ApproxED: Approximate exploitability descent via learned best responses

• URL: http://arxiv.org/abs/2301.08830v3
• Date: Wed, 12 Jun 2024 22:39:58 GMT
• Title: ApproxED: Approximate exploitability descent via learned best responses
• Authors: Carlos Martin, Tuomas Sandholm,
• Abstract summary: We study the problem of finding an approximate Nash equilibrium of games with continuous action sets. We propose two new methods that minimize an approximation of exploitability with respect to the strategy profile.
• Score: 61.17702187957206
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: There has been substantial progress on finding game-theoretic equilibria. Most of that work has focused on games with finite, discrete action spaces. However, many games involving space, time, money, and other fine-grained quantities have continuous action spaces (or are best modeled as having such). We study the problem of finding an approximate Nash equilibrium of games with continuous action sets. The standard measure of closeness to Nash equilibrium is exploitability, which measures how much players can benefit from unilaterally changing their strategy. We propose two new methods that minimize an approximation of exploitability with respect to the strategy profile. The first method uses a learned best-response function, which takes the current strategy profile as input and outputs candidate best responses for each player. The strategy profile and best-response functions are trained simultaneously, with the former trying to minimize exploitability while the latter tries to maximize it. The second method maintains an ensemble of candidate best responses for each player. In each iteration, the best-performing elements of each ensemble are used to update the current strategy profile. The strategy profile and ensembles are simultaneously trained to minimize and maximize the approximate exploitability, respectively. We evaluate our methods on various continuous games and GAN training, showing that they outperform prior methods.

Related papers

• Leading the Pack: N-player Opponent Shaping [52.682734939786464]
  We extend Opponent Shaping (OS) methods to environments involving multiple co-players and multiple shaping agents. We find that when playing with a large number of co-players, OS methods' relative performance reduces, suggesting that in the limit OS methods may not perform well.
  arXiv  Detail & Related papers  (2023-12-19T20:01:42Z)
• Finding mixed-strategy equilibria of continuous-action games without gradients using randomized policy networks [83.28949556413717]
  We study the problem of computing an approximate Nash equilibrium of continuous-action game without access to gradients. We model players' strategies using artificial neural networks. This paper is the first to solve general continuous-action games with unrestricted mixed strategies and without any gradient information.
  arXiv  Detail & Related papers  (2022-11-29T05:16:41Z)
• Provably Efficient Fictitious Play Policy Optimization for Zero-Sum Markov Games with Structured Transitions [145.54544979467872]
  We propose and analyze new fictitious play policy optimization algorithms for zero-sum Markov games with structured but unknown transitions. We prove tight $widetildemathcalO(sqrtK)$ regret bounds after $K$ episodes in a two-agent competitive game scenario. Our algorithms feature a combination of Upper Confidence Bound (UCB)-type optimism and fictitious play under the scope of simultaneous policy optimization.
  arXiv  Detail & Related papers  (2022-07-25T18:29:16Z)
• Anytime Optimal PSRO for Two-Player Zero-Sum Games [17.821479538423155]
  Policy Space Response Oracles (PSRO) is a reinforcement learning algorithm for games that can handle continuous actions. AODO is a double oracle algorithm for 2-player zero-sum games that converges to a Nash equilibrium. We show that our methods achieve far lower exploitability than DO and PSRO and never increase exploitability.
  arXiv  Detail & Related papers  (2022-01-19T16:34:11Z)
• On the Impossibility of Convergence of Mixed Strategies with No Regret Learning [10.515544361834241]
  We study convergence properties of the mixed strategies that result from a general class of optimal no regret learning strategies. We consider the class of strategies whose information set at each step is the empirical average of the opponent's realized play.
  arXiv  Detail & Related papers  (2020-12-03T18:02:40Z)
• Faster Algorithms for Optimal Ex-Ante Coordinated Collusive Strategies in Extensive-Form Zero-Sum Games [123.76716667704625]
  We focus on the problem of finding an optimal strategy for a team of two players that faces an opponent in an imperfect-information zero-sum extensive-form game. In that setting, it is known that the best the team can do is sample a profile of potentially randomized strategies (one per player) from a joint (a.k.a. correlated) probability distribution at the beginning of the game. We provide an algorithm that computes such an optimal distribution by only using profiles where only one of the team members gets to randomize in each profile.
  arXiv  Detail & Related papers  (2020-09-21T17:51:57Z)
• Efficient Competitive Self-Play Policy Optimization [20.023522000925094]
  We propose a new algorithmic framework for competitive self-play reinforcement learning in two-player zero-sum games. Our method trains several agents simultaneously, and intelligently takes each other as opponent based on simple adversarial rules. We prove theoretically that our algorithm converges to an approximate equilibrium with high probability in convex-concave games.
  arXiv  Detail & Related papers  (2020-09-13T21:01:38Z)
• Efficient exploration of zero-sum stochastic games [83.28949556413717]
  We investigate the increasingly important and common game-solving setting where we do not have an explicit description of the game but only oracle access to it through gameplay. During a limited-duration learning phase, the algorithm can control the actions of both players in order to try to learn the game and how to play it well. Our motivation is to quickly learn strategies that have low exploitability in situations where evaluating the payoffs of a queried strategy profile is costly.
  arXiv  Detail & Related papers  (2020-02-24T20:30:38Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.