Behavior Prior Representation learning for Offline Reinforcement Learning.
Hongyu ZangXin LiJie YuChen LiuRiashat IslamRemi Tachet des CombesRomain LarochePublished in: ICLR (2023)
Keyphrases
- reinforcement learning
- learning process
- learning algorithm
- learning systems
- knowledge acquisition
- prior knowledge
- function approximation
- mobile learning
- learned knowledge
- model free
- learning tasks
- supervised learning
- background knowledge
- learning classifier systems
- dynamic programming
- visual representation
- real robot
- active learning
- reinforcement learning methods
- autonomous learning
- eligibility traces