Multiple source domain adaptation: A bound using p-norm covering numbers.
Jianwei LiuJiajia ZhouXionglin LuoPublished in: TAAI (2015)
Keyphrases
- domain adaptation
- uniform convergence
- reproducing kernel hilbert space
- multiple sources
- vc dimension
- sample complexity
- upper bound
- target domain
- labeled data
- semi supervised
- transfer learning
- generalization error
- semi supervised learning
- unlabeled data
- cross domain
- data sources
- training data
- lower bound
- sentiment classification
- co training
- data points
- test data
- data sets
- supervised learning
- learning problems
- prior knowledge
- training set
- mistake bound
- learning algorithm