Towards an Information Theoretic Framework of Context-Based Offline Meta-Reinforcement Learning.
Lanqing LiHai ZhangXinyu ZhangShatong ZhuJunqiao ZhaoPheng-Ann HengPublished in: CoRR (2024)
Keyphrases
- reinforcement learning
- function approximation
- meta level
- state space
- reinforcement learning algorithms
- meta reasoning
- multi agent
- optimal policy
- machine learning
- temporal difference
- context sensitive
- neural network
- temporal difference learning
- action selection
- learning process
- real world
- real time
- optimal control
- transfer learning
- dynamic programming
- learning agents
- multi agent reinforcement learning
- relational reinforcement learning
- search results clustering