Please use this identifier to cite or link to this item:
Title: Online prediction of time series data with recurrent kernels
Authors: Xu, Zhao
Song, Qing
Haijin, Fan
Wang, Danwei
Keywords: DRNTU::Engineering::Electrical and electronic engineering
Issue Date: 2012
Source: 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).
Abstract: 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.
DOI: 10.1109/IJCNN.2012.6252747
Rights: © 2012 IEEE.
Fulltext Permission: none
Fulltext Availability: No Fulltext
Appears in Collections:EEE Conference Papers

Google ScholarTM




Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.