The goal of this project is to approach the problem of finding a symmetric Nash Equilibrium in a symmetric two-player normal form game as a function minimization problem.
The algorithm used is simulated annealing. The motivation behind choosing this particular method is that the objective function does not guarantee all minima to be global, i.e. equal 0 as that is the game-theoretic interpretation here. Hence, stochastic approach is used to find the global minimum of the objective function which is the symmetric Nash Equilibrium of the given game.