Warm Start Active Learning with Proxy Labels & Selection via Semi-Supervised Fine-Tuning.
Vishwesh NathDong YangHolger R. RothDaguang XuPublished in: CoRR (2022)
Keyphrases
- fine tuning
- semi supervised
- active learning
- semi supervised learning
- label propagation
- labeled examples
- unlabeled examples
- pairwise
- unlabeled samples
- pairwise constraints
- annotation effort
- labeled data
- fully supervised
- partially labeled data
- unlabeled instances
- unlabeled data
- training examples
- label information
- partially labeled
- fine tune
- labeled samples
- supervised learning
- fine tuned
- viable alternative
- multi label classification
- label noise
- training set
- labeling effort
- multi view
- semi supervised classification
- unsupervised learning
- fully labeled
- labeled and unlabeled data
- training data
- class labels
- learning strategies
- learning process
- semi supervised clustering
- co training
- incremental learning
- transfer learning
- active learning framework
- machine learning
- e learning
- text categorization
- labeled instances
- sample selection
- selective sampling
- multiple instance learning