Tw. Lee et al., Independent component analysis using an extended infomax algorithm for mixed subgaussian and supergaussian sources, NEURAL COMP, 11(2), 1999, pp. 417-441
Citations number
52
Categorie Soggetti
Neurosciences & Behavoir","AI Robotics and Automatic Control
An extension of the infomax algorithm of Bell and Sejnowski (1995) is prese
nted that is able blindly to separate mixed signals with sub- and supergaus
sian source distributions. This was achieved by using a simple type of lear
ning rule first derived by Girolami (1997) by choosing negentropy as a proj
ection pursuit index. Parameterized probability distributions that have sub
- and supergaussian regimes were used to derive a general learning rule tha
t preserves the simple architecture proposed by Bell and Sejnowski (1995),
is optimized using the natural gradient by Amari (1998), and uses the stabi
lity analysis of Cardoso and Laheld (1996) to switch between sub- and super
gaussian regimes. We demonstrate that the extended infomax algorithm is abl
e to separate 20 sources with a variety of source distributions easily. App
lied to high-dimensional data from electroencephalographic recordings, it i
s effective at separating artifacts such as eye blinks and line noise from
weaker electrical signals that arise from sources in the brain.