Optimal Policy Characterization Enhanced Proximal Policy Optimization for Multitask Scheduling in Cloud Computing.
Jiangliang JinYunjian XuPublished in: IEEE Internet Things J. (2022)
Keyphrases
- optimal policy
- cloud computing
- multi task
- markov decision processes
- dynamic programming
- state space
- reinforcement learning
- finite horizon
- data center
- infinite horizon
- data management
- state dependent
- average reward
- long run
- utility computing
- service providers
- scheduling problem
- learning tasks
- control policies
- average cost
- markov decision process
- learning problems
- cloud computing platform
- scheduling algorithm
- transfer learning
- sufficient conditions
- markov decision problems
- cloud services
- stochastic demand
- database systems