Summary
International Symposium on Nonlinear Theory and Its Applications
2022
Session Number:C3L-D
Session:
Number:C3L-D-02
Synthesis of Permutation Binary Neural Networks
Kento Saka , Toshimichi Saito,
pp.487-490
Publication Date:12/12/2022
Online ISSN:2188-5079
DOI:10.34385/proc.71.C3L-D-02
PDF download (1.7MB)
Summary:
This paper studied a simple synthesis method of permutation binary neural networks. The network is characterized by local binary connection, global permutation connection, and the signum activation function. Depending on the connection, the network generates various periodic orbits of binary vectors. The synthesis method is based on the genetic algorithm. Performing basic numerical experiments, efficiency of the synthesis method is confirmed.