RAPTOR: End-to-end Risk-Aware MDP Planning and Policy Learning by Backpropagation.
Noah PattonJihwan JeongMichael GimelfarbScott SannerPublished in: CoRR (2021)
Keyphrases
- end to end
- back propagation
- learning algorithm
- neural network
- artificial neural networks
- cascade correlation
- error back propagation
- feedforward neural networks
- neural nets
- action selection
- reinforcement learning
- partially observable
- feed forward
- feed forward neural networks
- congestion control
- optimal policy
- video coding
- admission control
- support vector machine
- state space
- control system
- expert systems
- markov decision problems
- data mining
- real time