Enregistré dans:
Détails bibliographiques
Auteurs principaux: Neto, Afonso de Sá Delgado, Egger, Maximilian, Bakshi, Mayank, Bitar, Rawad
Format: Preprint
Publié: 2024
Sujets:
Accès en ligne:https://arxiv.org/abs/2406.14362
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866929392354066432
author Neto, Afonso de Sá Delgado
Egger, Maximilian
Bakshi, Mayank
Bitar, Rawad
author_facet Neto, Afonso de Sá Delgado
Egger, Maximilian
Bakshi, Mayank
Bitar, Rawad
contents We introduce CYBER-0, the first zero-order optimization algorithm for memory-and-communication efficient Federated Learning, resilient to Byzantine faults. We show through extensive numerical experiments on the MNIST dataset and finetuning RoBERTa-Large that CYBER-0 outperforms state-of-the-art algorithms in terms of communication and memory efficiency while reaching similar accuracy. We provide theoretical guarantees on its convergence for convex loss functions.
format Preprint
id arxiv_https___arxiv_org_abs_2406_14362
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Communication-Efficient Byzantine-Resilient Federated Zero-Order Optimization
Neto, Afonso de Sá Delgado
Egger, Maximilian
Bakshi, Mayank
Bitar, Rawad
Machine Learning
Artificial Intelligence
We introduce CYBER-0, the first zero-order optimization algorithm for memory-and-communication efficient Federated Learning, resilient to Byzantine faults. We show through extensive numerical experiments on the MNIST dataset and finetuning RoBERTa-Large that CYBER-0 outperforms state-of-the-art algorithms in terms of communication and memory efficiency while reaching similar accuracy. We provide theoretical guarantees on its convergence for convex loss functions.
title Communication-Efficient Byzantine-Resilient Federated Zero-Order Optimization
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2406.14362