Genetic algorithms used to build curves

Question

Suppose I have an unknown function that I want to approximate using genetic algorithms. For this case, I assume that this is y = 2x.

I would have DNA consisting of 5 elements, one y for each x, from x = 0 to x = 4, in which, after many tests and calculations, I would come closer to something like a form:

best_adn = [0, 2, 4, 6, 8]

Keep in mind that I don't know in advance if it is a linear function, a polynomial, or something even uglier. Also, my goal is not to deduce from best_adn what a function type is, I just want these points, so I can use them later.

This was just an example of a problem. In my case, instead of having only 5 points in the DNA, I have something like 50 or 100. What is the best approach with GA to find the best set of points?

• Population generation 100, drop the worst 20%
• Recover the remaining 80%? How? Cutting them at a random point and then combining the first part of the ADN father with the second part of the ADN mother?
• Mutation, how do I identify problems in this type of mutation?
• Should elitism be used?
• Any other simple idea worth using around?

thanks

Answer 1

. , Mathematica Global Optimization - , , .

:

, . , , . / , . (IE , , ), .

, .

Answer 2

... , GA, GA.

, , . , , , , ?

, .

:

2. 80%? ? , ?

. , - . , 10 . X 1 10 x..10-rand% 10 + 1.. ... .

3.Mutation, ? , , - . , , , ( ), ... .

4. ? .