Sparse optimal control of networks with multiplicative noise via policy gradient.
Benjamin GravellYi GuoTyler H. SummersPublished in: CoRR (2019)
Keyphrases
- optimal control
- policy gradient
- multiplicative noise
- actor critic
- dynamic programming
- control problems
- reinforcement learning
- control strategy
- infinite horizon
- additive noise
- image denoising
- noisy images
- mathematical model
- total variation
- synthetic aperture radar
- probability distribution
- noise reduction
- function approximation
- markov chain
- variance reduction
- state space