Safety robustness of reinforcement learning policies: A view from robust control.
Hao XiongXiumin DiaoPublished in: Neurocomputing (2021)
Keyphrases
- reinforcement learning
- control policy
- optimal policy
- control problems
- high robustness
- control policies
- optimal control
- digital image watermarking
- control strategies
- action selection
- policy search
- computationally efficient
- control system
- function approximation
- real time
- state space
- strong robustness
- hierarchical reinforcement learning
- machine learning
- reinforcement learning algorithms
- markov decision processes
- robot control
- markov decision process
- learning algorithm