A Trajectory Perspective on the Role of Data Sampling Techniques in Offline Reinforcement Learning.
Jinyi LiuYi MaJianye HaoYujing HuYan ZhengTangjie LvChangjie FanPublished in: AAMAS (2024)
Keyphrases
- data sets
- data collection
- data processing
- reinforcement learning
- complex data
- image data
- training data
- data sources
- real time
- synthetic data
- data analysis
- high quality
- trajectory data
- sensor data
- high dimensional data
- data quality
- knowledge discovery
- database
- prior knowledge
- data structure
- multi agent
- small number
- input data
- end users
- mobile robot
- statistical analysis
- spatio temporal
- video sequences
- data objects
- neural network
- sampled data
- spatial and temporal resolution