Synthesis of maximum margin and multiview learning using unlabeled data.
Sándor SzedmákJohn Shawe-TaylorPublished in: Neurocomputing (2007)
Keyphrases
- unlabeled data
- learning algorithm
- supervised learning
- active learning
- semi supervised learning
- maximum margin
- co training
- labeled data
- labeled and unlabeled data
- semi supervised
- learning problems
- prior knowledge
- sample selection bias
- learning tasks
- support vector
- learning process
- unsupervised learning
- text categorization
- reinforcement learning
- machine learning
- decision trees
- class labels
- background knowledge
- transfer learning
- text classification
- learning models
- multi task
- generalization error
- small number