Equilibria in load balancing games Article First Online: 08 September 2009 Received: 08 May 2009 Revised: 08 July 2009 DOI:
Cite this article as: Chen, B. Acta Math. Appl. Sin. Engl. Ser. (2009) 25: 723. doi:10.1007/s10255-009-8832-8 Abstract
A Nash equilibrium (NE) in a multi-agent game is a strategy profile that is resilient to unilateral deviations. A strong Nash equilibrium (SE) is one that is stable against coordinated deviations of any coalition. We show that, in the load balancing games, NEs approximate SEs in the sense that the benefit of each member of any coalition from coordinated deviations is well limited. Furthermore, we show that an easily recognizable special subset of NEs exhibit even better approximation of SEs.
Keywords Nash equilibrium load balancing approximation
Supported by “Taishan Scholar” Project in Applied Mathematics, Qufu Normal University.
Andelman, N., Feldman, M., Mansour, Y. Strong price of anarchy. In: Proceeding of 18th Annual ACM-SIAM Symposium on Discrete Algorithms, 2007, 189–198
Aumann, R. Acceptable Points in General Cooperative
n-Person Games. In: Contributions to the Theory of Games IV, Annals of Mathematics 40, ed. by R.D. Luce, A.W. Tucker, 1959, 287–324
Feldman, M., Tamir, T. Approximate Strong Equilibrium in Job Scheduling Games. To appear in: Journal of Artificial Intelligence Research
Fotakis, D., Kontogiannis, S., Mavronicolas, M., Spiraklis, P. The structure and complexity of Nash equilibria for a selfish routing game. In: Proceedings of the 29th International Colloquium on Automata, Languages and Programming, 2002, 510–519
Graham, R. Bounds on multiprocessing timing anomalies.
SIAM J. Applied Mathematics
, 17: 263–269 (1969)
Korte, B., Vygen, J. Combinatorial Optimization: Theory and Algorithms, 4th ed. Springer, 2008
Monien, B., Schroeder, U.-P., editors. Proceedings of the 1st International Symposium on Algorithmic Game Theory, Vol 4997 of Lecture Notes in Computer Science. Springer, 2008
Wolfram Research. Mathematica: Technical and Scientific Software.
http://www.wolfram.com/ Copyright information
© Institute of Applied Mathematics, Academy of Mathematics and System Sciences, Chinese Academy of Sciences and Springer Berlin Heidelberg 2009