[mlpack-git] [mlpack/mlpack] Neural evolution algorithms implementation (CNE, NEAT, HyperNEAT) (#686)

[Excalibur]

> +  void InitPopulation() {
> +    aPopulation = Population(aSeedGenome, aPopulationSize);
> +  }
> +
> +  // Reproduce next generation of population.
> +  void Reproduce() {
> +    // keep best genome.
> +    Genome lastBestGenome = aPopulation.BestGenome();
> +
> +    // Remove stale species.
> +    if (aPopulation.aSpecies.size() > 10) {
> +      RemoveStaleSpecies(aPopulation);
> +    }
> +
> +    // Remove weak genomes in each species.
> +    CullSpecies(aPopulation, aCullSpeciesPercentage);

I referred to this implementation which seems performs good on Mario game:
http://pastebin.com/ZZmSNaHX
And truncating species by 50%, or by only one left are also come from this.
Actually I am now in the progress of reviewing other implementation ideas, such as the neat-python. The NEAT's implementation seems are all quite different for different packages. I am trying to come up with a progress which combines the goodness of different packages.

---
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/mlpack/mlpack/pull/686/files/48e15c45ab1a9c65dbd8fe3fe6f07e368c965883#r70904440
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.cc.gatech.edu/pipermail/mlpack-git/attachments/20160714/3e20f151/attachment-0001.html>