A brief survey of differential evolution on Graphic Processing Units

Author

Kromer, Pavel ; Platos, Jan ; Snasel, Vaclav

Author_Institution

Dept. of Comput. Sci., VrB - Tech. Univ. of Ostrava, Ostrava, Czech Republic

fYear

2013

fDate

16-19 April 2013

Firstpage

157

Lastpage

164

Abstract

Differential evolution is an efficient populational meta-heuristic optimization algorithm that has been applied to many difficult real world problems. Due to the relative simplicity of its operations and real encoded data structures, it is suitable for a parallel implementation on multicore systems and on Graphic Processing Units (GPUs) that nowadays reach peak performance of hundreds and thousands of giga FLOPS (floating-point operations per second). This study provides a brief survey of current state-of-the-art research on differential evolution in the GPU environment.

Keywords

data structures; evolutionary computation; graphics processing units; multiprocessing systems; parallel processing; FLOPS; GPU environment; differential evolution; floating-point operation per second; graphic processing units; multicore systems; parallel implementation; populational metaheuristic optimization algorithm; real encoded data structures; real world problems; relative simplicity; Evolutionary computation; Graphics processing units; Instruction sets; Optimization; Sociology; Statistics; Vectors; applications; differential evolution; general purpose GPUs; research; survey;

fLanguage

English

Publisher

ieee

Conference_Titel

Differential Evolution (SDE), 2013 IEEE Symposium on

Conference_Location

Singapore

Type

conf

DOI

10.1109/SDE.2013.6601456

Filename

6601456