As the name suggests, the SAFE method is about the co-evolution of the solution and the fitness function, which guides the solution search optimization. The SAFE method is built around the commensalistic co-evolution strategy of two populations:
- The population of potential solutions, which evolve to solve the problem at hand
- The population of objective function candidates, which evolve to guide the evolution of the solution population
In this book, we have already discussed several search optimization strategies that can be used to guide the evolution of potential solution candidates. These strategies are objective-based fitness optimization and Novelty Search optimization. The former optimization strategy is perfect in situations when we have a plain fitness function landscape and can concentrate our optimization search on the ultimate goal. In this case, we can...
