Summary
the 2014 International Symposium on Nonlinear Theory and its Applications
2014
Session Number:C2L-C
Session:
Number:C2L-C2
Back Propagation Learning Based on an IDL Model
Yuta HORIUCHI, Yoshihiro HAYAKAWA, Takeshi ONOMI, Koji NAKAJIMA,
pp.512-515
Publication Date:2014/9/14
Online ISSN:2188-5079
Summary:
The Inverse function Delayed (ID) model was proposed as one of novel neural models. The ID model has an ability of oscillation, and this model can solve some local minimum problem in combinatorial optimization problems. However, it needs large calculation cost, and it is difficult to apply for large size combinational optimization problems. This problem was solved by Inverse function Delay-Less (IDL) model in combinational optimization problems. But learning process of IDL model has not been discussed yet. This study is to build a hierarchical network by using IDL model, and to derive a back propagation learning with IDL model. Finally, we discuss the performance of back propagation learning based on an IDL model.