Home // INTELLI 2013, The Second International Conference on Intelligent Systems and Applications // View article

A new Heuristic of Fish School Segregation for Multi-Solution Optimization of Multimodal Problems

Authors:
Marcelo G. P. Lacerda
Fernando B. de L. Neto

Keywords: Heuristic Search; Multi-Solution Optimization; Multimodal Problems; Fish School Search.

Abstract:
This work presents a new heuristic for fish school segregation applied to the Fish School Search algorithm (FSS), aiming to serve as a basis for the creation of a new multi-solution optimization method for multimodal problems. In this new approach, the weight of the fish is used as a population segregation element, allowing the heaviest fishes, i.e the most successful ones, to swim,i.e. to move in the search space, more independently and the lightest ones to be guided by the heaviest ones. The obtained results showed that this new approach is able to find a good amount of solutions in the search space, overcoming the three techniques used for comparison in 6 of 7 benchmark functions. Moreover, it can be seen that the new approach requires less computational effort to obtain excellent results. Another advantage of the new approach is that there is no need for an addition of operators in the original FSS. Even though this new version of multimodal FSS does not have an ideal coverage, which causes the return of many “extra” solutions, the sole use of the weight of the fishes, i.e. a readily available information, as a population segregation operator is an economical and good alternative to be considered upon multi-solution problems. This specially taking into account the expediency of the method and that the detected candidate solutions are mostly false-positives, which can be more easily pruned than the addition of false-negatives.

Pages: 115 to 121

Copyright: Copyright (c) IARIA, 2013

Publication date: April 21, 2013

Published in: conference

ISSN: 2308-4065

ISBN: 978-1-61208-269-1

Location: Venice, Italy

Dates: from April 21, 2013 to April 26, 2013