Safety Margins for Reinforcement Learning.
Alexander GrushinWalt WoodsAlvaro VelasquezSimon KhanPublished in: CoRR (2023)
Keyphrases
- reinforcement learning
- function approximation
- reinforcement learning algorithms
- robotic control
- markov decision processes
- learning algorithm
- optimal control
- direct policy search
- real time
- soft margin
- temporal difference
- state space
- machine learning
- real world
- dynamical systems
- optimal policy
- generalization error
- learning classifier systems
- model free
- support vector
- learning agent
- training data