Adversarial Domain Adaptation via Category Transfer
Adversarial Domain Adaptation via Category Transfer
复制标题
DOI：
10.1109/ijcnn.2019.8851925
复制
发表时间：
2019-07
期刊：
影响因子：
--
通讯作者：
Lusi Li;Haibo He;Jie Li;Guang Yang
中科院分区：
文献类型：
--
作者：
Lusi Li;Haibo He;Jie Li;Guang Yang
对抗性领域自适应在学习可转移特征表示和减少源域和目标域之间的分布差异方面取得了一定的成功。然而，现有的方法主要集中在全局源分布和目标分布的对齐上，而没有考虑不同分布下的类别中的复杂结构，导致领域混淆和可区分结构的混合。本文提出了一种基于类别转移的对抗性领域自适应方法(ADACT)，用于无监督领域自适应(UDA)。ADACT首先通过训练源和目标特征生成器以及标签预测器捕获多类别信息。其次，它使用多类别领域批评者网络来估计跨域的Wasserstein距离。然后，它通过将不同的数据分布与估计的Wasserstein距离进行精细匹配来学习类别不变的特征表示。通过这种两步迭代的标准反向传播训练方法，可以实现自适应。ADACT的有效性得到了验证，因为它在公共领域适应数据集上的性能优于几种最先进的UDA方法。
Adversarial domain adaptation has achieved some success in learning transferable feature representations and reducing distribution discrepancy between source and target domains. However, existing approaches mainly focus on alignment of global source and target distributions without considering complex structures in categories underlying different distributions, resulting in domain confusion and the mix of distinguishable structures. In this paper, we propose an adversarial domain adaptation via category transfer (ADACT) approach for unsupervised domain adaptation (UDA). ADACT first captures multi-category information through training source and target feature generators as well as a label predictor. Secondly, it uses multi-category domain critic networks to category-wisely estimate Wasserstein distances across domains. Then it learns category-invariant feature representations by finely-grained matching different data distributions with the estimated Wasserstein distances. The adaptation can be achieved by the standard back-propagation training approach with this two-step iteration. The effectiveness of ADACT is demonstrated since it outperforms several state-of-the-art UDA methods on common domain adaptation datasets.