Efficient Computation of Behavior Strategies

By Bernhard von Stengel. 1996

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

Deep Reinforcement Learning from Self-Play in Imperfect-Information Games (2016, neutral): “Neural Fitted Q Iteration (NFQ) (Riedmiller, 2005) and Deep Q Network (DQN) (Mnih et al., 2015) are extensions of FQI that use neural network function approximation with batch and online updates respectively.”
Fictitious Self-Play in Extensive-Form Games (2015, methodology): “Throughout this paper, we use small Greek letters for behavioural strategies of the extensive-form and large Greek letters for pure and mixed strategies of a game’s normalform.”
Abstraction pathologies in extensive games (2009, methodology): “Given a sequence form game and a strategy x, one can compute by(x), y’s best response value to x, with the following linear program: by(x) = min u uT f subject to uT F ≥ −xT A (13) Following the best response linear program, we can derive a linear program for computing a Nash equilibrium strategy fo”
A Competitive Texas Hold'em Poker Player via Automated Abstraction and Real-Time Equilibrium Computation (2006, methodology): “Thus, a minimax solution1 for reasonablysized two-player zero-sum games can be computed using this method (von Stengel 1996; Koller, Megiddo, & von Stengel 1996; Koller & Pfeffer 1997).2 That approach alone scales to games with around a million nodes (Gilpin & Sandholm 2005), but Texas Hold’em with…”
Gradient-Based Algorithms for Finding Nash Equilibria in Extensive Form Games (2007, methodology): “For two-player zero-sum sequential games of imperfect information, the Nash equilibrium problem can be formulated using the sequence form representation [1,2,3] as the following saddle-point problem:”
Smoothing Techniques for Computing Nash Equilibria of Sequential Games (2010, background): “The latter is an immediate consequence of the sequence form formulation of Nash equilibria for sequential games, as detailed in [ 13 , 14].”
An Exact Double-Oracle Algorithm for Zero-Sum Extensive-Form Games with Imperfect Information (2014, methodology): “The leading exact algorithm in the literature uses the compact sequence-form representation and linear programming optimization techniques to solve games of this type (Koller, Megiddo, & von Stengel, 1996; von Stengel, 1996).”