Real-Time Scheduling Based on Simulation and Deep Reinforcement Learning with Featured Action Space.
Shufang XieTao ZhangOliver RosePublished in: WSC (2022)
Keyphrases
- action space
- reinforcement learning
- state space
- markov decision processes
- state and action spaces
- real valued
- continuous state
- state action
- continuous state spaces
- function approximation
- reinforcement learning methods
- control policies
- action selection
- single agent
- function approximators
- optimal policy
- reinforcement learning algorithms
- markov decision process
- stochastic processes
- learning algorithm
- markov decision problems
- objective function
- model free
- average reward
- mobile robot
- decision theoretic
- multi agent
- dynamic programming