Early Failure Detection of Deep End-to-End Control Policy by Reinforcement Learning.
Keuntaek LeeKamil SaigolEvangelos A. TheodorouPublished in: ICRA (2019)
Keyphrases
- control policy
- end to end
- failure detection
- reinforcement learning
- admission control
- approximate dynamic programming
- long run
- control policies
- fault detection
- wireless ad hoc networks
- congestion control
- multipath
- model free
- function approximation
- state space
- scalable video
- highly reliable
- optimal policy
- control strategies
- action space
- real time
- application layer
- markov decision processes
- multistage
- data mining