A novel population initialization method for accelerating evolutionary algorithms

  • Authors:

  • Shahryar Rahnamayan;Hamid R. Tizhoosh;Magdy M. A. Salama

  • Affiliations:

  • Medical Instrument Analysis and Machine Intelligence Research Group, Faculty of Engineering, University of Waterloo, Waterloo, Ontario, N2L 3G1, Canada;Medical Instrument Analysis and Machine Intelligence Research Group, Faculty of Engineering, University of Waterloo, Waterloo, Ontario, N2L 3G1, Canada;Medical Instrument Analysis and Machine Intelligence Research Group, Faculty of Engineering, University of Waterloo, Waterloo, Ontario, N2L 3G1, Canada

  • Venue:
  • Computers & Mathematics with Applications
  • Year:
  • 2007

Quantified Score

Hi-index 0.09

Visualization

Abstract

Population initialization is a crucial task in evolutionary algorithms because it can affect the convergence speed and also the quality of the final solution. If no information about the solution is available, then random initialization is the most commonly used method to generate candidate solutions (initial population). This paper proposes a novel initialization approach which employs opposition-based learning to generate initial population. The conducted experiments over a comprehensive set of benchmark functions demonstrate that replacing the random initialization with the opposition-based population initialization can accelerate convergence speed.