Model-Free Safe Reinforcement Learning Through Neural Barrier Certificate.
Yujie YangYuxuan JiangYichen LiuJianyu ChenShengbo Eben LiPublished in: IEEE Robotics Autom. Lett. (2023)
Keyphrases
- model free
- reinforcement learning
- reinforcement learning algorithms
- fitted q iteration
- function approximation
- temporal difference
- neural network
- rl algorithms
- state space
- markov decision processes
- reinforcement learning methods
- policy evaluation
- policy iteration
- average reward
- learning tasks
- action selection
- markov chain
- transfer learning
- temporal difference learning
- image classification
- multi agent