Lyapunov-based uncertainty-aware safe reinforcement learning.
Ashkan B. JeddiNariman L. DehghaniAbdollah ShafieezadehPublished in: CoRR (2021)
Keyphrases
- reinforcement learning
- adaptive control
- partial observability
- control scheme
- state space
- dynamical systems
- function approximation
- learning process
- learning algorithm
- stability analysis
- model free
- supervised learning
- sufficient conditions
- uncertain data
- dynamic programming
- transfer learning
- multi agent
- control law
- control theory
- machine learning
- class of nonlinear systems
- real time
- decision theory
- decision theoretic
- closed loop
- markov decision processes