Reusing the Task-specific Classifier as a Discriminator: Discriminator-free Adversarial Domain Adaptation

被引:56
|
作者
Chen, Lin [1 ]
Chen, Huaian [1 ]
Wei, Zhixiang [1 ]
Jin, Xin [1 ]
Tan, Xiao [1 ]
Jin, Yi [1 ]
Chen, Enhong [1 ]
机构
[1] Univ Sci & Technol China, Hefei, Peoples R China
基金
中国国家自然科学基金;
关键词
D O I
10.1109/CVPR52688.2022.00704
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Adversarial learning has achieved remarkable performances for unsupervised domain adaptation (UDA). Existing adversarial UDA methods typically adopt an additional discriminator to play the min-max game with a feature extractor. However, most of these methods failed to effectively leverage the predicted discriminative information, and thus cause mode collapse for generator. In this work, we address this problem from a different perspective and design a simple yet effective adversarial paradigm in the form of a discriminator-free adversarial learning network (DALN), wherein the category classifier is reused as a discriminator, which achieves explicit domain alignment and category distinguishment through a unified objective, enabling the DALN to leverage the predicted discriminative information for sufficient feature alignment. Basically, we introduce a Nuclear-norm Wasserstein discrepancy (NWD) that has definite guidance meaning for performing discrimination. Such NWD can be coupled with the classifier to serve as a discriminator satisfying the K-Lipschitz constraint without the requirements of additional weight clipping or gradient penalty strategy. Without bells and whistles, DALN compares favorably against the existing state-of-the-art (SOTA) methods on a variety of public datasets. Moreover, as a plug-and-play technique, NWD can be directly used as a generic regularizer to benefit existing UDA algorithms. Code is available at https://github.com/xiaoachen98/DALN.
引用
收藏
页码:7171 / 7180
页数:10
相关论文
共 50 条
  • [21] EEG emotion Enhancement using Task-specific Domain Adversarial Neural Network
    Ding, Ke-Ming
    Kimura, Tsukasa
    Fukui, Ken-ichi
    Numao, Masayuki
    2021 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS (IJCNN), 2021,
  • [22] Implicit Discriminator Domain Adversarial Residual Network for Cross Domain Rolling Bearing Fault Diagnosis
    Li, Zhuorui
    Ma, Jun
    Wu, Jiande
    Li, Xiang
    Wang, Xiaodong
    IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, 2024, 73
  • [23] Dual Classifier-Discriminator Adversarial Networks for Open Set Fault Diagnosis of Train Bearings
    Ren, He
    Wang, Jun
    Shen, Changqing
    Huang, Weiguo
    Zhu, Zhongkui
    IEEE SENSORS JOURNAL, 2023, 23 (18) : 22040 - 22050
  • [24] Domain adaptation of time series via contrastive learning with task-specific consistency
    Wu, Tao
    Chen, Qiushu
    Zhao, Dongfang
    Wang, Jinhua
    Jiang, Linhua
    APPLIED INTELLIGENCE, 2024, 54 (23) : 12576 - 12588
  • [25] Adaptation to delayed visual feedback is task-specific
    de la Malla, C.
    Lopez-Moliner, J.
    Brenner, E.
    PERCEPTION, 2013, 42 : 232 - 232
  • [26] Task-specific sensorimotor adaptation to reversing prisms
    Marotta, JJ
    Keith, GP
    Crawford, JD
    JOURNAL OF NEUROPHYSIOLOGY, 2005, 93 (02) : 1104 - 1110
  • [27] Task-specific adaptation of speech recognition models
    Sankar, A
    Kannan, A
    Shahshahani, B
    Jackson, E
    ASRU 2001: IEEE WORKSHOP ON AUTOMATIC SPEECH RECOGNITION AND UNDERSTANDING, CONFERENCE PROCEEDINGS, 2001, : 433 - 436
  • [28] UDAMA: Unsupervised Domain Adaptation through Multi-discriminator Adversarial Training with Noisy Labels Improves Cardio-fitness Prediction
    Wu, Yu
    Spathis, Dimitris
    Jia, Hong
    Perez-Pozuelo, Ignacio
    Gonzales, Tomas I.
    Brage, Soren
    Wareham, Nicholas
    Mascolo, Cecilia
    MACHINE LEARNING FOR HEALTHCARE CONFERENCE, VOL 219, 2023, 219
  • [29] Unsupervised Real-World Super Resolution with Cycle Generative Adversarial Network and Domain Discriminator
    Kim, Gwantae
    Park, Jaihyun
    Lee, Kanghyu
    Lee, Junyeop
    Min, Jeongki
    Lee, Bokyeung
    Han, David K.
    Ko, Hanseok
    2020 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION WORKSHOPS (CVPRW 2020), 2020, : 1862 - 1871
  • [30] Bi-Discriminator Domain Adversarial Neural Networks With Class-Level Gradient Alignment
    Zhao, Chuang
    Zhao, Hongke
    Zhu, Hengshu
    Huang, Zhenya
    Feng, Nan
    Chen, Enhong
    Xiong, Hui
    IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS, 2024, 54 (09): : 5283 - 5295