Cellular Automata Ants

Bitsakidis, Nikolaos P. ; Dourvas, Nikolaos I. ; Chatzichristofis, Savvas A. ; Sirakoulis, Georgios Ch. (2017)

Book chapter

During the last decades much attention was given to bio-inspired techniques able to successfully handle really complex algorithmic problems. As such Ant Colony Optimization (ACO) algorithms have been introduced as a metaheuristic optimization technique arriving from the swarm intelligence methods family and applied to several computational and combinatorial optimization problems. However, long before ACO, Cellular Automata (CA) have been proposed as a powerful parallel computational tool where space and time are discrete and interactions are local. It has been proven that CA are ubiquitous: they are mathematical models of computation and computer models of natural systems and their research in interdisciplinary topics leads to new theoretical constructs, novel computational solutions and elegant powerful models. As a result, in this chapter we step forward presenting a combination of CA with ant colonies aiming at the introduction of an unconventional computational model, namely “Cellular Automata Ants”. This rather theoretical approach is stressed in rather competitive field, namely clustering. It is well known that the spread of data for almost all areas of life has rapidly increased during the last decades. Nevertheless, the overall process of discovering true knowledge from data demands more powerful clustering techniques to ensure that some of those data are useful and some are not. In this chapter it is presented that Cellular Automata Ants can provide efficient, robust and low cost solutions to data clustering problems using quite small amount of computational resources.

Except where otherwise noted, this item's license is described as http://creativecommons.org/licenses/by-nc-nd/4.0/