Assessment of Reward Functions for Reinforcement Learning Traffic Signal Control under Real-World Limitations.
Alvaro Cabrejas EgeaShaun HowellMaksis KnutinsColm ConnaughtonPublished in: SMC (2020)
Keyphrases
- reinforcement learning
- reward function
- traffic signal control
- policy search
- reinforcement learning algorithms
- markov decision processes
- state space
- optimal policy
- markov decision process
- inverse reinforcement learning
- function approximation
- machine learning
- transition model
- model free
- multi agent
- temporal difference
- multiple agents
- generative model
- learning algorithm
- data mining
- transition probabilities
- state variables
- optimal control
- markov decision problems
- continuous state
- multi objective