IAT Project : Implementing reinforcement learning Space Invaders agent.

Implementation of a Q-Learning function for an agent playing space Invaders. The objective of our project is to implement a reinforcement learning algorithm that will optimally play the Space Invaders game. With the different possible actions (moving the ship and firing the missile) and through the different states, our algorithm and its reward system will optimize the inputs and propose the best way to play after a learning phase.