This project is a JavaScript-based implementation of NEAT (Neuroevolution of Augmenting Topologies), an evolutionary algorithm developed by Kenneth O. Stanley and Risto Miikkulainen. Originally introduced in their 2002 paper, [Evolving Neural Networks Through Augmenting Topologies](https://nn.cs.utexas.edu/downloads/papers/stanley.ec02.pdf), NEAT presents a novel approach to evolving artificial neural networks by optimizing both network weights and structures over generations.

It runs in both NodeJS and browser environments allowing for some cool visual demo's:

Chrome Dino game: https://neat-javascript.org/examples/dino.html Self-driving car: https://neat-javascript.org/examples/self-driving-car.html Flappy Bird: https://neat-javascript.org/examples/flappy-bird.html Pole balancing: https://neat-javascript.org/examples/pole-balancing.html Snake: https://neat-javascript.org/examples/snake.html Ant simulator: https://neat-javascript.org/examples/ant-simulator.html XOR: https://neat-javascript.org/xor.html

Very happy to share this here:) Thanks

GitHub: https://github.com/joshuadam/NEAT-JavaScript Docs: https://neat-javascript.org