On the Robustness of Evolving Populations
Most theoretical work that studies the benefit of recombination focuses on the ability of crossover to speed up optimization time on specific search problems. In this paper, we take a slightly different perspective and investigate recombination in the context of evolving solutions that exhibit mutational robustness, i.e., they display insensitivity to small perturbations. Various models in population genetics have demonstrated that increasing the effective recombination rate promotes the evolution of robustness. We show this result also holds in the context of evolutionary computation by rigorously proving crossover promotes the evolution of robust solutions in the standard (\(\mu \) + 1) GA. Surprisingly, our results show that this effect is still present even when robust solutions are at a selective disadvantage due to lower fitness values.
KeywordsRecombination Rate Fitness Landscape Robust Solution Uniform Crossover Mutational Robustness
The research leading to these results has received funding from the European Union Seventh Framework Programme (FP7/2007-2013) under grant agreement no. 618091 (SAGE) and the German Research Foundation (DFG) under grant agreement no. FR 2988 (TOSU).
- 4.Janson, S.: Tail bounds for sums of geometric and exponential variables. (2014). http://www2.math.uu.se/svante/papers/sjN14.pdf
- 5.Kötzing, T., Sudholt, D., Theile, M.: How crossover helps in pseudo-Boolean optimization. In: GECCO, pp. 989–996 (2011)Google Scholar
- 6.Lehre, P.K., Witt, C.: General drift analysis with tail bounds. arXiv:1307.2559 [cs.NE] (2013)
- 9.Oliveto, P.S., Witt, C.: Erratum: simplified drift analysis for proving lower bounds in evolutionary computation. arXiv:1211.7184 [cs.NE] (2012)
- 10.Sudholt, D.: Crossover is provably essential for the Ising model on trees. In: GECCO, pp. 1161–1167 (2005)Google Scholar
- 11.Sudholt, D.: Crossover speeds up building-block assembly. In: GECCO, pp. 689–696 (2012)Google Scholar
- 12.Witt, C.: Runtime analysis of the (\(\mu \) + 1) EA on simple Pseudo-Boolean functions. Evol. Comput. 14, 65–86 (2006)Google Scholar