Title :
Fuzzy aggregation networks of hybrid neurons with generalized Yager operators
Author :
Keller, James M. ; Yang, Hong
Author_Institution :
Dept. of Electr. & Comput. Eng., Missouri Univ., Columbia, MO, USA
Abstract :
There have been several models of neural network structures which incorporate fuzzy set theoretic connectives in the activation functions of the nodes. Fuzzy aggregation networks (FANs) are a flexible and trainable class of such networks. To date, FANs have used exponentially weighted inputs in the operators. In this paper, we introduce a linearly weighted version of the FAN which shows the same excellent decision-making potential as the more complex exponentially weighted predecessors
Keywords :
fuzzy neural nets; fuzzy set theory; activation functions; decision-making potential; exponentially weighted inputs; fuzzy aggregation networks; fuzzy set theoretic connectives; generalized Yager operators; hybrid neurons; linearly weighted version; neural network structures; Computer networks; Decision making; Fans; Feedforward neural networks; Fuzzy neural networks; Fuzzy sets; Neural networks; Neurons; Pattern recognition; Uncertainty;
Conference_Titel :
Neural Networks, 1995. Proceedings., IEEE International Conference on
Conference_Location :
Perth, WA
Print_ISBN :
0-7803-2768-3
DOI :
10.1109/ICNN.1995.487715
