![](https://img-blog.csdnimg.cn/20201014180756928.png?x-oss-process=image/resize,m_fixed,h_64,w_64)
Domain Adaptation
文章平均质量分 95
无CCFA就不改名
这个作者很懒,什么都没留下…
展开
-
DA-5-Source-Free Domain Adaptation via Distribution Estimation
Source-Free Domain Adaptation via Distribution Estimation翻译 2022-06-02 12:57:06 · 791 阅读 · 1 评论 -
DA-4-FixBi: Bridging Domain Spaces for Unsupervised Domain Adaptation
FixBi: Bridging Domain Spaces for Unsupervised Domain Adaptation, 2021CVPRMethodFixed Ratio-based MixupConfidence-based Learning正伪标签的双向匹配(Bidirectional Matching with positive pseudo-labels)带有负伪标签的自我惩罚(Self-penalization with negative pseudo-labels)一致性正则化(Co翻译 2021-10-27 14:51:50 · 852 阅读 · 3 评论 -
DA-3-Contrastive Adaptation Network for Unsupervised Domain Adaptation
Contrastive Adaptation Network for Unsupervised Domain Adaptation方法最大均值差异(MMD, Maximun Mean Discreoancy)对比域差异(Contrastive Domain Discrepancy)对比自适应网络(Contrastive Adaptation Network)CAN优化Alternative optimization (AO)Class-aware Sampling(CAS)原文链接(2019 CVPR)翻译 2021-10-25 11:16:14 · 687 阅读 · 0 评论 -
DA-2-Unsupervised Multi-source Domain Adaptation Without Access to Source Data
Unsupervised Multi-source Domain Adaptation Without Access to Source DataRelated worksMethodologyCVPR 2021 · Sk Miraj Ahmed, Dripta S. Raychaudhuri, Sujoy Paul, Samet Oymak, Amit K. Roy-Chowdhury ·原文链接:https://openaccess.thecvf.com/content/CVPR2021/pap翻译 2021-09-22 21:04:57 · 657 阅读 · 0 评论 -
DA-1-再生核Hilbert空间
Reproducing Kernel Hilbert Space,再生核希尔伯特空间Eigen Decomposition翻译的原文链接本文主要讨论和函数核再生核希尔伯特空间(reproducing kernel Hilbert space, RKHS)。核方法的动机在于映射一个Rn\mathcal{R}^{n}Rn空间中的向量为特征空间中的另一个向量。如下图所示,红色和蓝色的数据点在Rn\mathcal{R}^{n}Rn空间中是不易分的。但如果将这些数据点映射到高维特征空间,我们也许可以使其容易分翻译 2021-09-08 23:01:19 · 243 阅读 · 0 评论