多尺度信息融合与分层注意力聚合的子图联邦学习算法

王若禹, 丁世飞, 郭丽丽

doi:10.3969/j.issn.1007-130X.2026.05.016

计算机工程与科学 >

2026 , Vol. 48 >Issue 5: 925 - 935

DOI: https://doi.org/10.3969/j.issn.1007-130X.2026.05.016

人工智能与数据挖掘

多尺度信息融合与分层注意力聚合的子图联邦学习算法

展开

(1.中国矿业大学计算机科学与技术学院/人工智能学院,江苏徐州 221116;
2.中国矿业大学矿山数字化教育部工程研究中心,江苏徐州 221116)

收稿日期: 2025-07-29

修回日期: 2025-09-11

网络出版日期: 2026-05-21

基金资助

国家自然科学基金(62276265,61976216)

收起

Multi-scale information fusion and layered attention aggregation for subgraph federated learning algorithm

Expand

(1.School of Computer Science and Technology/School of Artificial Intelligence,
China University of Mining and Technology,Xuzhou 221116；
2.Mine Digitization Engineering Research Center of the Ministry of Education,
China University of Mining and Technology,Xuzhou 221116,China)

Received date: 2025-07-29

Revised date: 2025-09-11

Online published: 2026-05-21

Fold

摘要

子图联邦学习通过图卷积网络在本地客户端处理全局图的子图，并通过服务器更新这些参数，从而保护用户隐私。现有方法缺乏对某些任务或图结构中重要节点的关注，这可能会降低节点嵌入的效率。提出了一种新的图联邦学习算法FedMFG，引入了多尺度信息融合卷积整合节点特征与邻居信息，从而提升节点特征表示能力。该算法在预训练阶段传递参数以减少通信成本，并在服务器端应用注意力机制动态调整权重，以实现全局参数的更好聚合。在标准参考数据集上的实验结果表明，FedMFG相较于先前的算法具有更高的准确度、稳定性和更低的通信成本。

关键词： 子图联邦学习; 多尺度信息融合卷积; 图卷积网络（GCN）模型; 注意力机制

本文引用格式

王若禹, 丁世飞, 郭丽丽 . 多尺度信息融合与分层注意力聚合的子图联邦学习算法[J]. 计算机工程与科学, 2026 , 48(5) : 925 -935 . DOI: 10.3969/j.issn.1007-130X.2026.05.016

Abstract

Subgraph federated learning trains on subgraphs of a global graph on local clients using graph convolutional networks and updates these parameters on the server, thereby safeguarding user privacy. Existing methods lack attention to important nodes in specific tasks or graph structures, which may reduce the efficiency of node embedding. This paper proposes a novel graph federated learning algorithm, called FedMFG, which introduces multi-scale information fusion convolution to integrate node features with neighbor information, thereby enhancing node feature representation capabilities. The algorithm passes parameters during the pre-training phase to reduce communication costs and applies an attention mechanism at the server to dynamically adjust weights for better aggregation of global parameters. Experimental results on standard benchmark datasets demonstrate that FedMFG achieves higher accuracy, higher stability, and lower communication costs than previous algorithms.

Key words： subgraph federated learning; multi-scale information fusion convolution; graph convolutional network(GCN) model; attention mechanism

Options

文章导航

模态框（Modal）标题

摘要

本文引用格式

Abstract