Uncertainty-based Out-of-Distribution Classification in Deep Reinforcement Learning.
Andreas SedlmeierThomas GaborThomy PhanLenz BelznerClaudia Linnhoff-PopienPublished in: ICAART (2) (2020)
Keyphrases
- reinforcement learning
- classification accuracy
- machine learning
- pattern recognition
- classification systems
- decision trees
- feature extraction
- automatic classification
- pattern classification
- supervised learning
- benchmark datasets
- classification rules
- classification method
- classification algorithm
- class labels
- neural network
- machine learning algorithms
- random variables
- decision rules
- support vector machine svm
- text classification
- markov decision processes
- image classification
- feature vectors
- density estimation
- optimal control
- training set
- confidence intervals
- feature selection
- partial observability