paper-with-me

홈 › Papers

HiFlash: Communication-Efficient Hierarchical Federated Learning with Adaptive Staleness Control and Heterogeneity-aware Client-Edge Association

2023-01-16 · Qiong Wu, Xu Chen, Tao Ouyang, Zhi Zhou, Xiaoxi Zhang, Shusen Yang, Junshan Zhang

Federated learning (FL) is a promising paradigm that enables collaboratively learning a shared model across massive clients while keeping the training data locally. However, for many existing FL systems, clients need to frequently exchange model parameters of large data size with the remote cloud server directly via wide-area networks (WAN), leading to significant communication overhead and long transmission time. To mitigate the communication bottleneck, we resort to the hierarchical federated learning paradigm of HiFL, which reaps the benefits of mobile edge computing and combines synchronous client-edge model aggregation and asynchronous edge-cloud model aggregation together to greatly reduce the traffic volumes of WAN transmissions. Specifically, we first analyze the convergence bound of HiFL theoretically and identify the key controllable factors for model performance improvement. We then advocate an enhanced design of HiFlash by innovatively integrating deep reinforcement learning based adaptive staleness control and heterogeneity-aware client-edge association strategy to boost the system efficiency and mitigate the staleness effect without compromising model accuracy. Extensive experiments corroborate the superior performance of HiFlash in model accuracy, communication reduction, and system efficiency.

📄 PDF Abstract BibTeX arXiv:2301.06447

Code (0)

등록된 구현이 없습니다.

Tasks

Deep Reinforcement LearningEdge-computingFederated Learning

Similar Papers 제목 키워드 기반

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 communicat…

Federated LearningPrivacy Preserving

Asynchronous Hierarchical Federated Learning

2022-05-31 · Xing Wang, Yijun Wang

Federated Learning is a rapidly growing area of research and with various benefits and industry applications. Typical federated patterns have some intrinsic issues such as heavy server traffic, long periods of convergenc…

Federated Learningimage-classificationImage Classification

MAB-Based Channel Scheduling for Asynchronous Federated Learning in Non-Stationary Environments

2025-03-03 · Zhiyin Li, Yubo Yang, Tao Yang, Xiaofeng Wu 외

Federated learning enables distributed model training across clients under central coordination without raw data exchange. However, in wireless implementations, frequent parameter updates between the server and clients c…

FairnessFederated LearningScheduling

Federated Stochastic Gradient Descent Begets Self-Induced Momentum

2022-02-17 · Howard H. Yang, Zuozhu Liu, Yaru Fu, Tony Q. S. Quek 외

Federated learning (FL) is an emerging machine learning method that can be applied in mobile edge systems, in which a server and a host of clients collaboratively train a statistical model utilizing the data and computat…

Federated Learning

Revisiting Gradient Staleness: Evaluating Distance Metrics for Asynchronous Federated Learning Aggregation

2026-03-09 · Patrick Wilhelm, Odej Kao arxiv

In asynchronous federated learning (FL), client devices send updates to a central server at varying times based on their computational speed, often using stale versions of the global model. This staleness can degrade the…

Federated Learning