Title :
OWA neurons: a new class of fuzzy neurons
Author :
Yager, Ronald R.
Author_Institution :
Machine Intelligence Inst., Iona Coll., New Rochelle, NY, USA
Abstract :
The OWA (ordered weighted averaging) aggregation operators are described. It is shown how these operators provide aggregations which are said to be guided by linguistic quantifiers. There is a close connection between these quantifier-guided aggregations and the process of neuron aggregation. This connection is used to suggest a novel family of neurons called OWA neurons. Situations where these OWA neurons have more desirable characteristics than the classic neurons are discussed. It is shown that the OWA neurons are amenable to the currently popular learning paradigms such as backpropagation
Keywords :
backpropagation; fuzzy logic; fuzzy set theory; neural nets; OWA neurons; aggregation operators; backpropagation; fuzzy neurons; learning; linguistic quantifiers; neuron aggregation; ordered weighted averaging; Backpropagation; Dispersion; Educational institutions; Entropy; Logic; Machine intelligence; Mathematical programming; Neural networks; Neurons; Open wireless architecture;
Conference_Titel :
Neural Networks, 1992. IJCNN., International Joint Conference on
Conference_Location :
Baltimore, MD
Print_ISBN :
0-7803-0559-0
DOI :
10.1109/IJCNN.1992.287131
