Abstract: in the paper new hybrid system of computational intelligence called the Cascade Neo-Fuzzy? Neural Network (CNFNN) is introduced. This architecture has the similar structure with the Cascade-Correlation? Learning Architecture proposed by S.E. Fahlman and C. Lebiere, but differs from it in type of artificial neurons. CNFNN contains neo-fuzzy neurons, which can be adjusted using high-speed linear learning procedures. Proposed CNFNN is characterized by high learning rate, low size of learning sample and its operations can be described by fuzzy linguistic “if-then” rules providing “transparency” of received results, as compared with conventional neural networks. Using of cubic-spline membership functions instead of conventional triangular functions allows increasing accuracy of smooth functions approximation.

Keywords: artificial neural networks, constructive approach, fuzzy inference, hybrid systems, neo-fuzzy neuron, cubic-spline functions.

ACM Classification Keywords: I.2.6 Learning – Connectionism and neural nets.

Link:

THE CASCADE NEO-FUZZY ARCHITECTURE USING CUBIC–SPKINE ACTIVATION FUNCTIONS

Yevgeniy Bodyanskiy, Yevgen Viktorov

http://foibg.com/ijita/vol16/IJITA16-3-p05.pdf