FedBAT: Communication-Efficient Federated Learning via Learnable Binarization

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Li, Shiwei, Xu, Wenchao, Wang, Haozhao, Tang, Xing, Qi, Yining, Xu, Shijie, Luo, Weihong, Li, Yuhua, He, Xiuqiang, Li, Ruixuan
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866911979757633536
author Li, Shiwei
Xu, Wenchao
Wang, Haozhao
Tang, Xing
Qi, Yining
Xu, Shijie
Luo, Weihong
Li, Yuhua
He, Xiuqiang
Li, Ruixuan
author_facet Li, Shiwei
Xu, Wenchao
Wang, Haozhao
Tang, Xing
Qi, Yining
Xu, Shijie
Luo, Weihong
Li, Yuhua
He, Xiuqiang
Li, Ruixuan
contents Federated learning is a promising distributed machine learning paradigm that can effectively exploit large-scale data without exposing users' privacy. However, it may incur significant communication overhead, thereby potentially impairing the training efficiency. To address this challenge, numerous studies suggest binarizing the model updates. Nonetheless, traditional methods usually binarize model updates in a post-training manner, resulting in significant approximation errors and consequent degradation in model accuracy. To this end, we propose Federated Binarization-Aware Training (FedBAT), a novel framework that directly learns binary model updates during the local training process, thus inherently reducing the approximation errors. FedBAT incorporates an innovative binarization operator, along with meticulously designed derivatives to facilitate efficient learning. In addition, we establish theoretical guarantees regarding the convergence of FedBAT. Extensive experiments are conducted on four popular datasets. The results show that FedBAT significantly accelerates the convergence and exceeds the accuracy of baselines by up to 9\%, even surpassing that of FedAvg in some cases.
format Preprint
id arxiv_https___arxiv_org_abs_2408_03215
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle FedBAT: Communication-Efficient Federated Learning via Learnable Binarization
Li, Shiwei
Xu, Wenchao
Wang, Haozhao
Tang, Xing
Qi, Yining
Xu, Shijie
Luo, Weihong
Li, Yuhua
He, Xiuqiang
Li, Ruixuan
Machine Learning
Distributed, Parallel, and Cluster Computing
Federated learning is a promising distributed machine learning paradigm that can effectively exploit large-scale data without exposing users' privacy. However, it may incur significant communication overhead, thereby potentially impairing the training efficiency. To address this challenge, numerous studies suggest binarizing the model updates. Nonetheless, traditional methods usually binarize model updates in a post-training manner, resulting in significant approximation errors and consequent degradation in model accuracy. To this end, we propose Federated Binarization-Aware Training (FedBAT), a novel framework that directly learns binary model updates during the local training process, thus inherently reducing the approximation errors. FedBAT incorporates an innovative binarization operator, along with meticulously designed derivatives to facilitate efficient learning. In addition, we establish theoretical guarantees regarding the convergence of FedBAT. Extensive experiments are conducted on four popular datasets. The results show that FedBAT significantly accelerates the convergence and exceeds the accuracy of baselines by up to 9\%, even surpassing that of FedAvg in some cases.
title FedBAT: Communication-Efficient Federated Learning via Learnable Binarization
topic Machine Learning
Distributed, Parallel, and Cluster Computing
url https://arxiv.org/abs/2408.03215