期刊
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
卷 43, 期 11, 页码 3918-3930出版社
IEEE COMPUTER SOC
DOI: 10.1109/TPAMI.2020.2991050
关键词
Measurement; Training; Kernel; Task analysis; Adaptation models; Benchmark testing; Games; Domain adaptation; transfer learning; adversarial learning
资金
- National Key R&D Program of China [2018YFE0203900]
- National Natural Science Foundation of China [61806039, 61832001]
- Sichuan Department of Science and Technology [20ZDYF2771]
This paper proposes a new domain adaptation method named ATM, which aims to reduce domain divergence by minimizing inter-domain divergence and maximizing intra-class density, combining adversarial training and metric learning. Experimental results show that the proposed ATM achieves state-of-the-art performance on multiple benchmark datasets.
Unsupervised domain adaptation addresses the problem of transferring knowledge from a well-labeled source domain to an unlabeled target domain where the two domains have distinctive data distributions. Thus, the essence of domain adaptation is to mitigate the distribution divergence between the two domains. The state-of-the-art methods practice this very idea by either conducting adversarial training or minimizing a metric which defines the distribution gaps. In this paper, we propose a new domain adaptation method named adversarial tight match (ATM) which enjoys the benefits of both adversarial training and metric learning. Specifically, at first, we propose a novel distance loss, named maximum density divergence (MDD), to quantify the distribution divergence. MDD minimizes the inter-domain divergence (match in ATM) and maximizes the intra-class density (tight in ATM). Then, to address the equilibrium challenge issue in adversarial domain adaptation, we consider leveraging the proposed MDD into adversarial domain adaptation framework. At last, we tailor the proposed MDD as a practical learning loss and report our ATM. Both empirical evaluation and theoretical analysis are reported to verify the effectiveness of the proposed method. The experimental results on four benchmarks, both classical and large-scale, show that our method is able to achieve new state-of-the-art performance on most evaluations.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据