Image sequence recognition with active learning using uncertainty sampling.
Masatoshi MinakawaBisser RaytchevToru TamakiKazufumi KanedaPublished in: IJCNN (2013)
Keyphrases
- active learning
- uncertainty sampling
- cost sensitive
- image sequences
- experimental design
- random sampling
- query by committee
- learning algorithm
- supervised learning
- machine learning
- ensemble members
- semi supervised learning
- training examples
- misclassification costs
- learning strategies
- training set
- maximum entropy
- generalization error
- semi supervised
- feature extraction
- probability estimates
- learning process
- sample complexity
- machine learning algorithms
- unlabeled data
- class imbalance
- classification accuracy
- minimize total
- labeled data
- empirical studies
- multi class