Multi-label category enhancement fusion distillation based on variational estimation

被引:1
作者
Li, Li [1 ]
Xu, Jingzhou [1 ]
机构
[1] Beijing Univ Posts & Telecommun, Beijing Lab Adv Informat Networks, Beijing Key Lab Network Syst Architecture & Conve, Beijing 100876, Peoples R China
关键词
Knowledge distillation; Multi-label; Variational estimation; Category enhancement;
D O I
10.1016/j.knosys.2024.112092
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
One of the pivotal challenges in multi -label image classification lies in the fact that each image is often tagged with multiple semantic labels, without the aggregate prediction probabilities being bound to unity. This aspect complicates the straightforward application of conventional single -label image classification algorithms to multi -label contexts. To tackle this challenge, this paper introduces a variational estimation -based multilabel category enhancement fusion distillation technique. The devised loss function focuses on maximizing the biochemical mutual information, thereby enhancing category recognition capabilities. The goal is to adeptly extract and capitalize on the pivotal features of multi -label image scores and structural information, thus elevating the accuracy and efficiency of classification endeavors. This paper not only furnishes a thorough exposition of the issues tackled and the comprehensive architecture of the proposed algorithm but also delineates its operational principles and design rationale via an exhaustive analysis of each critical step within the algorithm. Through an array of experiments across diverse network architectures and datasets, coupled with comparative analyses against extant models and empirical validations, this paper unequivocally validates the efficacy of the suggested algorithm and markedly augments the performance of multi -label classification tasks.
引用
收藏
页数:12
相关论文
共 48 条
  • [1] Aguilar G, 2020, AAAI CONF ARTIF INTE, V34, P7350
  • [2] Variational Information Distillation for Knowledge Transfer
    Ahn, Sungsoo
    Hu, Shell Xu
    Damianou, Andreas
    Lawrence, Neil D.
    Dai, Zhenwen
    [J]. 2019 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2019), 2019, : 9155 - 9163
  • [3] Training Behavior of Sparse Neural Network Topologies
    Alford, Simon
    Robinett, Ryan
    Milechin, Lauren
    Kepner, Jeremy
    [J]. 2019 IEEE HIGH PERFORMANCE EXTREME COMPUTING CONFERENCE (HPEC), 2019,
  • [4] Ba LJ, 2014, ADV NEUR IN, V27
  • [5] Barber D, 2004, ADV NEUR IN, V16, P201
  • [6] Bucilua C., 2006, P 12 ACM SIGKDD INT, P535, DOI DOI 10.1145/1150402.1150464
  • [7] Changyong S, 2019, Arxiv, DOI arXiv:1904.05100
  • [8] Distilling Knowledge via Knowledge Review
    Chen, Pengguang
    Liu, Shu
    Zhao, Hengshuang
    Jia, Jiaya
    [J]. 2021 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR 2021, 2021, : 5006 - 5015
  • [9] Knowledge-Guided Multi-Label Few-Shot Learning for General Image Recognition
    Chen, Tianshui
    Lin, Liang
    Chen, Riquan
    Hui, Xiaolu
    Wu, Hefeng
    [J]. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2022, 44 (03) : 1371 - 1384
  • [10] The PASCAL Visual Object Classes Challenge: A Retrospective
    Everingham, Mark
    Eslami, S. M. Ali
    Van Gool, Luc
    Williams, Christopher K. I.
    Winn, John
    Zisserman, Andrew
    [J]. INTERNATIONAL JOURNAL OF COMPUTER VISION, 2015, 111 (01) : 98 - 136