Multiple source domain adaptation: A sharper bound using weighted Rademacher complexity.
Jianwei LiuJiajia ZhouXionglin LuoPublished in: TAAI (2015)
Keyphrases
- domain adaptation
- rademacher complexity
- reproducing kernel hilbert space
- multiple sources
- data dependent
- risk bounds
- target domain
- generalization error
- error bounds
- cross domain
- semi supervised
- data sources
- semi supervised learning
- kernel methods
- transfer learning
- labeled data
- learning algorithm
- sentiment classification
- training data
- user behavior
- text classification
- support vector
- reinforcement learning