Explore influence of differential operator in DE mutation with unrestrained method to generate mutant vector

  • Authors:

  • Hao Liu;Han Huang;Shusen Liu

  • Affiliations:

  • School of Software Engineering, South China University of Technology, Guangzhou, P.R. China;School of Software Engineering, South China University of Technology, Guangzhou, P.R. China and Department of Management Sciences, College of Business City, University of Hong Kong, Hong Kong;School of Software Engineering, South China University of Technology, Guangzhou, P.R. China

  • Venue:
  • SIDE'12 Proceedings of the 2012 international conference on Swarm and Evolutionary Computation
  • Year:
  • 2012

Quantified Score

Hi-index 0.00

Visualization

Abstract

Differential Evolution (DE) is an efficient optimizer in current use. Although many new DE mutant vectors have been proposed by alter the differential operator, there are few works studying the differential operator's effect in DE algorithm. This paper proposes a correlation between the DE performance and the mutant vector. That is, for a particular mutant vector, increase the number of differential operator would influence the performance of the algorithm linearly. These mutant vectors are evaluated by 23 benchmarks selected from Congress on Evolutionary Computation (CEC) competition. Additionally, this paper proposes an unrestrained method to generate mutant vector. Unlike the old method selects mutually exclusive individuals, the new method allows same individuals appear repeatedly to generate mutant vector. This new method could enhance the potential diversity of the population and improve the performance of DE in general. abstract environment.