Title :
Hamiltonian Neural Networks Based Classifiers and Mappings
Author :
Sieñko, W. ; Zamojski, D.
Author_Institution :
Member, IEEE, Department of Electrical Engineering, Gdynia Maritime University, Poland, e-mail: sienko@am.gdynia.pl
Abstract :
We point out that Hamiltonian Neural Networks (HNN) and based on HNN orthogonal filters can be used as universal signal processors. To illustrate it, we propose a procedure for design of large scale mappings, classifiers and supervised learning systems.
Keywords :
Adaptive systems; Artificial intelligence; Artificial neural networks; Equations; Filters; Large-scale systems; Neural networks; Neurons; Signal processing; Supervised learning;
Conference_Titel :
Neural Networks, 2006. IJCNN '06. International Joint Conference on
Print_ISBN :
0-7803-9490-9
DOI :
10.1109/IJCNN.2006.246765
