Publication detail

A projection-based Rao-Blackwellized particle filter to estimate parameters in conditionally conjugate state-space models

PAPEŽ, M.

Original Title

A projection-based Rao-Blackwellized particle filter to estimate parameters in conditionally conjugate state-space models

English Title

A projection-based Rao-Blackwellized particle filter to estimate parameters in conditionally conjugate state-space models

Type

conference paper

Language

en

Original Abstract

Particle filters constitute today a well-established class of techniques for state filtering in non-linear state-space models. However, online estimation of static parameters under the same framework represents a difficult problem. The solution can be found to some extent within a category of state-space models allowing us to perform parameter estimation in an analytically tractable manner, while still considering non-linearities in data evolution equations. Nevertheless, the well-known particle path degeneracy problem complicates the computation of the statistics that are required to estimate the parameters. The present paper proposes a simple and efficient method which is experimentally shown to suffer less from this issue.

English abstract

Particle filters constitute today a well-established class of techniques for state filtering in non-linear state-space models. However, online estimation of static parameters under the same framework represents a difficult problem. The solution can be found to some extent within a category of state-space models allowing us to perform parameter estimation in an analytically tractable manner, while still considering non-linearities in data evolution equations. Nevertheless, the well-known particle path degeneracy problem complicates the computation of the statistics that are required to estimate the parameters. The present paper proposes a simple and efficient method which is experimentally shown to suffer less from this issue.

Keywords

Sequential Monte Carlo; particle filtering; conditionally conjugate state-space models; Rao-Blackwellization

Released

10.06.2018

Publisher

Institute of Electrical and Electronics Engineers (IEEE)

ISBN

978-1-5386-1570-6

Book

Proceedings of the 20th Statistical Signal Processing Workshop (SSP)

Pages from

268

Pages to

272

Pages count

5

Documents

BibTex


@inproceedings{BUT148842,
  author="Milan {Papež}",
  title="A projection-based Rao-Blackwellized particle filter to estimate parameters in conditionally conjugate state-space models",
  annote="Particle filters constitute today a well-established class of techniques for state filtering in non-linear state-space models. However, online estimation of static parameters under the same framework represents a difficult problem. The solution can be found to some extent within a category of state-space models allowing us to perform parameter estimation in an analytically tractable manner, while still considering non-linearities in data evolution equations. Nevertheless, the well-known particle path degeneracy problem complicates the computation of the statistics that are required to estimate the parameters. The present paper proposes a simple and efficient method which is experimentally shown to suffer less from this issue.",
  address="Institute of Electrical and Electronics Engineers (IEEE)",
  booktitle="Proceedings of the 20th Statistical Signal Processing Workshop (SSP)",
  chapter="148842",
  doi="10.1109/SSP.2018.8450730",
  howpublished="online",
  institution="Institute of Electrical and Electronics Engineers (IEEE)",
  year="2018",
  month="june",
  pages="268--272",
  publisher="Institute of Electrical and Electronics Engineers (IEEE)",
  type="conference paper"
}