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