Summary
International Symposium on Nonlinear Theory and its Applications
2010
Session Number:C1L-C
Session:
Number:C1L-C1
Basic Learning Characteristics of Digital Spike Maps
Takashi Ogawa, Toshimichi Saito,
pp.484-487
Publication Date:2010/9/5
Online ISSN:2188-5079
DOI:10.34385/proc.44.C1L-C1
PDF download (307.8KB)
Summary:
This paper studies learning algorithm of the digital spike maps. The map is equivalent to a simple one-dimensional cellular automaton and can generate various digital spike-trains. In order to approximate a class of spike-trains, we present a learning algorithm with selforganizing function. Performing a basic numerical experiment, we have clarified that the map can learn a typical class of teacher signals. The results contribute to bridge between spiking neural systems and digital dynamical systems.