Fighting Uncertainty with Gradients: Offline Reinforcement Learning via Diffusion Score Matching.
H. J. Terry SuhGlen ChouHongkai DaiLujie YangAbhishek GuptaRuss TedrakePublished in: CoRR (2023)
Keyphrases
- reinforcement learning
- function approximation
- matching algorithm
- matching score
- partial observability
- real time
- model free
- matching process
- pattern matching
- graph matching
- diffusion process
- feature points
- keypoints
- shape matching
- belief functions
- control problems
- uncertain data
- optimal control
- supervised learning
- feature matching
- mobile robot
- temporal difference
- learning process
- learning algorithm