Fast Convergence of Fictitious Play for Diagonal Payoff Matrices
Jacob D. Abernethy, Kevin A. Lai, Andre Wibisono
摘要
Fictitious Play (FP) is a simple and natural dynamic for repeated play in zero-sum games. Proposed by Brown in 1949, FP was shown to converge to a Nash Equilibrium by Robinson in 1951, albeit at a slow rate that may depend on the dimension of the problem. In 1959, Karlin conjectured that FP converges at the more natural rate of O(1/ √ t). However, Daskalakis and Pan disproved a version of this conjecture in 2014, showing that a slow rate can occur, although their result relies on adversarial tie-breaking. In this paper, we show that Karlin's conjecture is indeed correct for the class of diagonal payoff matrices, as long as ties are broken lexicographically. Specifically, we show that FP converges at a O(1/ √ t) rate in the case when the payoff matrix is diagonal. We also prove this bound is tight by showing a matching lower bound in the identity payoff case under the lexicographic tie-breaking assumption.
问问这篇 Paper
智能体会读完全文。
Lune 把这篇 Paper 索引到了每一个公式,引用它的顶会 Paper 也一样。你提问,回答直接引用原文。
引用它的顶会 Paper6
- Learning to Hash Robustly, GuaranteedAlexandr Andoni, Daniel BeagleholeICML 2022 · 被引用 12 次
- Optimism Without Regularization: Constant Regret in Zero-Sum GamesJohn Lazarsfeld, Georgios Piliouras, Ryann Sim, Stratis SkoulakisNeurIPS 2025 · 被引用 7 次
- Online Learning and Equilibrium Computation with Ranking FeedbackMingyang Liu, Yongshan Chen, Zhiyuan Fan, Gabriele Farina 等ICLR 2026 · 被引用 2 次
- Exponential Lower Bounds for Fictitious Play in Potential GamesIoannis Panageas, Nikolas Patris, Stratis Skoulakis, Volkan CevherNeurIPS 2023 · 被引用 1 次
- Perturbing Best Responses in Zero-Sum GamesAdam Dziwoki, Rostislav HorcíkAAAI 2026
相关 Paper
- Fictitious Play and Best-Response Dynamics in Identical Interest and Zero-Sum Stochastic GamesLucas Baudin, Rida LarakiICML 2022 · 被引用 20 次
- Team-Fictitious Play for Reaching Team-Nash Equilibrium in Multi-team GamesAhmed Said Donmez, Yuksel Arslantas, Muhammed Omer SayinNeurIPS 2024 · 被引用 2 次
- (Doubly) Exponential Lower Bounds for Follow the Regularized Leader in Potential GamesIoannis Anagnostides, Ioannis Panageas, Nikolas Patris, Tuomas SandholmICML 2026 · 被引用 2 次
- Last-Iterate Convergence for Generalized Frank-Wolfe in Monotone Variational InequalitiesZaiwei Chen, Eric MazumdarNeurIPS 2024 · 被引用 7 次
- Learning While Playing in Mean-Field Games: Convergence and OptimalityQiaomin Xie, Zhuoran Yang, Zhaoran Wang, Andreea MincaICML 2021 · 被引用 45 次
