Summary
International Symposium on Nonlinear Theory and its Applications
2008
Session Number:B4L-E
Session:
Number:B4L-E4
A Heuristic Approach to Graph Coloring Problems Using a Complex-Valued Neural Network
Gouhei Tanaka, Kazuyuki Aihara,
pp.-
Publication Date:2008/9/7
Online ISSN:2188-5079
DOI:10.34385/proc.42.B4L-E4
PDF download (244.5KB)
Summary:
Complex-valued neural networks are widely used to process not only complex-valued information but also multivalued one. For instance, multivalued representation with a complex-valued neuron is more suitable for dealing with multicolor images than binary representation with a real-valued neuron, because less neurons are needed. In seeking for another beneficial application of a complex-valued neuron, we demonstrate that a complex-valued neural network can provide a heuristic approach to graph coloring problems.