Theoretical Analysis of Self-Training with Deep Networks on Unlabeled Data.
Colin WeiKendrick ShenYining ChenTengyu MaPublished in: ICLR (2021)
Keyphrases
- unlabeled data
- theoretical analysis
- semi supervised learning
- co training
- semi supervised
- labeled data
- semi supervised classification
- training set
- active learning
- labeled examples
- supervised learning
- text classification
- labeled and unlabeled data
- training data
- text categorization
- machine learning
- class labels
- domain adaptation
- small set of labeled
- abundant unlabeled data
- training examples
- labeled training data
- data points
- transfer learning
- single view
- positive examples
- prior knowledge
- pairwise
- unsupervised learning
- learning algorithm
- class distribution
- multi view
- multi view learning
- semi supervised learning algorithms
- support vector machine
- decision boundary
- named entities
- label propagation
- unlabeled instances
- object recognition