<?xml version="1.0" encoding="utf-8"?>
<TEI xmlns="http://www.tei-c.org/ns/1.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:hal="http://hal.archives-ouvertes.fr/" xmlns:gml="http://www.opengis.net/gml/3.3/" xmlns:gmlce="http://www.opengis.net/gml/3.3/ce" version="1.1" xsi:schemaLocation="http://www.tei-c.org/ns/1.0 http://api.archives-ouvertes.fr/documents/aofr-sword.xsd">
  <teiHeader>
    <fileDesc>
      <titleStmt>
        <title>HAL TEI export of hal-04457596</title>
      </titleStmt>
      <publicationStmt>
        <distributor>CCSD</distributor>
        <availability status="restricted">
          <licence target="https://creativecommons.org/publicdomain/zero/1.0/">CC0 1.0 - Universal</licence>
        </availability>
        <date when="2026-05-03T13:40:01+02:00"/>
      </publicationStmt>
      <sourceDesc>
        <p part="N">HAL API Platform</p>
      </sourceDesc>
    </fileDesc>
  </teiHeader>
  <text>
    <body>
      <listBibl>
        <biblFull>
          <titleStmt>
            <title xml:lang="en">Distributed Collapsed Gibbs Sampler for Dirichlet Process Mixture Models in Federated Learning</title>
            <author role="aut">
              <persName>
                <forename type="first">Reda</forename>
                <surname>Khoufache</surname>
              </persName>
              <email type="md5">3da83d9de3aed2a6b390321d0c69895d</email>
              <email type="domain">uvsq.fr</email>
              <idno type="idhal" notation="numeric">1352101</idno>
              <idno type="halauthorid" notation="string">3055336-1352101</idno>
              <affiliation ref="#struct-456060"/>
            </author>
            <author role="aut">
              <persName>
                <forename type="first">Mustapha</forename>
                <surname>Lebbah</surname>
              </persName>
              <email type="md5">102c24ce25d4a3d513afe52500f1104d</email>
              <email type="domain">uvsq.fr</email>
              <idno type="idhal" notation="string">mustapha-lebbah</idno>
              <idno type="idhal" notation="numeric">735055</idno>
              <idno type="halauthorid" notation="string">4777-735055</idno>
              <idno type="ORCID">https://orcid.org/0000-0001-7245-6371</idno>
              <idno type="IDREF">https://www.idref.fr/144970759</idno>
              <affiliation ref="#struct-456060"/>
            </author>
            <author role="aut">
              <persName>
                <forename type="first">Hanene</forename>
                <surname>Azzag</surname>
              </persName>
              <email type="md5">554162c83d9070b4f834ee98ae2d6d23</email>
              <email type="domain">univ-paris13.fr</email>
              <idno type="idhal" notation="numeric">1351727</idno>
              <idno type="halauthorid" notation="string">435407-1351727</idno>
              <affiliation ref="#struct-1000994"/>
            </author>
            <author role="aut">
              <persName>
                <forename type="first">Etienne</forename>
                <surname>Goffinet</surname>
              </persName>
              <email type="md5">0c59cc9889c20dae0e1690bcb2e62955</email>
              <email type="domain">tii.ae</email>
              <idno type="idhal" notation="numeric">1352102</idno>
              <idno type="halauthorid" notation="string">2369037-1352102</idno>
              <affiliation ref="#struct-1085162"/>
            </author>
            <author role="aut">
              <persName>
                <forename type="first">Djamel</forename>
                <surname>Bouchaffra</surname>
              </persName>
              <email type="md5">531cdaf4c7755ce42bec4ac1b418c622</email>
              <email type="domain">gmail.com</email>
              <idno type="idhal" notation="numeric">1352103</idno>
              <idno type="halauthorid" notation="string">3054409-1352103</idno>
              <affiliation ref="#struct-42013"/>
            </author>
            <editor role="depositor">
              <persName>
                <forename>Reda</forename>
                <surname>Khoufache</surname>
              </persName>
              <email type="md5">3da83d9de3aed2a6b390321d0c69895d</email>
              <email type="domain">uvsq.fr</email>
            </editor>
          </titleStmt>
          <editionStmt>
            <edition n="v1" type="current">
              <date type="whenSubmitted">2024-02-14 14:29:14</date>
              <date type="whenModified">2025-12-03 14:48:03</date>
              <date type="whenReleased">2024-02-15 08:20:22</date>
              <date type="whenProduced">2024-04-18</date>
              <date type="whenEndEmbargoed">2024-02-14</date>
              <ref type="file" target="https://hal.science/hal-04457596v1/document">
                <date notBefore="2024-02-14"/>
              </ref>
