Summary
International Symposium on Nonlinear Theory and Its Applications
2015
Session Number:B3L-D
Session:
Number:B3L-D-4
GPU Acceleration of Method for Solving Optimization Problems Using Chaotic Neural Networks
Toshihiro Tachibana, Masaharu Adachi,
pp.598-601
Publication Date:2015/12/1
Online ISSN:2188-5079
DOI:10.34385/proc.47.B3L-D-4
PDF download (324.3KB)
Summary:
Recently, graphics processing units (GPUs) are used for general purpose computing which is called GPGPU. A feature of GPGPU is executing parallel computation by many GPU cores. In this paper, GPGPU calculation is used for several methods of optimization problems. We show a reduction of computation time by applying GPGPU in the execution of a method for quadratic assignment problems using chaotic neurodynamics and of a method for multi-objective problems using particle swarm optimization and chaotic neurodynamics.