Zero-Shot Feature Selection via Transferring Supervised Knowledge

被引:0
|
作者
Wang, Zheng [1 ]
Wang, Qiao [2 ]
Zhao, Tingzhang [1 ]
Wang, Chaokun [2 ]
Ye, Xiaojun [2 ]
机构
[1] Univ Sci & Technol Beijing, Dept Comp Sci & Technol, Beijing, Peoples R China
[2] Tsinghua Univ, Sch Software, Beijing, Peoples R China
基金
中国国家自然科学基金; 国家重点研发计划; 中国博士后科学基金;
关键词
Data Mining; Feature Selection;
D O I
10.4018/IJDWM.2021040101
中图分类号
TP31 [计算机软件];
学科分类号
081202 ; 0835 ;
摘要
Feature selection, an effective technique for dimensionality reduction, plays an important role in many machine learning systems. Supervised knowledge can significantly improve the performance. However, faced with the rapid growth of newly emerging concepts, existing supervised methods might easily suffer from the scarcity and validity of labeled data for training. In this paper, the authors study the problem of zero-shot feature selection (i.e., building a feature selection model that generalizes well to "unseen" concepts with limited training data of "seen" concepts). Specifically, they adopt class-semantic descriptions (i.e., attributes) as supervision for feature selection, so as to utilize the supervised knowledge transferred from the seen concepts. For more reliable discriminative features, they further propose the center-characteristic loss which encourages the selected features to capture the central characteristics of seen concepts. Extensive experiments conducted on various real-world datasets demonstrate the effectiveness of the method.
引用
收藏
页码:1 / 20
页数:20
相关论文
共 50 条
  • [31] Semantics-Conditioned Generative Zero-Shot Learning via Feature Refinement
    Chen, Shiming
    Hong, Ziming
    You, Xinge
    Shao, Ling
    INTERNATIONAL JOURNAL OF COMPUTER VISION, 2025,
  • [32] Speech Enhancement with Zero-Shot Model Selection
    Zezario, Ryandhimas E.
    Fuh, Chiou-Shann
    Wang, Hsin-Min
    Tsao, Yu
    29TH EUROPEAN SIGNAL PROCESSING CONFERENCE (EUSIPCO 2021), 2021, : 491 - 495
  • [33] Model Selection for Generalized Zero-Shot Learning
    Zhang, Hongguang
    Koniusz, Piotr
    COMPUTER VISION - ECCV 2018 WORKSHOPS, PT II, 2019, 11130 : 198 - 204
  • [34] Relational Knowledge Transfer for Zero-Shot Learning
    Wang, Donghui
    Li, Yanan
    Lin, Yuetan
    Zhuang, Yueting
    THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE, 2016, : 2145 - 2151
  • [35] Zero-Shot Knowledge Distillation in Deep Networks
    Nayak, Gaurav Kumar
    Mopuri, Konda Reddy
    Shaj, Vaisakh
    Babu, R. Venkatesh
    Chakraborty, Anirban
    INTERNATIONAL CONFERENCE ON MACHINE LEARNING, VOL 97, 2019, 97
  • [36] Zero-shot Voice Conversion via Self-supervised Prosody Representation Learning
    Wang, Shijun
    Borth, Damian
    2022 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS (IJCNN), 2022,
  • [37] Zero-Shot Entity Typing in Knowledge Graphs
    Zhou, Shengye
    Wang, Zhe
    Wang, Kewen
    Zhuang, Zhiqiang
    DATABASE SYSTEMS FOR ADVANCED APPLICATIONS. DASFAA 2023 INTERNATIONAL WORKSHOPS, BDMS 2023, BDQM 2023, GDMA 2023, BUNDLERS 2023, 2023, 13922 : 238 - 250
  • [38] Zero-Shot Learning Based on Knowledge Sharing
    Zeng, Ting
    Xiang, Hongxin
    Xie, Cheng
    Yang, Yun
    Liu, Qing
    PROCEEDINGS OF THE 2021 IEEE 24TH INTERNATIONAL CONFERENCE ON COMPUTER SUPPORTED COOPERATIVE WORK IN DESIGN (CSCWD), 2021, : 643 - 648
  • [39] Hand and Pose-Based Feature Selection for Zero-Shot Sign Language Recognition
    Ozcan, Giray Sercan
    Bilge, Yunus Can
    Sumer, Emre
    IEEE ACCESS, 2024, 12 : 107757 - 107768
  • [40] Zero-Shot Text Normalization via Cross-Lingual Knowledge Distillation
    Wang, Linqin
    Huang, Xiang
    Yu, Zhengtao
    Peng, Hao
    Gao, Shengxiang
    Mao, Cunli
    Huang, Yuxin
    Dong, Ling
    Yu, Philip S.
    IEEE-ACM TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSING, 2024, 32 : 4631 - 4646