Genetic Algorithm Used in Training Nanocells

The first step in a GA is to generate an initial population. The NanoCell GA uses a random population of 25 individuals. Tournament selection with a tournament size of 5 is used to choose parents. If two chosen parents have the same fitness, then a new parent is chosen. Both uniform crossover and 5-point crossover are used with comparable success. The 

As mentioned previously, a balance must be maintained between population diversity and preference to the highly fit. In this NanoCell GA preference is given by using tournament selection and by saving the two best 

We further experimented with training NanoCells using another Monte-Carlo search algorithm, simulated annealing. Simulated annealing produced essentially the same results as the GA. It is our opinion that in this case, the particular base search algorithm is not nearly as important as the manner in which it is adapted to the NanoCell problem. 

---