PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

How to optimize sample in active learning : Dispersion, an optimum criterion for classification ?
Benoit Gandar, Gaëlle Loosli and Guillaume Deffuant
In: ENBIS-EMSE 2009, 01-03 July 2009, Saint Etienne, France.

Abstract

We want generate learning data appropriated to classification problems. First, we show that theorical results about low discrepancy sequences in regression problems are not adequate for classification problems. Then, we show with theorical and experimental arguments that minimising the dispersion of the sample is a relevant strategy to optimize performance of classification learning.

EPrint Type:Conference or Workshop Item (Poster)
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Computational, Information-Theoretic Learning with Statistics
Learning/Statistics & Optimisation
ID Code:6040
Deposited By:Benoît Gandar
Deposited On:08 March 2010