Meta-Learning Linear Quadratic Regulators: A Policy Gradient MAML Approach for the Model-free LQR.
Leonardo F. TosoDonglin ZhanJames AndersonHan WangPublished in: CoRR (2024)
Keyphrases
- linear quadratic
- meta learning
- policy gradient
- model free
- optimal control
- reinforcement learning
- reinforcement learning algorithms
- function approximation
- learning tasks
- closed loop
- average reward
- model selection
- machine learning
- dynamical systems
- policy iteration
- vector valued
- machine learning algorithms
- temporal difference
- reinforcement learning methods
- state space
- transfer learning
- learning algorithm
- dynamic programming
- gaussian model
- decision trees
- learning problems
- supervised learning
- control strategy
- data mining
- rl algorithms
- feature selection
- function approximators
- least squares
- multi agent
- partially observable markov decision processes
- optimal policy