GPU-based Parallel Differential Evolution with Local PatternSearch on Function Optimization

摘要

This paper presents a parallel Differential Evolution (DE) algorithm with local search for function optimizationproblems, with graphics hardware acceleration. Differential Evolution is population-based meta-heuristic, originallydesigned for continuous function optimization. Graphics Processing Units (GPU) is an emerging desktop parallelcomputing technology that is getting popular with its widespread adoption. In this paper, the classical DE isimplemented in a GPU platform with CUDA? technology and a local Pattern Search is added to enhance its searchability. The test results show great savings in computation times and demonstrate a promising direction for highspeed optimization on a desktop computing setting.