Policy Optimization for Markov Games: Unified Framework and Faster Convergence.
Runyu ZhangQinghua LiuHuan WangCaiming XiongNa LiYu BaiPublished in: CoRR (2022)
Keyphrases
- faster convergence
- unified framework
- markov games
- markov decision process
- convergence speed
- markov decision processes
- global optimum
- step size
- multiagent reinforcement learning
- global optimization
- reinforcement learning algorithms
- pso algorithm
- optimal policy
- convergence rate
- state space
- control problems
- probabilistic model
- reinforcement learning
- bayesian networks
- particle swarm optimization
- initial state
- dynamic programming
- search space
- cooperative
- objective function