Towards a Local Separation Performances Estimator Using Common ICA Contrast Functions ?

Vrins, Frédéric;Archambeau, Cédric;Verleysen, Michel
(2004) European Symposium on Artificial Neural Networks (ESANN 2004) — Location: Bruges, Belgium (28.April.2004)

Files

93-TowardsaLocalSeparationPerformancesEstimatorUsingCommonICAContrastFunctions.pdf
  • Restricted Access
  • Adobe PDF
  • 854 KB
Request a copy
93-TowardsaLocalSeparationPerformancesEstimatorUsingCommonICAContrastFunctions.pdf
  • Restricted Access
  • Adobe PDF
  • 854 KB
Request a copy

Details

Authors
Abstract
In most ICA algorithms, the separation performances are estimated through the evaluation of a contrast function , used in the update rule of elements of the unmixing matrix. In particular situations, optimizing does not lead to optimize the extraction of each source, one by one. However, in some applications, one can be interested in quantifying the ex- traction performances of a specic signal. In this paper, we emphasize that none of the usual 's could be directly applied, without further precautions, to evaluate the performances of the local separation.
Affiliations

Citations

  • APA
  • Chicago
  • FWB

Vrins, F., Archambeau, C., & Verleysen, M. (2004). Towards a Local Separation Performances Estimator Using Common ICA Contrast Functions ? Proceedings of ESANN 2004, 211-216. https://hdl.handle.net/2078.5/253944