ML Anthology

Natural Gradient Learning for Over- and Under-Complete Bases in ICA

Shun-ichi Amari
NeCo 1999 pp. 1875-1883
doi:10.1162/089976699300015990 /neco/1999/amari1999neco-natural/

Abstract

Independent component analysis or blind source separation is a new technique of extracting independent signals from mixtures. It is applicable even when the number of independent sources is unknown and is larger or smaller than the number of observed mixture signals. This article extends the natural gradient learning algorithm to be applicable to these overcomplete and undercomplete cases. Here, the observed signals are assumed to be whitened by preprocessing, so that we use the natural Riemannian gradient in Stiefel manifolds.

NeCo Semantic Scholar

Cite

Text

Amari. "Natural Gradient Learning for Over- and Under-Complete Bases in ICA." Neural Computation, 1999. doi:10.1162/089976699300015990

Markdown

[Amari. "Natural Gradient Learning for Over- and Under-Complete Bases in ICA." Neural Computation, 1999.](https://mlanthology.org/neco/1999/amari1999neco-natural/) doi:10.1162/089976699300015990

BibTeX

@article{amari1999neco-natural,
  title     = {{Natural Gradient Learning for Over- and Under-Complete Bases in ICA}},
  author    = {Amari, Shun-ichi},
  journal   = {Neural Computation},
  year      = {1999},
  pages     = {1875-1883},
  doi       = {10.1162/089976699300015990},
  volume    = {11},
  url       = {https://mlanthology.org/neco/1999/amari1999neco-natural/}
}