Policy Gradient-based Deep Reinforcement Learning for Deadline-aware Transfer over Wide Area Networks.
Kohei ShiomotoTakashi KurimotoPublished in: NetSoft (2021)
Keyphrases
- wide area network
- reinforcement learning
- optimal policy
- policy search
- transfer learning
- model free
- action selection
- markov decision process
- distributed databases
- policy evaluation
- function approximators
- action space
- markov decision processes
- reward function
- state space
- replicated databases
- function approximation
- reinforcement learning algorithms
- response time
- temporal difference
- policy gradient
- tcp ip
- resource constraints
- neural network
- scheduling algorithm