View in EDS PDF Full Text

Optimal Re-Materialization Strategies for Heterogeneous Chains: How to Train Deep Neural Networks with Limited Memory.

Saved in:
Bibliographic Details
Title: Optimal Re-Materialization Strategies for Heterogeneous Chains: How to Train Deep Neural Networks with Limited Memory.
Authors: BEAUMONT, OLIVIER1, olivier.beaumont@inria.fr, EYRAUD-DUBOIS, LIONEL1, lionel.eyraud-dubois@inria.fr, HERRMANN, JULIEN2, julien.herrmann@irit.fr, JOLY, ALEXIS3, alexis.joly@inria.fr, SHILOVA, ALENA4, Alena.Shilova@inria.fr
Source: ACM Transactions on Mathematical Software; Jun2024, Vol. 50 Issue 2, p1-38, 38p
Database: Applied Science & Technology Source
FullText Links:
  – Type: pdflink
Text:
  Availability: 0
Header DbId: aci
DbLabel: Applied Science & Technology Source
An: 178145971
AccessLevel: 2
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Optimal Re-Materialization Strategies for Heterogeneous Chains: How to Train Deep Neural Networks with Limited Memory.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AU" term="%22BEAUMONT%2C+OLIVIER%22">BEAUMONT, OLIVIER</searchLink><relatesTo>1</relatesTo>, <i>olivier.beaumont@inria.fr</i><br /><searchLink fieldCode="AU" term="%22EYRAUD-DUBOIS%2C+LIONEL%22">EYRAUD-DUBOIS, LIONEL</searchLink><relatesTo>1</relatesTo>, <i>lionel.eyraud-dubois@inria.fr</i><br /><searchLink fieldCode="AU" term="%22HERRMANN%2C+JULIEN%22">HERRMANN, JULIEN</searchLink><relatesTo>2</relatesTo>, <i>julien.herrmann@irit.fr</i><br /><searchLink fieldCode="AU" term="%22JOLY%2C+ALEXIS%22">JOLY, ALEXIS</searchLink><relatesTo>3</relatesTo>, <i>alexis.joly@inria.fr</i><br /><searchLink fieldCode="AU" term="%22SHILOVA%2C+ALENA%22">SHILOVA, ALENA</searchLink><relatesTo>4</relatesTo>, <i>Alena.Shilova@inria.fr</i>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="JN" term="%22ACM+Transactions+on+Mathematical+Software%22">ACM Transactions on Mathematical Software</searchLink>; Jun2024, Vol. 50 Issue 2, p1-38, 38p
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=aci&AN=178145971
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1145/3648633
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 38
        StartPage: 1
    Titles:
      – TitleFull: Optimal Re-Materialization Strategies for Heterogeneous Chains: How to Train Deep Neural Networks with Limited Memory.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: BEAUMONT, OLIVIER
      – PersonEntity:
          Name:
            NameFull: EYRAUD-DUBOIS, LIONEL
      – PersonEntity:
          Name:
            NameFull: HERRMANN, JULIEN
      – PersonEntity:
          Name:
            NameFull: JOLY, ALEXIS
      – PersonEntity:
          Name:
            NameFull: SHILOVA, ALENA
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 06
              Text: Jun2024
              Type: published
              Y: 2024
          Identifiers:
            – Type: issn-print
              Value: 00983500
          Numbering:
            – Type: volume
              Value: 50
            – Type: issue
              Value: 2
          Titles:
            – TitleFull: ACM Transactions on Mathematical Software
              Type: main
ResultId 1