Faster independent component analysis by preconditioning with Hessian approximations

Abstract : Independent Component Analysis (ICA) is a technique for unsupervised exploration of multi-channel data that is widely used in observational sciences. In its classic form, ICA relies on modeling the data as linear mixtures of non-Gaussian independent sources. The maximization of the corresponding likelihood is a challenging problem if it has to be completed quickly and accurately on large sets of real data. We introduce the Preconditioned ICA for Real Data (Picard) algorithm, which is a relative L-BFGS algorithm preconditioned with sparse Hessian approximations. Extensive numerical comparisons to several algorithms of the same class demonstrate the superior performance of the proposed technique, especially on real data, for which the ICA model does not necessarily hold.
Document type :
Preprints, Working Papers, ...
Complete list of metadatas

Cited literature [46 references]  Display  Hide  Download

https://hal.inria.fr/hal-01552340
Contributor : Pierre Ablin <>
Submitted on : Friday, September 8, 2017 - 6:50:18 PM
Last modification on : Monday, May 27, 2019 - 11:44:03 AM

File

faster-ica-arxiv (1).pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01552340, version 2
  • ARXIV : 1706.08171

Citation

Pierre Ablin, Jean-François Cardoso, Alexandre Gramfort. Faster independent component analysis by preconditioning with Hessian approximations. 2017. ⟨hal-01552340v2⟩

Share

Metrics

Record views

571

Files downloads

1302