您好，欢迎访问江苏省农业科学院 机构知识库！
江苏省农业科学院机构知识库

全部

  • 全部
  • 标题
  • 学者
  • 机构
  • 关键词

QuAsyncFL: Asynchronous Federated Learning With Quantization for CloudEdgeTerminal Collaboration Enabled AIoT

文献类型： 外文期刊

作者： Liu, Ye 1 ; Huang, Peishan 3 ; Yang, Fan 4 ; Huang, Kai 5 ; Shu, Lei 6 ;

作者机构： 1.Anhui Higher Educ Inst, Key Lab Architectural Acoust Environm, Hefei 230601, Peoples R China

2.Nanjing Agr Univ, Coll Artificial Intelligence, Nanjing 210031, Peoples R China

3.Macau Univ Sci & Technol, Sch Comp Sci & Engn, Macau, Peoples R China

4.Jiangsu Normal Univ, Sch Math & Stat, Xuzhou 221116, Peoples R China

5.Jiangsu Acad Agr Sci, Inst Agr Facil & Equipment, Nanjing 210014, Peoples R China

6.Nanjing Agr Univ, Nanjing 210031, Peoples R China

7.Univ Lincoln, Sch Engn, Coll Sci, Lincoln LN6 7TS, England

关键词： Artificial Intelligence of Things (AIoT); asynchronous federated learning; cloud-edge-terminal collaboration; communication efficiency; quantization

期刊名称：IEEE INTERNET OF THINGS JOURNAL （ 影响因子：8.2； 五年影响因子：9.0 ）

ISSN： 2327-4662

年卷期： 2024 年 11 卷 1 期

页码：

收录情况： SCI

摘要： Federated Learning is a promising technique that facilitates cloud-edge-terminal collaboration in Artificial Intelligence of Things (AIoT). It will enable model training without centralizing data, addressing privacy, and security concerns. However, when applied to AIoT, this technique faces several challenges, such as low communication efficiency among terminal devices, edges, and cloud platforms. In this article, we propose a novel approach called asynchronous federated learning with quantization (QuAsyncFL), which combines asynchronous federated learning with an unbiased nonuniform quantizer to address the issue of low communication efficiency. Moreover, we provide a detailed theoretical analysis of convergence with quantized gradients proving that the model could converge to a certain bound. Our experiments demonstrate that QuAsyncFL outperforms the original approach, achieving significant improvements in terms of communication efficiency. The research results represent a further step toward developing cloud-edge-terminal collaboration enabled AIoT.

  • 相关文献
作者其他论文 更多>>
置顶
购物车
反馈

© 京ICP备09089781号-29 主办单位：江苏省农业科学院

学术资源： 中国农业科学院 农业专业知识服务系统 农科联盟资源共建共享平台 中国农业科技文献与信息服务平台 中国农业期刊集成服务平台