A modular Boolean architecture for pattern recognition
André C. P. L. F. de Carvalho, M.C. Fairhurst, D.L. Bisset · 2002
This paper describes a modular neural structure where a self-organising Boolean neural network architecture (SOFT) is used as a front-end processor to a feedforward neural architecture based on goal-seeking principles (GSN/sup f/). The characteristics of this architecture are illustrated by considering its application to a character recognition problem.>