An Efficient PTAS for Two-Strategy Anonymous Games

  • Authors:
  • Constantinos Daskalakis

  • Affiliations:
  • Microsoft Research,

  • Venue:
  • WINE '08 Proceedings of the 4th International Workshop on Internet and Network Economics
  • Year:
  • 2008

Quantified Score

Hi-index 0.00

Visualization

Abstract

We present a novel polynomial time approximation scheme fortwo-strategy anonymous games, in which the players’ utilityfunctions, although potentially different, do not differentiateamong the identities of the other players. Our algorithm computesan ε-approximate Nash equilibrium of an n-player 2-strategyanonymous game in time $\text{poly}(n) \cdot(1/\epsilon)^{O(1/\epsilon^2)}$, which significantly improves uponthe running time $n^{O(1/\epsilon^2)}$ required by the algorithm ofDaskalakis & Papadimitriou, 2007. The improved running time isbased on a new structural understanding of approximate Nashequilibria: We show that, for any ε, there exists anε-approximate Nash equilibrium in which either onlyO(1/ε 3) players randomize, or all players who randomizeuse the same mixed strategy. To show this result we employ toolsfrom the literature on Stein’s Method.