Self-configuring Ensemble of Multimodal Genetic Algorithms

Ensemble methods are widely used to improve decision making in the field of statistics and machine learning. On average, the collective solution of multiple algorithms provides better performance than could be obtained from any of the constituent algorith

  • PDF / 2,174,755 Bytes
  • 19 Pages / 439.37 x 666.142 pts Page_size
  • 64 Downloads / 191 Views

DOWNLOAD

REPORT


Abstract. Ensemble methods are widely used to improve decision making in the field of statistics and machine learning. On average, the collective solution of multiple algorithms provides better performance than could be obtained from any of the constituent algorithms. The ensemble concept can be also used in the field of evolutionary algorithms. The main idea is to include many search algorithms in the ensemble and to design effective control of interaction of algorithms. Such interaction is implemented in different forms of island models, coevolutionary schemes, population-based algorithm portfolios and others. In this paper, a metaheuristic for designing multi-strategy genetic algorithm for multimodal optimization is proposed. Multimodal optimization is the problem of finding many or all global and local optima. In recent years many efficient multimodal techniques have been proposed in the field of population-based nature-inspired search algorithms. The majority of techniques are designed for real-valued problems. At the same time many real-world problems contain variables of many different types, including integer, rank, binary and others. In this case, a binary representation is used. There is a lack of efficient approaches for problems with binary representation. Moreover, binary and binarized problems are usually “black-box” optimization problems, thus there exists the problem of choosing a suitable algorithm and fine tuning it for a certain problem. The proposed approach contains many different multimodal genetic algorithms, which implement different search strategies. The metaheuristic adaptively controls the interactions of many search techniques and leads to the self-configuring solving of problems with a priori unknown structure. We present the results of numerical experiments for classical binary benchmark problems and benchmark problems from the CEC 2013 competition on multimodal optimization. We also present the results for some real-world problems. Keywords: Metaheuristic  Multimodal optimization Niching  Self-configuration



Genetic algorithm



1 Introduction Many real-world problems have more than one optimal solution, or there exists only one global optimum and several local optima in the feasible solution space. Such problems are called multimodal. The goal of multimodal optimization (MMO) is to find all optima (global and local) or a representative subset of all optima. © Springer International Publishing AG 2017 J.J. Merelo et al. (eds.), Computational Intelligence, Studies in Computational Intelligence 669, DOI 10.1007/978-3-319-48506-5_4

Self-configuring Ensemble of Multimodal Genetic Algorithms

57

Evolutionary and genetic algorithms (EAs and GAs) demonstrate good performance for many complex optimization problems. EAs and GAs are also efficient in the multimodal environment as they use a stochastic population-based search instead of the individual search in conventional algorithms. At the same time, traditional EAs and GAs have a tendency to converge to the best-found optimum losing population d