MARS: Malleable Actor-Critic Reinforcement Learning Scheduler.
Betis BaheriJacob TrongeBo FangAng LiVipin ChaudharyQiang GuanPublished in: IPCCC (2022)
Keyphrases
- actor critic
- reinforcement learning
- policy gradient
- temporal difference
- reinforcement learning algorithms
- optimal control
- approximate dynamic programming
- neuro fuzzy
- gradient method
- function approximation
- state space
- policy iteration
- learning algorithm
- dynamic programming
- rl algorithms
- action selection
- evaluation function
- partially observable
- natural actor critic
- model free
- optimal policy
- multi agent
- machine learning
- linear program
- markov decision processes
- fuzzy rules
- dynamical systems
- control problems
- control policy
- average reward
- supervised learning
- reinforcement learning methods
- least squares
- search space
- neural network