Network-Structured Firefly Algorithm and its Behavior
Haruna Matsushita, Daiki Matsumoto, Yoshifumi Nishio · IEICE Proceeding Series · 2014
Abstract—This paper proposes a network-structured firefly algorithm (NS-FA). The standard firefly algorithm (FA) consists of multiple fireflies. An attractiveness of the firefly is proportional to its brightness, and for any two fire-flies, the less brighter one will be attracted by the brighter one at every generation step. However, the fireflies of the NS-FA have a network structure that changes with gener-ation step, and they move depending on its network struc-ture. That is to say, the firefly of the NS-FA is not affected by brighter firefly if there is no connection between the two fireflies. In other words, even if there is brighter firefly in the firefly swarm, the firefly is not always attracted to the brighter firefly. We apply the NS-FA to various optimiza-tion benchmarks and confirm its effectiveness. 1.