Fast Online Reinforcement Learning of Distributed Optimal Controller for Large-Scale Network Systems.
Tomoki HoshiyaTomonori SadamotoPublished in: CCTA (2021)
Keyphrases
- reinforcement learning
- distributed systems
- distributed network
- optimal control
- computer networks
- data intensive
- peer to peer
- multi agent
- cooperative
- dynamic programming
- communication cost
- fault management
- highly distributed
- control policy
- communication overhead
- network traffic
- computer systems
- online learning
- massive scale
- mobile sensor
- single point of failure
- real time
- local area network
- network management
- complex networks
- learning algorithm
- neural network