NEAT (NeuroEvolution of Augmenting Topologies) is a method developed by Kenneth O. Stanley for evolving arbitrary neural networks. This project started with initial implementation of neat-python. Society algorithm use similar ideas as proposed by the NEAT algorithm, yet have different behavior torwards how individuals are treated in the evolution. It was forked from the excellent project by @MattKallada and continued by @CodeReclaimers. Major updates on how the population evoloves is the key changes in the algorithm.
For further information regarding general concepts and theory, please see Selected Publications on Stanley's website.
If you want to try neat-python, please check out the repository, start playing with the examples (XOR, single pole balancing, or double pole balancing) and then try creating your own experiment.
The documentation, which is still a work in progress, is available on Read The Docs.