An improved analysis of the Rademacher data-dependent bound using its self bounding property.
Identifieur interne : 000234 ( PubMed/Checkpoint ); précédent : 000233; suivant : 000235An improved analysis of the Rademacher data-dependent bound using its self bounding property.
Auteurs : Luca Oneto [Italie] ; Alessandro Ghio ; Davide Anguita ; Sandro RidellaSource :
- Neural networks : the official journal of the International Neural Network Society ; 2013.
English descriptors
- KwdEn :
- MESH :
- methods : Statistics as Topic.
- trends : Statistics as Topic.
- Artificial Intelligence.
Abstract
The problem of assessing the performance of a classifier, in the finite-sample setting, has been addressed by Vapnik in his seminal work by using data-independent measures of complexity. Recently, several authors have addressed the same problem by proposing data-dependent measures, which tighten previous results by taking in account the actual data distribution. In this framework, we derive some data-dependent bounds on the generalization ability of a classifier by exploiting the Rademacher Complexity and recent concentration results: in addition of being appealing for practical purposes, as they exploit empirical quantities only, these bounds improve previously known results.
DOI: 10.1016/j.neunet.2013.03.017
PubMed: 23587720
Affiliations:
Links toward previous steps (curation, corpus...)
Links to Exploration step
pubmed:23587720Le document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en">An improved analysis of the Rademacher data-dependent bound using its self bounding property.</title><author><name sortKey="Oneto, Luca" sort="Oneto, Luca" uniqKey="Oneto L" first="Luca" last="Oneto">Luca Oneto</name><affiliation wicri:level="1"><nlm:affiliation>DITEN - University of Genova, Via Opera Pia 11A, I-16145 Genova, Italy. Luca.Oneto@unige.it</nlm:affiliation><country xml:lang="fr">Italie</country><wicri:regionArea>DITEN - University of Genova, Via Opera Pia 11A, I-16145 Genova</wicri:regionArea><wicri:noRegion>I-16145 Genova</wicri:noRegion></affiliation></author><author><name sortKey="Ghio, Alessandro" sort="Ghio, Alessandro" uniqKey="Ghio A" first="Alessandro" last="Ghio">Alessandro Ghio</name></author><author><name sortKey="Anguita, Davide" sort="Anguita, Davide" uniqKey="Anguita D" first="Davide" last="Anguita">Davide Anguita</name></author><author><name sortKey="Ridella, Sandro" sort="Ridella, Sandro" uniqKey="Ridella S" first="Sandro" last="Ridella">Sandro Ridella</name></author></titleStmt><publicationStmt><idno type="wicri:source">PubMed</idno><date when="2013">2013</date><idno type="doi">10.1016/j.neunet.2013.03.017</idno><idno type="RBID">pubmed:23587720</idno><idno type="pmid">23587720</idno><idno type="wicri:Area/PubMed/Corpus">000228</idno><idno type="wicri:Area/PubMed/Curation">000228</idno><idno type="wicri:Area/PubMed/Checkpoint">000234</idno></publicationStmt><sourceDesc><biblStruct><analytic><title xml:lang="en">An improved analysis of the Rademacher data-dependent bound using its self bounding property.</title><author><name sortKey="Oneto, Luca" sort="Oneto, Luca" uniqKey="Oneto L" first="Luca" last="Oneto">Luca Oneto</name><affiliation wicri:level="1"><nlm:affiliation>DITEN - University of Genova, Via Opera Pia 11A, I-16145 Genova, Italy. Luca.Oneto@unige.it</nlm:affiliation><country xml:lang="fr">Italie</country><wicri:regionArea>DITEN - University of Genova, Via Opera Pia 11A, I-16145 Genova</wicri:regionArea><wicri:noRegion>I-16145 Genova</wicri:noRegion></affiliation></author><author><name sortKey="Ghio, Alessandro" sort="Ghio, Alessandro" uniqKey="Ghio A" first="Alessandro" last="Ghio">Alessandro Ghio</name></author><author><name sortKey="Anguita, Davide" sort="Anguita, Davide" uniqKey="Anguita