RL-OPRA: Reinforcement Learning for Online and Proactive Resource Allocation of crowdsourced live videos.
Emna BaccourAiman ErbadAmr MohamedFatima HaouariMohsen GuizaniMounir HamdiPublished in: Future Gener. Comput. Syst. (2020)
Keyphrases
- resource allocation
- reinforcement learning
- function approximation
- reinforcement learning algorithms
- resource allocation problems
- state space
- rl algorithms
- model free
- resource management
- real time
- allocation problems
- online learning
- resource requirements
- video sequences
- resource allocation decisions
- optimal policy
- temporal difference
- multi agent
- learning process
- video annotation
- learning algorithm
- partially observable domains
- resource allocation and scheduling
- optimal resource allocation
- exploration exploitation tradeoff
- markov decision processes
- dynamic programming
- cooperative
- resource usage
- actor critic
- allocation strategies
- combinatorial auctions
- scarce resources
- machine learning
- trading agents
- function approximators
- markov decision process
- resource availability
- grid environment