Probabilistic Aggregation of Classifiers for Incremental Learning
Identifieur interne : 000E09 ( Main/Exploration ); précédent : 000E08; suivant : 000E10Probabilistic Aggregation of Classifiers for Incremental Learning
Auteurs : Patricia Trejo [Chili] ; Ricardo Anculef [Chili] ; Héctor Allende [Chili] ; Claudio Moraga [Espagne, Allemagne]Source :
- Lecture Notes in Computer Science [ 0302-9743 ] ; 2007.
Abstract
Abstract: We work with a recently proposed algorithm where an ensemble of base classifiers, combined using weighted majority voting, is used for incremental classification of data. To successfully accommodate novel information without compromising previously acquired knowledge this algorithm requires an adequate strategy to determine the voting weights. Given an instance to classify, we propose to define each voting weight as the posterior probability of the corresponding hypothesis given the instance. By operating with priors and the likelihood models the obtained weights can take into account the location of the instance in the different class-specific feature spaces but also the coverage of each class k given the classifier and the quality of the learned hypothesis. This approach can provide important improvements in the generalization performance of the resulting classifier and its ability to control the stability/plasticity tradeoff. Experiments are carried out with three real classification problems already introduced to test incremental algorithms.
Url:
DOI: 10.1007/978-3-540-73007-1_17
Affiliations:
Links toward previous steps (curation, corpus...)
- to stream Istex, to step Corpus: 001A21
- to stream Istex, to step Curation: 001916
- to stream Istex, to step Checkpoint: 000824
- to stream Main, to step Merge: 000E22
- to stream Main, to step Curation: 000E09
Le document en format XML
<record><TEI wicri:istexFullTextTei="biblStruct"><teiHeader><fileDesc><titleStmt><title xml:lang="en">Probabilistic Aggregation of Classifiers for Incremental Learning</title><author><name sortKey="Trejo, Patricia" sort="Trejo, Patricia" uniqKey="Trejo P" first="Patricia" last="Trejo">Patricia Trejo</name></author><author><name sortKey=" Anculef, Ricardo" sort=" Anculef, Ricardo" uniqKey=" Anculef R" first="Ricardo" last=" Anculef">Ricardo Anculef</name></author><author><name sortKey="Allende, Hector" sort="Allende, Hector" uniqKey="Allende H" first="Héctor" last="Allende">Héctor Allende</name></author><author><name sortKey="Moraga, Claudio" sort="Moraga, Claudio" uniqKey="Moraga C" first="Claudio" last="Moraga">Claudio Moraga</name></author></titleStmt><publicationStmt><idno type="wicri:source">ISTEX</idno><idno type="RBID">ISTEX:5BEB131D9A195594978A9B4E1F871BD8C01C1A89</idno><date when="2007" year="2007">2007</date><idno type="doi">10.1007/978-3-540-73007-1_17</idno><idno type="url">https://api.istex.fr/document/5BEB131D9A195594978A9B4E1F871BD8C01C1A89/fulltext/pdf</idno><idno type="wicri:Area/Istex/Corpus">001A21</idno><idno type="wicri:Area/Istex/Curation">001916</idno><idno type="wicri:Area/Istex/Checkpoint">000824</idno><idno type="wicri:doubleKey">0302-9743:2007:Trejo P:probabilistic:aggregation:of</idno><idno type="wicri:Area/Main/Merge">000E22</idno><idno type="wicri:Area/Main/Curation">000E09</idno><idno type="wicri:Area/Main/Exploration">000E09</idno></publicationStmt><sourceDesc><biblStruct><analytic><title level="a" type="main" xml:lang="en">Probabilistic Aggregation of Classifiers for Incremental Learning</title><author><name sortKey="Trejo, Patricia" sort="Trejo, Patricia" uniqKey="Trejo P" first="Patricia" last="Trejo">Patricia Trejo</name><affiliation wicri:level="1"><country xml:lang="fr">Chili</country><wicri:regionArea>Universidad Técnica Federico Santa María, Departamento de Informática, CP 110-V Valparaíso</wicri:regionArea><wicri:noRegion>CP 110-V Valparaíso</wicri:noRegion></affiliation><affiliation wicri:level="1"><country wicri:rule="url">Chili</country></affiliation></author><author><name sortKey=" Anculef, Ricardo" sort=" Anculef, Ricardo" uniqKey=" Anculef R" first="Ricardo" last=" Anculef">Ricardo Anculef</name><affiliation wicri:level="1"><country xml:lang="fr">Chili</country><wicri:regionArea>Universidad Técnica Federico Santa María, Departamento de Informática, CP 110-V Valparaíso</wicri:regionArea><wicri:noRegion>CP 110-V Valparaíso</wicri:noRegion></affiliation><affiliation wicri:level="1"><country wicri:rule="url">Chili</country></affiliation></author><author><name sortKey="Allende, Hector" sort="Allende, Hector" uniqKey="Allende H" first="Héctor" last="Allende">Héctor Allende</name><affiliation wicri:level="1"><country xml:lang="fr">Chili</country><wicri:regionArea>Universidad Técnica Federico Santa María, Departamento de Informática, CP 110-V Valparaíso</wicri:regionArea><wicri:noRegion>CP 