D" first="Davide" last="Anguita">Davide Anguita</name></author><author><name sortKey="Ridella, Sandro" sort="Ridella, Sandro" uniqKey="Ridella S" first="Sandro" last="Ridella">Sandro Ridella</name></author></analytic><series><title level="j">Neural networks : the official journal of the International Neural Network Society</title><idno type="e-ISSN">1879-2782</idno><imprint><date when="2013" type="published">2013</date></imprint></series></biblStruct></sourceDesc></fileDesc><profileDesc><textClass><keywords scheme="KwdEn" xml:lang="en"><term>Artificial Intelligence</term><term>Statistics as Topic (methods)</term><term>Statistics as Topic (trends)</term></keywords><keywords scheme="MESH" qualifier="methods" xml:lang="en"><term>Statistics as Topic</term></keywords><keywords scheme="MESH" qualifier="trends" xml:lang="en"><term>Statistics as Topic</term></keywords><keywords scheme="MESH" xml:lang="en"><term>Artificial Intelligence</term></keywords></textClass></profileDesc></teiHeader><front><div type="abstract" xml:lang="en">The problem of assessing the performance of a classifier, in the finite-sample setting, has been addressed by Vapnik in his seminal work by using data-independent measures of complexity. Recently, several authors have addressed the same problem by proposing data-dependent measures, which tighten previous results by taking in account the actual data distribution. In this framework, we derive some data-dependent bounds on the generalization ability of a classifier by exploiting the Rademacher Complexity and recent concentration results: in addition of being appealing for practical purposes, as they exploit empirical quantities only, these bounds improve previously known results.</div></front></TEI><pubmed><MedlineCitation Owner="NLM" Status="MEDLINE"><PMID Version="1">23587720</PMID><DateCreated><Year>2013</Year><Month>06</Month><Day>03</Day></DateCreated><DateCompleted><Year>2014</Year><Month>02</Month><Day>13</Day></DateCompleted><Article PubModel="Print-Electronic"><Journal><ISSN IssnType="Electronic">1879-2782</ISSN><JournalIssue CitedMedium="Internet"><Volume>44</Volume><PubDate><Year>2013</Year><Month>Aug</Month></PubDate></JournalIssue><Title>Neural networks : the official journal of the International Neural Network Society</Title><ISOAbbreviation>Neural Netw</ISOAbbreviation></Journal><ArticleTitle>An improved analysis of the Rademacher data-dependent bound using its self bounding property.</ArticleTitle><Pagination><MedlinePgn>107-11</MedlinePgn></Pagination><ELocationID EIdType="doi" ValidYN="Y">10.1016/j.neunet.2013.03.017</ELocationID><ELocationID EIdType="pii" ValidYN="Y">S0893-6080(13)00102-0</ELocationID><Abstract><AbstractText>The problem of assessing the performance of a classifier, in the finite-sample setting, has been addressed by Vapnik in his seminal work by using data-independent measures of complexity. Recently, several authors have addressed the same problem by proposing data-dependent measures, which tighten previous results by taking in account the actual data distribution. In this framework, we derive some data-dependent bounds on the generalization ability of a classifier by exploiting the Rademacher Complexity and recent concentration results: in addition of being appealing for practical purposes, as they exploit empirical quantities only, these bounds improve previously known results.</AbstractText><CopyrightInformation>Copyright © 2013 Elsevier Ltd. All rights reserved.