A Distributed Model-Free Ride-Sharing Approach for Joint Matching, Pricing, and Dispatching Using Deep Reinforcement Learning.
Marina HaliemGanapathy ManiVaneet AggarwalBharat K. BhargavaPublished in: IEEE Trans. Intell. Transp. Syst. (2021)
Keyphrases
- model free
- reinforcement learning
- reinforcement learning algorithms
- function approximation
- temporal difference
- multi agent
- policy iteration
- state space
- optimal policy
- rl algorithms
- policy evaluation
- reinforcement learning methods
- dynamic programming
- temporal difference learning
- machine learning
- scheduling problem
- partially observable
- average reward
- fixed point
- transfer learning
- impedance control