离异图引导消歧的偏标记学习
DOI:
作者:
作者单位:

重庆邮电大学

作者简介:

通讯作者:

中图分类号:

TP181

基金项目:

国家重点研发计划项目 (2018YFC0832102),重庆市高校与中国科学院所属院所重点合作项目 (No.HZ2021008),重庆市自然科学基金 (cstc2021jcyj-msxmX0849)


Introducing Outlier Graph to Disambiguate for Partial Label Learning
Author:
Affiliation:

Chongqing University of Posts and Telecommunications

Fund Project:

the National Key R & D Plan Project (2018yfc0832102), the Key Cooperation Project between Chongqing Universities and Institutes Affiliated to the Chinese Academy of Sciences (No. hz2021008), the Chongqing Natural Science Foundation (cstc2021jcyj-msxmx0849)

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    偏标记学习是一种弱监督学习框架,它试图从样本的多个候选标签中选择唯一正确的标签.消歧是偏标记学习中的一种重要手段,主要通过算法判别潜在的真实标签.目前,研究人员普遍采用单一的特征空间或者标签空间进行消歧,容易导致算法受到不准确先验知识的引导而陷入鞍点.针对消歧过程中特征相似样本易受到异类样本影响从而影响消歧效果这一问题,本文定义了样本离异点和离异图;在此基础上,提出了一种离异图引导消歧的偏标记学习方法.该方法利用标签空间的差异构建离异图,可以有效结合特征空间的相似性和标签空间的差异性,以降低离异点为消歧过程带来的潜在风险.实验结果表明,与PLKNN、IPAL、SURE、PL-AGGD、SDIM、PL-BLC、PRODEN等方法相比较,文中提出的算法在偏标签学习方法中表现更好,且取得了良好的消歧效果.

    Abstract:

    Partial label learning is a weakly supervised learning framework, which attempts to select the only correct label from multiple candidate labels of the sample. Disambiguation is an important means in partial tag learning, which mainly uses algorithms to identify potential real tags. At present, researchers generally use a single feature space or label space for disambiguation, which is easy to lead the algorithm to fall into saddle point under the guidance of inaccurate prior knowledge. In order to solve the problem that similar samples are easy to be affected by abnormal samples in the process of disambiguation, sample outlier and outlier graph are defined in this paper; On this basis, a partial marker learning method for dissociation graph guided disambiguation is proposed. The algorithm uses the difference of label space to construct the divorced point map, which can effectively combine the prior knowledge of feature space and label space, and restrict each other to resist the potential high risk brought by divorced samples to the disambiguation process. Experimental results show that compared with PLKNN, IPAL, SURE, PL-AGGD, SDIM, PL-BLC and PRODEN, the proposed algorithm performs better in partial label learning method and achieves good disambiguation effect.

    参考文献
    相似文献
    引证文献
引用本文
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2021-10-27
  • 最后修改日期:2022-03-02
  • 录用日期:2022-03-15
  • 在线发布日期:
  • 出版日期: