View in EDS

Class-incremental continual graph learning with adversarial graph condensation.

Saved in:
Bibliographic Details
Title: Class-incremental continual graph learning with adversarial graph condensation.
Authors: Yuan, QiAo1,2 (AUTHOR) qiao.yuan17@student.xjtlu.edu.cn, Zhu, Boxuan1,2 (AUTHOR) boxuan.zhu@liverpool.ac.uk, Guan, Sheng-Uei1,2 (AUTHOR) Steven.Guan@xjtlu.edu.cn, Man, Ka Lok2 (AUTHOR) Ka.Man@xjtlu.edu.cn, Wong, Prudence1 (AUTHOR) pwong@liverpool.ac.uk
Source: Neurocomputing. Jun2026, Vol. 682, pN.PAG-N.PAG. 1p.
Subjects: Graph neural networks, Statistical learning, Graph connectivity
Abstract: Continual Graph Learning (CGL) enables models to incrementally learn from streaming graph-structured data without forgetting previously acquired knowledge. Experience replay is a common solution that reuses a subset of past samples during training. However, it may lead to information loss and privacy risks. Generative replay addresses these concerns by synthesizing informative subgraphs for rehearsal. Existing generative replay approaches often rely on graph condensation via distribution matching, which faces two key challenges: (1) the use of random feature encodings may fail to capture the characteristic kernel of the discrepancy metric, weakening distribution alignment; and (2) matching over a fixed small subgraph cannot guarantee low risk on previous tasks, as indicated by domain adaptation theory. To overcome these limitations, we propose an Adversarial Condensation based Generative Replay (ACGR) framework. It reformulates graph condensation as a min-max optimization problem to achieve better distribution matching. Moreover, instead of learning a single subgraph, we learn its distribution, allowing for the generation of multiple samples and improved empirical risk minimization. Experiments on three benchmark datasets demonstrate that ACGR outperforms existing methods in both accuracy and stability. [ABSTRACT FROM AUTHOR]
Copyright of Neurocomputing is the property of Elsevier B.V. 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
FullText Text:
  Availability: 0
Header DbId: egs
DbLabel: Engineering Source
An: 193091240
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Class-incremental continual graph learning with adversarial graph condensation.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Yuan%2C+QiAo%22">Yuan, QiAo</searchLink><relatesTo>1,2</relatesTo> (AUTHOR)<i> qiao.yuan17@student.xjtlu.edu.cn</i><br /><searchLink fieldCode="AR" term="%22Zhu%2C+Boxuan%22">Zhu, Boxuan</searchLink><relatesTo>1,2</relatesTo> (AUTHOR)<i> boxuan.zhu@liverpool.ac.uk</i><br /><searchLink fieldCode="AR" term="%22Guan%2C+Sheng-Uei%22">Guan, Sheng-Uei</searchLink><relatesTo>1,2</relatesTo> (AUTHOR)<i> Steven.Guan@xjtlu.edu.cn</i><br /><searchLink fieldCode="AR" term="%22Man%2C+Ka+Lok%22">Man, Ka Lok</searchLink><relatesTo>2</relatesTo> (AUTHOR)<i> Ka.Man@xjtlu.edu.cn</i><br /><searchLink fieldCode="AR" term="%22Wong%2C+Prudence%22">Wong, Prudence</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> pwong@liverpool.ac.uk</i>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="JN" term="%22Neurocomputing%22">Neurocomputing</searchLink>. Jun2026, Vol. 682, pN.PAG-N.PAG. 1p.
– Name: Subject
  Label: Subjects
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Graph+neural+networks%22">Graph neural networks</searchLink><br /><searchLink fieldCode="DE" term="%22Statistical+learning%22">Statistical learning</searchLink><br /><searchLink fieldCode="DE" term="%22Graph+connectivity%22">Graph connectivity</searchLink>
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: Continual Graph Learning (CGL) enables models to incrementally learn from streaming graph-structured data without forgetting previously acquired knowledge. Experience replay is a common solution that reuses a subset of past samples during training. However, it may lead to information loss and privacy risks. Generative replay addresses these concerns by synthesizing informative subgraphs for rehearsal. Existing generative replay approaches often rely on graph condensation via distribution matching, which faces two key challenges: (1) the use of random feature encodings may fail to capture the characteristic kernel of the discrepancy metric, weakening distribution alignment; and (2) matching over a fixed small subgraph cannot guarantee low risk on previous tasks, as indicated by domain adaptation theory. To overcome these limitations, we propose an Adversarial Condensation based Generative Replay (ACGR) framework. It reformulates graph condensation as a min-max optimization problem to achieve better distribution matching. Moreover, instead of learning a single subgraph, we learn its distribution, allowing for the generation of multiple samples and improved empirical risk minimization. Experiments on three benchmark datasets demonstrate that ACGR outperforms existing methods in both accuracy and stability. [ABSTRACT FROM AUTHOR]
– Name: AbstractSuppliedCopyright
  Label:
  Group: Ab
  Data: <i>Copyright of Neurocomputing is the property of Elsevier B.V. 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=193091240
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1016/j.neucom.2026.133420
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 1
        StartPage: N.PAG
    Subjects:
      – SubjectFull: Graph neural networks
        Type: general
      – SubjectFull: Statistical learning
        Type: general
      – SubjectFull: Graph connectivity
        Type: general
    Titles:
      – TitleFull: Class-incremental continual graph learning with adversarial graph condensation.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Yuan, QiAo
      – PersonEntity:
          Name:
            NameFull: Zhu, Boxuan
      – PersonEntity:
          Name:
            NameFull: Guan, Sheng-Uei
      – PersonEntity:
          Name:
            NameFull: Man, Ka Lok
      – PersonEntity:
          Name:
            NameFull: Wong, Prudence
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 14
              M: 06
              Text: Jun2026
              Type: published
              Y: 2026
          Identifiers:
            – Type: issn-print
              Value: 09252312
          Numbering:
            – Type: volume
              Value: 682
          Titles:
            – TitleFull: Neurocomputing
              Type: main
ResultId 1