《MUTUALMEAN-TEACHING:PSEUDO LABEL REFINERY FOR UNSUPERVISED DOMAIN ADAPTATI ON ONPERSON》ICLR2020
聚類算法: 爲無標籤數據賦予僞標籤 優點： 1、原理比較簡單，實現也是很容易，收斂速度快。 2、當結果簇是密集的，而簇與簇之間區別明顯時, 它的效果較好。 3、主要需要調參的參數僅僅是簇數k。 缺點： 1、K值需要預先給定，很多情況下K值的估計是非常困難的。 2、K-Means算法對初始選取的質心點是敏感的，不同的隨機種子點得到的聚類結果完全不同 ，對結果影響很大。 3、對噪音和異常點比較的敏感。
相關文章
- 1. Pseudo Label Based on Multiple Clustering for Unsupervised Cross-Domain Person Re-Identification
- 2. Unsupervised Domain Adaptation by Backpropagation(2015)
- 3. Maximum Classifier Discrepancy for Unsupervised Domain Adaptation
- 4. Unsupervised domain adaptation with residual transfer networks(NIPS 2016)
- 5. Dynamic Label Graph Matching for Unsupervised Video Re-Identification
- 6. Geodesic flow kernel for unsupervised domain adaptation
- 7. [MICCAI2019]Unsupervised Domain Adaptation via Disentangled Representations
- 8. Re-weighted adversarial adaptation network for unsupervised domain adaptation
- 9. 《Unsupervised Domain Adaptation with Random Walks on Target Labelings》閱讀理解
- 10. AAAI2020|Domain Adaptative Attention Learning for Unsupervised Cross-Domain Person Re-Identification
- 更多相關文章...
- • ASP.NET Label 控件 - ASP.NET 教程
- • XSL-FO list-item-label 對象 - XSL-FO 教程
- • 委託模式
- • RxJava操作符（一）Creating Observables
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. Pseudo Label Based on Multiple Clustering for Unsupervised Cross-Domain Person Re-Identification
- 2. Unsupervised Domain Adaptation by Backpropagation(2015)
- 3. Maximum Classifier Discrepancy for Unsupervised Domain Adaptation
- 4. Unsupervised domain adaptation with residual transfer networks(NIPS 2016)
- 5. Dynamic Label Graph Matching for Unsupervised Video Re-Identification
- 6. Geodesic flow kernel for unsupervised domain adaptation
- 7. [MICCAI2019]Unsupervised Domain Adaptation via Disentangled Representations
- 8. Re-weighted adversarial adaptation network for unsupervised domain adaptation
- 9. 《Unsupervised Domain Adaptation with Random Walks on Target Labelings》閱讀理解
- 10. AAAI2020|Domain Adaptative Attention Learning for Unsupervised Cross-Domain Person Re-Identification