Maximal-discrepancy bounds for regularized classifiers

  • Authors:
  • Sergio Decherchi;Paolo Gastaldo;Judith Redi;Rodolfo Zunino

  • Affiliations:
  • Department of Biophysical and Electronic Engineering, University of Genoa, Genova, Italy;Department of Biophysical and Electronic Engineering, University of Genoa, Genova, Italy;Department of Biophysical and Electronic Engineering, University of Genoa, Genova, Italy;Department of Biophysical and Electronic Engineering, University of Genoa, Genova, Italy

  • Venue:
  • IJCNN'09 Proceedings of the 2009 international joint conference on Neural Networks
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

Regularized Classifiers such as SVM or RLS are among the most used and successful classifiers in machine learning. The theory and the empirical evaluation of the associate generalization bounds are of paramount importance; bounds based on the Maximal-Discrepancy approach proved quite effective. The paper shows an efficient, iterative procedure to evaluate Maximal-Discrepancy bounds for this kind of classifiers. Empirical results on UCI datasets show that this approach can attain tighter bounds to the run-time classification error.