Evolution strategies for continuous optimization : A survey of the state-of-the-art

Research output: Journal Publications and Reviews (RGC: 21, 22, 62)21_Publication in refereed journal

1 Scopus Citations
View graph of relations

Author(s)

Related Research Unit(s)

Detail(s)

Original languageEnglish
Article number100694
Journal / PublicationSwarm and Evolutionary Computation
Volume56
Online published19 Apr 2020
Publication statusPublished - Aug 2020

Abstract

Evolution strategies are a class of evolutionary algorithms for black-box optimization and achieve state-of-the-art performance on many benchmarks and real-world applications. Evolution strategies typically evolve a Gaussian distribution to approach the optimum. In this paper, we present a survey of recent advances in evolution strategies. We summarize the techniques, extensions, and practical considerations of evolution strategies for various optimization problems. We discuss some important open questions and promising topics that desire further research. Many of the discussed techniques and principles are applicable to other algorithms.

Research Area(s)

  • Black-box optimization, Evolution strategies, Covariance matrix adaptation, Evolution path, Natural gradient