Alternating group lasso for block-term tensor decomposition with application to ECG source separation

Abstract : In some applications, blind source separation can be performed by computing an approximate block-term tensor decomposition (BTD), under much milder constraints than matrix-based techniques. However, choosing the BTD model structure (i.e., the number of blocks and their ranks) is a difficult problem, and the standard least-squares formulation can be ill-posed. This paper proposes an alternating group lasso algorithm to compute approximate low-rank BTDs. It solves, in a provably convergent manner, a well-posed mixed-norm regularized tensor approximation problem which allows jointly estimating the model parameters and its structure. A variant is also put forward for dealing with linearly constrained blocks, motivated by the problem of blind separation of sums of complex exponentials, which can be cast as a low-rank Hankel-structured block-term tensor approximation problem. An experimental comparison with a standard nonlinear least-squares algorithm on synthetic tensor data indicates that the proposed algorithm is much more robust with respect to initialization. We also apply the constrained variant to the extraction of atrial activity from semi-synthetic and real-world electrocardiogram recordings during atrial fibrillation episodes. Our results show its ability to consistently select an adequate structure and to extract multiple signals which can be physiologically interpreted as atrial fibrillation patterns.
Document type :
Preprints, Working Papers, ...
Liste complète des métadonnées

https://hal.archives-ouvertes.fr/hal-01899469
Contributor : José Henrique de Morais Goulart <>
Submitted on : Thursday, November 22, 2018 - 1:01:49 PM
Last modification on : Tuesday, February 12, 2019 - 8:02:04 AM
Document(s) archivé(s) le : Saturday, February 23, 2019 - 2:27:16 PM

File

double.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01899469, version 1

Citation

José Henrique de Morais Goulart, Pedro Marinho R. de Oliveira, Rodrigo Cabral Farias, Vicente Zarzoso, Pierre Comon. Alternating group lasso for block-term tensor decomposition with application to ECG source separation. 2018. ⟨hal-01899469⟩

Share

Metrics

Record views

236

Files downloads

77