Deep Policy Iteration for high-dimensional mean field games.
Mouhcine AssouliBadr MissaouiPublished in: Appl. Math. Comput. (2024)
Keyphrases
- decision problems
- policy iteration
- optimal policy
- high dimensional
- markov decision processes
- sample path
- reinforcement learning
- fixed point
- state space
- stochastic games
- infinite horizon
- finite state
- average reward
- dynamic programming
- policy evaluation
- game theory
- markov decision process
- sufficient conditions
- markov decision problems
- markov random field
- nash equilibria
- least squares
- feature space
- model free
- average cost
- reward function
- bayesian inference
- probability distribution
- temporal difference
- nash equilibrium
- convergence rate
- monte carlo
- em algorithm
- markov chain
- linear programming
- decision making