Duplicate poker is a mechanism that allows for lower variance evaluation of poker agents. Using this technique has allowed the competition to better evaluate the different agents that enter using limited computational resources. Some form of this mechanism has been used since the first competition in 2006.
A duplicate match consists of an ordinary heads-up match of k hands between two players. The cards played are recorded (or the seed used to shuffle is saved) and then the players' memories are reset, they switch positions and play a second k hands.
Introduced in the 2009 competition, multiplayer duplicate generalizes the heads-up duplicate format for the 3 player matches. If we consider that there are 3 possible seats that each bot can sit in, and 2 different relative orderings of the other 2 bots given the position of one bot, then there are six total possible configurations of 3 players at a given table. If we choose to play N hands per match then the following system will assure all players rotate through all possible seats and relative orderings:
In order to accurately evaluate how well an agent would fare against a given opponent (or set of opponents) in a match of a specified length, we would like to have repeated duplicate matches. To do this, the agents must remember nothing between each duplicate match. Therefore, in a series of duplicate matches, each agent's memory is reset after every duplicate match.