              <ref type="file" subtype="author" n="1" target="https://hal.science/hal-04457596v1/file/paper2.pdf" id="file-4457596-3868901">
                <date notBefore="2024-02-14"/>
              </ref>
            </edition>
            <respStmt>
              <resp>contributor</resp>
              <name key="1328454">
                <persName>
                  <forename>Reda</forename>
                  <surname>Khoufache</surname>
                </persName>
                <email type="md5">3da83d9de3aed2a6b390321d0c69895d</email>
                <email type="domain">uvsq.fr</email>
              </name>
            </respStmt>
          </editionStmt>
          <publicationStmt>
            <distributor>CCSD</distributor>
            <idno type="halId">hal-04457596</idno>
            <idno type="halUri">https://hal.science/hal-04457596</idno>
            <idno type="halBibtex">khoufache:hal-04457596</idno>
            <idno type="halRefHtml">&lt;i&gt;SIAM International Conference on Data Mining (SDM24)&lt;/i&gt;, Apr 2024, Houston, Texas, United States. pp.815 - 823, &lt;a target="_blank" href="https://dx.doi.org/10.1137/1.9781611978032.93"&gt;&amp;#x27E8;10.1137/1.9781611978032.93&amp;#x27E9;&lt;/a&gt;</idno>
            <idno type="halRef">SIAM International Conference on Data Mining (SDM24), Apr 2024, Houston, Texas, United States. pp.815 - 823, &amp;#x27E8;10.1137/1.9781611978032.93&amp;#x27E9;</idno>
            <availability status="restricted">
              <licence target="https://about.hal.science/hal-authorisation-v1/">HAL Authorization<ref corresp="#file-4457596-3868901"/></licence>
            </availability>
          </publicationStmt>
          <seriesStmt>
            <idno type="stamp" n="UNIV-PARIS13">Université Paris-Nord - Paris XIII </idno>
            <idno type="stamp" n="CNRS">CNRS - Centre national de la recherche scientifique</idno>
            <idno type="stamp" n="GRID5000">Grid'5000</idno>
            <idno type="stamp" n="LIPN" corresp="UNIV-PARIS13">Laboratoire d'Informatique de Paris-Nord</idno>
            <idno type="stamp" n="UVSQ">Université de Versailles Saint-Quentin-en-Yvelines</idno>
            <idno type="stamp" n="DAVID">Données et algorithmes pour une ville intelligente et durable</idno>
            <idno type="stamp" n="UNIV-PARIS-SACLAY">Université Paris-Saclay</idno>
            <idno type="stamp" n="SORBONNE-PARIS-NORD">Université Sorbonne Paris Nord</idno>
            <idno type="stamp" n="UVSQ-UPSACLAY" corresp="UNIV-PARIS-SACLAY">UVSQ-UPSACLAY</idno>
            <idno type="stamp" n="GS-HUMANITIES-HERITAGE">Graduate School Humanités - Sciences du Patrimoine</idno>
            <idno type="stamp" n="GS-COMPUTER-SCIENCE">Graduate School Computer Science</idno>
            <idno type="stamp" n="CONFRENCE-NATIONALE-SUR-LES-APPLICATIONS-PRATIQUES-DE-LINTELLIGENCE-ARTIFICIELLE" corresp="CNRS">Conférence Nationale sur les Applications Pratiques de l’Intelligence Artificielle </idno>
            <idno type="stamp" n="ACT-R" corresp="UNIV-PARIS13">Act'R </idno>
            <idno type="stamp" n="SLICES-FR">Publications from users of the SILECS research infrastructure</idno>
          </seriesStmt>
          <notesStmt>
            <note type="audience" n="2">International</note>
            <note type="invited" n="0">No</note>
            <note type="popular" n="0">No</note>
            <note type="peer" n="1">Yes</note>
            <note type="proceedings" n="0">No</note>
          </notesStmt>
          <sourceDesc>
            <biblStruct>
              <analytic>
                <title xml:lang="en">Distributed Collapsed Gibbs Sampler for Dirichlet Process Mixture Models in Federated Learning</title>
                <author role="aut">
                  <persName>
                    <forename type="first">Reda</forename>
                    <surname>Khoufache</surname>
                  </persName>
                  <email type="md5">3da83d9de3aed2a6b390321d0c69895d</email>
                  <email type="domain">uvsq.fr</email>
                  <idno type="idhal" notation="numeric">1352101</idno>
                  <idno type="halauthorid" notation="string">3055336-1352101</idno>
                  <affiliation ref="#struct-456060"/>
                </author>
                <author role="aut">
                  <persName>
                    <forename type="first">Mustapha</forename>
                    <surname>Lebbah</surname>
                  </persName>
                  <email type="md5">102c24ce25d4a3d513afe52500f1104d</email>
                  <email type="domain">uvsq.fr</email>
                  <idno type="idhal" notation="string">mustapha-lebbah</idno>
                  <idno type="idhal" notation="numeric">735055</idno>
                  <idno type="halauthorid" notation="string">4777-735055</idno>
                  <idno type="ORCID">https://orcid.org/0000-0001-7245-6371</idno>
