Multi-Layer Perceptrons as Nonlinear Generative Models for Unsupervised Learning: a Bayesian Treatment

Harri Lappalainen
Neural Networks Research Centre
Helsinki University of Technology, Finland
Harri.Lappalainen@hut.fi

Xavier Giannakopoulos
IDSIA
Lugano, Switzerland
xavier@idsia.ch

Post Script version (126 kb)

Abstract:

In this paper, multi-layer perceptrons are used as nonlinear generative models. The problem of indeterminacy of the models is resolved using a recently developed Bayesian method called ensemble learning. Using a Bayesian approach, models can be compared according to their probabilities. In simulations with artificial data, the network is able to find the underlying causes of the observations despite the strong nonlinearities of the data.