arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1909.07618cs.CVcs.MM
收藏

循环一致条件对抗转移网络

Cycle-consistent Conditional Adversarial Transfer Networks

University of Electronic Science and Technology of China(电子科技大学);The University of Queensland(昆士兰大学);Indiana University-Purdue University Indianapolis(印第安纳大学与普渡大学印第安纳波利斯联合分校);Shandong Normal University(山东师范大学)

Jingjing Li, Erpeng Chen, Zhengming Ding, Lei Zhu, Ke Lu, Zi Huang

摘要:

领域自适应研究的是跨领域知识转移问题,其中标记的源领域和未标记的目标领域具有不同的数据分布。近年来,对抗训练已成功应用于领域自适应,并取得了最新的性能。然而,目前的对抗模型仍然存在一个致命的弱点,这是由对抗训练的均衡挑战提出的。具体地说,尽管大多数现有方法能够混淆域匹配,但它们不能保证源域和目标域足够相似。在本文中,我们提出了一种新的方法命名为{ \it循环一致的条件对抗转移网络}(3CATN)来处理这个问题。我们的方法通过利用对抗训练来处理域对齐。具体来说,我们用学习特征和分类器预测的互协方差来调节对抗网络,以捕获数据分布的多模态结构。然而,由于分类器预测不是确定性信息,因此当预测不准确时,具有预测的强条件是有风险的。因此,我们进一步提出,真正的域不变的功能应该能够从一个域到其他翻译。为此,我们引入了两个特征翻译损失和一个周期一致性损失的条件对抗领域自适应网络。在经典数据集和大规模数据集上进行的大量实验验证了我们的模型能够超越以前的最先进技术,并有显着的改进。

英文摘要:

Domain adaptation investigates the problem of cross-domain knowledge transfer where the labeled source domain and unlabeled target domain have distinctive data distributions. Recently, adversarial training have been successfully applied to domain adaptation and achieved state-of-the-art performance. However, there is still a fatal weakness existing in current adversarial models which is raised from the equilibrium challenge of adversarial training. Specifically, although most of existing methods are able to confuse the domain discriminator, they cannot guarantee that the source domain and target domain are sufficiently similar. In this paper, we propose a novel approach named {\it cycle-consistent conditional adversarial transfer networks} (3CATN) to handle this issue. Our approach takes care of the domain alignment by leveraging adversarial training. Specifically, we condition the adversarial networks with the cross-covariance of learned features and classifier predictions to capture the multimodal structures of data distributions. However, since the classifier predictions are not certainty information, a strong condition with the predictions is risky when the predictions are not accurate. We, therefore, further propose that the truly domain-invariant features should be able to be translated from one domain to the other. To this end, we introduce two feature translation losses and one cycle-consistent loss into the conditional adversarial domain adaptation networks. Extensive experiments on both classical and large-scale datasets verify that our model is able to outperform previous state-of-the-arts with significant improvements.

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。

↑