Learning from Data Streams Using Kernel Adaptive Filtering

27 Pages Posted: 8 Jan 2019

See all articles by Sergio Garcia-Vega

Sergio Garcia-Vega

The University of Manchester - School of Computer Science

Xiao-Jun Zeng

The University of Manchester

John Keane

The University of Manchester - School of Computer Science

Date Written: December 24, 2018

Abstract

A learning task is sequential if its data samples become available over time. Kernel adaptive filters (KAF) are sequential learning algorithms. There are two main challenges in KAF: (1) the lack of an effective method to determine the kernel-sizes in the online learning context; (2) how to tune the step-size parameter. We propose a framework for online prediction using KAF which does not require a predefined set of kernel-sizes; rather, the kernel-sizes are both created and updated in an online sequential way. Further, to improve convergence time, we propose an online technique to optimize the step-size parameter. The framework is tested on two real-world data sets, i.e., internet traffic and foreign exchange market. Results show that, without any specific hyperparameter tuning, our proposal converges faster to relatively low values of mean squared error and achieves better accuracy.

Keywords: Learning From Data Streams, Sequence Prediction, Kernel Adaptive Filters, Kernel Least Mean Square

Suggested Citation

Garcia-Vega, Sergio and Zeng, Xiao-Jun and Keane, John, Learning from Data Streams Using Kernel Adaptive Filtering (December 24, 2018). Available at SSRN: https://ssrn.com/abstract=3306245 or http://dx.doi.org/10.2139/ssrn.3306245

Sergio Garcia-Vega (Contact Author)

The University of Manchester - School of Computer Science ( email )

Oxford Road
Manchester, M13 9PL
United Kingdom

Xiao-Jun Zeng

The University of Manchester ( email )

Oxford Road
Manchester, N/A M13 9PL
United Kingdom

John Keane

The University of Manchester - School of Computer Science ( email )

Kilburn Building, Oxford Road
Manchester M13 9GH, M13 9PL
United Kingdom

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
80
Abstract Views
625
Rank
502,856
PlumX Metrics