Section: New Results

Axis 2: Decentralized learning with budgeted network load using Gaussian copulas and classifier ensembles

Participant: Benjamin Guedj

We examine a network of learners which address the same classification task but must learn from different data sets. The learners cannot share data but instead share their models. Models are shared only one time so as to preserve the network load. We introduce DELCO (standing for Decentralized Ensemble Learning with COpulas), a new approach allowing to aggregate the predictions of the classifiers trained by each learner. 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 in case of dependent classifiers. A companion python implementation is available online.

Joint work with John Klein, Olivier Colot, Mahmoud Albardan (Université de Lille).

This work has been published in the proceedings of ECML-PKDD 2019, as part (oral presentation) of the workshop Decentralized Machine Learning at the Edge. Published: [34]