Cumulative step length adaptation for evolution strategies using negative recombination weights

Authors:
Dirk V. Arnold;D. C. Scott Van Wart
Affiliations:
Faculty of Computer Science, Dalhousie University, Halifax, Nova Scotia, Canada;Faculty of Computer Science, Dalhousie University, Halifax, Nova Scotia, Canada
Venue:
Evo'08 Proceedings of the 2008 conference on Applications of evolutionary computing
Year:
2008

Citing 10
Cited 5

Evolution and Optimum Seeking: The Sixth Generation

Evolution and Optimum Seeking: The Sixth Generation
Noisy Local Optimization with Evolution Strategies

Noisy Local Optimization with Evolution Strategies
Evolution strategies –A comprehensive introduction

Natural Computing: an international journal
Step-Size Adaption Based on Non-Local Use of Selection Information

PPSN III Proceedings of the International Conference on Evolutionary Computation. The Third Conference on Parallel Problem Solving from Nature: Parallel Problem Solving from Nature
Reducing the time complexity of the derandomized evolution strategy with covariance matrix adaptation (CMA-ES)

Evolutionary Computation
Completely Derandomized Self-Adaptation in Evolution Strategies

Evolutionary Computation
Hierarchically organised evolution strategies on the parabolic ridge

Proceedings of the 8th annual conference on Genetic and evolutionary computation
Reconsidering the progress rate theory for evolution strategies in finite dimensions

Proceedings of the 8th annual conference on Genetic and evolutionary computation
Weighted multirecombination evolution strategies

Theoretical Computer Science - Foundations of genetic algorithms
Optimal weighted recombination

FOGA'05 Proceedings of the 8th international conference on Foundations of Genetic Algorithms

Stochastic Local Search Techniques with Unimodal Continuous Distributions: A Survey

EvoWorkshops '09 Proceedings of the EvoWorkshops 2009 on Applications of Evolutionary Computing: EvoCOMNET, EvoENVIRONMENT, EvoFIN, EvoGAMES, EvoHOT, EvoIASP, EvoINTERACTION, EvoMUSART, EvoNUM, EvoSTOC, EvoTRANSLOG
Stochastic local search in continuous domains: questions to be answered when designing a novel algorithm

Proceedings of the 12th annual conference companion on Genetic and evolutionary computation
Mirrored sampling and sequential selection for evolution strategies

PPSN'10 Proceedings of the 11th international conference on Parallel problem solving from nature: Part I
Log(λ) modifications for optimal parallelism

PPSN'10 Proceedings of the 11th international conference on Parallel problem solving from nature: Part I
Bandit-based estimation of distribution algorithms for noisy optimization: rigorous runtime analysis

LION'10 Proceedings of the 4th international conference on Learning and intelligent optimization

Quantified Score

Hi-index	0.00

Visualization

Abstract

Cumulative step length adaptation is a mutation strength control mechanism commonly employed with evolution strategies. When using weighted recombination with negative weights it can be observed to be prone to failure, often leading to divergent behaviour in low-dimensional search spaces. This paper traces the reasons for this breakdown of step length control. It then proposes a novel variant of the algorithm that reliably results in convergent behaviour for the test functions considered. The influence of the dimensionality as well as of the degree of ill-conditioning on optimisation performance are evaluated in computer experiments. Implications for the use of weighted recombination with negative weights are discussed.