M-DRL: Deep Reinforcement Learning Based Coflow Traffic Scheduler with MLFQ Threshold Adaption.
Tianba ChenWei LiYuKang SunYunchun LiPublished in: Int. J. Parallel Program. (2021)
Keyphrases
- reinforcement learning
- resource utilization
- function approximation
- model free
- traffic flow
- network traffic
- markov decision processes
- machine learning
- temporal difference
- action selection
- scheduling algorithm
- dynamic programming
- multi agent
- learning algorithm
- transfer learning
- policy search
- optimal control
- learning process
- real time
- traffic congestion
- network resources
- robotic control
- adaptive threshold
- threshold selection
- traffic signal
- traffic management
- function approximators
- telecommunication networks
- reinforcement learning algorithms
- response time
- learning problems