Recovering the Propensity Score from Biased Positive Unlabeled Data.
Walter GerychThomas HartvigsenLuke BuquicchioEmmanuel AguElke A. RundensteinerPublished in: AAAI (2022)
Keyphrases
- unlabeled data
- labeled data
- semi supervised learning
- semi supervised
- sample selection bias
- active learning
- co training
- text categorization
- learning algorithm
- training data
- labeled and unlabeled data
- semi supervised classification
- text classification
- supervised learning
- labeled examples
- data points
- labeled training data
- class labels
- positive and negative
- number of labeled examples
- training set
- small set of labeled
- domain adaptation
- supervised learning algorithms
- semisupervised learning
- labeled data for training
- transfer learning
- label propagation
- data mining
- unsupervised learning
- machine learning
- decision boundary
- learning tasks
- unlabeled instances
- training examples
- pairwise
- supervised and semi supervised
- positive and unlabeled examples
- similarity measure