Temporal Difference Learning for Nondeterministic Board Games.
Glenn F. Matthews, Khaled Rasheed · International Conference on Artificial Intelligence · 2008
We use temporal difference (TD) learning to train neural networks for four nondeterministic board games: backgammon, hypergammon, pachisi, and Parcheesi. We investigate the influence of two variables on the development of these networks: first, the source of training data, either learner-vs.self or learner-vs.-other game play; second, the choice of attributes used: a simple encoding of the board layout, a set of derived features, or a combination of these. Experimental results show that the TD learning approach is viable for all four games, that learner-vs.-self play can provide highly effective training data, and that the combination of raw and smart features allows for the development of stronger players.