Orthogonal Maximum Correntropy Learning

Mingfei Lu; Badong Chen

doi:10.1109/MLSP55214.2022.9943496

Orthogonal Maximum Correntropy Learning

Mingfei Lu
, Badong Chen

Faculty of Electronic and Information Engineering

Xi'an Jiaotong University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

One-hidden-layer networks are very efficient to solve structure-unknown system identification/modeling problems. But it is difficult and often redundant to choose a basis without prior knowledge. We present in this paper a new algorithm named orthogonal maximum correntropy (OMC), with which a robust and parsimonious system-identification/model-selection process is reached. Moreover, the proposed method is capable of suppressing non-Gaussian, especially heavy-tailed noises that may be encountered in the datasets, benefiting from the maximum correntropy criterion. Experiments on some common models such as polynomial NARx, RBF networks, and wavelet networks, demonstrate the effectiveness of the proposed method.

Original language	English
Title of host publication	2022 IEEE 32nd International Workshop on Machine Learning for Signal Processing, MLSP 2022
Publisher	IEEE Computer Society
ISBN (Electronic)	9781665485470
DOIs	https://doi.org/10.1109/MLSP55214.2022.9943496
State	Published - 2022
Event	32nd IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2022 - Xi'an, China Duration: 22 Aug 2022 → 25 Aug 2022

Publication series

Name	IEEE International Workshop on Machine Learning for Signal Processing, MLSP
Volume	2022-August
ISSN (Print)	2161-0363
ISSN (Electronic)	2161-0371

Conference

Conference	32nd IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2022
Country/Territory	China
City	Xi'an
Period	22/08/22 → 25/08/22

Keywords

model selection
non-Gaussian noise
orthogonal maximum correntropy
system identification

Access to Document

10.1109/MLSP55214.2022.9943496

Cite this

@inproceedings{ef34b11cb5754263aa9404fa8a0345be,

title = "Orthogonal Maximum Correntropy Learning",

abstract = "One-hidden-layer networks are very efficient to solve structure-unknown system identification/modeling problems. But it is difficult and often redundant to choose a basis without prior knowledge. We present in this paper a new algorithm named orthogonal maximum correntropy (OMC), with which a robust and parsimonious system-identification/model-selection process is reached. Moreover, the proposed method is capable of suppressing non-Gaussian, especially heavy-tailed noises that may be encountered in the datasets, benefiting from the maximum correntropy criterion. Experiments on some common models such as polynomial NARx, RBF networks, and wavelet networks, demonstrate the effectiveness of the proposed method.",

keywords = "model selection, non-Gaussian noise, orthogonal maximum correntropy, system identification",

author = "Mingfei Lu and Badong Chen",

note = "Publisher Copyright: {\textcopyright} 2022 IEEE.; 32nd IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2022 ; Conference date: 22-08-2022 Through 25-08-2022",

year = "2022",

doi = "10.1109/MLSP55214.2022.9943496",

language = "英语",

series = "IEEE International Workshop on Machine Learning for Signal Processing, MLSP",

publisher = "IEEE Computer Society",

booktitle = "2022 IEEE 32nd International Workshop on Machine Learning for Signal Processing, MLSP 2022",

}

Lu, M & Chen, B 2022, Orthogonal Maximum Correntropy Learning. in 2022 IEEE 32nd International Workshop on Machine Learning for Signal Processing, MLSP 2022. IEEE International Workshop on Machine Learning for Signal Processing, MLSP, vol. 2022-August, IEEE Computer Society, 32nd IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2022, Xi'an, China, 22/08/22. https://doi.org/10.1109/MLSP55214.2022.9943496

Orthogonal Maximum Correntropy Learning. / Lu, Mingfei; Chen, Badong.
2022 IEEE 32nd International Workshop on Machine Learning for Signal Processing, MLSP 2022. IEEE Computer Society, 2022. (IEEE International Workshop on Machine Learning for Signal Processing, MLSP; Vol. 2022-August).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Orthogonal Maximum Correntropy Learning

AU - Lu, Mingfei

AU - Chen, Badong

PY - 2022

Y1 - 2022

N2 - One-hidden-layer networks are very efficient to solve structure-unknown system identification/modeling problems. But it is difficult and often redundant to choose a basis without prior knowledge. We present in this paper a new algorithm named orthogonal maximum correntropy (OMC), with which a robust and parsimonious system-identification/model-selection process is reached. Moreover, the proposed method is capable of suppressing non-Gaussian, especially heavy-tailed noises that may be encountered in the datasets, benefiting from the maximum correntropy criterion. Experiments on some common models such as polynomial NARx, RBF networks, and wavelet networks, demonstrate the effectiveness of the proposed method.

AB - One-hidden-layer networks are very efficient to solve structure-unknown system identification/modeling problems. But it is difficult and often redundant to choose a basis without prior knowledge. We present in this paper a new algorithm named orthogonal maximum correntropy (OMC), with which a robust and parsimonious system-identification/model-selection process is reached. Moreover, the proposed method is capable of suppressing non-Gaussian, especially heavy-tailed noises that may be encountered in the datasets, benefiting from the maximum correntropy criterion. Experiments on some common models such as polynomial NARx, RBF networks, and wavelet networks, demonstrate the effectiveness of the proposed method.

KW - model selection

KW - non-Gaussian noise

KW - orthogonal maximum correntropy

KW - system identification

UR - https://www.scopus.com/pages/publications/85142737138

U2 - 10.1109/MLSP55214.2022.9943496

DO - 10.1109/MLSP55214.2022.9943496

M3 - 会议稿件

AN - SCOPUS:85142737138

T3 - IEEE International Workshop on Machine Learning for Signal Processing, MLSP

BT - 2022 IEEE 32nd International Workshop on Machine Learning for Signal Processing, MLSP 2022

PB - IEEE Computer Society

T2 - 32nd IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2022

Y2 - 22 August 2022 through 25 August 2022

ER -

Orthogonal Maximum Correntropy Learning

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this