MimiC: Combating Client Dropouts in Federated Learning by Mimicking Central Updates

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Sun, Yuchang, Mao, Yuyi, Zhang, Jun
Natura: Preprint
Pubblicazione: 2023
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866913303288086528
author Sun, Yuchang
Mao, Yuyi
Zhang, Jun
author_facet Sun, Yuchang
Mao, Yuyi
Zhang, Jun
contents Federated learning (FL) is a promising framework for privacy-preserving collaborative learning, where model training tasks are distributed to clients and only the model updates need to be collected at a server. However, when being deployed at mobile edge networks, clients may have unpredictable availability and drop out of the training process, which hinders the convergence of FL. This paper tackles such a critical challenge. Specifically, we first investigate the convergence of the classical FedAvg algorithm with arbitrary client dropouts. We find that with the common choice of a decaying learning rate, FedAvg oscillates around a stationary point of the global loss function, which is caused by the divergence between the aggregated and desired central update. Motivated by this new observation, we then design a novel training algorithm named MimiC, where the server modifies each received model update based on the previous ones. The proposed modification of the received model updates mimics the imaginary central update irrespective of dropout clients. The theoretical analysis of MimiC shows that divergence between the aggregated and central update diminishes with proper learning rates, leading to its convergence. Simulation results further demonstrate that MimiC maintains stable convergence performance and learns better models than the baseline methods.
format Preprint
id arxiv_https___arxiv_org_abs_2306_12212
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle MimiC: Combating Client Dropouts in Federated Learning by Mimicking Central Updates
Sun, Yuchang
Mao, Yuyi
Zhang, Jun
Machine Learning
Distributed, Parallel, and Cluster Computing
Federated learning (FL) is a promising framework for privacy-preserving collaborative learning, where model training tasks are distributed to clients and only the model updates need to be collected at a server. However, when being deployed at mobile edge networks, clients may have unpredictable availability and drop out of the training process, which hinders the convergence of FL. This paper tackles such a critical challenge. Specifically, we first investigate the convergence of the classical FedAvg algorithm with arbitrary client dropouts. We find that with the common choice of a decaying learning rate, FedAvg oscillates around a stationary point of the global loss function, which is caused by the divergence between the aggregated and desired central update. Motivated by this new observation, we then design a novel training algorithm named MimiC, where the server modifies each received model update based on the previous ones. The proposed modification of the received model updates mimics the imaginary central update irrespective of dropout clients. The theoretical analysis of MimiC shows that divergence between the aggregated and central update diminishes with proper learning rates, leading to its convergence. Simulation results further demonstrate that MimiC maintains stable convergence performance and learns better models than the baseline methods.
title MimiC: Combating Client Dropouts in Federated Learning by Mimicking Central Updates
topic Machine Learning
Distributed, Parallel, and Cluster Computing
url https://arxiv.org/abs/2306.12212