A better metric in kernel adaptive filtering

Airi Takeuchi, Masahiro Yukawa, Klaus Muller

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

The metric in the reproducing kernel Hilbert space (RKHS) is known to be given by the Gram matrix (which is also called the kernel matrix). It has been reported that the metric leads to a decorrelation of the kernelized input vector because its autocorrelation matrix can be approximated by the (down scaled) squared Gram matrix subject to some condition. In this paper, we derive a better metric (a best one under the condition) based on the approximation, and present an adaptive algorithm using the metric. Although the algorithm has quadratic complexity, we present its linear-complexity version based on a selective updating strategy. Numerical examples validate the approximation in a practical scenario, and show that the proposed metric yields fast convergence and tracking performance.

Original languageEnglish
Title of host publication2016 24th European Signal Processing Conference, EUSIPCO 2016
PublisherEuropean Signal Processing Conference, EUSIPCO
Pages1578-1582
Number of pages5
Volume2016-November
ISBN (Electronic)9780992862657
DOIs
Publication statusPublished - 2016 Nov 28
Event24th European Signal Processing Conference, EUSIPCO 2016 - Budapest, Hungary
Duration: 2016 Aug 282016 Sep 2

Other

Other24th European Signal Processing Conference, EUSIPCO 2016
CountryHungary
CityBudapest
Period16/8/2816/9/2

Fingerprint

Adaptive filtering
Hilbert spaces
Adaptive algorithms
Autocorrelation

ASJC Scopus subject areas

  • Signal Processing
  • Electrical and Electronic Engineering

Cite this

Takeuchi, A., Yukawa, M., & Muller, K. (2016). A better metric in kernel adaptive filtering. In 2016 24th European Signal Processing Conference, EUSIPCO 2016 (Vol. 2016-November, pp. 1578-1582). [7760514] European Signal Processing Conference, EUSIPCO. https://doi.org/10.1109/EUSIPCO.2016.7760514

A better metric in kernel adaptive filtering. / Takeuchi, Airi; Yukawa, Masahiro; Muller, Klaus.

2016 24th European Signal Processing Conference, EUSIPCO 2016. Vol. 2016-November European Signal Processing Conference, EUSIPCO, 2016. p. 1578-1582 7760514.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Takeuchi, A, Yukawa, M & Muller, K 2016, A better metric in kernel adaptive filtering. in 2016 24th European Signal Processing Conference, EUSIPCO 2016. vol. 2016-November, 7760514, European Signal Processing Conference, EUSIPCO, pp. 1578-1582, 24th European Signal Processing Conference, EUSIPCO 2016, Budapest, Hungary, 16/8/28. https://doi.org/10.1109/EUSIPCO.2016.7760514
Takeuchi A, Yukawa M, Muller K. A better metric in kernel adaptive filtering. In 2016 24th European Signal Processing Conference, EUSIPCO 2016. Vol. 2016-November. European Signal Processing Conference, EUSIPCO. 2016. p. 1578-1582. 7760514 https://doi.org/10.1109/EUSIPCO.2016.7760514
Takeuchi, Airi ; Yukawa, Masahiro ; Muller, Klaus. / A better metric in kernel adaptive filtering. 2016 24th European Signal Processing Conference, EUSIPCO 2016. Vol. 2016-November European Signal Processing Conference, EUSIPCO, 2016. pp. 1578-1582
@inproceedings{8eb94058c0d1408a8591612963eeda69,
title = "A better metric in kernel adaptive filtering",
abstract = "The metric in the reproducing kernel Hilbert space (RKHS) is known to be given by the Gram matrix (which is also called the kernel matrix). It has been reported that the metric leads to a decorrelation of the kernelized input vector because its autocorrelation matrix can be approximated by the (down scaled) squared Gram matrix subject to some condition. In this paper, we derive a better metric (a best one under the condition) based on the approximation, and present an adaptive algorithm using the metric. Although the algorithm has quadratic complexity, we present its linear-complexity version based on a selective updating strategy. Numerical examples validate the approximation in a practical scenario, and show that the proposed metric yields fast convergence and tracking performance.",
author = "Airi Takeuchi and Masahiro Yukawa and Klaus Muller",
year = "2016",
month = "11",
day = "28",
doi = "10.1109/EUSIPCO.2016.7760514",
language = "English",
volume = "2016-November",
pages = "1578--1582",
booktitle = "2016 24th European Signal Processing Conference, EUSIPCO 2016",
publisher = "European Signal Processing Conference, EUSIPCO",

}

TY - GEN

T1 - A better metric in kernel adaptive filtering

AU - Takeuchi, Airi

AU - Yukawa, Masahiro

AU - Muller, Klaus

PY - 2016/11/28

Y1 - 2016/11/28

N2 - The metric in the reproducing kernel Hilbert space (RKHS) is known to be given by the Gram matrix (which is also called the kernel matrix). It has been reported that the metric leads to a decorrelation of the kernelized input vector because its autocorrelation matrix can be approximated by the (down scaled) squared Gram matrix subject to some condition. In this paper, we derive a better metric (a best one under the condition) based on the approximation, and present an adaptive algorithm using the metric. Although the algorithm has quadratic complexity, we present its linear-complexity version based on a selective updating strategy. Numerical examples validate the approximation in a practical scenario, and show that the proposed metric yields fast convergence and tracking performance.

AB - The metric in the reproducing kernel Hilbert space (RKHS) is known to be given by the Gram matrix (which is also called the kernel matrix). It has been reported that the metric leads to a decorrelation of the kernelized input vector because its autocorrelation matrix can be approximated by the (down scaled) squared Gram matrix subject to some condition. In this paper, we derive a better metric (a best one under the condition) based on the approximation, and present an adaptive algorithm using the metric. Although the algorithm has quadratic complexity, we present its linear-complexity version based on a selective updating strategy. Numerical examples validate the approximation in a practical scenario, and show that the proposed metric yields fast convergence and tracking performance.

UR - http://www.scopus.com/inward/record.url?scp=85006042953&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85006042953&partnerID=8YFLogxK

U2 - 10.1109/EUSIPCO.2016.7760514

DO - 10.1109/EUSIPCO.2016.7760514

M3 - Conference contribution

AN - SCOPUS:85006042953

VL - 2016-November

SP - 1578

EP - 1582

BT - 2016 24th European Signal Processing Conference, EUSIPCO 2016

PB - European Signal Processing Conference, EUSIPCO

ER -