Towards Effective Context for Meta-Reinforcement Learning: an Approach based on Contrastive Learning.
Haotian FuHongyao TangJianye HaoChen ChenXidong FengDong LiWulong LiuPublished in: AAAI (2021)
Keyphrases
- reinforcement learning
- learning algorithm
- learning process
- supervised learning
- learning systems
- learning problems
- machine learning
- knowledge acquisition
- evolutionary learning
- effective learning
- learning mechanism
- learning tasks
- transfer learning
- prior knowledge
- markov decision processes
- background knowledge
- context sensitive
- optimal control
- context dependent
- higher education
- stochastic games
- data sets