paper-with-me

홈 › Papers

Towards Efficient and Stable K-Asynchronous Federated Learning with Unbounded Stale Gradients on Non-IID Data

2022-03-02 · ZiHao Zhou, Yanan Li, Xuebin Ren, Shusen Yang

Federated learning (FL) is an emerging privacy-preserving paradigm that enables multiple participants collaboratively to train a global model without uploading raw data. Considering heterogeneous computing and communication capabilities of different participants, asynchronous FL can avoid the stragglers effect in synchronous FL and adapts to scenarios with vast participants. Both staleness and non-IID data in asynchronous FL would reduce the model utility. However, there exists an inherent contradiction between the solutions to the two problems. That is, mitigating the staleness requires to select less but consistent gradients while coping with non-IID data demands more comprehensive gradients. To address the dilemma, this paper proposes a two-stage weighted $K$ asynchronous FL with adaptive learning rate (WKAFL). By selecting consistent gradients and adjusting learning rate adaptively, WKAFL utilizes stale gradients and mitigates the impact of non-IID data, which can achieve multifaceted enhancement in training speed, prediction accuracy and training stability. We also present the convergence analysis for WKAFL under the assumption of unbounded staleness to understand the impact of staleness and non-IID data. Experiments implemented on both benchmark and synthetic FL datasets show that WKAFL has better overall performance compared to existing algorithms.

📄 PDF Abstract BibTeX arXiv:2203.01214

Code (0)

등록된 구현이 없습니다.

Tasks

Federated LearningPrivacy Preserving

Similar Papers 제목 키워드 기반

Unbounded Gradients in Federated Leaning with Buffered Asynchronous Aggregation

2022-10-03 · Mohammad Taha Toghani, César A. Uribe

Synchronous updates may compromise the efficiency of cross-device federated learning once the number of active clients increases. The \textit{FedBuff} algorithm (Nguyen et al., 2022) alleviates this problem by allowing a…

Federated Learning

AFBS:Buffer Gradient Selection in Semi-asynchronous Federated Learning

2025-06-15 · Chaoyi Lu, Yiding Sun, Jinqian Chen, Zhichuan Yang 외

Asynchronous federated learning (AFL) accelerates training by eliminating the need to wait for stragglers, but its asynchronous nature introduces gradient staleness, where outdated gradients degrade performance. Existing…

Federated Learning

FedGSM: Efficient Federated Learning for LEO Constellations with Gradient Staleness Mitigation

2023-04-17 · Lingling Wu, Jingjing Zhang

Recent advancements in space technology have equipped low Earth Orbit (LEO) satellites with the capability to perform complex functions and run AI applications. Federated Learning (FL) on LEO satellites enables collabora…

Federated Learning

AsyncFedED: Asynchronous Federated Learning with Euclidean Distance based Adaptive Weight Aggregation

2022-05-27 · Qiyuan Wang, Qianqian Yang, Shibo He, Zhiguo Shi 외

In an asynchronous federated learning framework, the server updates the global model once it receives an update from a client instead of waiting for all the updates to arrive as in the synchronous setting. This allows he…

Federated Learning

Take History as a Mirror in Heterogeneous Federated Learning

2023-12-16 · Xiaorui Jiang, Hengwei Xu, Yu Gao, Yong Liao 외

Federated Learning (FL) allows several clients to cooperatively train machine learning models without disclosing the raw data. In practice, due to the system and statistical heterogeneity among devices, synchronous FL of…

Federated Learning