arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2103.10583cs.CV
收藏

多源领域自适应的动态传输

Dynamic Transfer for Multi-Source Domain Adaptation

UC San Diego(加州大学圣迭戈分校);Microsoft(微软公司)

Yunsheng Li, Lu Yuan, Yinpeng Chen, Pei Wang, Nuno Vasconcelos

摘要:

多源领域自适应的最新研究工作集中在学习领域无关的模型,其参数是静态的。然而,这样的静态模型难以处理跨多个域的冲突,并且在源域和目标域中遭受性能降级。在本文中,我们提出了动态转移,以解决域冲突,模型参数是适应样本。关键的见解是,跨域调整模型是通过跨样本调整模型来实现的。因此,它打破了源域障碍,将多源域变成了单源域。这也简化了源域和目标域之间的对齐,因为它只需要目标域与源域联合的任何部分对齐。此外,我们发现动态传输可以简单地通过聚集残差矩阵和静态卷积矩阵来建模。实验结果表明,在不使用域标签的情况下,我们的动态转移比最先进的方法高出3%以上的多源领域自适应数据集- DomainNet。源代码位于https://github.com/liyunsheng13/DRT。

英文摘要:

Recent works of multi-source domain adaptation focus on learning a domain-agnostic model, of which the parameters are static. However, such a static model is difficult to handle conflicts across multiple domains, and suffers from a performance degradation in both source domains and target domain. In this paper, we present dynamic transfer to address domain conflicts, where the model parameters are adapted to samples. The key insight is that adapting model across domains is achieved via adapting model across samples. Thus, it breaks down source domain barriers and turns multi-source domains into a single-source domain. This also simplifies the alignment between source and target domains, as it only requires the target domain to be aligned with any part of the union of source domains. Furthermore, we find dynamic transfer can be simply modeled by aggregating residual matrices and a static convolution matrix. Experimental results show that, without using domain labels, our dynamic transfer outperforms the state-of-the-art method by more than 3% on the large multi-source domain adaptation datasets -- DomainNet. Source code is at https://github.com/liyunsheng13/DRT.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