Multivariate Linear Regression on Classifier Outputs: a Capacity Study

Yann Guermeur (1) , Hélène Paugam-Moisy , Patrick Gallinari (1)

1 APA - Apprentissage et Acquisition des connaissances

We consider the problem of combining the outputs of severed classifiers trained independently to perform a discrimination task, in order to improve the prediction accuracy of individual classifiers. We briefly describe the multivariate linear regression model which has already been implemented successfully for that purpose and we study its capacity, using generalizations of the notion of VC dimension.

Domaines: Informatique [cs]

Yann Guermeur, Hélène Paugam-Moisy, Patrick Gallinari. Multivariate Linear Regression on Classifier Outputs: a Capacity Study. ICANN 1998 - 8th International Conference of Artificial Neural Networks, Sep 1998, Skövde, Sweden. pp.693-698, ⟨10.1007/978-1-4471-1599-1_106⟩. ⟨hal-01617494⟩