%0 Conference Paper
%F Oral 
%T A pragmatic approach to multi-class classification
%+ Hochschule Ruhr West (HRW)
%+ Robotique et Vision (RV)
%+ Flowing Epigenetic Robots and Systems (Flowers)
%A Kopinski, Thomas
%A Magand, Stéphane
%A Handmann, Uwe
%A Gepperth, Alexander
%< avec comité de lecture
%B European Symposium on artificial neural networks (ESANN)
%C Bruges, Belgium
%8 2015-04
%D 2015
%Z 1601.01121
%R 10.1109/IJCNN.2015.7280768
%Z Computer Science [cs]/Machine Learning [cs.LG]Conference papers
%X We present a novel hierarchical approach to multi-class classification which is generic in that it can be applied to different classification models (e.g., support vector machines, perceptrons), and makes no explicit assumptions about the probabilistic structure of the problem as it is usually done in multi-class classification. By adding a cascade of additional classifiers, each of which receives the previous classifier's output in addition to regular input data, the approach harnesses unused information that manifests itself in the form of, e.g., correlations between predicted classes. Using multilayer perceptrons as a classification model, we demonstrate the validity of this approach by testing it on a complex ten-class 3D gesture recognition task.
%G English
%2 https://hal-ensta-paris.archives-ouvertes.fr/hal-01251382/document
%2 https://hal-ensta-paris.archives-ouvertes.fr/hal-01251382/file/root.pdf
%L hal-01251382
%U https://hal-ensta-paris.archives-ouvertes.fr/hal-01251382
%~ ENSTA
%~ INRIA
%~ INRIA-BORDEAUX
%~ INRIA_TEST
%~ TESTALAIN1
%~ ENSTA_U2IS
%~ TESTBORDEAUX
%~ TESTBORDEAUX2
%~ INRIA2