fine grained tournament selection

[vladofilipovic]

The ratio between exploration and exploitation governs the search process. Level of exploration(looking for new solutions) and exploitation (using previously acquired knowledge) is determined by the tournament size. Very often, the search process converges too slow with smaller tournament size and too fast with bigger tournament size. Fine Grained Tournament Selection (denoted as FGTS) preserves good features of the tournament selection and (at the same time) allows that setting of the ratio between exploration and exploitation becomes more precise. FGTS is controlled by real value parameter fgtournsize (the desired average tournament size) instead of the integer parameter. Similarly to the tournament selection, an individual is chosen if it is the best individual on the tournament. However, unlike tournament selection, size of the tournament is not unique in the whole population, i.e., tournaments with different number of competitors can be held within one step of the selection. (see [Filipovic2003fgts]_) .. [Filipovic2003fgts] Filipovic, 2003, Fine-grained Tournament Selection Operator in Genetic Algorithms. Available from: https://www.cai.sk/ojs/index.php/cai/article/view/452

[fmder]

Thanks for the contribution. Could you please add a couple of unit tests to the new selection?