Dealing with the Unknown: Pessimistic Offline Reinforcement Learning.
Jinning LiChen TangMasayoshi TomizukaWei ZhanPublished in: CoRR (2021)
Keyphrases
- reinforcement learning
- initially unknown
- function approximation
- real time
- temporal difference learning
- state space
- temporal difference
- reinforcement learning algorithms
- databases
- social networks
- information systems
- dynamic programming
- optimal control
- action selection
- relational reinforcement learning
- control problems
- partially observable
- supervised learning
- learning algorithm
- machine learning
- real world