A novel double-mGBDT-based Q-learning.
Qiming FuShuai MaDawei TianJianping ChenZhen GaoShan ZhongPublished in: Int. J. Model. Identif. Control. (2021)
Keyphrases
- reinforcement learning
- cooperative
- multi agent
- function approximation
- state space
- learning algorithm
- optimal policy
- reinforcement learning algorithms
- action selection
- stochastic approximation
- learning rate
- model free
- multi agent reinforcement learning
- potential field
- policy iteration
- real time
- temporal difference learning
- continuous state and action spaces