December 2019
Intermediate to advanced
368 pages
11h 10m
English
Having defined the two co-evolving populations, we need to create functions to evaluate the fitness scores of the individuals in each population. As we have already mentioned, the fitness scores of the individuals in the maze-solver population depend on the outputs produced by the population of objective function candidates. At the same time, the fitness score of each objective function candidate is wholly determined by the novelty score of that individual.
Thus, we have two different approaches to the evaluation of fitness scores, and we need to implement two different functions. Hereafter, we discuss both implementations.
Read now
Unlock full access