Accéder directement au contenu Accéder directement à la navigation
Communication dans un congrès

Comment générer les meilleurs échantillons a faible dispersion pour l'apprentissage actif en classification ?

Abstract : We consider a problem of active learning classification: we suppose we can determine, with an oracle, the label of any point in a given compact set, and we want to generate a sample of a given size which will allow us to get the best approximation of the oracle function. It's well known that the more numerous the data are, the best quality the modelling is. However obtaining data can be expensive or destructive in consequence we want to get the best value from this investment. We have to choose the best learning set. The first contribution of this paper is to state that dispersion is the most relevant criterion for generating samples in active classification leanring whereas discrepance is the relevant criterion for active regression learning. However low dispersion samples are not easy to generate. The second contribution consists then in making a study of different ways to proceed and in proposing a new algorithm.
Type de document :
Communication dans un congrès
Liste complète des métadonnées

https://hal.inrae.fr/hal-02594341
Déposant : Migration Irstea Publications <>
Soumis le : vendredi 15 mai 2020 - 18:24:18
Dernière modification le : mercredi 14 octobre 2020 - 03:56:16

Identifiants

  • HAL Id : hal-02594341, version 1
  • IRSTEA : PUB00030736

Collections

Citation

Benoît Gandar, G. Loosli, Guillaume Deffuant. Comment générer les meilleurs échantillons a faible dispersion pour l'apprentissage actif en classification ?. Active Learning and Experimental Design Workshop AISTATS, May 2010, Sardaigne, France. pp.16. ⟨hal-02594341⟩

Partager

Métriques

Consultations de la notice

14