I'm exploring the NEAT algorithm, and among all of my questions on how to optimize my models, I was wondering how should I evaluate the stagnation of species in order to eventually extinct it.

Currently, I'm using the mean fitness of all members of each species, as it seems like the best choice. But, even though the mean fitness of each species is not evolving, the best individual of the generation is in fact evolving sporadically.

In this case, the training might come to an end if the mean fitness of each generation doesn't evolve for long enough to hit stagnation, making all species extinct.

Other options to evaluate stagnation are the maximum and median value of each species.

Is there a rule of thumb when choosing this function?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.