Title :
A low-complexity fuzzy activation function for artificial neural networks
Author :
Soria-Olivas, E. ; Martin-Guerrero, J.D. ; Camps-Valls, G. ; Serrano-Lopez, A.J. ; Calpe-Maravilla, J. ; Gomez-Chova, L.
Author_Institution :
Dept. of Enginyeria Electronica, Univ. de Valencia, Spain
Abstract :
A novel fuzzy-based activation function for artificial neural networks is proposed. This approach provides easy hardware implementation and straightforward interpretability in the basis of IF-THEN rules. Backpropagation learning with the new activation function also has low computational complexity. Several application examples ( XOR gate, chaotic time-series prediction, channel equalization, and independent component analysis) support the potential of the proposed scheme.
Keywords :
backpropagation; computational complexity; fuzzy logic; multilayer perceptrons; transfer functions; artificial neural network; backpropagation learning; computational complexity; fuzzy logic; hardware implementation; if-then rules; low-complexity fuzzy activation function; rule extraction; Artificial neural networks; Backpropagation; Chaos; Computational complexity; Fuzzy logic; Fuzzy neural networks; Hardware; Independent component analysis; Neural networks; Neurons;
Journal_Title :
Neural Networks, IEEE Transactions on
DOI :
10.1109/TNN.2003.820444
