A Policy-Based Reinforcement Learning Approach for High-Speed Railway Timetable Rescheduling.
Yin WangYisheng LvJianying ZhouZhiming YuanQi ZhangMin ZhouPublished in: ITSC (2021)
Keyphrases
- reinforcement learning
- high speed railway
- optimal policy
- policy search
- markov decision process
- modeling method
- action selection
- control policy
- function approximators
- state space
- function approximation
- markov decision processes
- key technologies
- reward function
- policy gradient
- action space
- temporal difference
- reinforcement learning algorithms
- high speed train
- model free
- timetabling problem
- agent learns
- partially observable markov decision processes
- virtual reality
- learning algorithm
- puts forward
- image processing
- computer vision
- genetic algorithm
- machine learning