AIM-GP and parallelism

Many machine learning tasks are just too hard to be solved with a single processor machine, no matter how efficient the algorithms are and how fast our hardware is. Luckily genetic programming is well suited for parallelization compared to standard serial algorithms. The paper describes the first parallel implementation of an AIM-GP system, creating the potential for an extremely fast system. The system is tested on three problems and several variants of demes and migration are evaluated. Most of the results are applicable to both linear and tree based systems.