Independent component analysis using an extended infomax algorithm for mixed subgaussian and supergaussian sources

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.