                  <idno type="IDREF">https://www.idref.fr/144970759</idno>
                  <affiliation ref="#struct-456060"/>
                </author>
                <author role="aut">
                  <persName>
                    <forename type="first">Hanene</forename>
                    <surname>Azzag</surname>
                  </persName>
                  <email type="md5">554162c83d9070b4f834ee98ae2d6d23</email>
                  <email type="domain">univ-paris13.fr</email>
                  <idno type="idhal" notation="numeric">1351727</idno>
                  <idno type="halauthorid" notation="string">435407-1351727</idno>
                  <affiliation ref="#struct-1000994"/>
                </author>
                <author role="aut">
                  <persName>
                    <forename type="first">Etienne</forename>
                    <surname>Goffinet</surname>
                  </persName>
                  <email type="md5">0c59cc9889c20dae0e1690bcb2e62955</email>
                  <email type="domain">tii.ae</email>
                  <idno type="idhal" notation="numeric">1352102</idno>
                  <idno type="halauthorid" notation="string">2369037-1352102</idno>
                  <affiliation ref="#struct-1085162"/>
                </author>
                <author role="aut">
                  <persName>
                    <forename type="first">Djamel</forename>
                    <surname>Bouchaffra</surname>
                  </persName>
                  <email type="md5">531cdaf4c7755ce42bec4ac1b418c622</email>
                  <email type="domain">gmail.com</email>
                  <idno type="idhal" notation="numeric">1352103</idno>
                  <idno type="halauthorid" notation="string">3054409-1352103</idno>
                  <affiliation ref="#struct-42013"/>
                </author>
              </analytic>
              <monogr>
                <meeting>
                  <title>SIAM International Conference on Data Mining (SDM24)</title>
                  <date type="start">2024-04-18</date>
                  <date type="end">2024-04-20</date>
                  <settlement>Houston, Texas</settlement>
                  <country key="US">United States</country>
                </meeting>
                <imprint>
                  <biblScope unit="pp">815 - 823</biblScope>
                </imprint>
              </monogr>
              <idno type="doi">10.1137/1.9781611978032.93</idno>
              <ref type="publisher">https://www.siam.org/conferences/cm/conference/sdm24</ref>
            </biblStruct>
          </sourceDesc>
          <profileDesc>
            <langUsage>
              <language ident="en">English</language>
            </langUsage>
            <textClass>
              <keywords scheme="author">
                <term xml:lang="en">Markov Chain Monte Carlo</term>
                <term xml:lang="en">Bayesian non-parametric</term>
                <term xml:lang="en">Federated learning</term>
                <term xml:lang="en">Distributed computing</term>
                <term xml:lang="en">Dirichlet process mixture models</term>
              </keywords>
              <classCode scheme="halDomain" n="stat.ml">Statistics [stat]/Machine Learning [stat.ML]</classCode>
              <classCode scheme="halTypology" n="COMM">Conference papers</classCode>
              <classCode scheme="halOldTypology" n="COMM">Conference papers</classCode>
              <classCode scheme="halTreeTypology" n="COMM">Conference papers</classCode>
            </textClass>
            <abstract xml:lang="en">
              <p>Dirichlet Process Mixture Models (DPMMs) are widely used to address clustering problems. Their main advantage lies in their ability to automatically estimate the number of clusters during the inference process through the Bayesian non-parametric framework. However, the inference becomes considerably slow as the dataset size increases. This paper proposes a new distributed Markov Chain Monte Carlo (MCMC) inference method for DPMMs (DisCGS) using sufficient statistics. Our approach uses the collapsed Gibbs sampler and is specifically designed to work on distributed data across independent and heterogeneous machines, which habilitates its use in horizontal federated learning. Our method achieves highly promising results and notable scalability. For instance, with a dataset of 100K data points, the centralized algorithm requires approximately 12 hours to complete 100 iterations while our approach achieves the same number of iterations in just 3 minutes, reducing the execution time by a factor of 200 without compromising clustering performance. The code source is publicly available at https://github. com/redakhoufache/DisCGS.</p>
            </abstract>
          </profileDesc>
        </biblFull>
      </listBibl>
    </body>
