An Island Model based on Stigmergy to solve optimization problems
- PDF / 4,974,836 Bytes
- 29 Pages / 595.276 x 790.866 pts Page_size
- 12 Downloads / 213 Views
(0123456789().,-volV)(0123456789().,-volV)
An Island Model based on Stigmergy to solve optimization problems Grasiele Regina Duarte1,2 • Afonso Celso de Castro Lemonge3 Beatriz Souza Leite Pires de Lima2
•
Leonardo Goliatt da Fonseca3
•
Accepted: 4 November 2020 Springer Nature B.V. 2020
Abstract Island Model (IM) is an alternative often used to parallel Evolutionary Algorithms (EA). In IM, the population is distributed between islands that evolve their solutions in parallel, connected by a topology. Periodically, solutions migrate between islands according to a migration policy. The IM can be seen as an ideal structure to combine different algorithms to be used in an organized and cooperative way to solve a problem. Motivated by the number and distinction of EAs proposed in the last decades, in terms of performance and evolutionary behavior, this work proposes a hybrid configuration for IM, called Stigmergy Island Model (Stgm-IM), inspired by the natural phenomenon of stigmergy. Stigmergy is present in groups of some social species, and, by it, their agents organize themselves and maintain a level of cooperation through indirect communication. The Stgm-IM was evaluated regarding its evolutionary behavior and its performance on a benchmark suite of fifteen optimization problems, showing expected results. Keywords Island model Evolutionary algorithms Stigmergy Migration policy
1 Introduction In the last decades, several EAs have been proposed in the literature. When applied to optimization problems, those algorithms perform the search differently according to their set of operations. It is common that EAs present trends in their evolutionary behavior. Regarding their performance, they can depend strongly on the problem under consideration. However, the problem characteristics in which a
& Grasiele Regina Duarte [email protected] Afonso Celso de Castro Lemonge [email protected] Leonardo Goliatt da Fonseca [email protected] Beatriz Souza Leite Pires de Lima [email protected] 1
Graduate Program of Computational Modeling, Federal University of Juiz de Fora, Juiz de Fora, Brazil
2
Civil Engineering Program, COPPE/Federal University of Rio de Janeiro, Rio de Janeiro, Brazil
3
Department of Applied and Computational Mechanics, Federal University of Juiz de Fora, Juiz de Fora, Brazil
specific EA can be more or less efficient, generally are not known. Besides, the implementation of an EA can vary through its parameters. This information illustrates how difficult it can be for the user, to choose an EA to solve a problem. On the other hand, EAs have common characteristics. One of them is that they can be implemented relatively easy to run in parallel computing environments. Often, these resources are used to reduce run times of EAs, motivated by their high computational cost, which is another common characteristic between them. The IM is one of the alternatives to implement EAs to run in parallel environments. Usually, in addition to promoting speedup, the IM improves the solu
Data Loading...
