Playing Games with Computational Intelligence

Games provide an ideal environment in which to study computational intelligence, offering a range of challenging and engaging problems. This talk will begin with an overview of the field including some sample applications and an introduction to the main learning algorithms (evolution and temporal difference learning) in the context of games. Despite each of these having a long history, there is still little agreement on which works best when, and why. I'll offer some insights into by showing the dependence of the algorithms on the choice of function approximator, and also show how information theory can help to guide the choice of algorithm.