 |
    |
Stopping Criteria for Multimodal Optimization
Simon Wessing1, Mike Preuss2, and Heike Trautmann2
1Department of Computer Science, TU Dortmund, Germany
simon.wessing@tu-dortmund.de
2Information Systems and Statistics Group, University of Münster, Germany
mike.preuss@uni-muenster.de
trautmann@uni-muenster.de
Abstract. Multimodal optimization requires maintenance of a good search space coverage and approximation of several optima at the same time. We analyze two constitutive optimization algorithms and show that in many cases, a phase transition occurs at some point, so that either diversity collapses or optimization stagnates. But how to derive suitable stopping criteria for multimodal optimization? Experimental results indicate that an algorithm’s population contains sufficient information to estimate the point in time when several performance indicators reach their optimum. Thus, stopping criteria are formulated based on summary characteristics employing objective values and mutation strength.
Keywords: Multimodal optimization, global optimization, multiobjective selection, convergence detection, stopping criteria
LNCS 8672, p. 141 ff.
lncs@springer.com
© Springer International Publishing Switzerland 2014