Few-shot classification with Fork Attention Adapter

被引:0
|
作者
Sun, Jieqi [1 ]
Li, Jian [1 ]
机构
[1] Shaanxi Univ Sci & Technol, Sch Elect & Control Engn, Sch Math & Data Sci, Xian 710021, Peoples R China
关键词
Few-shot Classification; Meta-Learning; Attention mechanism; Dense feature similarity;
D O I
10.1016/j.patcog.2024.110805
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Few-shot learning aims to transfer the knowledge learned from seen categories to unseen categories with a few references. It is also an essential challenge to bridge the gap between humans and deep learning models in real- world applications. Despite extensive previous efforts to tackle this problem by finding an appropriate similarity function, we emphasize that most existing methods have merely considered a single low-resolution representation pair utilized in similarity calculations between support and query samples. Such representational limitations could induce the instability of category predictions. To better achieve metric learning stabilities, we present a novel method dubbed Fork Attention Adapter (FA-adapter), which can seamlessly establish the dense feature similarity with the newly generated nuanced features. The utility of the proposed method is more performant and efficient via the two-stage training phase. Extensive experiments demonstrate consistent and substantial accuracy gains on the fine-grained CUB, Aircraft, non-fine-grained mini-ImageNet, and tiered-ImageNet benchmarks. By comprehensively studying and visualizing the learned knowledge from different source domains, we further present an extension version termed FA-adapter++ ++ to boost the performance in fine-grained scenarios.
引用
收藏
页数:10
相关论文
共 50 条
  • [21] An Investigation of Few-Shot Learning in Spoken Term Classification
    Chen, Yangbin
    Ko, Tom
    Shang, Lifeng
    Chen, Xiao
    Jiang, Xin
    Li, Qing
    INTERSPEECH 2020, 2020, : 2582 - 2586
  • [22] Attention Relational Network for Few-Shot Learning
    Shuai, Jia
    Chen, JiaMing
    Yang, Meng
    INTELLIGENCE SCIENCE AND BIG DATA ENGINEERING: BIG DATA AND MACHINE LEARNING, PT II, 2019, 11936 : 163 - 174
  • [23] META-LEARNING WITH ATTENTION FOR IMPROVED FEW-SHOT LEARNING
    Hou, Zejiang
    Walid, Anwar
    Kung, Sun-Yuan
    2021 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP 2021), 2021, : 2725 - 2729
  • [24] Graph Neural Networks With Triple Attention for Few-Shot Learning
    Cheng, Hao
    Zhou, Joey Tianyi
    Tay, Wee Peng
    Wen, Bihan
    IEEE TRANSACTIONS ON MULTIMEDIA, 2023, 25 : 8225 - 8239
  • [25] Dual Attention Relation Network With Fine-Tuning for Few-Shot EEG Motor Imagery Classification
    An, Sion
    Kim, Soopil
    Chikontwe, Philip
    Park, Sang Hyun
    IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2024, 35 (11) : 15479 - 15493
  • [26] A Image Enhancement Method for Few-shot Classification
    Wu, Benze
    Wu, Yirui
    Wan, Shaohua
    2021 IEEE 19TH INTERNATIONAL CONFERENCE ON EMBEDDED AND UBIQUITOUS COMPUTING (EUC 2021), 2021, : 201 - 207
  • [27] Diversified Contrastive Learning For Few-Shot Classification
    Lu, Guangtong
    Li, Fanzhang
    ARTIFICIAL NEURAL NETWORKS AND MACHINE LEARNING, ICANN 2023, PT I, 2023, 14254 : 147 - 158
  • [28] Few-Shot Learning for Medical Image Classification
    Cai, Aihua
    Hu, Wenxin
    Zheng, Jun
    ARTIFICIAL NEURAL NETWORKS AND MACHINE LEARNING, ICANN 2020, PT I, 2020, 12396 : 441 - 452
  • [29] Few-shot relation classification by context attention-based prototypical networks with BERT
    Bei Hui
    Liang Liu
    Jia Chen
    Xue Zhou
    Yuhui Nian
    EURASIP Journal on Wireless Communications and Networking, 2020
  • [30] Few-Shot Malware Classification via Attention-Based Transductive Learning Network
    Deng, Liting
    Yu, Chengli
    Wen, Hui
    Xin, Mingfeng
    Sun, Yue
    Sun, Limin
    Zhu, Hongsong
    MOBILE NETWORKS & APPLICATIONS, 2024, : 1690 - 1704