Training Confidence-calibrated Classifiers for Detecting Out-of-Distribution Samples.
Kimin LeeHonglak LeeKibok LeeJinwoo ShinPublished in: CoRR (2017)
Keyphrases
- training samples
- training set
- training examples
- test set
- training data
- sample set
- training and test data
- training and testing data
- supervised learning
- number of training samples
- test data
- trained classifiers
- learning algorithm
- training set size
- active learning
- optimum path forest
- data sets
- discriminative classifiers
- decision boundary
- training dataset
- test sample
- training process
- feature space
- supervised training
- naive bayes
- feature selection
- boosted classifiers
- decision trees
- support vector
- classification performances
- avoid overfitting
- machine learning
- small sample
- high dimensional
- data samples
- random variables
- data distribution
- class distribution
- roc curve
- sufficient training data
- support vector machine
- feature set
- class labels
- classification algorithm
- confidence measure
- confidence scores
- density estimates
- linear classifiers
- support vectors
- labeled samples
- unlabeled samples
- labeled training data