MCDAL: Maximum Classifier Discrepancy for Active Learning.
Jae-Won ChoDong-Jin KimYunjae JungIn So KweonPublished in: IEEE Trans. Neural Networks Learn. Syst. (2023)
Keyphrases
- active learning
- training examples
- label noise
- training set
- training data
- learning algorithm
- query by committee
- decision trees
- support vector
- classifier systems
- learning strategies
- representative samples
- training samples
- random sampling
- neural network
- class labels
- classification scheme
- selective sampling
- svm classifier
- feature selection
- semi supervised
- feature set
- unlabeled data
- classification method
- transfer learning
- ensemble learning
- classification rate
- random selection
- learning process
- machine learning
- feature space
- labeled instances
- multiple classifiers
- multi class
- binary classifiers
- labeled examples
- classification process
- learning classifier systems
- experimental design
- linear classifiers
- generalization error
- relevance feedback
- supervised learning