Mostrar el registro sencillo del ítem
dc.contributor.author
Pulido, Manuel Arturo
![Se ha confirmado la validez de este valor de autoridad por un usuario](/themes/CONICETDigital/images/authority_control/invisible.gif)
dc.contributor.author
van Leeuwen, Peter Jan
dc.date.available
2022-04-29T15:17:33Z
dc.date.issued
2018
dc.identifier.citation
Kernel embedding of maps for Bayesian inference: the variational mapping particle filter; EGU General Assembly; Vienna; Austria; 2018; 1-1
dc.identifier.issn
1029-7006
dc.identifier.uri
http://hdl.handle.net/11336/156122
dc.description.abstract
Data assimilation for high-dimensional highly nonlinear systems is becoming crucial for several geosciences applications. In this work, a novel particle filter is introduced which aims to an efficient sampling of the posterior pdf in high-dimensional state spaces considering a limited number of particles. Particles are mapped from the proposal to the posterior density using the principles of optimal transport. The Kullback-Leibler divergence between the posterior density and the proposal divergence is minimised using variational principles, leading to an iterative gradient-descent like algorithm. A key ingredient of the mapping is that the transformations are embedded in a reproducing kernel Hilbert space which constrains the dimensions of the space for the optimal transport to the number of particles. Gradient information of the Kullback-Leibler divergence allows a quick convergence using well known gradient-based optimization algorithms from machine learning, adadelta and adam, which do not require cost function calculations. Evaluation of the method and comparison with a SIR filter is conducted as a proof-of-concept in the Lorenz-63 system, where the exact solution is known. No resampling is required even for long recursive implementations. The number of effective particles remains close to the total number of particles in all the recursions. Hence, the mapping particle filter does not suffer from sample impoverishment, even in highly nonlinear settings. Finally, results from experiments on a high-dimensional turbulent geophysical system will be presented, and the performance of the new method compared to other existing method will be discussed.
dc.format
application/pdf
dc.language.iso
eng
dc.publisher
Copernicus Publications
![Se ha confirmado la validez de este valor de autoridad por un usuario](/themes/CONICETDigital/images/authority_control/invisible.gif)
dc.rights
info:eu-repo/semantics/openAccess
dc.rights.uri
https://creativecommons.org/licenses/by-nc-sa/2.5/ar/
dc.subject
STEIN GRADIENT
dc.subject
MONTE CARLO SEQUENTIAL
dc.subject.classification
Meteorología y Ciencias Atmosféricas
![Se ha confirmado la validez de este valor de autoridad por un usuario](/themes/CONICETDigital/images/authority_control/invisible.gif)
dc.subject.classification
Ciencias de la Tierra y relacionadas con el Medio Ambiente
![Se ha confirmado la validez de este valor de autoridad por un usuario](/themes/CONICETDigital/images/authority_control/invisible.gif)
dc.subject.classification
CIENCIAS NATURALES Y EXACTAS
![Se ha confirmado la validez de este valor de autoridad por un usuario](/themes/CONICETDigital/images/authority_control/invisible.gif)
dc.title
Kernel embedding of maps for Bayesian inference: the variational mapping particle filter
dc.type
info:eu-repo/semantics/publishedVersion
dc.type
info:eu-repo/semantics/conferenceObject
dc.type
info:ar-repo/semantics/documento de conferencia
dc.date.updated
2022-03-16T20:29:28Z
dc.journal.volume
20
dc.journal.pagination
1-1
dc.journal.pais
Austria
![Se ha confirmado la validez de este valor de autoridad por un usuario](/themes/CONICETDigital/images/authority_control/invisible.gif)
dc.journal.ciudad
Vienna
dc.description.fil
Fil: Pulido, Manuel Arturo. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Conicet - Mar del Plata. Instituto de Investigaciones En Biodiversidad y Biotecnología. Grupo de Investigación en Química Analítica y Modelado Molecular; Argentina. University of Reading; Reino Unido
dc.description.fil
Fil: van Leeuwen, Peter Jan. University of Reading; Reino Unido
dc.relation.alternativeid
info:eu-repo/semantics/altIdentifier/url/https://meetingorganizer.copernicus.org/EGU2018/EGU2018-3750.pdf
dc.relation.alternativeid
info:eu-repo/semantics/altIdentifier/url/https://www.geophysical-research-abstracts.net/egu2018.html
dc.conicet.rol
Autor
![Se ha confirmado la validez de este valor de autoridad por un usuario](/themes/CONICETDigital/images/authority_control/invisible.gif)
dc.conicet.rol
Autor
![Se ha confirmado la validez de este valor de autoridad por un usuario](/themes/CONICETDigital/images/authority_control/invisible.gif)
dc.coverage
Internacional
dc.type.subtype
Conferencia
dc.description.nombreEvento
EGU General Assembly
dc.date.evento
2018-04-08
dc.description.ciudadEvento
Vienna
dc.description.paisEvento
Austria
![Se ha confirmado la validez de este valor de autoridad por un usuario](/themes/CONICETDigital/images/authority_control/invisible.gif)
dc.type.publicacion
Journal
dc.description.institucionOrganizadora
European Geosciences Union
dc.source.revista
Geophysical Research Abstracts
dc.date.eventoHasta
2018-04-13
dc.type
Conferencia
Archivos asociados