OntoZSL: Ontology-enhanced Zero-shot Learning

被引:34
|
作者
Geng, Yuxia [1 ]
Chen, Jiaoyan [2 ]
Chen, Zhuo [1 ]
Pan, Jeff Z. [3 ]
Ye, Zhiquan [4 ]
Yuan, Zonggang [5 ]
Jia, Yantao [6 ]
Chen, Huajun [7 ,8 ]
机构
[1] Zhejiang Univ, Hangzhou, Peoples R China
[2] Univ Oxford, Oxford, England
[3] Univ Edinburgh, Edinburgh, Midlothian, Scotland
[4] Zhejiang Univ, Coll Comp Sci, Hangzhou, Peoples R China
[5] Huawei Technol Co Ltd, NAIE CTO Off, Nanjing, Peoples R China
[6] Huawei Technol Co Ltd, Poisson Lab, Beijing, Peoples R China
[7] Zhejiang Univ, Coll Comp Sci, AZFT Knowledge Engine Lab, Hangzhou, Peoples R China
[8] Zhejiang Univ, HIC, AZFT Knowledge Engine Lab, Hangzhou, Peoples R China
关键词
Zero-shot Learning; Ontology; Generative Adversarial Networks; Image Classification; Knowledge Graph Completion; CLASSIFICATION;
D O I
10.1145/3442381.3450042
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Zero-shot Learning (ZSL), which aims to predict for those classes that have never appeared in the training data, has arisen hot research interests. The key of implementing ZSL is to leverage the prior knowledge of classes which builds the semantic relationship between classes and enables the transfer of the learned models (e.g., features) from training classes (i.e., seen classes) to unseen classes. However, the priors adopted by the existing methods are relatively limited with incomplete semantics. In this paper, we explore richer and more competitive prior knowledge to model the inter-class relationship for ZSL via ontology-based knowledge representation and semantic embedding. Meanwhile, to address the data imbalance between seen classes and unseen classes, we developed a generative ZSL framework with Generative Adversarial Networks (GANs). Our main findings include: (i) an ontology-enhanced ZSL framework that can be applied to different domains, such as image classification (IMGC) and knowledge graph completion (KGC); (ii) a comprehensive evaluation with multiple zero-shot datasets from different domains, where our method often achieves better performance than the state-of-the-art models. In particular, on four representative ZSL baselines of IMGC, the ontology-based class semantics outperform the previous priors e.g., the word embeddings of classes by an average of 12.4 accuracy points in the standard ZSL across two example datasets (see Figure 4).
引用
收藏
页码:3325 / 3336
页数:12
相关论文
共 50 条
  • [1] Disentangled Ontology Embedding for Zero-shot Learning
    Geng, Yuxia
    Chen, Jiaoyan
    Zhang, Wen
    Xu, Yajing
    Chen, Zhuo
    Pan, Jeff Z.
    Huang, Yufeng
    Xiong, Feiyu
    Chen, Huajun
    PROCEEDINGS OF THE 28TH ACM SIGKDD CONFERENCE ON KNOWLEDGE DISCOVERY AND DATA MINING, KDD 2022, 2022, : 443 - 453
  • [2] Combining ontology and reinforcement learning for zero-shot classification
    Liu, Bin
    Yao, Li
    Ding, Zheyuan
    Xu, Junyi
    Wu, Junfeng
    KNOWLEDGE-BASED SYSTEMS, 2018, 144 : 42 - 50
  • [3] Ontology-enhanced Prompt-tuning for Few-shot Learning
    Ye, Hongbin
    Zhang, Ningyu
    Deng, Shumin
    Chen, Xiang
    Chen, Hui
    Xiong, Feiyu
    Chen, Xi
    Chen, Huajun
    PROCEEDINGS OF THE ACM WEB CONFERENCE 2022 (WWW'22), 2022, : 778 - 787
  • [4] Ontology-guided Semantic Composition for Zero-Shot Learning
    Chen, Jiaoyan
    Lecue, Freddy
    Geng, Yuxia
    Pan, Jeff Z.
    Chen, Huajun
    KR2020: PROCEEDINGS OF THE 17TH INTERNATIONAL CONFERENCE ON PRINCIPLES OF KNOWLEDGE REPRESENTATION AND REASONING, 2020, : 850 - 854
  • [5] ENCYCLOPEDIA ENHANCED SEMANTIC EMBEDDING FOR ZERO-SHOT LEARNING
    Jia, Zhen
    Zhang, Junge
    Huang, Kaiqi
    Tan, Tieniu
    2017 24TH IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING (ICIP), 2017, : 1287 - 1291
  • [6] Ordinal Zero-Shot Learning
    Huo, Zengwei
    Geng, Xin
    PROCEEDINGS OF THE TWENTY-SIXTH INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE, 2017, : 1916 - 1922
  • [7] Zero-Shot Kernel Learning
    Zhang, Hongguang
    Koniusz, Piotr
    2018 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2018, : 7670 - 7679
  • [8] Zero-shot causal learning
    Nilforoshan, Hamed
    Moor, Michael
    Roohani, Yusuf
    Chen, Yining
    Surina, Anja
    Yasunaga, Michihiro
    Oblak, Sara
    Leskovec, Jure
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 36 (NEURIPS 2023), 2023,
  • [9] Zero-shot Metric Learning
    Xu, Xinyi
    Cao, Huanhuan
    Yang, Yanhua
    Yang, Erkun
    Deng, Cheng
    PROCEEDINGS OF THE TWENTY-EIGHTH INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE, 2019, : 3996 - 4002
  • [10] Active Zero-Shot Learning
    Xie, Sihong
    Wang, Shaoxiong
    Yu, Philip S.
    CIKM'16: PROCEEDINGS OF THE 2016 ACM CONFERENCE ON INFORMATION AND KNOWLEDGE MANAGEMENT, 2016, : 1889 - 1892