NONLINEAR INDEPENDENT COMPONENT ANALYSIS USING ENSEMBLE LEARNING: THEORY
Email: Harri.Valpola@hut.fi URL: http://www.cis.hut.fi/
A nonlinear version of independent component analysis
is presented. The mapping from sources to observations
is modelled by a multilayer perceptron network and
the distributions of sources are modelled by mixtures
of Gaussians. The posterior probability of all the un
known parameters is estimated by ensemble learning.
In this paper, we present the theory of the method, and
in a companion paper experimental results.