Physics – Condensed Matter – Disordered Systems and Neural Networks
Scientific paper
2001-05-03
Physics
Condensed Matter
Disordered Systems and Neural Networks
17 pages, 3 figures. To appear in Neural Computation
Scientific paper
Previous analytical studies of on-line Independent Component Analysis (ICA) learning rules have focussed on asymptotic stability and efficiency. In practice the transient stages of learning will often be more significant in determining the success of an algorithm. This is demonstrated here with an analysis of a Hebbian ICA algorithm which can find a small number of non-Gaussian components given data composed of a linear mixture of independent source signals. An idealised data model is considered in which the sources comprise a number of non-Gaussian and Gaussian sources and a solution to the dynamics is obtained in the limit where the number of Gaussian sources is infinite. Previous stability results are confirmed by expanding around optimal fixed points, where a closed form solution to the learning dynamics is obtained. However, stochastic effects are shown to stabilise otherwise unstable sub-optimal fixed points. Conditions required to destabilise one such fixed point are obtained for the case of a single non-Gaussian component, indicating that the initial learning rate \eta required to successfully escape is very low (\eta = O(N^{-2}) where N is the data dimension) resulting in very slow learning typically requiring O(N^3) iterations. Simulations confirm that this picture holds for a finite system.
No associations
LandOfFree
Stochastic trapping in a solvable model of on-line independent component analysis does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.
If you have personal experience with Stochastic trapping in a solvable model of on-line independent component analysis, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Stochastic trapping in a solvable model of on-line independent component analysis will most certainly appreciate the feedback.
Profile ID: LFWR-SCP-O-477654