tic tac toe - university of wisconsin–madison.docVIP

Tic Tac Toe An Artificial Neural Network approach CS539 Artificial Neural Network and Fuzzy Logic Final Project December 20, 2005 Justin Herbrand Introduction Getting games to react back to the user of a game has always been long hard question for game programmers. Because, lets just face it, a good game that doesn’t challenge the user’s ability to play the game doesn’t keep the user around very long. This idea can be applied to any form of game that is out there. Board games are never fun when the opponent that he or she is playing doesn’t learn or catches on. With today’s computers always advancing, programmers are always looking for new ways to make a video game more interesting and challenging for the user. And one solution that programmers are looking towards are different forms of Artificial Intelligences. The idea behind artificial intelligences is that they are suppose to learn when the user is starting to master the game, they are suppose increase the difficulty of the game to make it harder for the player to advance on in the game. Many people that buy games normally don’t look for games that are going to take a few minutes to beat and then throw it on their shelf. People normally want a game that is going to entertain them for months so when they have free time to spare later on, they will pick up the game and be challenged and entertained by this game. One way people are building artificial intelligences is neural networks. Neural networks are showing up everywhere from entertainment, gambling, to security. The basic idea behind a neural network is shown in the picture above. What you see here is one neuron. It has some form of way to take in input so it can compute what it is see through its inputs and then output the result through its output from what it was computed. But people don’t just build one single neuron to solve complex expression, we network a bunch of neurons together to have it compute a desired output. See picture be