Hyperparameter-free sparse regression of grouped variables

Ted Kronvall, Stefan Ingi Adalbjornsson, Santhosh Nadig, Andreas Jakobsson

Forskningsoutput: Kapitel i bok/rapport/Conference proceedingKonferenspaper i proceedingPeer review

Sammanfattning

In this paper, we introduce a novel framework for semi-parametric estimation of an unknown number of signals, each parametrized by a group of components. Via a reformulation of the covariance fitting criteria, we formulate a convex optimization problem over a grid of candidate representations, promoting solutions with only a few active groups. Utilizing the covariance fitting allows for a hyperparameter-free estimation procedure, highly robust against coherency between candidates, while still allowing for a computationally efficient implementation. Numerical simulations illustrate how the proposed method offers a performance similar to the group-LASSO for incoherent dictionaries, and superior performance for coherent dictionaries.

Originalspråkengelska
Titel på värdpublikationConference Record of the 50th Asilomar Conference on Signals, Systems and Computers, ACSSC 2016
FörlagIEEE Computer Society
Sidor394-398
Antal sidor5
ISBN (elektroniskt)9781538639542
DOI
StatusPublished - 2017 mars 1
Evenemang50th Asilomar Conference on Signals, Systems and Computers, ACSSC 2016 - Pacific Grove, USA
Varaktighet: 2016 nov. 62016 nov. 9

Konferens

Konferens50th Asilomar Conference on Signals, Systems and Computers, ACSSC 2016
Land/TerritoriumUSA
OrtPacific Grove
Period2016/11/062016/11/09

Ämnesklassifikation (UKÄ)

  • Sannolikhetsteori och statistik
  • Signalbehandling

Fingeravtryck

Utforska forskningsämnen för ”Hyperparameter-free sparse regression of grouped variables”. Tillsammans bildar de ett unikt fingeravtryck.

Citera det här