[mlpack] Neuroevolution Algorithms Implementation - Week 1

[bang liu]

Dear all,

During the last week, I finished the implementation of gene, genome (the
critical part) for CNE algorithm. Now I am going to do a PR for the two
classes after making sure there is no error.

Before submitting my code to mlpack repository, I update it frequently on
my local forked repository. If you are interested, you can check it at
https://github.com/BangLiu/mlpack/tree/ne/src/mlpack/methods/ne

This week I will focusing on come up with a good abstraction of CNE
algorithm and test it with XOR test.

Best,
Bang

2016-05-30 17:23 GMT-06:00 bang liu <lbldtb at gmail.com>:

> Dear all,
>
> I am working on the project "Neuroevolution Algorithms Implementation".
>
> This week, I am working on the implementation of Conventional
> Neuro-evolution (CNE): weight evolution on topologically fixed neural
> networks. As the first step, I am implementing classes corresponding to the
> concepts in NE algorithms, including: gene, genome, population and CNE.
> Currently, the implementation of gene is finished and tested, and the other
> classes' implementation are in progress.
>
> The main reference papers for the implementation of CNE includes:
> [1] "Training Feedforward Neural Networks Using Genetic Algorithms
> <http://www.ijcai.org/Proceedings/89-1/Papers/122.pdf>"
> [2] "Evolving Artificial Neural Networks
> <http://www.cs.bham.ac.uk/~axk/evoNN.pdf>"
>
> I will continue implementing the CNE algorithm in this week and hopefully
> submit my first PR to mlpack. After finished the implementation of a
> specific algorithm, I will summarize the implementation by mlpack blog.
>
> Best,
> Bang
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.cc.gatech.edu/pipermail/mlpack/attachments/20160606/79398059/attachment-0002.html>