The Implementation of Machine Learning in the Game of Checkers

Most games have a set algorithm that does not change. This means that these programs cannot adapt to a situation or learn from mis- takes that it makes. However if a machine could learn, then it could adapt to new situations and would have a nearly boundless skill level. Machine Learning programs can be beaten once, but against an oppo- nent that does not change, it eventually will be able to beat it. The project that I am writing will learn how to play the game of checkers as it plays, by modifying itself after ever game played. It will review its play and if it played well it will play that way more often, if it did not it will avoid that way of playing.