Pattern Learning and Recognition on Statistical Manifolds: An Information-Geometric Review
Identifieur interne : 001521 ( Main/Exploration ); précédent : 001520; suivant : 001522Pattern Learning and Recognition on Statistical Manifolds: An Information-Geometric Review
Auteurs : Frank Nielsen [Japon]Source :
- Lecture Notes in Computer Science [ 0302-9743 ]
Abstract
Abstract: We review the information-geometric framework for statistical pattern recognition: First, we explain the role of statistical similarity measures and distances in fundamental statistical pattern recognition problems. We then concisely review the main statistical distances and report a novel versatile family of divergences. Depending on their intrinsic complexity, the statistical patterns are learned by either atomic parametric distributions, semi-parametric finite mixtures, or non-parametric kernel density distributions. Those statistical patterns are interpreted and handled geometrically in statistical manifolds either as single points, weighted sparse point sets or non-weighted dense point sets. We explain the construction of the two prominent families of statistical manifolds: The Rao Riemannian manifolds with geodesic metric distances, and the Amari-Chentsov manifolds with dual asymmetric non-metric divergences. For the latter manifolds, when considering atomic distributions from the same exponential families (including the ubiquitous Gaussian and multinomial families), we end up with dually flat exponential family manifolds that play a crucial role in many applications. We compare the advantages and disadvantages of these two approaches from the algorithmic point of view. Finally, we conclude with further perspectives on how “geometric thinking” may spur novel pattern modeling and processing paradigms.
Url:
DOI: 10.1007/978-3-642-39140-8_1
Affiliations:
Links toward previous steps (curation, corpus...)
- to stream Istex, to step Corpus: 001B74
- to stream Istex, to step Curation: 001B53
- to stream Istex, to step Checkpoint: 000147
- to stream Main, to step Merge: 001533
- to stream Main, to step Curation: 001521
Le document en format XML
<record><TEI wicri:istexFullTextTei="biblStruct"><teiHeader><fileDesc><titleStmt><title xml:lang="en">Pattern Learning and Recognition on Statistical Manifolds: An Information-Geometric Review</title>
<author><name sortKey="Nielsen, Frank" sort="Nielsen, Frank" uniqKey="Nielsen F" first="Frank" last="Nielsen">Frank Nielsen</name>
</author>
</titleStmt>
<publicationStmt><idno type="wicri:source">ISTEX</idno>
<idno type="RBID">ISTEX:77ECD0DEF000EEEF43CA1A6F4C4FEC696F8A1ABA</idno>
<date when="2013" year="2013">2013</date>
<idno type="doi">10.1007/978-3-642-39140-8_1</idno>
<idno type="url">https://api.istex.fr/ark:/67375/HCB-51Z96TLC-5/fulltext.pdf</idno>
<idno type="wicri:Area/Istex/Corpus">001B74</idno>
<idno type="wicri:explorRef" wicri:stream="Istex" wicri:step="Corpus" wicri:corpus="ISTEX">001B74</idno>
<idno type="wicri:Area/Istex/Curation">001B53</idno>
<idno type="wicri:Area/Istex/Checkpoint">000147</idno>
<idno type="wicri:explorRef" wicri:stream="Istex" wicri:step="Checkpoint">000147</idno>
<idno type="wicri:doubleKey">0302-9743:2013:Nielsen F:pattern:learning:and</idno>
<idno type="wicri:Area/Main/Merge">001533</idno>
<idno type="wicri:Area/Main/Curation">001521</idno>
<idno type="wicri:Area/Main/Exploration">001521</idno>
</publicationStmt>
<sourceDesc><biblStruct><analytic><title level="a" type="main" xml:lang="en">Pattern Learning and Recognition on Statistical Manifolds: An Information-Geometric Review</title>
<author><name sortKey="Nielsen, Frank" sort="Nielsen, Frank" uniqKey="Nielsen F" first="Frank" last="Nielsen">Frank Nielsen</name>
<affiliation wicri:level="3"><country xml:lang="fr">Japon</country>
<wicri:regionArea>Sony Computer Science Laboratories, Inc., Tokyo</wicri:regionArea>
<placeName><settlement type="city">Tokyo</settlement>
<region type="région">Région de Kantō</region>
</placeName>
</affiliation>
<affiliation></affiliation>
</author>
</analytic>
<monogr></monogr>
<series><title level="s" type="main" xml:lang="en">Lecture Notes in Computer Science</title>
<idno type="ISSN">0302-9743</idno>
<idno type="eISSN">1611-3349</idno>
<idno type="ISSN">0302-9743</idno>
</series>
</biblStruct>
</sourceDesc>
<seriesStmt><idno type="ISSN">0302-9743</idno>
</seriesStmt>
</fileDesc>
<profileDesc><textClass></textClass>
</profileDesc>
</teiHeader>
<front><div type="abstract" xml:lang="en">Abstract: We review the information-geometric framework for statistical pattern recognition: First, we explain the role of statistical similarity measures and distances in fundamental statistical pattern recognition problems. We then concisely review the main statistical distances and report a novel versatile family of divergences. Depending on their intrinsic complexity, the statistical patterns are learned by either atomic parametric distributions, semi-parametric finite mixtures, or non-parametric kernel density distributions. Those statistical patterns are interpreted and handled geometrically in statistical manifolds either as single points, weighted sparse point sets or non-weighted dense point sets. We explain the construction of the two prominent families of statistical manifolds: The Rao Riemannian manifolds with geodesic metric distances, and the Amari-Chentsov manifolds with dual asymmetric non-metric divergences. For the latter manifolds, when considering atomic distributions from the same exponential families (including the ubiquitous Gaussian and multinomial families), we end up with dually flat exponential family manifolds that play a crucial role in many applications. We compare the advantages and disadvantages of these two approaches from the algorithmic point of view. Finally, we conclude with further perspectives on how “geometric thinking” may spur novel pattern modeling and processing paradigms.</div>
</front>
</TEI>
<affiliations><list><country><li>Japon</li>
</country>
<region><li>Région de Kantō</li>
</region>
<settlement><li>Tokyo</li>
</settlement>
</list>
<tree><country name="Japon"><region name="Région de Kantō"><name sortKey="Nielsen, Frank" sort="Nielsen, Frank" uniqKey="Nielsen F" first="Frank" last="Nielsen">Frank Nielsen</name>
</region>
</country>
</tree>
</affiliations>
</record>
Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Wicri/Lorraine/explor/InforLorV4/Data/Main/Exploration
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 001521 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Main/Exploration/biblio.hfd -nk 001521 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien |wiki= Wicri/Lorraine |area= InforLorV4 |flux= Main |étape= Exploration |type= RBID |clé= ISTEX:77ECD0DEF000EEEF43CA1A6F4C4FEC696F8A1ABA |texte= Pattern Learning and Recognition on Statistical Manifolds: An Information-Geometric Review }}
This area was generated with Dilib version V0.6.33. |