Fiche du document
21 mars 2022
- 2203.10774
Mots-clés
Computer Science - Computer Science and Game Theory Computer Science - Artificial Intelligence Computer Science - Multiagent Systems Economics - Theoretical EconomicsCiter ce document
Sam Ganzfried, « Fictitious Play with Maximin Initialization », arXiv - économie
Partage / Export
Résumé
Fictitious play has recently emerged as the most accurate scalable algorithm for approximating Nash equilibrium strategies in multiplayer games. We show that the degree of equilibrium approximation error of fictitious play can be significantly reduced by carefully selecting the initial strategies. We present several new procedures for strategy initialization and compare them to the classic approach, which initializes all pure strategies to have equal probability. The best-performing approach, called maximin, solves a nonconvex quadratic program to compute initial strategies and results in a nearly 75% reduction in approximation error compared to the classic approach when 5 initializations are used.