A Domain Data Pattern Randomization based Deep Reinforcement Learning method for Sim-to-Real transfer.
Peng GongDianxi ShiChao XueXucan ChenPublished in: ICIAI (2021)
Keyphrases
- synthetic data
- significant improvement
- data sets
- detection method
- missing data
- data analysis
- original data
- noisy data
- reinforcement learning
- statistical methods
- test data
- preprocessing
- data collection
- input data
- prior knowledge
- similarity measure
- database
- image data
- correlation analysis
- segmentation method
- prior information
- probabilistic model
- pairwise
- high accuracy
- data structure
- data streams
- machine learning
- model free
- data quality
- training data
- high quality
- objective function
- missing values
- domain experts
- pattern matching
- active learning
- dynamic programming
- clustering method
- training samples