A parallel scheduling algorithm for reinforcement learning in large state space.
Quan LiuXudong YangLing JingJin LiJiao LiPublished in: Frontiers Comput. Sci. (2012)
Keyphrases
- scheduling algorithm
- state space
- reinforcement learning
- response time
- load balance
- reinforcement learning algorithms
- optimal policy
- markov decision processes
- scheduling strategy
- markov chain
- action space
- dynamic programming
- model free
- heuristic search
- partially observable
- planning problems
- continuous state spaces
- learning algorithm
- power saving
- state abstraction
- grid environment
- machine learning
- optimal scheduling
- temporal difference
- state variables
- function approximation
- dynamical systems
- markov decision process
- particle filter
- reward function
- reward shaping
- scheduling policies
- belief state
- action selection
- continuous queries
- shared memory
- database systems