View in EDS HTML Full Text PDF Full Text

Regularized Stein Variational Gradient Flow.

Saved in:
Bibliographic Details
Title: Regularized Stein Variational Gradient Flow.
Authors: He, Ye1 (AUTHOR) yhe367@gatech.edu, Balasubramanian, Krishnakumar2 (AUTHOR) kbala@ucdavis.edu, Sriperumbudur, Bharath K.3 (AUTHOR) bks18@psu.edu, Lu, Jianfeng4 (AUTHOR) jianfeng@math.duke.edu
Source: Foundations of Computational Mathematics. Aug2025, Vol. 25 Issue 4, p1199-1257. 59p.
Subjects: Particle methods (Numerical analysis), Sampling (Process), Global asymptotic stability, Quantitative research
Abstract: The stein variational gradient descent (SVGD) algorithm is a deterministic particle method for sampling. However, a mean-field analysis reveals that the gradient flow corresponding to the SVGD algorithm (i.e., the Stein Variational Gradient Flow) only provides a constant-order approximation to the Wasserstein gradient flow corresponding to the KL-divergence minimization. In this work, we propose the Regularized Stein Variational Gradient Flow, which interpolates between the Stein Variational Gradient Flow and the Wasserstein gradient flow. We establish various theoretical properties of the Regularized Stein Variational Gradient Flow (and its time-discretization) including convergence to equilibrium, existence and uniqueness of weak solutions, and stability of the solutions. We provide preliminary numerical evidence of the improved performance offered by the regularization. [ABSTRACT FROM AUTHOR]
Copyright of Foundations of Computational Mathematics 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: 187625243
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Regularized Stein Variational Gradient Flow.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22He%2C+Ye%22">He, Ye</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> yhe367@gatech.edu</i><br /><searchLink fieldCode="AR" term="%22Balasubramanian%2C+Krishnakumar%22">Balasubramanian, Krishnakumar</searchLink><relatesTo>2</relatesTo> (AUTHOR)<i> kbala@ucdavis.edu</i><br /><searchLink fieldCode="AR" term="%22Sriperumbudur%2C+Bharath+K%2E%22">Sriperumbudur, Bharath K.</searchLink><relatesTo>3</relatesTo> (AUTHOR)<i> bks18@psu.edu</i><br /><searchLink fieldCode="AR" term="%22Lu%2C+Jianfeng%22">Lu, Jianfeng</searchLink><relatesTo>4</relatesTo> (AUTHOR)<i> jianfeng@math.duke.edu</i>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="JN" term="%22Foundations+of+Computational+Mathematics%22">Foundations of Computational Mathematics</searchLink>. Aug2025, Vol. 25 Issue 4, p1199-1257. 59p.
– Name: Subject
  Label: Subjects
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Particle+methods+%28Numerical+analysis%29%22">Particle methods (Numerical analysis)</searchLink><br /><searchLink fieldCode="DE" term="%22Sampling+%28Process%29%22">Sampling (Process)</searchLink><br /><searchLink fieldCode="DE" term="%22Global+asymptotic+stability%22">Global asymptotic stability</searchLink><br /><searchLink fieldCode="DE" term="%22Quantitative+research%22">Quantitative research</searchLink>
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: The stein variational gradient descent (SVGD) algorithm is a deterministic particle method for sampling. However, a mean-field analysis reveals that the gradient flow corresponding to the SVGD algorithm (i.e., the Stein Variational Gradient Flow) only provides a constant-order approximation to the Wasserstein gradient flow corresponding to the KL-divergence minimization. In this work, we propose the Regularized Stein Variational Gradient Flow, which interpolates between the Stein Variational Gradient Flow and the Wasserstein gradient flow. We establish various theoretical properties of the Regularized Stein Variational Gradient Flow (and its time-discretization) including convergence to equilibrium, existence and uniqueness of weak solutions, and stability of the solutions. We provide preliminary numerical evidence of the improved performance offered by the regularization. [ABSTRACT FROM AUTHOR]
– Name: AbstractSuppliedCopyright
  Label:
  Group: Ab
  Data: <i>Copyright of Foundations of Computational Mathematics 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=187625243
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1007/s10208-024-09663-w
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 59
        StartPage: 1199
    Subjects:
      – SubjectFull: Particle methods (Numerical analysis)
        Type: general
      – SubjectFull: Sampling (Process)
        Type: general
      – SubjectFull: Global asymptotic stability
        Type: general
      – SubjectFull: Quantitative research
        Type: general
    Titles:
      – TitleFull: Regularized Stein Variational Gradient Flow.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: He, Ye
      – PersonEntity:
          Name:
            NameFull: Balasubramanian, Krishnakumar
      – PersonEntity:
          Name:
            NameFull: Sriperumbudur, Bharath K.
      – PersonEntity:
          Name:
            NameFull: Lu, Jianfeng
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 08
              Text: Aug2025
              Type: published
              Y: 2025
          Identifiers:
            – Type: issn-print
              Value: 16153375
          Numbering:
            – Type: volume
              Value: 25
            – Type: issue
              Value: 4
          Titles:
            – TitleFull: Foundations of Computational Mathematics
              Type: main
ResultId 1