PP-QADMM: A Dual-Driven Perturbation and Quantized ADMM for Privacy Preserving and Communication-Efficient Federated Learning

This article presents Privacy-Preserving and Quantized ADMM (PP-QADMM), a novel federated learning (FL) algorithm that is both privacy-preserving and communication-efficient, built upon the Alternating Direction Method of Multipliers (ADMM). PP-QADMM enhances privacy through three core mechanisms. F...

Descrizione completa

Dettagli Bibliografici

Autore principale:	Anis Elgabli
Natura:	Artigo
Lingua:	Inglês
Pubblicazione:	IEEE 2025-01-01
Serie:	IEEE Open Journal of the Communications Society
Soggetti:	Communication-efficient federated learning differential privacy ADMM stochastic quantization secure aggregation
Accesso online:	https://ieeexplore.ieee.org/document/10982181/
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!