AI and Game Theory
As the years have progressed, so have the studies of game theory and progression of artificial intelligence. They are seemingly unrelated topics yet there can be some very useful overlap which this article discusses. From a surface level view, many people may think that game theory can just be utilized for simple models like we have discussed in lecture. If extended to artificial intelligence, this assumption would extend to thinking that game theory would only be used for simple extensions of the concepts such as games like chess. However, the success of the two seems to be much more directly tied for many reasons. There is a ton of overlap with concepts of game theory that can help further artificial intelligence, like participant and mechanism design, as well as understanding many different types of games.
As we have discussed game theory has many applications and this is just another impressive utilization. More specifically artificial intelligence can use concepts regarding players and complex games discussed during class to extend to teaching AIs. Some of these game models are single move games in which the AI doesn’t know the move of the other player. For this case the more we understand about the processes and logic behind these decisions the easier it is to translate this to AI development.