Random-Sampling Multipath Hypothesis Propagation for Cost Approximation in Long-Horizon Optimal Control.
Shankarachary RagiHans D. MittelmannPublished in: CCTA (2020)
Keyphrases
- random sampling
- optimal control
- multipath
- end to end
- active learning
- dynamic programming
- sampling algorithm
- sample size
- control strategy
- antenna array
- multiple paths
- routing algorithm
- reinforcement learning
- optimal control problems
- bit error rate
- infinite horizon
- sliding window
- random sample
- total cost
- fading channels
- multipath routing
- data sets
- optimal policy
- data streams
- training data
- real time