagonalization of a matrix set can solve the linear Blind Source Separation problem. If the data possesses a bilinear structure, for example a spatio-temporal structure, transformations such as tensor decomposition can be applied. In this paper we show how the linear and bilinear joint diagonalization can be applied for extracting sources according to a composite model where some of the sources have a linear structure and other a bilinear structure. This is the case of Event Related Potentials (ERPs). The proposed model achieves higher performance in term of shape and robustness for the estimation of ERP sources in a Brain Computer Interface experiment.
Complete list of metadatas

Cited literature [16 references]  Display  Hide  Download
Contributor : Louis Korczowski <>
Submitted on : Monday, August 29, 2016 - 2:55:33 PM
Last modification on : Thursday, July 30, 2020 - 3:49:20 AM
Document(s) archivé(s) le : Wednesday, November 30, 2016 - 2:29:18 PM


Files produced by the author(s)


  • HAL Id : hal-01357245, version 1



Louis Korczowski, Florent Bouchard, Christian Jutten, Marco Congedo. Mining the Bilinear Structure of Data with Approximate Joint Diagonalization. 24th European Signal Processing Conference (EUSIPCO 2016), EURASIP, Aug 2016, Budapest, Hungary. pp.667-671. ⟨hal-01357245⟩



Record views


Files downloads