Online prediction of time series data with recurrent kernels

We propose a robust recurrent kernel online learning (RRKOL) algorithm which allows the exploitation of the kernel trick in an online fashion. The novel RRKOL algorithm achieves guaranteed weight convergence with regularized risk management through the recurrent hyper-parameters for a superior gener...

Full description

Saved in:

Bibliographic Details
Main Authors:	Xu, Zhao, Song, Qing, Haijin, Fan, Wang, Danwei
Other Authors:	School of Electrical and Electronic Engineering
Format:	Conference or Workshop Item
Language:	English
Published:	2013
Subjects:	DRNTU::Engineering::Electrical and electronic engineering
Online Access:	https://hdl.handle.net/10356/98301 http://hdl.handle.net/10220/12420
Tags:	Add Tag No Tags, Be the first to tag this record!
Institution:	Nanyang Technological University
Language:	English

id	sg-ntu-dr.10356-98301
record_format	dspace
spelling	sg-ntu-dr.10356-983012020-03-07T13:24:48Z Online prediction of time series data with recurrent kernels Xu, Zhao Song, Qing Haijin, Fan Wang, Danwei School of Electrical and Electronic Engineering International Joint Conference on Neural Networks (2012 : Brisbane, Australia) DRNTU::Engineering::Electrical and electronic engineering We propose a robust recurrent kernel online learning (RRKOL) algorithm which allows the exploitation of the kernel trick in an online fashion. The novel RRKOL algorithm achieves guaranteed weight convergence with regularized risk management through the recurrent hyper-parameters for a superior generalization performance. To select useful data to be learned and remove redundant ones, a sparcification procedure is developed based on the stability analysis of the system. Two time-series prediction examples are presented. 2013-07-29T03:18:43Z 2019-12-06T19:53:21Z 2013-07-29T03:18:43Z 2019-12-06T19:53:21Z 2012 2012 Conference Paper Xu, Z., Song, Q., Haijin, F., & Wang, D. (2012). Online prediction of time series data with recurrent kernels. The 2012 International Joint Conference on Neural Networks (IJCNN). https://hdl.handle.net/10356/98301 http://hdl.handle.net/10220/12420 10.1109/IJCNN.2012.6252747 en © 2012 IEEE.
institution	Nanyang Technological University
building	NTU Library
country	Singapore
collection	DR-NTU
language	English
topic	DRNTU::Engineering::Electrical and electronic engineering
spellingShingle	DRNTU::Engineering::Electrical and electronic engineering Xu, Zhao Song, Qing Haijin, Fan Wang, Danwei Online prediction of time series data with recurrent kernels
description	We propose a robust recurrent kernel online learning (RRKOL) algorithm which allows the exploitation of the kernel trick in an online fashion. The novel RRKOL algorithm achieves guaranteed weight convergence with regularized risk management through the recurrent hyper-parameters for a superior generalization performance. To select useful data to be learned and remove redundant ones, a sparcification procedure is developed based on the stability analysis of the system. Two time-series prediction examples are presented.
author2	School of Electrical and Electronic Engineering
author_facet	School of Electrical and Electronic Engineering Xu, Zhao Song, Qing Haijin, Fan Wang, Danwei
format	Conference or Workshop Item
author	Xu, Zhao Song, Qing Haijin, Fan Wang, Danwei
author_sort	Xu, Zhao
title	Online prediction of time series data with recurrent kernels
title_short	Online prediction of time series data with recurrent kernels
title_full	Online prediction of time series data with recurrent kernels
title_fullStr	Online prediction of time series data with recurrent kernels
title_full_unstemmed	Online prediction of time series data with recurrent kernels
title_sort	online prediction of time series data with recurrent kernels
publishDate	2013
url	https://hdl.handle.net/10356/98301 http://hdl.handle.net/10220/12420
_version_	1681043959875371008

Online prediction of time series data with recurrent kernels

Similar Items