Explore new Frontiers of Innovation

Evolution Strategies

Evolution Strategies (ES) are optimization technique based on ideas of adaptation and evolution. In this sense they are similar to GA, but here the main search procedure is a smart mutation operator.

1P1-ES is the single-objective (1+1)-Evolution Strategy performing robust local optimization with accelerated step-size control. It is a very fast evolutionary algorithm, and it is robust even with slightly noisy functions and in presence of discontinuities.

DES stands for Derandomized Evolution Strategy: it is a robust local search procedure that works with a fast deterministic (derandomized) adaptation of the multivariate mutation distribution. It is a fast single-objective optimizer in continuous search space.

MMES stands for Multi-membered Multiobjective Evolution Strategy: it allows to exploit the full power of global optimization using ES. It supports multiobjective Pareto optimization, and can deal with discrete and continuous variables.