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

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.
Document type :
Conference papers
Complete list of metadatas

Cited literature [15 references]  Display  Hide  Download

https://hal.inria.fr/inria-00524718
Contributor : Fabien Torre <>
Submitted on : Tuesday, October 19, 2010 - 9:50:53 PM
Last modification on : Thursday, February 21, 2019 - 10:52:49 AM
Long-term archiving on : Friday, December 2, 2016 - 1:08:42 AM

File

MTboostICMLA2010.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : inria-00524718, version 2

Collections

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. ⟨inria-00524718v2⟩

Share

Metrics

Record views

1680

Files downloads

648