DOMAIN-AGNOSTIC META-LEARNING FOR CROSS-DOMAIN FEW-SHOT CLASSIFICATION

被引:1
|
作者
Lee, Wei-Yu [1 ]
Wang, Jheng-Yu [1 ]
Wang, Yu-Chiang Frank [2 ]
机构
[1] MOXA Inc, Technol & Res Corp Div, Taipei, Taiwan
[2] Natl Taiwan Univ, Dept Elect Engn, Taipei, Taiwan
关键词
Meta-learning; Few-shot classification;
D O I
10.1109/ICASSP43922.2022.9746025
中图分类号
O42 [声学];
学科分类号
070206 ; 082403 ;
摘要
Few-shot classification requires one to classify instances of novel classes, given only a few examples of each class. Although promising meta-learning methods have been proposed recently, there is no guarantee that existing solutions would generalize to novel classes from an unseen domain. In this paper, we tackle the challenging task of cross-domain few-shot classification and propose Domain-Agnostic Meta-Learning (DAML) algorithm. Our DAML, serving as an optimization strategy, learns to adapt the model to novel classes in both seen and unseen domains by data sampled from multiple domains with desirable task settings. In our experiments, we apply DAML on three popular metric-based models under cross-domain settings. Experiments on several benchmarks (mini-ImageNet, CUB, Cars, Places, Plantae and META-DATASET) show that DAML significantly improves the generalization ability of learning models, and addresses cross-domain few-shot classification with promising results.
引用
收藏
页码:1715 / 1719
页数:5
相关论文
共 50 条
  • [31] Meta-collaborative comparison for effective cross-domain few-shot learning
    Zhou, Fei
    Wang, Peng
    Zhang, Lei
    Wei, Wei
    Zhang, Yanning
    PATTERN RECOGNITION, 2024, 156
  • [32] Perspectives of Calibrated Adaptation for Few-Shot Cross-Domain Classification
    Kong, Dechen
    Yang, Xi
    Wang, Nannan
    Gao, Xinbo
    IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY, 2025, 35 (03) : 2410 - 2421
  • [33] Adversarial Feature Augmentation for Cross-domain Few-Shot Classification
    Hu, Yanxu
    Ma, Andy J.
    COMPUTER VISION, ECCV 2022, PT XX, 2022, 13680 : 20 - 37
  • [34] Few-Shot Cross-Domain Object Detection With Instance-Level Prototype-Based Meta-Learning
    Zhang, Lin
    Zhang, Bo
    Shi, Botian
    Fan, Jiayuan
    Chen, Tao
    IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY, 2024, 34 (10) : 9078 - 9089
  • [35] HybridPrompt: Domain-Aware Prompting for Cross-Domain Few-Shot Learning
    Wu, Jiamin
    Zhang, Tianzhu
    Zhang, Yongdong
    INTERNATIONAL JOURNAL OF COMPUTER VISION, 2024, 132 (12) : 5681 - 5697
  • [36] HybridPrompt: Domain-Aware Prompting for Cross-Domain Few-Shot Learning
    Wu, Jiamin
    Zhang, Tianzhu
    Zhang, Yongdong
    International Journal of Computer Vision, 132 (12): : 5681 - 5697
  • [37] Feature extractor stacking for cross-domain few-shot learning
    Wang, Hongyu
    Frank, Eibe
    Pfahringer, Bernhard
    Mayo, Michael
    Holmes, Geoffrey
    MACHINE LEARNING, 2024, 113 (01) : 121 - 158
  • [38] Ranking Distance Calibration for Cross-Domain Few-Shot Learning
    Li, Pan
    Gong, Shaogang
    Wang, Chengjie
    Fu, Yanwei
    2022 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2022, : 9089 - 9098
  • [39] Spectral Decomposition and Transformation for Cross-domain Few-shot Learning
    Liu, Yicong
    Zou, Yixiong
    Li, Ruixuan
    Li, Yuhua
    NEURAL NETWORKS, 2024, 179
  • [40] Relevance equilibrium network for cross-domain few-shot learning
    Ji, Zhong
    Kong, Xiangyu
    Wang, Xuan
    Liu, Xiyao
    INTERNATIONAL JOURNAL OF MULTIMEDIA INFORMATION RETRIEVAL, 2024, 13 (02)