In this paper, we study the influence of the selective pressure on theperformance of cellular genetic algorithms. Cellular genetic algorithms aregenetic algorithms where the population is embedded on a toroidal grid. Thisstructure makes the propagation of the best so far individual slow down, andallows to keep in the population potentially good solutions. We present twoselective pressure reducing strategies in order to slow down even more the bestsolution propagation. We experiment these strategies on a hard optimizationproblem, the quadratic assignment problem, and we show that there is a valuefor of the control parameter for both which gives the best performance. Thisoptimal value does not find explanation on only the selective pressure,measured either by take over time and diversity evolution. This study makes usconclude that we need other tools than the sole selective pressure measures toexplain the performances of cellular genetic algorithms.
展开▼
