MetaCARD: Meta-Reinforcement Learning with Task Uncertainty Feedback via Decoupled Context-Aware Reward and Dynamics Components.
Min WangXin LiLeiji ZhangMingzhong WangPublished in: AAAI (2024)
Keyphrases
- context aware
- reinforcement learning
- contextual information
- context awareness
- ubiquitous computing
- mobile devices
- ubiquitous learning
- mobile computing
- ambient intelligence
- state space
- mobile users
- reward function
- markov decision processes
- learning algorithm
- context aware systems
- smart home
- multi agent
- optimal policy
- smart environments
- learning process
- ubiquitous environment
- mobile applications
- machine learning
- transfer learning
- distributed systems
- reinforcement learning algorithms
- location aware
- learning agent
- current context
- context aware services
- context aware computing
- personalized services
- context sensitive
- uncertain data