用于交通信号灯控制的特征表示近似Q学习 (Q-learning with Feature-based Approximation for Traffic Light Control).
Minshuo LiMinghai YaoPublished in: 计算机科学 (2018)
Keyphrases
- cooperative
- reinforcement learning
- multi agent
- learning algorithm
- closed form
- reinforcement learning algorithms
- function approximation
- learning rate
- approximation error
- approximation algorithms
- error bounds
- state space
- model free
- potential field
- action selection
- queueing networks
- relative error
- temporal difference learning
- stochastic approximation