Distributed Disaggregated Communications via Reinforcement Learning and Backpressure (D2CRaB).
Mu-Cheng WangPaul C. HersheyPublished in: SysCon (2023)
Keyphrases
- reinforcement learning
- multi agent
- distributed network
- cooperative
- distributed systems
- function approximation
- lightweight
- peer to peer
- robotic control
- fault tolerant
- temporal difference
- reinforcement learning algorithms
- machine learning
- linear programming relaxation
- heterogeneous environments
- group communication
- multi agent reinforcement learning
- optimal policy
- computing environments
- distributed environment
- markov decision processes
- database
- linear programming
- state space
- dynamic programming
- search algorithm
- learning algorithm
- data mining