On handling negative transfer and imbalanced distributions in multiple source transfer learning.
Liang GeJing GaoHung Q. NgoKang LiAidong ZhangPublished in: Stat. Anal. Data Min. (2014)
Keyphrases
- transfer learning
- target domain
- domain adaptation
- multiple sources
- training and test data
- source domain
- cross domain
- knowledge transfer
- labeled data
- learning tasks
- target class
- machine learning
- transfer knowledge
- active learning
- transductive transfer learning
- multi task learning
- reinforcement learning
- multi task
- collaborative filtering
- probability distribution
- learning algorithm
- semi supervised learning
- text classification
- machine learning algorithms
- text categorization
- unlabeled data
- database
- general purpose
- supervised learning
- similarity measure
- class distribution
- transferring knowledge
- random variables