期刊
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
卷 32, 期 1, 页码 172-182出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TCSVT.2021.3056208
关键词
Handheld computers; Training; Standards; Task analysis; Deep learning; Convolution; Adaptive systems; Partial domain adaptation; unsupervised domain adaptation; graph convolutional networks
资金
- Inha University Research Grant
- Institute for Information and Communications Technology Planning and Evaluation (IITP) - Korean Government (MSIT) (MSIT) (Artificial Intelligence Convergence Research Center, Inha University) [2020-0-01389]
- National Research Foundation of Korea [5199991014250] Funding Source: Korea Institute of Science & Technology Information (KISTI), National Science & Technology Information Service (NTIS)
This article addresses the challenges of Partial Domain Adaptation (PDA) and proposes an Adaptive Graph Adversarial Networks (AGAN) method to overcome the limitations of existing approaches. Experimental results show that the proposed method outperforms state-of-the-art methods on public PDA benchmark datasets.
This article tackles Partial Domain Adaptation (PDA) where the target label set is a subset of the source label set. A key challenging issue in PDA is to prevent negative transfer by isolating source-private classes. Since there is no label information for a target domain, PDA methods require to estimate a label commonness score between source and target domains. Existing approaches use either class-level or sample-level commonness to alleviate the negative transfer issue. However, class-level methods assign the same label commonness to all samples of the same class without considering each sample's characteristics. Also, the recently introduced sample-level approaches show better performance but they still suffer from negative transfer due to non-trivial anomaly samples. To address these limitations, we propose Adaptive Graph Adversarial Networks (AGAN) consisting of two specialized modules. The adaptive class-relational graph module is designed to utilize the intra- and inter-domain structures through adaptive feature propagation. Complementarily, the sample-level commonness predictor computes a commonness score of each sample. Extensive experimental results on public PDA benchmark datasets demonstrate that our structure-aware method outperforms state-of-the-art methods.
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