Is there an optimal number of species for NEAT?

Since too low and too high is bad, I am thinking about adjusting the threshold of the distance function at runtime in order to have the number of species always between some bounds. Does this make sense? Is there an optimal range?



You must log in to answer this question.