Sliding window online kernel-based classification by projection mappings

Konstantinos Slavakis, Sergios Theodoridis

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Very recently, an adaptive projection algorithm was introduced for the online classification task with sparsification in Reproducing Kernel Hilbert Spaces (RKHS). This paper presents another sparsification method for the projection-based approach by generating a sequence of linear subspaces in RKHS. Projection mappings give a geometrical flavor to the design; classification is performed by metric projection mappings, sparsification is achieved by orthogonal projections, while the online system's memory and tracking requirements are attained by oblique projections. The resulting sparsification scheme shows strong similarities with the classical sliding window adaptive schemes. Validation is performed by considering the adaptive equalization problem of a nonlinear communication channel. Although here the classification scheme is considered, the method is readily extended to regression tasks. Furthermore its generality allows for a number of cost functions including non-differentiable ones.

Original languageEnglish (US)
Title of host publication2008 IEEE International Symposium on Circuits and Systems, ISCAS 2008
Pages49-52
Number of pages4
DOIs
StatePublished - Sep 19 2008
Event2008 IEEE International Symposium on Circuits and Systems, ISCAS 2008 - Seattle, WA, United States
Duration: May 18 2008May 21 2008

Other

Other2008 IEEE International Symposium on Circuits and Systems, ISCAS 2008
CountryUnited States
CitySeattle, WA
Period5/18/085/21/08

Keywords

  • Kernel methods
  • Online algorithms
  • Projections
  • Sliding window
  • Sparsification

Fingerprint Dive into the research topics of 'Sliding window online kernel-based classification by projection mappings'. Together they form a unique fingerprint.

Cite this