Approach of Acceleration of Genetic Algorithm on CUDA platform
S. V. Skvortsov, Tatiana Fetisova, Dmitry Alexandrovich Perepelkin, Dmitry V. Fetisov · 2019
In this paper, the genetic algorithm is considered as a whole, as well as options for the implementation of its individual stages. The multi-threaded model is developed that contains all the necessary iterations and data parallelization elements. The possibility of multi-threaded implementation of the genetic algorithm on graphical processors of NVIDIA with the use of CUDA technology is described. Analysis of the relevance and applicability of algorithms in solving various applications are considered.