Valet: A Standardized Testbed of Traditional Imperfect-Information Card Games
Valet is a comprehensive and diverse testbed of 21 traditional imperfect-information card games. These games span multiple genres, cultures, player counts, deck structures, mechanics, winning conditions, and methods of hiding and revealing information.
AI algorithms for imperfect-information games are typically compared using performance metrics on individual games, making it difficult to assess robustness across game choices. Card games are a natural domain for imperfect information due to hidden hands and stochastic draws. Valet facilitates comparative research on imperfect-information game-playing algorithms and general game systems.
The rules for each game in Valet are coded in RECYCLE, a card game description language, to encourage standardized implementations across different systems. Each game has a page linked below with further details, including a summary of the rules and a diagram of the information flow in the game.
Valet was curated by Mark Goadrich, Achielle Morenville, and Eric Piette, funded by a COST Action: CA22145 STSM Grant.
