Connecting sufficient conditions for domain adaptation: source-guided uncertainty, relaxed divergences and discrepancy localization.
Sofien DhouibSetareh MaghsudiPublished in: CoRR (2022)
Keyphrases
- sufficient conditions
- domain adaptation
- multiple sources
- target domain
- source domain
- cross domain
- transfer learning
- semi supervised
- labeled data
- semi supervised learning
- test data
- domain specific
- exponential stability
- unlabeled data
- text classification
- training data
- asymptotic stability
- sentiment classification
- learning tasks
- training samples
- linear matrix inequality
- databases
- decision trees
- feature selection
- neural network