查看文章

The increasing interest of researchers in using low cost GPUs for applications requiring intensive parallel computing is due to the ability of these devices to solve parallelizable problems much faster than traditional sequential processors. The first applications of evolutionary algorithms (EAs) on GPUs have been developed to solve specific image processing problems; later, general purpose genetic algorithms (GA) have been implemented. However, those implementations used texture rendering for the encoding and evaluation of individuals, while, most of the times, still executing on the CPU tasks like pseudo-random number generation or selection. This project presents an implementation of a GA within the nVIDIA CUDA TM environment, which avoids the use of textures as data structures and performs all evolution on the GPU, reducing as much as possible the exchange of data with the CPU.