Researchers from Kanazawa University and the University of Toyama in Japan have proposed a scale-free mechanism to guide an artificial bee colony (ABC) algorithm's exploration process.
In the algorithm, employed bees search for food sources and share the information with onlooker bees, who then select a food source to leverage; scout bees randomly search for new food sources, whose positions represent possible solutions to an optimization problem.
To overcome the need for many iterations to reach a solution, the researchers designed the scale-free mechanism and analyzed how scale-free network properties—specifically the power law distribution and low degree-degree correlation coefficient—shape the optimization process.
The mechanism allows each employed bee to learn more effective information from its neighbors. This improves the algorithm's exploitation ability by preventing the information of high-quality employed bees from rapidly overtaking the entire population.
From Kanazawa University
View Full Article
Abstracts Copyright © 2018 Information Inc., Bethesda, Maryland, USA
No entries found