PAC-Bayesian Domain Adaptation Bounds for Multi-view learning.
Mehdi HennequinKhalid BenabdeslemHaytham ElghazelPublished in: CoRR (2024)
Keyphrases
- pac bayesian
- domain adaptation
- distribution free
- co training
- semi supervised learning
- semi supervised
- transfer learning
- unlabeled data
- generalization bounds
- data dependent
- multi view
- labeled data
- multiple sources
- cross domain
- text classification
- sentiment classification
- target domain
- reproducing kernel hilbert space
- machine learning
- learning problems
- document classification
- model selection
- training data
- test data
- kernel function
- text categorization
- supervised learning
- active learning
- pairwise
- three dimensional
- learning algorithm
- data sets