Boosting galactic swarm optimization with ABC
- PDF / 2,727,705 Bytes
- 19 Pages / 595.276 x 790.866 pts Page_size
- 59 Downloads / 249 Views
ORIGINAL ARTICLE
Boosting galactic swarm optimization with ABC Ersin Kaya1 · Sait Ali Uymaz1 · Baris Kocer1 Received: 17 October 2017 / Accepted: 19 September 2018 © Springer-Verlag GmbH Germany, part of Springer Nature 2018
Abstract Galactic swarm optimization (GSO) is a new global metaheuristic optimization algorithm. It manages multiple sub-populations to explore search space efficiently. Then superswarm is recruited from the best-found solutions. Actually, GSO is a framework. In this framework, search method in both sub-population and superswarm can be selected differently. In the original work, particle swarm optimization is used as the search method in both phases. In this work, performance of the state of the art and well known methods are tested under GSO framework. Experiments show that performance of artificial bee colony algorithm under the GSO framework is the best among the other algorithms both under GSO framework and original algorithms. Keywords Galactic swarm optimization · Artificial bee colony algorithm · Swarm intelligence · Metaheuristic optimization algorithm
1 Introduction The aim of the optimization is determining the parameters which maximizes or minimizes an objective function. These parameter may have discrete or continuous values. Real world optimization problems are hard because they may have enormous search space because of lots of parameters and lots of local minima points that make it more difficult to find global optima. Also some problems can’t be implemented by linear functions. Examining all possible solutions for that kind of problems requires too many computing and so almost impossible. Metaheuristic methods are developed to search best possible solution in reasonable time. These methods are developed by inspiring different disciplines like biology, physics and sociology. Most of the today’s successful metaheuristic methods are inspired from swarm behavior of animals, biological systems and natural phenomena. The most important bio-inspired metaheuristic search algorithm genetic algorithms is developed in 1975 and inspired from evolution theory [12]. It is a fundamental algorithm that sometimes used to solve other * Ersin Kaya [email protected] 1
Department of Computer Engineering, Faculty of Engineering and Natural Sciences, Konya Technical University, Konya, Turkey
algorithms problems like inverse problem of the support vector machines which is about splitting a given dataset into two clusters [36] and training extreme learning machines [1]. Researchers attempted to in 1983 collective behavior of the ant swarms give researchers an inspiration to develop ant colony optimization algorithm (ACO) [4]. Another similar algorithm particle swarm optimization [16] is inspired from collective behavior of fish schools and bird flocks in 1995 by Eberhart and Keneddy. Bees are another motivation source for metaheuristic search algorithm. For example artificial bee colony algorithm (ABC) [14] simulates the intelligence food gathering behavior of the bumble bees. Although the
Data Loading...
