1
$\begingroup$

Suppose an AI is to play the game flappy bird. And the fitness function is how long the bird has traveled before the game ends.

Would we have multiple neural networks initialized at the beginning with random weights (as in each bird has its own network) and then we determine the neural networks that have lasted the longest for the game and then we perform a selection of weights from the "better" neural networks followed by mutation? Those will then be used as the new weights of a brand new neural network (ie the offspring from two "better" neural networks?)?

If that is the case, does that mean there is no backpropagation because there isn't a cost function?

$\endgroup$
1
$\begingroup$

I found my answer in a different post: How to evolve weights of a neural network in Neuroevolution?. Note that the genetic algorithm is a subcategory of the neuroevolution algorithm. Short answer, my original thoughts were correct.

| improve this answer | |
$\endgroup$
  • $\begingroup$ There are other combinations possible with GAs and NNs, but I think that is the most usual way to see them combined with the GA searching for ideal weights and hyperparameters of the NN. $\endgroup$ – Neil Slater Nov 29 '19 at 16:12
  • $\begingroup$ I don't think that genetic algorithms are a subcategory of neuroevolution. $\endgroup$ – nbro Dec 1 '19 at 2:37

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.