On a New Class of Framelet Kernels for Support Vector Regression and Regularization Networks

  • Wei-Feng Zhang
  • Dao-Qing Dai
  • Hong Yan
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4426)

Abstract

Kernel-based machine learning techniques, such as support vector machines, regularization networks, have been widely used in pattern analysis. Kernel function plays an important role in the design of such learning machines. The choice of an appropriate kernel is critical in order to obtain good performance. This paper presents a new class of kernel functions derived from framelet. Framelet is a wavelet frame constructed via multiresolution analysis, and has both the merit of frame and wavelet. The usefulness of the new kernels is demonstrated through simulation experiments.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Springer Berlin Heidelberg 2007

Authors and Affiliations

  • Wei-Feng Zhang
    • 1
  • Dao-Qing Dai
    • 1
  • Hong Yan
    • 2
  1. 1.Center for Computer Vision and Department of Mathematics, Sun Yat-Sen (Zhongshan) University, Guangzhou 510275China
  2. 2.Department of Electronic Engineering, City University of Hong Kong, 83 Tat Chee Avenue, KowloonHong Kong

Personalised recommendations