Advanced

Online Group-Sparse Regression Using the Covariance Fitting Criterion

Kronvall, Ted LU ; Adalbjörnsson, Stefan Ingi LU ; Nadig, Santhosh and Jakobsson, Andreas LU (2017) In European Signal Processing Conference (EUSIPCO) CFP1740S-USB.
Abstract
In this paper, we present a time-recursive implementation of a recent hyperparameter-free group-sparse estimation technique. This is achieved byr eformulating the original method, termed group-SPICE, as a square-root group-LASSO with a suitable regularization level, for which a time-recursive implementation is derived. Using a proximal gradient step for lowering the computational cost, the proposed method may effectively cope with data sequences consisting of both stationary and non-stationary signals, such as transients, and/or amplitude modulated signals. Numerical examples illustrates the efficacy of the proposed method for both coherent Gaussian dictionaries and for the multi-pitch estimation problem.
Please use this url to cite or link to this publication:
author
organization
publishing date
type
Chapter in Book/Report/Conference proceeding
publication status
published
subject
host publication
Proceedings of the 25th European Signal Processing Conference (EUSIPCO)
series title
European Signal Processing Conference (EUSIPCO)
volume
CFP1740S-USB
pages
5 pages
publisher
EURASIP
ISSN
2076-1465
ISBN
978-0-9928626-8-8
language
English
LU publication?
yes
id
40f6dfcc-75fe-4384-931d-d17494c3d55e
alternative location
http://www.eurasip.org/Proceedings/Eusipco/Eusipco2017/papers/1570347373.pdf
date added to LUP
2017-10-05 14:19:00
date last changed
2019-03-08 02:33:57
@inproceedings{40f6dfcc-75fe-4384-931d-d17494c3d55e,
  abstract     = {In this paper, we present a time-recursive implementation of a recent hyperparameter-free group-sparse estimation technique. This is achieved byr eformulating the original method, termed group-SPICE, as a square-root group-LASSO with a suitable regularization level, for which a time-recursive implementation is derived. Using a proximal gradient step for lowering the computational cost, the proposed method may effectively cope with data sequences consisting of both stationary and non-stationary signals, such as transients, and/or amplitude modulated signals. Numerical examples illustrates the efficacy of the proposed method for both coherent Gaussian dictionaries and for the multi-pitch estimation problem. },
  author       = {Kronvall, Ted and Adalbjörnsson, Stefan Ingi and Nadig, Santhosh and Jakobsson, Andreas},
  booktitle    = {European Signal Processing Conference (EUSIPCO)},
  isbn         = {978-0-9928626-8-8},
  issn         = {2076-1465},
  language     = {eng},
  pages        = {5},
  publisher    = {EURASIP},
  title        = {Online Group-Sparse Regression Using the Covariance Fitting Criterion},
  volume       = {CFP1740S-USB},
  year         = {2017},
}