A Communication-Efficient Multi-Agent Actor-Critic Algorithm for Distributed Reinforcement Learning.
Yixuan LinKaiqing ZhangZhuoran YangZhaoran WangTamer BasarRomeil SandhuJi LiuPublished in: CoRR (2019)
Keyphrases
- reinforcement learning
- multi agent
- actor critic
- dynamic programming
- learning algorithm
- objective function
- particle swarm optimization
- computational complexity
- simulated annealing
- optimal solution
- approximate dynamic programming
- average reward
- np hard
- multi agent systems
- linear programming
- state space
- markov chain
- mathematical model
- convergence rate
- temporal difference
- policy iteration
- cost function