IG-RL: Inductive Graph Reinforcement Learning for Massive-Scale Traffic Signal Control.
François-Xavier DevaillyDenis LarocqueLaurent CharlinPublished in: IEEE Trans. Intell. Transp. Syst. (2022)
Keyphrases
- reinforcement learning
- traffic signal control
- massive scale
- state space
- machine learning
- function approximation
- reinforcement learning algorithms
- model free
- markov decision processes
- low latency
- optimal policy
- rl algorithms
- learning algorithm
- temporal difference
- low cost
- multi agent
- action space
- genetic algorithm
- cooperative
- highly efficient