flappy-bird-policy-network

This is an attempt to train a flappy-bird style game agent using reinforment learning in the browser.

All code related to the game is in game.js.

All code related to policy network is in index.js.

Built using:

-- Flappy Bird Clone from Dan Shiffman / The Coding Train: https://github.com/CodingTrain/Flappy-Bird-Clone

-- Reinforcement Learning / Policy Gradient Training: https://github.com/tensorflow/tfjs-examples/tree/master/cart-pole

-- Tensorflow.js Machine Learning framework: https://www.tensorflow.org/js/