Qualitative and quantitative assessment of step size adaptation rules

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

We present a comparison of step size adaptation methods for evolution strategies, covering recent developments in the field. Following recent work by Hansen et al. we formulate a concise list of performance criteria: a) fast convergence of the mean, b) near-optimal fixed point of the normalized step size dynamics, and c) invariance to adding constant dimensions of the objective function. Our results show that algorithms violating these principles tend to underestimate the step size or are unreliable when the function does not fit to the algorithm's tuned hyperparameters. In contrast, we find that cumulative step size adaptation (CSA) and twopoint adaptation (TPA) provide reliable estimates of the optimal step size. We further find that removing the evolution path of CSA still leads to a reliable algorithm without the computational requirements of CSA.

Original languageEnglish
Title of host publicationProceedings of the 14th ACM/SIGEVO Conference on Foundations of Genetic Algorithms
Number of pages10
PublisherAssociation for Computing Machinery
Publication date2017
Pages139-148
ISBN (Electronic)978-1-4503-4651-1
DOIs
Publication statusPublished - 2017
Event14th ACM/SIGEVO Conference on Foundations of Genetic Algorithms - København, Denmark
Duration: 12 Jan 201715 Jan 2017
Conference number: 14

Conference

Conference14th ACM/SIGEVO Conference on Foundations of Genetic Algorithms
Nummer14
LandDenmark
ByKøbenhavn
Periode12/01/201715/01/2017

    Research areas

  • Comparison, Evolution strategies, Step size adaptation

ID: 179557726