An eco-inspired evolutionary algorithm applied to numerical optimization

The search for nature-inspired ideas, models and computational paradigms always was of great interest for computer scientists, particularly for those from the Natural Computing area. The concept of optimization is present in several natural processes as in the evolution of species, in the behavior of social groups, in the dynamics of the immune system, in the food search strategies and ecological relationships of different animal populations. This work uses the ecological concepts of habitats, ecological relationships and ecological successions to build an ecology-inspired optimization algorithm, named ECO. The proposed approach uses several populations of candidate solutions that cooperates and coevolves with each other, according to a given meta-heuristic. In this particular work, we used the Artificial Bee Colony (ABC) algorithm as the main meta-heuristic. Experiments were done for optimizing benchmarck mathematical functions. Results were compared with the ABC algorithm running without the ecology concepts previously mentioned. The ECO algorithm performed significantly better than the ABC, especially as the dimensionality of the functions increase, possibly thanks to the ecological interactions (intra and inter-habitats) that enabled the coevolution of populations. Results suggest that the eco-inspired algorithm can be an interesting alternative for numerical optimization.