M-ADDA: Unsupervised Domain Adaptation with Deep Metric Learning
基於域適應度量的對抗性判別神經網絡 1) 應用度量學習方式去訓練源域模型，通過the triplet loss（）進行優化。這樣的結果使具有相同標籤的類別相互靠近，不同類別的標籤相互遠離。 2)利用對抗的方式像ADDA從源域和目標域中提取特徵，同時，我們優化了一個新的損失函數，促進目標域數據集從聚類中嵌入。 神經網絡，通過最小化每個目標域樣本嵌入和它對應於源域嵌入的聚類中心。
相關文章
- 1. Unsupervised Domain Adaptation with Residual Transfer Networks（2017）
- 2. 【論文閱讀】Deep Cocktail Network: Multi-source Unsupervised Domain Adaptation with Category Shift
- 3. [cvpr2017]Unsupervised Pixel–Level Domain Adaptation with Generative Adversarial Networks
- 4. [cvpr2017]Deep Hashing Network for Unsupervised Domain Adaptation
- 5. Learning Transferable Features with Deep Adaptation Networks
- 6. Unsupervised Domain Adaptation by Backpropagation(2015)
- 7. Unsupervised domain adaptation with residual transfer networks(NIPS 2016)
- 8. Maximum Classifier Discrepancy for Unsupervised Domain Adaptation
- 9. 【2017_ICCV】Deep Metric Learning with Angular Loss
- 10. Re-weighted adversarial adaptation network for unsupervised domain adaptation
- 更多相關文章...
- • XSLT
- • XQuery 添加元素 和屬性 - XQuery 教程
- • Java Agent入門實戰（一）-Instrumentation介紹與使用
- • Java Agent入門實戰（三）-JVM Attach原理與使用
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. Unsupervised Domain Adaptation with Residual Transfer Networks（2017）
- 2. 【論文閱讀】Deep Cocktail Network: Multi-source Unsupervised Domain Adaptation with Category Shift
- 3. [cvpr2017]Unsupervised Pixel–Level Domain Adaptation with Generative Adversarial Networks
- 4. [cvpr2017]Deep Hashing Network for Unsupervised Domain Adaptation
- 5. Learning Transferable Features with Deep Adaptation Networks
- 6. Unsupervised Domain Adaptation by Backpropagation(2015)
- 7. Unsupervised domain adaptation with residual transfer networks(NIPS 2016)
- 8. Maximum Classifier Discrepancy for Unsupervised Domain Adaptation
- 9. 【2017_ICCV】Deep Metric Learning with Angular Loss
- 10. Re-weighted adversarial adaptation network for unsupervised domain adaptation