PhD Seminar Course on

A Localized Generalization Error Bound for Classifier Systems

Cagliari, 1 July 2009

Instructor: Prof. Daniel Yeung
South China University of Technology (Guangzhou, China)
Duration: 1 hour
Schedule: 1 July 2009
Venue: Mocci Classroom
Topics: Generalization error model provides a theoretical support for a classifier's performance in terms of prediction accuracy. However, existing models give very loose error bounds. This explains why classification systems generally rely on experimental validation for their claims on prediction accuracy. In this talk we will revisit this problem and explore the idea of developing a new generalization error model based on the assumption that only prediction accuracy on unseen points in a neighbourhood of a training point will be considered, since it will be unreasonable to require a classifier to accurately predict unseen points "far away" from training samples. The new error model makes use of the concept of sensitivity measure for the multiplayer feedforward neural networks (Multilayer Perceptrons or Radial Basis Function Neural Networks). Its application to resolve the problem of feature selection for single RBFNN classifiers will be demonstrated. The error model for the case of multiple classifier systems will also be presented and discussed.
Organizer: Prof. Fabio Roli
Dep. of Electrical and Electronic Engineering
University of Cagliari, Italy