</CopyrightInformation></Abstract><AuthorList CompleteYN="Y"><Author ValidYN="Y"><LastName>Oneto</LastName><ForeName>Luca</ForeName><Initials>L</Initials><AffiliationInfo><Affiliation>DITEN - University of Genova, Via Opera Pia 11A, I-16145 Genova, Italy. Luca.Oneto@unige.it</Affiliation></AffiliationInfo></Author><Author ValidYN="Y"><LastName>Ghio</LastName><ForeName>Alessandro</ForeName><Initials>A</Initials></Author><Author ValidYN="Y"><LastName>Anguita</LastName><ForeName>Davide</ForeName><Initials>D</Initials></Author><Author ValidYN="Y"><LastName>Ridella</LastName><ForeName>Sandro</ForeName><Initials>S</Initials></Author></AuthorList><Language>eng</Language><PublicationTypeList><PublicationType UI="D016428">Journal Article</PublicationType></PublicationTypeList><ArticleDate DateType="Electronic"><Year>2013</Year><Month>04</Month><Day>03</Day></ArticleDate></Article><MedlineJournalInfo><Country>United States</Country><MedlineTA>Neural Netw</MedlineTA><NlmUniqueID>8805018</NlmUniqueID><ISSNLinking>0893-6080</ISSNLinking></MedlineJournalInfo><CitationSubset>IM</CitationSubset><MeshHeadingList><MeshHeading><DescriptorName MajorTopicYN="Y" UI="D001185">Artificial Intelligence</DescriptorName></MeshHeading><MeshHeading><DescriptorName MajorTopicYN="N" UI="D013223">Statistics as Topic</DescriptorName><QualifierName MajorTopicYN="N" UI="Q000379">methods</QualifierName><QualifierName MajorTopicYN="Y" UI="Q000639">trends</QualifierName></MeshHeading></MeshHeadingList></MedlineCitation><PubmedData><History><PubMedPubDate PubStatus="received"><Year>2012</Year><Month>3</Month><Day>2</Day></PubMedPubDate><PubMedPubDate PubStatus="revised"><Year>2013</Year><Month>3</Month><Day>22</Day></PubMedPubDate><PubMedPubDate PubStatus="accepted"><Year>2013</Year><Month>3</Month><Day>22</Day></PubMedPubDate><PubMedPubDate PubStatus="aheadofprint"><Year>2013</Year><Month>4</Month><Day>3</Day></PubMedPubDate><PubMedPubDate PubStatus="entrez"><Year>2013</Year><Month>4</Month><Day>17</Day><Hour>6</Hour><Minute>0</Minute></PubMedPubDate><PubMedPubDate PubStatus="pubmed"><Year>2013</Year><Month>4</Month><Day>17</Day><Hour>6</Hour><Minute>0</Minute></PubMedPubDate><PubMedPubDate PubStatus="medline"><Year>2014</Year><Month>2</Month><Day>14</Day><Hour>6</Hour><Minute>0</Minute></PubMedPubDate></History><PublicationStatus>ppublish</PublicationStatus><ArticleIdList><ArticleId IdType="pii">S0893-6080(13)00102-0</ArticleId><ArticleId IdType="doi">10.1016/j.neunet.2013.03.017</ArticleId><ArticleId IdType="pubmed">23587720</ArticleId></ArticleIdList></PubmedData></pubmed><affiliations><list><country><li>Italie</li></country></list><tree><noCountry><name sortKey="Anguita, Davide" sort="Anguita, Davide" uniqKey="Anguita D" first="Davide" last="Anguita">Davide Anguita</name><name sortKey="Ghio, Alessandro" sort="Ghio, Alessandro" uniqKey="Ghio A" first="Alessandro" last="Ghio">Alessandro Ghio</name><name sortKey="Ridella, Sandro" sort="Ridella, Sandro" uniqKey="Ridella S" first="Sandro" last="Ridella">Sandro Ridella</name></noCountry><country name="Italie"><noRegion><name sortKey="Oneto, Luca" sort="Oneto, Luca" uniqKey="Oneto L" first="Luca" last="Oneto">Luca Oneto</name></noRegion></country></tree></affiliations></record>Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Wicri/Musique/explor/OperaV1/Data/PubMed/Checkpoint
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 000234 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/PubMed/Checkpoint/biblio.hfd -nk 000234 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien
|wiki= Wicri/Musique
|area= OperaV1
|flux= PubMed
|étape= Checkpoint
|type= RBID
|clé= pubmed:23587720
|texte= An improved analysis of the Rademacher data-dependent bound using its self bounding property.
}}
Pour générer des pages wiki
HfdIndexSelect -h $EXPLOR_AREA/Data/PubMed/Checkpoint/RBID.i -Sk "pubmed:23587720" \
| HfdSelect -Kh $EXPLOR_AREA/Data/PubMed/Checkpoint/biblio.hfd \
| NlmPubMed2Wicri -a OperaV1
| This area was generated with Dilib version V0.6.21. |  |