MSRL: Distributed Reinforcement Learning with Dataflow Fragments.
Huanzhou ZhuBo ZhaoGang ChenWeifeng ChenYijie ChenLiang ShiYaodong YangPeter R. PietzuchLei ChenPublished in: USENIX Annual Technical Conference (2023)
Keyphrases
- reinforcement learning
- multi agent
- distributed systems
- function approximation
- peer to peer
- lightweight
- computer networks
- cooperative
- state space
- data flow
- distributed environment
- learning algorithm
- robotic control
- temporal difference learning
- parallel execution
- optimal policy
- machine learning
- markov decision processes
- fault tolerant
- computing environments
- mobile agents
- distributed data
- reinforcement learning algorithms
- learning process
- database systems
- database machine