Ranking Variables Based on Goodness of Fits in Nonlinear Nonparametric System Identification

Er-Wei Bai; Changmin Cheng

doi:10.1109/MED.2018.8442944

Back

Conference proceeding

Ranking Variables Based on Goodness of Fits in Nonlinear Nonparametric System Identification

Er-Wei Bai and Changmin Cheng

2018 26th Mediterranean Conference on Control and Automation (MED), pp.1-6

06/2018

DOI: 10.1109/MED.2018.8442944

View Online

Abstract

AIdentification of a high dimensional nonlinear nonparametric system is costly. On the other hand for many real-world problems, they are sparse in the sense that not all variables contribute or contribute significantly. If these variables that do not contribute or contribute little can be identified and removed prior to system identification, the identification problem is of lower dimension. In this paper, methods to rank variables based on the Goodness of Fits are proposed without full scale identification.

Additives

Analysis of variance

Estimation

Fans

Input variables

Nonlinear systems

Optical fibers

Details

Title: Subtitle: Ranking Variables Based on Goodness of Fits in Nonlinear Nonparametric System Identification
Creators: Er-Wei Bai - University of Iowa
Changmin Cheng - Shanghai Jiao Tong University
Resource Type: Conference proceeding
Publication Details: 2018 26th Mediterranean Conference on Control and Automation (MED), pp.1-6
DOI: 10.1109/MED.2018.8442944
eISSN: 2473-3504
Publisher: IEEE
Language: English
Date published: 06/2018
Academic Unit: Electrical and Computer Engineering
Record Identifier: 9984197265202771

Metrics

15 Record Views