Scaling Laws and Local Minima in Hebbian ICA

We study the dynamics of a Hebbian ICA algorithm extracting a single non-Gaussian component from a high-dimensional Gaussian background. For both on-line and batch learning we find that a surprisingly large number of examples are required to avoid trapping in a sub-optimal state close to the initial conditions. To extract a skewed signal at least

• Type: Conference
• Year: 2001
• Where: NIPS
• Authors: Magnus Rattray, Gleb Basalyga
• Keywords: Hebbian ICA Algorithm, NIPS 2001, NIPS 2007, Single Non-Gaussian Component, Skewed Signal