DART: Domain-Adversarial Residual-Transfer networks for unsupervised cross-domain image classification.
Xianghong FangHaoli BaiZiyi GuoBin ShenSteven C. H. HoiZenglin XuPublished in: Neural Networks (2020)
Keyphrases
- cross domain
- image classification
- knowledge transfer
- transfer learning
- multi domain
- multiple domains
- domain adaptation
- sentiment classification
- target domain
- image representation
- text categorization
- knowledge sharing
- cross domain learning
- machine learning
- supervised learning
- learning tasks
- semi supervised
- image features
- unsupervised learning
- text mining
- source domain
- active learning