Boosting Multi-Task Weak Learners with Applications to Textual and Social Data

Jean-Baptiste Faddoul 1 Boris Chidlovskii 1 Fabien Torre 2, 3 Rémi Gilleron 2, 3, 4
1 Xerox Research Centre Europe [Meylan]
2 LIFL - Laboratoire d'Informatique Fondamentale de Lille
3 MOSTRARE - Modeling Tree Structures, Machine Learning, and Information Extraction
LIFL - Laboratoire d'Informatique Fondamentale de Lille, Inria Lille - Nord Europe
4 GRAppA - LIFL - Groupe de Recherche en Apprentissage Automatique
Abstract : Learning multiple related tasks from data simultaneously can improve predictive performance relative to learning these tasks independently. In this paper we propose a novel multi-task learning algorithm called MT-Adaboost: it extends Adaboost algorithm to the multi-task setting; it uses as multi-task weak classifier a multi-task decision stump. This allows to learn different dependencies between tasks for different regions of the learning space. Thus, we relax the conventional hypothesis that tasks behave similarly in the whole learning space. Moreover, MT-Adaboost can learn multiple tasks without imposing the constraint of sharing the same label set and/or examples between tasks. A theoretical analysis is derived from the analysis of the original Adaboost. Experiments for multiple tasks over large scale textual data sets with social context (Enron and Tobacco) give rise to very promising results.
Keywords : Textual and social data Multi-task learning Boosting Decision trees Textual and social data.
Type de document :
Communication dans un congrès
The Ninth International Conference on Machine Learning and Applications (ICMLA 2010), Dec 2010, Hayatt Regency Bethesda, Washington DC, United States. IEEE, 2010
Liste complète des métadonnées

Littérature citée [15 références]  Voir  Masquer  Télécharger
https://hal.inria.fr/inria-00524718
Contributeur : Fabien Torre <>
Soumis le : mardi 19 octobre 2010 - 21:50:53
Dernière modification le : jeudi 11 janvier 2018 - 06:22:13
Document(s) archivé(s) le : vendredi 2 décembre 2016 - 01:08:42

Fichier

MTboostICMLA2010.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : inria-00524718, version 2

Collections

INRIA | UNIV-LILLE3 | MOSTRARE | LIFL

Citation

Jean-Baptiste Faddoul, Boris Chidlovskii, Fabien Torre, Rémi Gilleron. Boosting Multi-Task Weak Learners with Applications to Textual and Social Data. The Ninth International Conference on Machine Learning and Applications (ICMLA 2010), Dec 2010, Hayatt Regency Bethesda, Washington DC, United States. IEEE, 2010. 〈inria-00524718v2〉

Exporter

BibTeX TEI DC DCterms EndNote

Partager

Métriques

Consultations de la notice

1166

Téléchargements de fichiers

327

Contact


archive-ouverte@inria.fr