Generalizable Task Representation Learning for Offline Meta-Reinforcement Learning with Data Limitations.
Renzhe ZhouChenxiao GaoZongzhang ZhangYang YuPublished in: AAAI (2024)
Keyphrases
- reinforcement learning
- learning algorithm
- prior knowledge
- data sets
- raw data
- data collection
- data sources
- human experts
- supervised learning
- image data
- data processing
- data analysis
- high quality
- machine learning
- database
- background knowledge
- synthetic data
- previously learned
- original data
- data mining techniques
- knowledge discovery
- data points
- learning process
- data structure
- domain knowledge
- training data
- neural network
- temporal abstractions
- real time