I am trying to implement a roulette wheel selection. I have understood this algorithm:
- Calculate the sum S of all chromosome fitnesses in population
- Generate a random number, r, from interval (0,S)
- Loop through the population and sum fitnesses from 0 till S, this is the partial sum, call it P.
- When the P > S: stop and return the corresponding chromosome.
What I don't understand is how this corresponds to doing this instead: Roulette wheel selection algorithm (the answer with 44 votes). This makes sense to me, but not the one above.
The following is done using the sum
the following is done by normalizing between 0 and 1