Sequential Monte Carlo for rare event estimation
hal.structure.identifier | Applications of interacting particle systems to statistics [ASPI] | |
hal.structure.identifier | Institut de Recherche Mathématique de Rennes [IRMAR] | |
dc.contributor.author | CÉROU, Frédéric | |
hal.structure.identifier | Advanced Learning Evolutionary Algorithms [ALEA] | |
dc.contributor.author | DEL MORAL, Pierre | |
hal.structure.identifier | Multimedia content-based indexing [TEXMEX] | |
dc.contributor.author | FURON, Teddy | |
hal.structure.identifier | Applications of interacting particle systems to statistics [ASPI] | |
hal.structure.identifier | Institut de Recherche Mathématique de Rennes [IRMAR] | |
dc.contributor.author | GUYADER, Arnaud | |
dc.date.issued | 2012 | |
dc.identifier.issn | 0960-3174 | |
dc.description.abstractEn | This paper discusses a novel strategy for simulating rare events and an associated Monte Carlo estimation of tail probabilities. Our method uses a system of interacting particles and exploits a Feynman-Kac representation of that system to analyze their fluctuations. Our precise analysis of the variance of a standard multilevel splitting algorithm reveals an opportunity for improvement. This leads to a novel method that relies on adaptive levels and produces, in the limit of an idealized version of the algorithm, estimates with optimal variance. The motivation for this theoretical work comes from problems occurring in watermarking and fin- gerprinting of digital contents, which represents a new field of applications of rare event simulation techniques. Some numerical results show performance close to the idealized version of our technique for these practical applications. | |
dc.description.sponsorship | Sécurité et fiabilité des techniques de tatouages - ANR-06-SETI-0009 | |
dc.language.iso | en | |
dc.publisher | Springer Verlag (Germany) | |
dc.subject.en | Rare event | |
dc.subject.en | Sequential importance sampling | |
dc.subject.en | Feynman-Kac formula | |
dc.subject.en | Metropolis-Hastings | |
dc.subject.en | Fingerprinting | |
dc.subject.en | Watermarking | |
dc.type | Article de revue | |
dc.identifier.doi | 10.1007/s11222-011-9231-6 | |
dc.subject.hal | Statistiques [stat]/Calcul [stat.CO] | |
bordeaux.journal | Statistics and Computing | |
bordeaux.page | 795-908 | |
bordeaux.volume | 22 | |
bordeaux.issue | 3 | |
bordeaux.peerReviewed | oui | |
hal.identifier | inria-00584352 | |
hal.version | 1 | |
hal.popular | non | |
hal.audience | Internationale | |
dc.title.it | Sequential Monte Carlo for rare event estimation | |
hal.origin.link | https://hal.archives-ouvertes.fr//inria-00584352v1 | |
bordeaux.COinS | ctx_ver=Z39.88-2004&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.jtitle=Statistics%20and%20Computing&rft.date=2012&rft.volume=22&rft.issue=3&rft.spage=795-908&rft.epage=795-908&rft.eissn=0960-3174&rft.issn=0960-3174&rft.au=C%C3%89ROU,%20Fr%C3%A9d%C3%A9ric&DEL%20MORAL,%20Pierre&FURON,%20Teddy&GUYADER,%20Arnaud&rft.genre=article |
Fichier(s) constituant ce document
Fichiers | Taille | Format | Vue |
---|---|---|---|
Il n'y a pas de fichiers associés à ce document. |