Abstract:

In fuzzy neural networks, the multi-variable continuous functions class is extended as a functions class of preserving maximum values on compact sets. The properties and the metric problems of the extended functions class on compact sets are studied by means of the concept of preserving maximum values. Finally, it is found that three-layer regular fuzzy neural networks possess approximation with respect to the extended fuzzy function class. Also, the approximation effect between the expected output and the real output of regular fuzzy neural networks is analysed and text is studied by an example.

Key words: regular fuzzy neural networks; preserving maximum value; support function; universal approximation