A Competitive Texas Hold'em Poker Player via Automated Abstraction and Real-Time Equilibrium Computation

By Andrew Gilpin, Tuomas Sandholm. 2006

115 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Time and Space: Why Imperfect Information Games are Hard (2018, background): “Abstraction has been popular in game-theoretic poker research, and one method of generating abstract poker games has involved grouping information sets together into a collection called a bin or a bucket [79, 4, 32].”
Evaluating state-space abstractions in extensive-form games (2013, background): “…events in a number of different ways, such as bucketing based on expert-chosen ranges of E[HS] values [2], bucketing based on E[HS] ranges chosen so as to contain an equal number of hands (called percentile bucketing) [21], or by merging hands whose E[HS] values differ by less than a threshold [4].”
A Practical Use of Imperfect Recall (2009, methodology): “To compute effective strategies in games of this size, typically one first employs an abstraction technique (Billings et al. 2003; Gilpin and Sandholm 2006; 2007 ) to create a much smaller game.”
Smoothing Techniques for Computing Nash Equilibria of Sequential Games (2010, methodology): “[2]; Gilpin and Sandholm [4, 5]; Shi and Littman [20]).”
Computer poker: A review (2011, methodology): “GS1 [31], developed by Andrew Gilpin and Thomas Sandholm from Carnegie Mellon University, also attempts to derive near-equilibrium solutions for limit hold’em.”
SOTOPIA: Interactive Evaluation for Social Intelligence in Language Agents (2023, background): “From a game theory perspective, leaking secrets often leads to a loss of utility (Gilpin & Sandholm, 2006).”
Endgame Solving in Large Imperfect-Information Games (2015, methodology): “For this reason, prior approaches for endgame solving have made the (significantly) simplifying assumption that the distributions are independent (Gilpin and Sandholm 2006; 2007).”
Regret Minimization in Games with Incomplete Information (2007, methodology): “Early poker abstractions [2, 4] involved limiting the possible sequences of bets, e.”
Solving Imperfect Information Games Using Decomposition (2013, background): “red no guarantees of correctness in imperfect information games, the temptation of using decomposition to (unsafely) use a larger abstraction has resulted in its application. Both PS-Opti [1] and GS1 [5] were strong computer poker agents for their time, and both chose to split the game, in an unsafe”
Gradient-Based Algorithms for Finding Nash Equilibria in Extensive Form Games (2007, background): “A recent and fruitful approach to finding strategies for sequential game s is to employ lossy abstractions [7, 8, 6, 9, 10] to approximate the Nash equilibrium.”
Abstraction pathologies in extensive games (2009, unknown): “In the first year of the event, teams used linear programming techniques to solve small games, containing only a subset of the betting rounds [1, 4 ]. These small abstract games were then pieced together to create an overall poker strategy.”
Safe and Nested Endgame Solving for Imperfect-Information Games (2017, background): “We ﬁrst review the most intuitive form of endgame solving, which we refer to as unsafe endgame solving (Billings et al. 2003; Gilpin and Sandholm 2006; 2007; Ganzfried and Sandholm 2015).”
Lossy stochastic game abstraction with bounds (2012, methodology): “When state abstraction is done for reason (i), it is nowadays usually done automatically using some abstraction algorithm [Gilpin and Sandholm 2007b; Shi and Littman 2002; Gilpin and Sandholm 2006a, 2007a; Gilpin et al. 2007; Zinke vich et al. 2007; Gilpin et al. 2008; Gilpin and Sandholm 2008].”
The State of Solving Large Incomplete-Information Games, and Application to Poker (2010, methodology): “Today’s best general-purpose LP solvers can solve games with up to 107 or 108 leaves in the game tree (corresponding to nonzero entries in A) (Gilpin and Sandholm 2006).”