Struct MeanFieldGameSolver 

pub struct MeanFieldGameSolver {
    pub num_states: usize,
    pub num_actions: usize,
    pub transitions: Vec<Vec<Vec<f64>>>,
    pub rewards: Vec<Vec<f64>>,
    pub discount: f64,
    pub tol: f64,
    pub max_iter: usize,
}

Mean field game solver via fixed-point iteration.

Iterates:

1. Given population distribution μ, solve individual optimal control → policy π.
2. Given policy π, simulate population dynamics → new distribution μ’.
3. Repeat until ||μ’ - μ|| < ε.

Fields

num_states: usize

Number of states.

num_actions: usize

Number of actions.

transitions: Vec<Vec<Vec<f64>>>

Transition kernel (state-dependent on population): T[s][a][s’].

rewards: Vec<Vec<f64>>

Reward (depends on state, action, mean population weight). reward[s][a] — simplified to not depend on μ for tractability.

discount: f64

Discount factor.

tol: f64

Convergence tolerance.

max_iter: usize

Maximum iterations.

Implementations

impl MeanFieldGameSolver

pub fn new( num_states: usize, num_actions: usize, transitions: Vec<Vec<Vec<f64>>>, rewards: Vec<Vec<f64>>, discount: f64, tol: f64, max_iter: usize, ) -> Self

Construct a mean field game solver.

pub fn solve(&self) -> (Vec<usize>, Vec<f64>)

Run the mean field game fixed-point iteration. Returns (equilibrium_policy, equilibrium_distribution).

Trait Implementations

impl Clone for MeanFieldGameSolver

fn clone(&self) -> MeanFieldGameSolver

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for MeanFieldGameSolver

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.