Decentralized Collaborative Learning of Personalized Models over Networks

Paul Vanhaesebrouck 1 Aurélien Bellet 1 Marc Tommasi 1, 2
1 MAGNET - Machine Learning in Information Networks
Inria Lille - Nord Europe, CRIStAL - Centre de Recherche en Informatique, Signal et Automatique de Lille (CRIStAL) - UMR 9189
2 Université de Lille
Abstract : We consider a set of learning agents in a col-laborative peer-to-peer network, where each agent learns a personalized model according to its own learning objective. The question addressed in this paper is: how can agents improve upon their locally trained model by communicating with other agents that have similar objectives? We introduce and analyze two asynchronous gossip algorithms running in a fully decentralized manner. Our first approach , inspired from label propagation, aims to smooth pre-trained local models over the network while accounting for the confidence that each agent has in its initial model. In our second approach, agents jointly learn and propagate their model by making iterative updates based on both their local dataset and the behavior of their neighbors. To optimize this challenging objective, our decentralized algorithm is based on ADMM.
Document type :
Conference papers
Complete list of metadatas

Cited literature [21 references]  Display  Hide  Download
https://hal.inria.fr/hal-01533182
Contributor : Team Magnet <>
Submitted on : Tuesday, June 6, 2017 - 9:29:21 AM
Last modification on : Tuesday, September 10, 2019 - 11:32:02 AM
Long-term archiving on : Thursday, September 7, 2017 - 12:32:31 PM

File

supplementary.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01533182, version 1
  • ARXIV : 1610.05202

Collections

INRIA | CRISTAL-MAGNET | CRISTAL | UNIV-LILLE3 | UNIV-LILLE

Citation

Paul Vanhaesebrouck, Aurélien Bellet, Marc Tommasi. Decentralized Collaborative Learning of Personalized Models over Networks. International Conference on Artificial Intelligence and Statistics (AISTATS), Apr 2017, Fort Lauderdale, Florida., United States. ⟨hal-01533182⟩

Export

BibTeX TEI DC DCterms EndNote

Share

Metrics

Record views

616

Files downloads

459

Contact


archive-ouverte@inria.fr