Policy Gradient Methods for the Noisy Linear Quadratic Regulator over a Finite Horizon.
Ben HamblyRenyuan XuHuining YangPublished in: CoRR (2020)
Keyphrases
- linear quadratic
- finite horizon
- optimal control
- infinite horizon
- policy gradient methods
- policy gradient
- average cost
- closed loop
- optimal policy
- markov decision processes
- vector valued
- dynamical systems
- dynamic programming
- control strategy
- reinforcement learning
- natural actor critic
- partially observable
- markov decision process
- multistage
- gaussian model
- real time
- machine learning
- probability distribution
- real valued
- steady state
- state space
- control system