Beyond OOD State Actions: Supported Cross-Domain Offline Reinforcement Learning.
Jinxin LiuZiqi ZhangZhenyu WeiZifeng ZhuangYachen KangSibo GaiDonglin WangPublished in: CoRR (2023)
Keyphrases
- cross domain
- reinforcement learning
- transfer learning
- state space
- action space
- knowledge transfer
- domain adaptation
- cross domain learning
- data sets
- sentiment classification
- text categorization
- learning tasks
- unsupervised learning
- text mining
- formal specification
- learning process
- knowledge base
- reward function
- information systems
- agent learns
- data mining