MARLIN: Soft Actor-Critic based Reinforcement Learning for Congestion Control in Real Networks.
Raffaele GallieraAlessandro MorelliRoberto FrontedduNiranjan SuriPublished in: NOMS (2023)
Keyphrases
- actor critic
- congestion control
- reinforcement learning
- end to end
- temporal difference
- policy gradient
- optimal control
- reinforcement learning algorithms
- approximate dynamic programming
- neuro fuzzy
- function approximation
- packet loss
- gradient method
- quality of service
- window size
- model free
- state space
- rate control
- policy iteration
- dynamic programming
- average reward
- markov decision processes
- network bandwidth
- action selection
- learning algorithm
- policy gradient methods
- partially observable
- supervised learning
- multi agent
- machine learning
- control policy
- evaluation function
- dynamical systems
- optimal policy
- reinforcement learning methods
- data mining