Detail výsledku

Design of variable exponential forgetting for estimation of the statistics of the Normal distribution

DOKOUPIL, J.; VÁCLAVEK, P. Design of variable exponential forgetting for estimation of the statistics of the Normal distribution. In 55th Conference on Decision and Control. IEEE, 2016. p. 1179-1184. ISBN: 978-1-5090-1837-6.
Typ
článek ve sborníku konference
Jazyk
anglicky
Autoři
Dokoupil Jakub, Ing., Ph.D., RG-2-02 (CEITEC VUT)
Václavek Pavel, prof. Ing., Ph.D., RG-2-02 (CEITEC VUT), UAMT (FEKT)
Abstrakt

A recursive algorithm for estimating the statistics of the Normal distribution is designed, making it adaptive in the sense that the forgetting factor is driven by data. A mechanism to suppress obsolete information is proposed, following the principles of Bayesian decision-making. Specifically, the best combination of two time-evolution model hypotheses in terms of the geometric mean is performed. The first hypothesis assumes no change in the parameter evolution, while the second one assumes that all parameter changes are equally admitted.
In order to provide data-driven forgetting, complementary probabilities assigned to each hypothesis are determined as the maximizers of the decision problem. Simulations, including a performance comparison with a recently proposed self-tuning estimator, are presented.

Klíčová slova

estimation; forgetting factor; Kullback-Leibler divergence; Normal distribution

URL
Rok
2016
Strany
1179–1184
Sborník
55th Conference on Decision and Control
Konference
55th IEEE Conference on Decision and Control
ISBN
978-1-5090-1837-6
Vydavatel
IEEE
DOI
UT WoS
000400048101057
EID Scopus
BibTeX
@inproceedings{BUT130677,
  author="Jakub {Dokoupil} and Pavel {Václavek}",
  title="Design of variable exponential forgetting for estimation of the statistics of the Normal distribution",
  booktitle="55th Conference on Decision and Control",
  year="2016",
  pages="1179--1184",
  publisher="IEEE",
  doi="10.1109/CDC.2016.7798426",
  isbn="978-1-5090-1837-6",
  url="http://ieeexplore.ieee.org/document/7798426/"
}
Pracoviště
Kybernetika pro materiálové vědy (RG-2-02)
Nahoru