English(EN) Variable Bound Tightening for Nash Equilibrium Computation in Multiplayer Imperfect-Information Games

新算法解决复杂多人博弈中的纳什均衡问题 · 跟踪3 个来源

作者 PulseAugur 编辑部 · [3 个来源] · 2026-06-24 16:08

研究人员开发了一种名为“投影可利用性下降”（Projected Exploitability Descent, PED）的新算法，用于近似计算具有不完美信息的复杂多人博弈中的纳什均衡。该算法最小化了可利用性函数的一个代理目标，这是一个非凸且不光滑的目标。虽然 PED 在长时间运行中表现出持续的改进，但最初的性能不如已有的方法，如虚构博弈（Fictitious Play, FP）和反事实遗憾最小化（Counterfactual Regret Minimization, CFR）。一种混合方法 FP-PED 结合了 FP 的初始效率和 PED 的长期优化能力，在三方库恩扑克等基准测试中表现出改进的性能。 AI

影响这项研究可能带来更具可扩展性和效率的游戏均衡计算方法，从而影响复杂战略环境中的 AI 代理。

排序理由该集群描述了一篇学术论文中提出的一种新算法，用于解决博弈论中的特定计算问题。

在 Hugging Face Daily Papers 阅读 →

AI 生成摘要 · Google Gemini · 来自 3 个来源。我们如何撰写摘要 →

报道来源 [3]

arXiv cs.AI TIER_1 English(EN) · Sam Ganzfried · 2026-06-30 04:00

Projected Exploitability Descent for Nash Equilibrium Computation in Multiplayer Imperfect-Information Games

arXiv:2606.29169v1 Announce Type: cross Abstract: Many important games have more than two players and imperfect information. Existing approaches for computing Nash equilibrium, the central game-theoretic solution concept, in such games either lack scalability or obtain poor perfo…
arXiv cs.MA (Multiagent) TIER_1 English(EN) · Sam Ganzfried · 2026-06-28 03:25

Projected Exploitability Descent for Nash Equilibrium Computation in Multiplayer Imperfect-Information Games

Many important games have more than two players and imperfect information. Existing approaches for computing Nash equilibrium, the central game-theoretic solution concept, in such games either lack scalability or obtain poor performance. In this paper we introduce a new algorithm…
Hugging Face Daily Papers TIER_1 English(EN) · 2026-06-24 16:08

Variable Bound Tightening for Nash Equilibrium Computation in Multiplayer Imperfect-Information Games

There has been significant recent progress in algorithms for approximation of Nash equilibrium in large two-player zero-sum imperfect-information games and exact computation of Nash equilibrium in multiplayer strategic-form games. While counterfactual regret minimization and fict…

报道来源 [3]

Projected Exploitability Descent for Nash Equilibrium Computation in Multiplayer Imperfect-Information Games

Projected Exploitability Descent for Nash Equilibrium Computation in Multiplayer Imperfect-Information Games

Variable Bound Tightening for Nash Equilibrium Computation in Multiplayer Imperfect-Information Games

相关实体

相关话题