Unsupervised domain adaptation aims to transfer knowledge from a labeled source domain to a related but unlabeled target domain. Most existing approaches either adversarially reduce the domain shift or use pseudo-labels to provide category information during adaptation. However, an adversarial training method may sacrifice the discriminability of the target data, since no category information is available. Moreover, a pseudo labeling method is difficult to produce high-confidence samples, since the classifier is often source-trained and there exists the domain discrepancy. Thus, it may have a negative influence on learning target representations. A potential solution is to make them compensate each other to simultaneously guarantee the feature transferability and discriminability, which are the two key criteria of feature representations in domain adaptation. In this paper, we propose a novel method named ATPL, which mutually promotes
A
dversarial
T
raining and
P
seudo
L
abeling for unsupervised domain adaptation. ATPL can produce high-confidence pseudo-labels by adversarial training. Accordingly, ATPL will use the pseudo-labeled information to improve the adversarial training process, which can guarantee the feature transferability by generating adversarial data to fill in the domain gap. Those pseudo-labels can also boost the feature discriminability. Extensive experiments on real datasets demonstrate that the proposed ATPL method outperforms state-of-the-art unsupervised domain adaptation methods.
中文翻译:
无监督域适应旨在将知识从标记的源域转移到相关但未标记的目标域。大多数现有方法要么对抗性地减少域转移,要么使用伪标签在适应期间提供类别信息。然而,对抗性训练方法可能会牺牲目标数据的可辨别性,因为没有可用的类别信息。此外,伪标记方法难以产生高置信度的样本,因为分类器通常是经过源训练的并且存在域差异。因此,它可能对学习目标表示产生负面影响。一个潜在的解决方案是让它们相互补偿以同时保证特征的可迁移性和可辨别性,这是域适应中特征表示的两个关键标准。在本文中,我们提出了一种名为 ATPL 的新方法,它可以相互促进
用于无监督域
适应
的
对抗
训练
和
伪
标签。ATPL 可以通过对抗训练产生高置信度的伪标签。因此,ATPL 将使用伪标记信息来改进对抗训练过程,通过生成对抗数据来填补领域空白,从而保证特征的可迁移性。这些伪标签还可以提高特征的可辨别性。对真实数据集的大量实验表明,所提出的 ATPL 方法优于最先进的无监督域适应方法。