We assess a player's long-term expected winnings or losses at the game of blackjack without knowing the strategy employed. We do this by comparing the player's moves to those of a baseline strategy with known expected winnings. This allows an accurate estimate of the player's expectation to be found hundreds of times faster than the naive approach of using the average winnings observed as an estimate. © Springer-Verlag Berlin Heidelberg 2003.
CITATION STYLE
Wolfe, D. (2003). Distinguishing gamblers from investors at the blackjack table. Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2883, 1–10. https://doi.org/10.1007/978-3-540-40031-8_1
Mendeley helps you to discover research relevant for your work.