<?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-02407890</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="2026-05-13T08:39:45+02:00"/>
      </publicationStmt>
      <sourceDesc>
        <p part="N">HAL API Platform</p>
      </sourceDesc>
    </fileDesc>
  </teiHeader>
  <text>
    <body>
      <listBibl>
        <biblFull>
          <titleStmt>
            <title xml:lang="en">Learning and forgetting using reinforced Bayesian change detection</title>
            <author role="aut">
              <persName>
                <forename type="first">Vincent</forename>
                <surname>Moens</surname>
              </persName>
              <idno type="halauthorid">1746727-0</idno>
            </author>
            <author role="aut">
              <persName>
                <forename type="first">Alexandre</forename>
                <surname>Zénon</surname>
              </persName>
              <email type="md5">c9dd59217383756004468f817e10ff51</email>
              <email type="domain">u-bordeaux.fr</email>
              <idno type="idhal" notation="string">alexandre-zenon</idno>
              <idno type="idhal" notation="numeric">741769</idno>
              <idno type="halauthorid" notation="string">1566161-741769</idno>
              <idno type="ORCID">https://orcid.org/0000-0001-7989-1261</idno>
              <idno type="GOOGLE SCHOLAR">https://scholar.google.com/citations?user=tBwSeA4AAAAJ&amp;hl=en</idno>
              <idno type="IDREF">https://www.idref.fr/227531558</idno>
              <affiliation ref="#struct-92863"/>
            </author>
            <editor role="depositor">
              <persName>
                <forename>Alexandre</forename>
                <surname>Zénon</surname>
              </persName>
              <email type="md5">c9dd59217383756004468f817e10ff51</email>
              <email type="domain">u-bordeaux.fr</email>
            </editor>
            <funder ref="#projanr-42319"/>
          </titleStmt>
          <editionStmt>
            <edition n="v1" type="current">
              <date type="whenSubmitted">2020-12-21 08:34:55</date>
              <date type="whenModified">2026-04-10 14:18:03</date>
              <date type="whenReleased">2021-01-06 14:55:59</date>
              <date type="whenProduced">2019-04-17</date>
              <date type="whenEndEmbargoed">2020-12-21</date>
              <ref type="file" target="https://hal.science/hal-02407890v1/document">
                <date notBefore="2020-12-21"/>
              </ref>
              <ref type="file" subtype="author" n="1" target="https://hal.science/hal-02407890v1/file/pcbi.1006713.pdf" id="file-3084297-2699171">
                <date notBefore="2020-12-21"/>
              </ref>
              <ref type="externalLink" target="https://doi.org/10.1371/journal.pcbi.1006713"/>
            </edition>
            <respStmt>
              <resp>contributor</resp>
              <name key="823594">
                <persName>
                  <forename>Alexandre</forename>
                  <surname>Zénon</surname>
                </persName>
                <email type="md5">c9dd59217383756004468f817e10ff51</email>
                <email type="domain">u-bordeaux.fr</email>
              </name>
            </respStmt>
          </editionStmt>
          <publicationStmt>
            <distributor>CCSD</distributor>
            <idno type="halId">hal-02407890</idno>
            <idno type="halUri">https://hal.science/hal-02407890</idno>
            <idno type="halBibtex">moens:hal-02407890</idno>
            <idno type="halRefHtml">&lt;i&gt;PLoS Computational Biology&lt;/i&gt;, 2019, 15 (4), pp.e1006713. &lt;a target="_blank" href="https://dx.doi.org/10.1371/journal.pcbi.1006713"&gt;&amp;#x27E8;10.1371/journal.pcbi.1006713&amp;#x27E9;&lt;/a&gt;</idno>
            <idno type="halRef">PLoS Computational Biology, 2019, 15 (4), pp.e1006713. &amp;#x27E8;10.1371/journal.pcbi.1006713&amp;#x27E9;</idno>
            <availability status="restricted">
              <licence target="https://about.hal.science/hal-authorisation-v1/">HAL Authorization<ref corresp="#file-3084297-2699171"/></licence>
            </availability>
          </publicationStmt>
          <seriesStmt>
            <idno type="stamp" n="SHS">Sciences de l'Homme et de la Société</idno>
            <idno type="stamp" n="ANR">ANR</idno>
          </seriesStmt>
          <notesStmt>
            <note type="audience" n="2">International</note>
            <note type="popular" n="0">No</note>
            <note type="peer" n="1">Yes</note>
          </notesStmt>
          <sourceDesc>
            <biblStruct>
              <analytic>
                <title xml:lang="en">Learning and forgetting using reinforced Bayesian change detection</title>
                <author role="aut">
                  <persName>
                    <forename type="first">Vincent</forename>
                    <surname>Moens</surname>
                  </persName>
                  <idno type="halauthorid">1746727-0</idno>
                </author>
                <author role="aut">
