Joint empirical risk minimization for instance-dependent positive-unlabeled data.
Wojciech RejchelPawel TeisseyreJan MielniczukPublished in: CoRR (2023)
Keyphrases
- unlabeled data
- empirical risk minimization
- labeled data
- semi supervised learning
- semi supervised
- active learning
- supervised learning
- labeled and unlabeled data
- co training
- text classification
- training data
- vc dimension
- labeled examples
- data points
- learning algorithm
- training set
- positive and negative
- training examples
- positive examples
- class labels
- phase transition
- domain adaptation
- transfer learning
- learning problems
- machine learning
- data sets
- information extraction
- prior knowledge
- learning tasks
- unsupervised learning
- support vector machine
- decision trees
- feature selection