A Boolean approach to construct neural networks for non-Boolean problems

Abstract

A neural network construction method for problems specified for data sets with input and/or output values in the continuous or discrete domain is described and evaluated. This approach is based on a Boolean approximation of the data set and is generic for various neural network architectures. The construction method takes advantage of a construction method for Boolean problems without increasing the dimensions of the input or output vectors, which is an advantage over approaches which work on a binarized version of the data set with an increased number of input and output elements. Further, the networks are pruned in a second phase in order to obtain very small networks.