eCite Digital Repository

Evolution strategies with thresheld convergence


Piad-Morffis, A and Estevez-Velarde, S and Bolufe-Rohler, A and Montgomery, J and Chen, S, Evolution strategies with thresheld convergence, Proceedings of 2015 IEEE Congress on Evolutionary Computation, 25-28 May, Sendai, Japan, pp. 2097-2104. ISBN 978-1-4799-7491-7 (2015) [Refereed Conference Paper]

Copyright Statement

Copyright 2015 IEEE

DOI: doi:10.1109/CEC.2015.7257143


When optimizing multi-modal spaces, effective search techniques must carefully balance two conflicting tasks: exploration and exploitation. The first refers to the process of identifying promising areas in the search space. The second refers to the process of actually finding the local optima in these areas. This balance becomes increasingly important in stochastic search, where the only knowledge about a functionís landscape relies on the relative comparison of random samples. Thresheld convergence is a technique designed to effectively separate the processes of exploration and exploitation. This paper addresses the design of thresheld convergence in the context of evolution strategies. We analyze the behavior of the standard (μ, λ)-ES on multi-modal landscapes and argue that part of itís shortcomings are due to an ineffective balance between exploration and exploitation. Afterwards we present a design for thresheld convergence tailored to ES, as a simple yet effective mechanism to increase the performance of (μ, λ)-ES on multimodal functions.

Item Details

Item Type:Refereed Conference Paper
Keywords:non-convex optimisation, evolution strategies, search heuristics
Research Division:Information and Computing Sciences
Research Group:Machine learning
Research Field:Neural networks
Objective Division:Expanding Knowledge
Objective Group:Expanding knowledge
Objective Field:Expanding knowledge in the information and computing sciences
UTAS Author:Montgomery, J (Dr James Montgomery)
ID Code:99241
Year Published:2015
Deposited By:Information and Communication Technology
Deposited On:2015-03-18
Last Modified:2017-11-20

Repository Staff Only: item control page