This is a neuroevolution simulation of an arbitrary game that I made up. Entities (blips) can choose to either move in the positive or negative x direction to avoid obstacles (bars) moving in the negative y direction. The neuroevolution algorithm begins with a random population of blips with randomly generated neural networks (brains). At the end of each generation the best blips are chosen to pass along characteristics of their neural nets to the next generation. Eventually after a few generations we are left with a population of blips that have mastered the game.
marcSoda/NeuroEvolution-NEAT
Folders and files
| Name | Name | Last commit date | ||
|---|---|---|---|---|