Publication detail

Recursive Variational Inference for Total Least-Squares

FRIML, D. VÁCLAVEK, P.

Original Title

Recursive Variational Inference for Total Least-Squares

Type

journal article in Web of Science

Language

English

Original Abstract

This letter analyzes methods for deriving credible intervals to facilitate errors-in-variables identification by expanding on Bayesian total least squares. The credible intervals are approximated employing Laplace and variational approximations of the intractable posterior density function. Three recursive identification algorithms providing an approximation of the credible intervals for inference with the Bingham and the Gaussian priors are proposed. The introduced algorithms are evaluated on numerical experiments, and a practical example of application on battery cell total capacity estimation compared to the state-of-the-art algorithms is presented.

Keywords

Bayes methods; parameter estimation; identification; variational methods

Authors

FRIML, D.; VÁCLAVEK, P.

Released

26. 6. 2023

Publisher

IEEE

Location

PISCATAWAY

ISBN

2475-1456

Periodical

IEEE Control Systems Letters

Year of study

7

Number

1

State

United States of America

Pages from

2839

Pages to

2844

Pages count

6

URL

https://ieeexplore.ieee.org/document/10163935

Full text in the Digital Library

http://hdl.handle.net/11012/244278

BibTex

@article{BUT184309,
  author="Dominik {Friml} and Pavel {Václavek}",
  title="Recursive Variational Inference for Total Least-Squares",
  journal="IEEE Control Systems Letters",
  year="2023",
  volume="7",
  number="1",
  pages="2839--2844",
  doi="10.1109/LCSYS.2023.3289608",
  issn="2475-1456",
  url="https://ieeexplore.ieee.org/document/10163935"
}