Gambit provides an array of programs for manipulating and analyzing games.
This chapter describes these programs in detail.
- Table of Contents
- gambit-enumpure -- Searches for pure-strategy Nash equilibria in a game.
- gambit-enumpoly -- Computes Nash equilibria in a game by solving polynomial systems
- gambit-enummixed -- Computes Nash equilibria in a two-player strategic game using extreme
point enumeration
- gambit-gnm -- Computes Nash equilibria in a strategic game using a global Newton method
- gambit-ipa -- Computes Nash equilibria in a strategic game using iterated polymatrix approximation
- gambit-lcp -- Computes Nash equilibria in a two-player game by solving a
linear complementarity program
- gambit-lp -- Computes Nash equilibria in a two-player constant-sum game by
solving a linear program
- gambit-liap -- Computes Nash equilibria in a game using a function minimization
approach
- gambit-simpdiv -- Computes approximations to Nash equilibria in a strategic game using
a simplicial subdivision approach
- gambit-logit -- Computes the principal branch of the (logit) quantal response
correspondence in a game
This section describes the implementations of various methods for
computing Nash equilibria provided by Gambit. For a general overview
of the formulations of the Nash equilibrium problem, see the survey
of McKelvey and McLennan (1996) [McKMcL96].
By convention, these programs take an extensive or strategic game file
on standard input, and output a list of equilibria computed. The
equilibria computed are presented as a list of comma-separated probabilities,
preceded by the tag "NE". Many of the programs optionally
output additional information about the operation of the algorithm;
these outputs have other, program-specific tags, described in the
individual program documentation.