110-V Valparaíso</wicri:noRegion></affiliation><affiliation wicri:level="1"><country wicri:rule="url">Chili</country></affiliation></author><author><name sortKey="Moraga, Claudio" sort="Moraga, Claudio" uniqKey="Moraga C" first="Claudio" last="Moraga">Claudio Moraga</name><affiliation wicri:level="1"><country xml:lang="fr">Espagne</country><wicri:regionArea>European Centre for Soft Computing 33600 Mieres, Asturias</wicri:regionArea><wicri:noRegion>Asturias</wicri:noRegion></affiliation><affiliation wicri:level="3"><country xml:lang="fr">Allemagne</country><wicri:regionArea>Dortmund University, 44221 Dortmund</wicri:regionArea><placeName><region type="land" nuts="1">Rhénanie-du-Nord-Westphalie</region><region type="district" nuts="2">District d'Arnsberg</region><settlement type="city">Dortmund</settlement></placeName></affiliation><affiliation><wicri:noCountry code="no comma">E-mail: mail@claudio-moraga.eu</wicri:noCountry></affiliation></author></analytic><monogr></monogr><series><title level="s">Lecture Notes in Computer Science</title><imprint><date>2007</date></imprint><idno type="ISSN">0302-9743</idno><idno type="eISSN">1611-3349</idno><idno type="ISSN">0302-9743</idno></series><idno type="istex">5BEB131D9A195594978A9B4E1F871BD8C01C1A89</idno><idno type="DOI">10.1007/978-3-540-73007-1_17</idno><idno type="ChapterID">17</idno><idno type="ChapterID">Chap17</idno></biblStruct></sourceDesc><seriesStmt><idno type="ISSN">0302-9743</idno></seriesStmt></fileDesc><profileDesc><textClass></textClass><langUsage><language ident="en">en</language></langUsage></profileDesc></teiHeader><front><div type="abstract" xml:lang="en">Abstract: We work with a recently proposed algorithm where an ensemble of base classifiers, combined using weighted majority voting, is used for incremental classification of data. To successfully accommodate novel information without compromising previously acquired knowledge this algorithm requires an adequate strategy to determine the voting weights. Given an instance to classify, we propose to define each voting weight as the posterior probability of the corresponding hypothesis given the instance. By operating with priors and the likelihood models the obtained weights can take into account the location of the instance in the different class-specific feature spaces but also the coverage of each class k given the classifier and the quality of the learned hypothesis. This approach can provide important improvements in the generalization performance of the resulting classifier and its ability to control the stability/plasticity tradeoff. Experiments are carried out with three real classification problems already introduced to test incremental algorithms.</div></front></TEI><affiliations><list><country><li>Allemagne</li><li>Chili</li><li>Espagne</li></country><region><li>District d'Arnsberg</li><li>Rhénanie-du-Nord-Westphalie</li></region><settlement><li>Dortmund</li></settlement></list><tree><country name="Chili"><noRegion><name sortKey="Trejo, Patricia" sort="Trejo, Patricia" uniqKey="Trejo P" first="Patricia" last="Trejo">Patricia Trejo</name></noRegion><name sortKey=" Anculef, Ricardo" sort=" Anculef, Ricardo" uniqKey=" Anculef R" first="Ricardo" last=" Anculef">Ricardo Anculef</name><name sortKey=" Anculef, Ricardo" sort=" Anculef, Ricardo" uniqKey=" Anculef R" first="Ricardo" last=" Anculef">Ricardo Anculef</name><name sortKey="Allende, Hector" sort="Allende, Hector" uniqKey="Allende H" first="Héctor" last="Allende">Héctor Allende</name><name sortKey="Allende, Hector" sort="Allende, Hector" uniqKey="Allende H" first="Héctor" last="Allende">Héctor Allende</name><name sortKey="Trejo, Patricia" sort="Trejo, Patricia" uniqKey="Trejo P" first="Patricia" last="Trejo">Patricia Trejo</name></country><country name="Espagne"><noRegion><name sortKey="Moraga, Claudio" sort="Moraga, Claudio" uniqKey="Moraga C" first="Claudio" last="Moraga">Claudio Moraga</name></noRegion></country><country name="Allemagne"><region name="Rhénanie-du-Nord-Westphalie"><name sortKey="Moraga, Claudio" sort="Moraga, Claudio" uniqKey="Moraga C" first="Claudio" last="Moraga">Claudio Moraga</name></region></country></tree></affiliations></record>Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Ticri/CIDE/explor/OcrV1/Data/Main/Exploration
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 000E09 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Main/Exploration/biblio.hfd -nk 000E09 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien
|wiki= Ticri/CIDE
|area= OcrV1
|flux= Main
|étape= Exploration
|type= RBID
|clé= ISTEX:5BEB131D9A195594978A9B4E1F871BD8C01C1A89
|texte= Probabilistic Aggregation of Classifiers for Incremental Learning
}}
| This area was generated with Dilib version V0.6.32. |  |