Genetic Programming with Genetic Regulatory Networks

Evolutionary Algorithms (EA) approach differently from nature the genotype - phenotype relationship, and this view is a recurrent issue among researchers. Recently, some researchers have started exploring computationally the new comprehension of the multitude of regulatory mechanisms that are fundamental in both processes of inheritance and of development in natural systems, by trying to include those mechanisms in the EAs. One of the first successful proposals was the Artificial Regulatory Network (ARN) model. Soon after some variants of the ARN, including different improvements over the base model, were tested. In this paper, we combine two of those alternatives, demonstrating experimentally how the resulting model can deal with complex problems, including those that have multiple outputs. The efficacy and efficiency of this variant are tested experimentally using two benchmark problems that show how we can evolve a controller or an artificial artist.