Using Co-solvability to Model and Exploit Synergetic Effects in Evolution

We define co-solvability as an ability of a solution to solve a pair of fitness cases (tests). The fitness function based on this concept grants rewards to solutions depending on their performance on all pairs of tests, in a way analogous to implicit fitness sharing. We prove that cosolvability fitness function is synergetic by definition and imposes selection pressure which is qualitatively different from that of standard fitness function or implicit fitness sharing. The results of experimental verification suggest that evolutionary runs driven by co-solvability fitness function usually produce well-performing solutions faster and are more likely to reach global optima. The idea in a nutshell Solutions are rewarded for correct processing of particular pairs of fitness cases. The reward for each pair is independently estimated based on population performance.