    <back>
      <listOrg type="structures">
        <org type="laboratory" xml:id="struct-456060" status="VALID">
          <idno type="IdRef">19191620X</idno>
          <idno type="RNSR">201622145F</idno>
          <idno type="ROR">https://ror.org/01xta2p78</idno>
          <idno type="Wikidata">Q51785421</idno>
          <orgName>Données et algorithmes pour une ville intelligente et durable - DAVID</orgName>
          <orgName type="acronym">DAVID</orgName>
          <date type="start">2015-07-01</date>
          <desc>
            <address>
              <addrLine>Bâtiment Descartes — Campus des SciencesUniversité de Versailles Saint-Quentin45 avenue des États-Unis78035 Versailles</addrLine>
              <country key="FR"/>
            </address>
            <ref type="url">http://www.david.uvsq.fr/</ref>
          </desc>
          <listRelation>
            <relation name="UR 7431" active="#struct-81173" type="direct"/>
          </listRelation>
        </org>
        <org type="laboratory" xml:id="struct-1000994" status="VALID">
          <idno type="IdRef">191907855</idno>
          <idno type="RNSR">200112433P</idno>
          <idno type="ROR">05g1zjw44</idno>
          <orgName>Laboratoire d'Informatique de Paris-Nord</orgName>
          <orgName type="acronym">LIPN</orgName>
          <date type="start">2020-01-01</date>
          <desc>
            <address>
              <addrLine>Institut Galilée, Université Paris 13, 99 avenue Jean-Baptiste Clément, F-93430, Villetaneuse</addrLine>
              <country key="FR"/>
            </address>
            <ref type="url">http://www-lipn.univ-paris13.fr/</ref>
          </desc>
          <listRelation>
            <relation name="UMR7030" active="#struct-441569" type="direct"/>
            <relation active="#struct-581146" type="direct"/>
          </listRelation>
        </org>
        <org type="institution" xml:id="struct-1085162" status="VALID">
          <idno type="ISNI">0000 0004 8306 7226</idno>
          <idno type="ROR">https://ror.org/001kv2y39</idno>
          <idno type="Wikidata">Q103813517</idno>
          <orgName>Technology Innovation Institute [Abu Dhabi]</orgName>
          <orgName type="acronym">TII</orgName>
          <date type="start">2020-01-01</date>
          <desc>
            <address>
              <addrLine>P.O.Box: 9639, Yas Island Abu Dhabi</addrLine>
              <country key="AE"/>
            </address>
            <ref type="url">https://www.tii.ae/</ref>
          </desc>
        </org>
        <org type="institution" xml:id="struct-42013" status="VALID">
          <orgName>Centre de Développement des Technologies Avancées</orgName>
          <orgName type="acronym">CDTA</orgName>
          <desc>
            <address>
              <addrLine>Cité 20 août 1956 Baba Hassen ALGER</addrLine>
              <country key="DZ"/>
            </address>
            <ref type="url">http://www.cdta.dz/</ref>
          </desc>
        </org>
        <org type="institution" xml:id="struct-81173" status="VALID">
          <idno type="IdRef">03082057X</idno>
          <idno type="ISNI">0000 0001 2323 0229 </idno>
          <idno type="ROR">https://ror.org/03mkjjy25</idno>
          <orgName>Université de Versailles Saint-Quentin-en-Yvelines</orgName>
          <orgName type="acronym">UVSQ</orgName>
          <date type="start">1991-07-22</date>
          <desc>
            <address>
              <addrLine>55 avenue de Paris - 78035 Versailles cedex</addrLine>
              <country key="FR"/>
            </address>
            <ref type="url">http://www.uvsq.fr/</ref>
          </desc>
        </org>
        <org type="regroupinstitution" xml:id="struct-441569" status="VALID">
          <idno type="IdRef">02636817X</idno>
          <idno type="ISNI">0000000122597504</idno>
          <idno type="ROR">https://ror.org/02feahw73</idno>
          <orgName>Centre National de la Recherche Scientifique</orgName>
          <orgName type="acronym">CNRS</orgName>
          <date type="start">1939-10-19</date>
          <desc>
            <address>
              <country key="FR"/>
            </address>
            <ref type="url">https://www.cnrs.fr/</ref>
          </desc>
        </org>
        <org type="institution" xml:id="struct-581146" status="VALID">
          <idno type="IdRef">02640463X</idno>
          <idno type="ISNI">0000000121496883</idno>
          <idno type="ROR">https://ror.org/0199hds37</idno>
          <orgName>Université Sorbonne Paris Nord</orgName>
          <date type="start">2020-01-01</date>
          <desc>
            <address>
              <addrLine>99, avenue Jean-Baptiste Clément, 93430 Villetaneuse</addrLine>
              <country key="FR"/>
            </address>
            <ref type="url">https://www.univ-paris13.fr/</ref>
          </desc>
        </org>
      </listOrg>
    </back>
  </text>
</TEI>