Evolving Specific Network Statistical Properties using a Gene Regulatory Network Model
Miguel Nicolau and Marc Schoenauer
Genetic and Evolutionary Computation Conference (Genetic and Evolutionary Computation Conference 2009
The generation of network topologies with specific, user-specified statistical properties is addressed using an Evolutionary Algorithm that is seeded by an Artificial Gene Regulatory Network Model. The work presented here extends
previous work where the proposed approach was demonstrated to be able to evolve scale-free topologies. The present results reinforce the applicability of the proposed method, showing that the evolution of small-world topologies is also possible, but requires a carefully crafted fitness function.