Consistency-Based Semi-Supervised Active Learning: Towards Minimizing Labeling Cost.
Mingfei GaoZizhao ZhangGuo YuSercan Ömer ArikLarry S. DavisTomas PfisterPublished in: CoRR (2019)
Keyphrases
- active learning
- semi supervised
- active learning strategies
- cost sensitive
- labeled data
- unlabeled data
- semi supervised learning
- supervised learning
- label propagation
- unsupervised learning
- cost sensitive learning
- labeled examples
- sample selection
- training set
- learning strategies
- learning algorithm
- random sampling
- machine learning
- relevance feedback
- training examples
- selective sampling
- active sampling
- expected loss
- partially labeled
- labeling effort
- query by committee
- pool based active learning
- co training
- learning process
- total cost
- transfer learning
- incremental learning
- data sets
- experimental design
- multi view
- unlabeled instances
- metric learning
- subspace learning
- semi supervised clustering
- class labels
- labeled and unlabeled data
- weakly supervised
- pairwise
- learning environment
- active learning framework
- decision trees
- batch mode active learning