NeuroEvolution of Augmenting Topologies Expressed in Dart! Create and evolve a neural net for solving all (or at least some) of life's hard problems.


  • Serialize your neural net to save and restore
  • Doesn't require the tracking of innovation numbers, meaning you can run disconnected and still crossover later
  • Includes the mutation of activation functions

Getting started

  1. Create a Neural Network with a certain number of inputs/outputs
  2. Generate a new generation
  3. Iterate through each genome of the generation
  4. Set the inputs and call update, then get the outputs
  5. Set the fitness of each genome and set it's fitness
  6. Go back to Step 2 and continue until the fitness is good enough


Here's an example of teaching a network to count from 0 to 9

var nn = NeuralNet(0, 1);
num currentFitness = 0;

for (int i=0; i<100; i++) {
  for (var g in nn.currentGeneration) {
    num fitness = 0;
    for (int i=0;i<10;i++) {
      var target = i;
      var result = g.getOutputs().toList().single;
      fitness += Activation.gaussian(result - target);
    } = fitness;