Intl . Joint Conf . on Neural Networks IJCNN ’ 98 , Anchorage

|Blackjack or twenty-one is a card game where the player attempts to beat the dealer, by obtaining a sum of card values that is equal to or less than 21 so that his total is higher than the dealer's. The probabilistic nature of the game makes it an interesting testbed problem for learning algorithms, though the problem of learning a good playing strategy is not obvious. Learning with a teacher systems are not very useful since the target outputs for a given stage of the game are not known. Instead, the learning system has to explore di erent actions and develop a certain strategy by selectively retaining the actions that maximize the player's performance. This paper explores the use of blackjack as a test bed for learning strategies in neural networks, and speci cally with reinforcement learning techniques. Furthermore, performance comparisons with previous related approaches are also reported. Keywords|Reinforcement learning, SARSA algorithm, Qlearning, Blackjack, learning strategies, arti cial neural net-