Go to content
UR Home

A Geometric Algorithm for Overcomplete Linear ICA

Theis, Fabian J. ; Lang, Elmar ; Puntonet, Carlos G.


We generalize geometric algorithms to overcomplete cases with more sources than sensors. With geometric ICA we get an efficient method for the matrix-recovery step in the framework of a two-step approach to the source separation problem. The second step - source-recovery - uses a maximum-likelihood approach. There we prove that the shortest-path algorithm as proposed by Bofill and Zibulevsky indeed solves the maximum-likelihood conditions.

Owner only: item control page
  1. Homepage UR

University Library

Publication Server


Publishing: oa@ur.de
0941 943 -4239 or -69394

Dissertations: dissertationen@ur.de
0941 943 -3904

Research data: datahub@ur.de
0941 943 -5707

Contact persons