PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

Minimising added classification error using walsh coefficients
Terry Windeatt and c zor
IEEE Transactions on Neural Networks Volume 22, Number 8, pp. 1334-1339, 2011.

Abstract

Two-class supervised learning in the context of a classifier ensemble may be formulated as learning an incompletely specified Boolean function, and the associated Walsh coefficients can be estimated without knowledge of the unspecified patterns. Using an extended version of the Tumer-Ghosh model, the relationship between Added Classification Error and second order Walsh coefficients is established. In this paper, the ensemble is composed of Multi-layer Perceptron (MLP) base classifiers, with the number of hidden nodes and epochs systematically varied. Experiments demonstrate that the mean second order coefficients peak at the same number of training epochs as ensemble test error reaches a minimum.

EPrint Type:Article
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Learning/Statistics & Optimisation
ID Code:9155
Deposited By:Terry Windeatt
Deposited On:21 February 2012