Pessimistic value iteration for multi-task data sharing in Offline Reinforcement Learning.
Chenjia BaiLingxiao WangJianye HaoZhuoran YangBin ZhaoZhen WangXuelong LiPublished in: Artif. Intell. (2024)
Keyphrases
- data sharing
- multi task
- reinforcement learning
- markov decision processes
- state space
- transfer learning
- learning problems
- optimal policy
- multi task learning
- learning tasks
- data integration
- data access
- multitask learning
- information sharing
- peer to peer
- learning algorithm
- multi class
- multiple tasks
- gaussian processes
- sparse learning
- feature selection
- supervised learning
- data management
- data sets
- text categorization
- labeled data
- prior knowledge
- pairwise
- metadata
- belief state
- machine learning
- databases