                  <persName>
                    <forename type="first">Alexandre</forename>
                    <surname>Zénon</surname>
                  </persName>
                  <email type="md5">c9dd59217383756004468f817e10ff51</email>
                  <email type="domain">u-bordeaux.fr</email>
                  <idno type="idhal" notation="string">alexandre-zenon</idno>
                  <idno type="idhal" notation="numeric">741769</idno>
                  <idno type="halauthorid" notation="string">1566161-741769</idno>
                  <idno type="ORCID">https://orcid.org/0000-0001-7989-1261</idno>
                  <idno type="GOOGLE SCHOLAR">https://scholar.google.com/citations?user=tBwSeA4AAAAJ&amp;hl=en</idno>
                  <idno type="IDREF">https://www.idref.fr/227531558</idno>
                  <affiliation ref="#struct-92863"/>
                </author>
              </analytic>
              <monogr>
                <idno type="halJournalId" status="VALID">1956</idno>
                <idno type="issn">1553-734X</idno>
                <idno type="eissn">1553-7358</idno>
                <title level="j">PLoS Computational Biology</title>
                <imprint>
                  <publisher>PLOS</publisher>
                  <biblScope unit="volume">15</biblScope>
                  <biblScope unit="issue">4</biblScope>
                  <biblScope unit="pp">e1006713</biblScope>
                  <date type="datePub">2019-04-17</date>
                </imprint>
              </monogr>
              <idno type="doi">10.1371/journal.pcbi.1006713</idno>
            </biblStruct>
          </sourceDesc>
          <profileDesc>
            <langUsage>
              <language ident="en">English</language>
            </langUsage>
            <textClass>
              <classCode scheme="halDomain" n="shs">Humanities and Social Sciences</classCode>
              <classCode scheme="halDomain" n="sdv">Life Sciences [q-bio]</classCode>
              <classCode scheme="halTypology" n="ART">Journal articles</classCode>
              <classCode scheme="halOldTypology" n="ART">Journal articles</classCode>
              <classCode scheme="halTreeTypology" n="ART">Journal articles</classCode>
            </textClass>
            <abstract xml:lang="en">
              <p>Agents living in volatile environments must be able to detect changes in contingencies while refraining to adapt to unexpected events that are caused by noise. In Reinforcement Learning (RL) frameworks, this requires learning rates that adapt to past reliability of the model. The observation that behavioural flexibility in animals tends to decrease following prolonged training in stable environment provides experimental evidence for such adaptive learning rates. However, in classical RL models, learning rate is either fixed or scheduled and can thus not adapt dynamically to environmental changes. Here, we propose a new Bayesian learning model, using variational inference, that achieves adaptive change detection by the use of Stabilized Forgetting, updating its current belief based on a mixture of fixed, initial priors and previous posterior beliefs. The weight given to these two sources is optimized alongside the other parameters, allowing the model to adapt dynamically to changes in environmental volatility and to unexpected observations. This approach is used to implement the "critic" of an actor-critic RL model, while the actor samples the resulting value distributions to choose which action to undertake. We show that our model can emulate different adaptation strategies to contingency changes, depending on its prior assumptions of environmental stability, and that model parameters can be fit to real data with high accuracy. The model also exhibits trade-offs between flexibility and computational costs that mirror those observed in real data. Overall, the proposed method provides a general framework to study learning flexibility and decision making in RL contexts.</p>
            </abstract>
          </profileDesc>
        </biblFull>
      </listBibl>
    </body>
    <back>
      <listOrg type="structures">
        <org type="institution" xml:id="struct-92863" status="VALID">
          <idno type="IdRef">026431297</idno>
          <idno type="ROR">https://ror.org/02495e989</idno>
          <orgName>Université Catholique de Louvain = Catholic University of Louvain</orgName>
          <orgName type="acronym">UCLouvain</orgName>
          <desc>
            <address>
              <addrLine>Place de l'Université 1, B-1348 Louvain-La-Neuve</addrLine>
              <country key="BE"/>
            </address>
            <ref type="url">https://www.uclouvain.be/fr</ref>
          </desc>
        </org>
      </listOrg>
      <listOrg type="projects">
        <org type="anrProject" xml:id="projanr-42319" status="VALID">
          <idno type="anr">ANR-10-IDEX-0003</idno>
          <orgName>IDEX BORDEAUX</orgName>
          <desc>Initiative d'excellence de l'Université de Bordeaux</desc>
          <date type="start">2010</date>
        </org>
      </listOrg>
    </back>
  </text>
</TEI>