Parallel Computation of Nash Equilibria in N-Player Games

—We propose a parallel algorithm for finding Nash equilibria in n-player noncooperative games. The algorithm is based on enumerating the supports of mixed strategies in parallel and solving for each support the corresponding multilinear equations giving the candidate equilibria. We implemented the proposed algorithm using MPI on a cluster of computers. We performed extensive experiments on random games to show the performance of the parallel algorithm. Keywords-game theory; Nash equilibrium; parallel algorithm;

• Type: Conference
• Year: 2009
• Where: CSE
• Authors: Jonathan Widger, Daniel Grosu
• Keywords: Computer Science And Engineering, CSE 2009, N-player Noncooperative Games, Nash Equilibria, Parallel Algorithm