Reinforcement Learning assignment in the Introduction to Machine Learning class at Reykjavik University

Flappy Bird is a continouse game that can go on forever until you hit a pipe. The environment is never the same, you can never expect the environment to te same as the previous game thus the environment is not predefined. We decided to use Q-Learning algorithm and think that fits best. Q-learning is a model-free reinforcement learning algorithm and the goal is to learn a policy that tells the agent what action to take under specific circumstances. This algorithm does not require a model of the enviroment and that fits our problem with Flappy Bird.