Adaptive Metro Service Schedule and Train Composition With a Proximal Policy Optimization Approach Based on Deep Reinforcement Learning.
Cheng-shuo YingAndy H. F. ChowYi-Hui WangKwai-Sang ChinPublished in: IEEE Trans. Intell. Transp. Syst. (2022)
Keyphrases
- reinforcement learning
- optimal policy
- action selection
- optimization algorithm
- policy search
- admission control
- markov decision process
- actor critic
- management system
- partially observable environments
- markov decision processes
- partially observable
- web service composition
- service providers
- web services
- action space
- policy iteration
- function approximators
- markov decision problems
- composition of web services
- function approximation
- service oriented
- web services composition
- traffic control
- reinforcement learning problems
- policy gradient
- policy evaluation
- optimization problems
- scheduling problem
- partially observable markov decision processes
- transport systems
- control policies
- control policy
- service selection
- machine learning
- learning capabilities
- service oriented architecture
- service composition
- dynamic programming
- learning algorithm