That all depends on what you are doing, and how you want to train it. If you want to train one large neural network with backprop or RPROP, there is really no way to do it. If you want to train many networks at a time, that should not be an issue. Genetic training can also be done so long as you implement some way to merge the various populations you have on multiple computers.
That all depends on what you
That all depends on what you are doing, and how you want to train it. If you want to train one large neural network with backprop or RPROP, there is really no way to do it. If you want to train many networks at a time, that should not be an issue. Genetic training can also be done so long as you implement some way to merge the various populations you have on multiple computers.