Decentralized learning with budgeted network load using Gaussian copulas and classifier ensembles - CNRS - Centre national de la recherche scientifique Accéder directement au contenu
Communication Dans Un Congrès Année : 2019

Decentralized learning with budgeted network load using Gaussian copulas and classifier ensembles

Résumé

We examine a network of learners which address the same classification task but must learn from different data sets. The learners can share a limited portion of their data sets so as to preserve the network load. We introduce DELCO (standing for Decentralized Ensemble Learning with COpulas), a new approach in which the shared data and the trained models are sent to a central machine that allows to build an ensemble of classifiers. The proposed method aggregates the base classifiers using a probabilistic model relying on Gaussian copulas. Experiments on logistic regressor ensembles demonstrate competing accuracy and increased robustness as compared to gold standard approaches. A companion python implementation can be downloaded at https://github.com/john-klein/DELCO
Fichier principal
Vignette du fichier
1804.10028.pdf (662.33 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01779989 , version 1 (09-12-2019)

Identifiants

Citer

John Klein, Mahmoud Albardan, Benjamin Guedj, Olivier Colot. Decentralized learning with budgeted network load using Gaussian copulas and classifier ensembles. ECML-PKDD, Decentralized Machine Learning at the Edge Workshop, Sep 2019, Wurzburg, Germany. ⟨10.1007/978-3-030-43823-4_26⟩. ⟨hal-01779989⟩
153 Consultations
217 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More