Minerva: A reinforcement learning-based technique for optimal scheduling and bottleneck detection in distributed factory operations.
Tara Elizabeth ThomasJinkyu KooSomali ChaterjiSaurabh BagchiPublished in: COMSNETS (2018)
Keyphrases
- optimal scheduling
- reinforcement learning
- distributed systems
- multi agent
- detection method
- learning algorithm
- automatic detection
- cooperative
- object detection
- scheduling problem
- peer to peer
- distributed data
- state space
- distributed environment
- event detection
- false alarms
- detection algorithm
- function approximation
- concurrent processes
- manufacturing systems
- detection accuracy
- reinforcement learning algorithms
- multi agent reinforcement learning
- single point of failure
- database
- computing environments
- scheduling algorithm
- fault tolerant
- computer networks
- detection rate
- markov decision processes
- false positives
- sensor networks
- multi agent systems
- neural network
- real time