Adaptive proportional fair parameterization based LTE scheduling using continuous actor-critic reinforcement learning.
Ioan Sorin ComsaSijing ZhangMehmet AydinJianping ChenPierre KuonenJean-Frédéric WagenPublished in: GLOBECOM (2014)
Keyphrases
- actor critic
- reinforcement learning
- temporal difference
- reinforcement learning algorithms
- policy gradient
- approximate dynamic programming
- neuro fuzzy
- optimal control
- policy iteration
- function approximation
- markov decision processes
- gradient method
- action space
- model free
- adaptive control
- resource allocation
- control problems
- evaluation function
- recursive least squares
- learning capabilities
- state space
- step size
- learning problems
- transfer learning
- rl algorithms
- multi agent
- machine learning
- natural actor critic
- reinforcement learning methods
- single agent
- linear program
- optimal policy
- dynamic programming
- learning algorithm