Supervised learning in fuzzy neural networks

Fuzzy neural networks are a connecting link between fuzzy logic and neurocomputing. The goal of this connection is to combine the advantages of both disciplines in order to process uncertain or vague information. Fuzzy neural networks are (layer by layer) fully interconnected feedforward networks whose processing elements - the formal neurons - operate on fuzzy numbers instead of real numbers. Our aim is to achieve supervised learning in this kind of neural networks. Our work is based on the traditional backpropagation algorithm. Therefore we will give a brief survey of the basics of fuzzy logic and neurocomputing first. (orig.)Available from TIB Hannover: RO 7057(1995,12) / FIZ - Fachinformationszzentrum Karlsruhe / TIB - Technische InformationsbibliothekSIGLEDEGerman