Ricerc@Sapienza

Finite mixture models are an important tool in the statistical analysis of data, for example in data clustering. The optimal parameters of a mixture model are usually computed by maximizing the log-likelihood functional via the Expectation-Maximization algorithm. We propose an alternative approach based on the theory of Mean Field Games, a class of differential games with an infinite number of agents. We show that the solution of a finite state space multi-population Mean Field Games system characterizes the critical points of the log-likelihood functional for a Bernoulli mixture.

In this work, we consider the discretization of some nonlinear Fokker-Planck-Kolmogorov equations. The scheme we propose preserves the nonnegativity of the solution, conserves the mass, and, as the discretization parameters tend to zero, has limit measure-valued trajectories which are shown to solve the equation. The main assumptions to obtain a convergence result are that the coefficients are continuous and satisfy a suitable linear growth property with respect to the space variable. In particular, we obtain a new proof of existence of solutions for such equations.

We propose a numerical method for stationary Mean Field Games defined on a network.
In this framework a correct approximation of the transition conditions at the vertices plays a crucial
role. We prove existence, uniqueness and convergence of the scheme and we also propose a least squares
method for the solution of the discrete system. Numerical experiments are carried out.