Genetic programming on GPGPU cards using EASEA

Abstract

Genetic programming is one of the most powerful evolutionary paradigms because it allows us to optimize not only the parameter space but also the structure of a solution. The search space explored by genetic programming is therefore huge and necessitates a very large computing power which is exactly what GPGPUs can provide. This chapter will show how Koza-like tree-based genetic programming can be efficiently ported onto GPGPU processors.