Fully Decentralized Joint Learning of Personalized Models and Collaboration Graphs - Archive ouverte HAL Access content directly
Conference Papers Year :

Fully Decentralized Joint Learning of Personalized Models and Collaboration Graphs

(1) , (2) , (2)
1
2

Abstract

We consider the fully decentralized machine learning scenario where many users with personal datasets collaborate to learn models through local peer-to-peer exchanges, without a central coordinator. We propose to train personalized models that leverage a collaboration graph describing the relationships between user personal tasks, which we learn jointly with the models. Our fully decentralized optimization procedure alternates between training nonlinear models given the graph in a greedy boosting manner, and updating the collaboration graph (with controlled sparsity) given the models. Throughout the process, users exchange messages only with a small number of peers (their direct neighbors when updating the models, and a few random users when updating the graph), ensuring that the procedure naturally scales with the number of users. Overall, our approach is communication-efficient and avoids exchanging personal data. We provide an extensive analysis of the convergence rate, memory and communication complexity of our approach, and demonstrate its benefits compared to competing techniques on synthetic and real datasets.
Fichier principal
Vignette du fichier
aistats20_graph.pdf (537.83 Ko) Télécharger le fichier
Vignette du fichier
aistats20_graph_supp.pdf (944.83 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-03100057 , version 1 (06-01-2021)

Identifiers

  • HAL Id : hal-03100057 , version 1

Cite

Valentina Zantedeschi, Aurélien Bellet, Marc Tommasi. Fully Decentralized Joint Learning of Personalized Models and Collaboration Graphs. AISTATS 2020 - The 23rd International Conference on Artificial Intelligence and Statistics, Aug 2020, Palerme / Virtual, Italy. ⟨hal-03100057⟩
55 View
115 Download

Share

Gmail Facebook Twitter LinkedIn More