A stable data-augmented reinforcement learning method with ensemble exploration and exploitation.
Guoyu ZuoZhipeng TianGao HuangPublished in: Appl. Intell. (2023)
Keyphrases
- synthetic data
- test data
- data sets
- reinforcement learning
- data collection
- input data
- data points
- data structure
- data analysis
- prior knowledge
- noisy data
- training data
- database
- missing data
- missing values
- significant improvement
- dynamic programming
- high dimensional
- feature set
- statistical methods
- pairwise
- prior information
- detection method
- decision trees
- temporal abstractions
- machine learning methods
- spectral clustering
- information loss
- function approximation
- support vector machine svm
- training samples
- data mining techniques
- knowledge discovery
- probabilistic model
- training set
- preprocessing
- similarity measure
- image segmentation
- data mining
- neural network