DARL: Distance-Aware Uncertainty Estimation for Offline Reinforcement Learning.
Hongchang ZhangJianzhun ShaoShuncheng HeYuhang JiangXiangyang JiPublished in: AAAI (2023)
Keyphrases
- reinforcement learning
- real time
- uncertain data
- parameter estimation
- learning algorithm
- partial observability
- distance function
- markov decision processes
- inherent uncertainty
- supervised learning
- function approximation
- distance measure
- estimation algorithm
- state space
- estimation process
- markov decision process
- partially observable
- estimation accuracy
- reinforcement learning algorithms
- model free
- estimation error
- euclidean distance
- optimal policy