Cooperation of optimization algorithms: A simple hierarchical model
Radka Poláková, Josef Tvrdík, Petr Bujok · 2015
A simple model for the cooperation of optimization evolutionary algorithms was proposed and tested on CEC 2015 benchmark suite. The four adaptive algorithms were chosen for this model, namely covariance matrix adaptation evolutionary strategy (CMA-ES) and three variants of adaptive differential evolution. Three algorithms with constant population size work in pseudo-parallel way and after stopping the whole populations migrate to the top algorithm with dynamic population reduction for final processing. The simple cooperative algorithm outperformed CMA-ES in 24 out of 60 test problems, which is promising for the development of more sophisticated cooperative algorithms for the global optimization.