Optimal Parameter Settings for the $(1+(λ, λ))$ Genetic Algorithm
This work provides a theoretical confirmation for practitioners using this algorithm, but it is incremental as it verifies existing intuitive choices rather than introducing new methods.
The authors determined that the previously proven asymptotic time complexity for the intuitive parameter settings of the (1+(λ,λ)) genetic algorithm is optimal across all possible parameter configurations, showing no improvement is possible.
The $(1+(λ,λ))$ genetic algorithm is one of the few algorithms for which a super-constant speed-up through the use of crossover could be proven. So far, this algorithm has been used with parameters based also on intuitive considerations. In this work, we rigorously regard the whole parameter space and show that the asymptotic time complexity proven by Doerr and Doerr (GECCO 2015) for the intuitive choice is best possible among all settings for population size, mutation probability, and crossover bias.