Offline Robot Reinforcement Learning with Uncertainty-Guided Human Expert Sampling.
Ashish KumarIlya KuzovkinPublished in: CoRR (2022)
Keyphrases
- human experts
- reinforcement learning
- robot control
- real robot
- mobile robot
- post processing
- autonomous learning
- perceptual aliasing
- autonomous robots
- robotic systems
- human robot interaction
- humanoid robot
- multi robot
- domain experts
- optimal policy
- state space
- markov decision processes
- machine learning
- learning algorithm
- supervised learning
- learning process
- feature vectors
- training set
- learning agent
- action space
- expert systems