Global Convergence of Localized Policy Iteration in Networked Multi-Agent Reinforcement Learning.
Yizhou ZhangGuannan QuPan XuYiheng LinZaiwei ChenAdam WiermanPublished in: CoRR (2022)
Keyphrases
- multi agent reinforcement learning
- global convergence
- policy iteration
- convergence rate
- reinforcement learning
- markov decision processes
- stochastic games
- convergence speed
- convergence analysis
- average reward
- model free
- global optimum
- step size
- multi agent
- optimal policy
- least squares
- temporal difference
- optimization methods
- fixed point
- multi agent systems
- markov decision process
- finite state
- infinite horizon
- reinforcement learning algorithms
- objective function
- learning algorithm
- optimal control
- function approximation
- hybrid algorithm
- search algorithm
- transfer learning
- multiagent systems
- linear programming
- supervised learning
- dynamic programming
- search space