Class-Incremental Domain Adaptation
類增長域自適應 摘要： 引入了CIDA範例，現存的DA方法能解決domain-shift問題但是不使用在學習目標域中新穎的類別，CI方法在源訓練數據缺失的情況下可以學習新的類別，但是不能解決無監督的domain-shift問題，本文就是解決CIDA問題，基於原型網絡可以識別shared-class和novel class(one-shot) Introduction: 現目前深度學習的弱點
相關文章
- 1. Open Set Domain Adaptation
- 2. domain adaptation 之 cycleGAN
- 3. Multi-Adversarial Domain Adaptation
- 4. Open Compound Domain Adaptation
- 5. ADDA: Adversarial Discriminative Domain Adaptation
- 6. DANN：Unsupervised Domain Adaptation by Backpropagation
- 7. Multi-source Distilling Domain Adaptation
- 8. Domain Adaptation總結（2017.9）
- 9. [cvpr2017]Adversarial Discriminative Domain Adaptation
- 10. Re-weighted adversarial adaptation network for unsupervised domain adaptation
- 更多相關文章...
- • RSS
元素 - RSS 教程 - • RSS
元素 - RSS 教程 - • Tomcat學習筆記（史上最全tomcat學習筆記）
相關標籤/搜索
每日一句
-
每一个你不满意的现在，都有一个你没有努力的曾经。
歡迎關注本站公眾號,獲取更多信息
相關文章
- 1. Open Set Domain Adaptation
- 2. domain adaptation 之 cycleGAN
- 3. Multi-Adversarial Domain Adaptation
- 4. Open Compound Domain Adaptation
- 5. ADDA: Adversarial Discriminative Domain Adaptation
- 6. DANN：Unsupervised Domain Adaptation by Backpropagation
- 7. Multi-source Distilling Domain Adaptation
- 8. Domain Adaptation總結（2017.9）
- 9. [cvpr2017]Adversarial Discriminative Domain Adaptation
- 10. Re-weighted adversarial adaptation network for unsupervised domain adaptation