Using reinforcement learning to minimize taxi idle times.
Kevin O'KeeffeSam AnklesariaPaolo SantiCarlo RattiPublished in: J. Intell. Transp. Syst. (2022)
Keyphrases
- reinforcement learning
- function approximation
- state space
- markov decision processes
- orders of magnitude
- multi agent
- response time
- reinforcement learning algorithms
- temporal difference
- autonomous learning
- temporal difference learning
- dynamic programming
- partially observable
- policy search
- machine learning
- reward function
- model free
- optimal control
- optimal policy
- case study
- knowledge base
- learning algorithm