Assessment of Reward Functions in Reinforcement Learning for Multi-Modal Urban Traffic Control under Real-World limitations.
Alvaro Cabrejas EgeaColm ConnaughtonPublished in: ITSC (2021)
Keyphrases
- multi modal
- reward function
- reinforcement learning
- reinforcement learning algorithms
- markov decision processes
- state space
- policy search
- optimal policy
- markov decision process
- transition model
- function approximation
- temporal difference
- high dimensional
- inverse reinforcement learning
- multi modality
- multiple agents
- generative model
- state variables
- multi agent
- state action
- cross modal
- model free
- dynamic programming
- transition probabilities
- markov decision problems
- linear programming
- multi agent systems
- uni modal