The Complexity of Games on Highly Regular Graphs

  • Konstantinos Daskalakis
  • Christos H. Papadimitriou
Conference paper

DOI: 10.1007/11561071_9

Volume 3669 of the book series Lecture Notes in Computer Science (LNCS)
Cite this paper as:
Daskalakis K., Papadimitriou C.H. (2005) The Complexity of Games on Highly Regular Graphs. In: Brodal G.S., Leonardi S. (eds) Algorithms – ESA 2005. ESA 2005. Lecture Notes in Computer Science, vol 3669. Springer, Berlin, Heidelberg

Abstract

We present algorithms and complexity results for the problem of finding equilibria (mixed Nash equilibria, pure Nash equilibria and correlated equilibria) in games with extremely succinct description that are defined on highly regular graphs such as the d-dimensional grid; we argue that such games are of interest in the modelling of large systems of interacting agents. We show that mixed Nash equilibria can be found in time exponential in the succinct representation by quantifier elimination, while correlated equilibria can be found in polynomial time by taking advantage of the game’s symmetries. Finally, the complexity of determining whether such a game on the d-dimensional grid has a pure Nash equilibrium depends on d and the dichotomy is remarkably sharp: it is solvable in polynomial time (in fact NL-complete) when d = 1, but it is NEXP-complete for d ≥ 2.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Springer-Verlag Berlin Heidelberg 2005

Authors and Affiliations

  • Konstantinos Daskalakis
    • 1
  • Christos H. Papadimitriou
    • 1
  1. 1.Computer Science DivisionUC BerkeleyBerkeley