Thirty years, 1993–2023, is a huge time frame in science. We address some major developmentsin the field of evolutionary algorithms, with applications in parameteroptimization, over these 30 years. These include the covariance matrix adaptationevolution strategy and some fast-growing fields such as multimodal optimization,surrogate-assisted optimization, multiobjective optimization, and automated algorithmdesign. Moreover, we also discuss particle swarm optimization and differentialevolution, which did not exist 30 years ago, either. One of the key arguments madein the paper is that we need fewer algorithms, not more, which, however, is the currenttrend through continuously claiming paradigms from nature that are suggestedto be useful as new optimization algorithms. Moreover, we argue that we need properbenchmarking procedures to sort out whether a newly proposed algorithm is usefulor not.We also briefly discuss automated algorithm design approaches, including configurablealgorithm design frameworks, as the proposed next step toward designingoptimization algorithms automatically, rather than by hand.
展开▼