Cultured Differential Evolution for Constrained Optimization


Ricardo Landa Becerra developed a cultured version of differential evolution for solving constrained (single-objective) optimization problems. This approach incorporates knowledge of the problem during the search (instead of requiring it a priori) in order to perform a more efficient search. Such knowledge incorporation allows a significant reduction in the number of fitness function evaluations required by the algorithm.


The source code of this approach is available here