View in EDS HTML Full Text PDF Full Text

Transfer and share: semi-supervised learning from long-tailed data.

Saved in:
Bibliographic Details
Title: Transfer and share: semi-supervised learning from long-tailed data.
Authors: Wei, Tong1 (AUTHOR) weit@seu.edu.cn, Liu, Qian-Yu2 (AUTHOR), Shi, Jiang-Xin2 (AUTHOR), Tu, Wei-Wei3 (AUTHOR), Guo, Lan-Zhe2 (AUTHOR) guolz@lamda.nju.edu.cn
Source: Machine Learning. Apr2024, Vol. 113 Issue 4, p1725-1742. 18p.
Subjects: Supervised learning, Secure Sockets Layer (Computer network protocol), Logits, Sharing
Abstract: Long-Tailed Semi-Supervised Learning (LTSSL) aims to learn from class-imbalanced data where only a few samples are annotated. Existing solutions typically require substantial cost to solve complex optimization problems, or class-balanced undersampling which can result in information loss. In this paper, we present the TRAS (TRAnsfer and Share) to effectively utilize long-tailed semi-supervised data. TRAS transforms the imbalanced pseudo-label distribution of a traditional SSL model via a delicate function to enhance the supervisory signals for minority classes. It then transfers the distribution to a target model such that the minority class will receive significant attention. Interestingly, TRAS shows that more balanced pseudo-label distribution can substantially benefit minority-class training, instead of seeking to generate accurate pseudo-labels as in previous works. To simplify the approach, TRAS merges the training of the traditional SSL model and the target model into a single procedure by sharing the feature extractor, where both classifiers help improve the representation learning. According to extensive experiments, TRAS delivers much higher accuracy than state-of-the-art methods in the entire set of classes as well as minority classes. Code for TRAS is available at https://github.com/Stomach-ache/TRAS. [ABSTRACT FROM AUTHOR]
Copyright of Machine Learning is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database: Engineering Source
Full text is not displayed to guests.
FullText Links:
  – Type: pdflink
Text:
  Availability: 1
Header DbId: egs
DbLabel: Engineering Source
An: 176338103
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Transfer and share: semi-supervised learning from long-tailed data.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Wei%2C+Tong%22">Wei, Tong</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> weit@seu.edu.cn</i><br /><searchLink fieldCode="AR" term="%22Liu%2C+Qian-Yu%22">Liu, Qian-Yu</searchLink><relatesTo>2</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Shi%2C+Jiang-Xin%22">Shi, Jiang-Xin</searchLink><relatesTo>2</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Tu%2C+Wei-Wei%22">Tu, Wei-Wei</searchLink><relatesTo>3</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Guo%2C+Lan-Zhe%22">Guo, Lan-Zhe</searchLink><relatesTo>2</relatesTo> (AUTHOR)<i> guolz@lamda.nju.edu.cn</i>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="JN" term="%22Machine+Learning%22">Machine Learning</searchLink>. Apr2024, Vol. 113 Issue 4, p1725-1742. 18p.
– Name: Subject
  Label: Subjects
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Supervised+learning%22">Supervised learning</searchLink><br /><searchLink fieldCode="DE" term="%22Secure+Sockets+Layer+%28Computer+network+protocol%29%22">Secure Sockets Layer (Computer network protocol)</searchLink><br /><searchLink fieldCode="DE" term="%22Logits%22">Logits</searchLink><br /><searchLink fieldCode="DE" term="%22Sharing%22">Sharing</searchLink>
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: Long-Tailed Semi-Supervised Learning (LTSSL) aims to learn from class-imbalanced data where only a few samples are annotated. Existing solutions typically require substantial cost to solve complex optimization problems, or class-balanced undersampling which can result in information loss. In this paper, we present the TRAS (TRAnsfer and Share) to effectively utilize long-tailed semi-supervised data. TRAS transforms the imbalanced pseudo-label distribution of a traditional SSL model via a delicate function to enhance the supervisory signals for minority classes. It then transfers the distribution to a target model such that the minority class will receive significant attention. Interestingly, TRAS shows that more balanced pseudo-label distribution can substantially benefit minority-class training, instead of seeking to generate accurate pseudo-labels as in previous works. To simplify the approach, TRAS merges the training of the traditional SSL model and the target model into a single procedure by sharing the feature extractor, where both classifiers help improve the representation learning. According to extensive experiments, TRAS delivers much higher accuracy than state-of-the-art methods in the entire set of classes as well as minority classes. Code for TRAS is available at https://github.com/Stomach-ache/TRAS. [ABSTRACT FROM AUTHOR]
– Name: AbstractSuppliedCopyright
  Label:
  Group: Ab
  Data: <i>Copyright of Machine Learning is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract.</i> (Copyright applies to all Abstracts.)
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=egs&AN=176338103
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1007/s10994-022-06247-z
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 18
        StartPage: 1725
    Subjects:
      – SubjectFull: Supervised learning
        Type: general
      – SubjectFull: Secure Sockets Layer (Computer network protocol)
        Type: general
      – SubjectFull: Logits
        Type: general
      – SubjectFull: Sharing
        Type: general
    Titles:
      – TitleFull: Transfer and share: semi-supervised learning from long-tailed data.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Wei, Tong
      – PersonEntity:
          Name:
            NameFull: Liu, Qian-Yu
      – PersonEntity:
          Name:
            NameFull: Shi, Jiang-Xin
      – PersonEntity:
          Name:
            NameFull: Tu, Wei-Wei
      – PersonEntity:
          Name:
            NameFull: Guo, Lan-Zhe
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 04
              Text: Apr2024
              Type: published
              Y: 2024
          Identifiers:
            – Type: issn-print
              Value: 08856125
          Numbering:
            – Type: volume
              Value: 113
            – Type: issue
              Value: 4
          Titles:
            – TitleFull: Machine Learning
              Type: main
ResultId 1