Independent Policy Gradient for Large-Scale Markov Potential Games: Sharper Rates, Function Approximation, and Game-Agnostic Convergence.
Dongsheng DingChen-Yu WeiKaiqing ZhangMihailo R. JovanovicPublished in: CoRR (2022)
Keyphrases
- function approximation
- policy gradient
- game theory
- game playing
- video games
- reinforcement learning
- game play
- nash equilibria
- game theoretic
- nash equilibrium
- actor critic
- temporal difference learning
- temporal difference
- stochastic games
- learning tasks
- reinforcement learning algorithms
- imperfect information
- markov chain
- function approximators
- radial basis function
- model free
- policy search
- convergence rate
- convergence speed
- gradient method
- learning experience