Teaching AI to Play Games using Neuroevolution of Augmenting Topologies

The paper is a demonstration of using the NEAT algorithm to allow artificial neural networks to evolve througha genetic algorithm to play the game of Flappy Bird. So it builds a most optimal network by itself by adding nodes or connections, or changing the weights of the connections, or enabling or disabling certain connections as and when required to accomplish the task at hand i.e the game of FlappyBird. We provide the inputs to the algorithm so as to create anoptimal neural network that completes the task, in our case thegame of Flappy bird.