Sparse Estimation with the Swept Approximated Message-Passing Algorithm

Abstract : Approximate Message Passing (AMP) has been shown to be a superior method for inference problems, such as the recovery of signals from sets of noisy, lower-dimensionality measurements, both in terms of reconstruction accuracy and in computational efficiency. However, AMP suffers from serious convergence issues in contexts that do not exactly match its assumptions. We propose a new approach to stabilizing AMP in these contexts by applying AMP updates to individual coefficients rather than in parallel. Our results show that this change to the AMP iteration can provide theoretically expected, but hitherto unobtainable, performance for problems on which the standard AMP iteration diverges. Additionally, we find that the computational costs of this swept coefficient update scheme is not unduly burdensome, allowing it to be applied efficiently to signals of large dimensionality.
Document type :
Preprints, Working Papers, ...
Liste complète des métadonnées

https://hal-cea.archives-ouvertes.fr/cea-01140814
Contributor : Emmanuelle De Laborderie <>
Submitted on : Thursday, April 9, 2015 - 3:37:40 PM
Last modification on : Wednesday, January 23, 2019 - 2:39:04 PM

Links full text

Identifiers

  • HAL Id : cea-01140814, version 1
  • ARXIV : 1406.4311

Citation

Andre Manoel, Florent Krzakala, Eric W. Tramel, Lenka Zdeborová. Sparse Estimation with the Swept Approximated Message-Passing Algorithm. t14/184. 11 pages, 3 figures, implementation available at https://github.com/eric-tramel/SwAMP-Demo. 2015. 〈cea-01140814〉

Share

Metrics

Record views

286