English

Analysing Equilibrium States for Population Diversity

Neural and Evolutionary Computing 2023-04-20 v1

Abstract

Population diversity is crucial in evolutionary algorithms as it helps with global exploration and facilitates the use of crossover. Despite many runtime analyses showing advantages of population diversity, we have no clear picture of how diversity evolves over time. We study how population diversity of (μ+1)(\mu+1) algorithms, measured by the sum of pairwise Hamming distances, evolves in a fitness-neutral environment. We give an exact formula for the drift of population diversity and show that it is driven towards an equilibrium state. Moreover, we bound the expected time for getting close to the equilibrium state. We find that these dynamics, including the location of the equilibrium, are unaffected by surprisingly many algorithmic choices. All unbiased mutation operators with the same expected number of bit flips have the same effect on the expected diversity. Many crossover operators have no effect at all, including all binary unbiased, respectful operators. We review crossover operators from the literature and identify crossovers that are neutral towards the evolution of diversity and crossovers that are not.

Keywords

Cite

@article{arxiv.2304.09690,
  title  = {Analysing Equilibrium States for Population Diversity},
  author = {Johannes Lengler and Andre Opris and Dirk Sudholt},
  journal= {arXiv preprint arXiv:2304.09690},
  year   = {2023}
}

Comments

To appear at GECCO 2023

R2 v1 2026-06-28T10:11:05.513Z