TABDeep: A two-level action branch architecture-based deep reinforcement learning for distributed sub-tree scheduling of online multicast sessions in EON.
Xia LiYuping WangPublished in: Comput. Networks (2024)
Keyphrases
- reinforcement learning
- distributed architecture
- distributed systems
- meeting scheduling
- hierarchical architecture
- real time
- distributed processing
- multi agent
- loosely coupled
- multi agent architecture
- action selection
- master slave
- layered architecture
- scalable distributed
- heterogeneous environments
- distributed environment
- online learning
- tree structure
- function approximation
- scheduling algorithm
- multicast tree
- resource allocation
- round robin
- model free
- distributed query processing
- reward shaping
- partially observable domains
- learning algorithm
- scheduling problem
- index structure
- application layer
- state space
- end to end
- multicast routing
- reinforcement learning algorithms
- quality of service
- machine learning
- peer to peer
- transition model
- optimal policy
- markov decision processes
- computational grids
- user sessions
- network topology
- network model
- network architecture