Accelerating likelihood optimization for ICA on real signals - Archive ouverte HAL Access content directly
Conference Papers Year :

Accelerating likelihood optimization for ICA on real signals

Acceleration de l'optimisation de la vraisemblance pour l'analyse en composantes indépendantes de signaux réels

(1, 2) , (3, 4) , (1, 2)


We study optimization methods for solving the maximum likelihood formulation of independent component analysis (ICA). We consider both the the problem constrained to white signals and the unconstrained problem. The Hessian of the objective function is costly to compute, which renders Newton's method impractical for large data sets. Many algorithms proposed in the literature can be rewritten as quasi-Newton methods, for which the Hessian approximation is cheap to compute. These algorithms are very fast on simulated data where the linear mixture assumption really holds. However, on real signals, we observe that their rate of convergence can be severely impaired. In this paper, we investigate the origins of this behavior, and show that the recently proposed Preconditioned ICA for Real Data (Picard) algorithm overcomes this issue on both constrained and unconstrained problems.
Fichier principal
Vignette du fichier
main.pdf (568.93 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-01822602 , version 1 (25-06-2018)



Pierre A Ablin, Jean-François Cardoso, Alexandre Gramfort. Accelerating likelihood optimization for ICA on real signals. LVA-ICA 2018, Jul 2018, Guildford, United Kingdom. ⟨hal-01822602⟩
149 View
212 Download



Gmail Facebook Twitter LinkedIn More