图联邦学习:问题、方法与挑战

doi:10.11896/jsjkx.240500118

摘要/Abstract

摘要： 图作为一种高效、灵活、通用的数据结构,在多个学科领域得到了广泛应用。近年来,基于图的深度学习算法不断涌现,并在社交网络、生物信息学、推荐系统等领域取得显著成效。尽管公开的图数据量在增加,但高质量的数据往往分散在不同的数据所有者手中。随着社会对数据隐私保护要求的提高,现有的图学习算法面临着许多挑战。图联邦学习作为一种有效的解决方案应运而生。文中系统回顾了图联邦学习领域近五年的研究进展,将该领域的核心问题划分为3个部分,并在结构上进行了垂直整合,在关系上进行了递进阐述,包括:1)原始图数据差异导致的结构异构性;2)图联邦特性导致的模型聚合问题;3)模型整体调优方面的挑战。针对每个问题,详细分析了代表性工作及其优缺点,并总结了图联邦学习领域的典型应用和未来挑战。

关键词: 联邦学习, 图神经网络, 图联邦学习, 隐私计算

Abstract: Graph has been widely used in various fields for many years as an efficient,flexible,and versatile data structure.In recent years,graph-based deep learning algorithms have emerged,achieving significant success in areas like social network,bioinformatics,and recommendation systems.Although publicly graph data online is increasing,high-quality data remains scattered among different owners.With society’s growing demand for data privacy protection,existing graph learning algorithms require enhancement.Graph federated learning is a novel approach to addresses this issue.This paper systematically reviews the research progress in the field of federated graph learning over the past five years.The core problems in the field are divided into three parts,and the structure is vertically integrated and the relationships are progressively explained:1)structural heterogeneity from differences in raw graph data;2)model aggregation issues due to federated graph learning characteristics;3)overall model tuning.For each section,it provides a detailed analysis of representative works and their advantages and disadvantages,summarizes the typical applications and future challenges in the field of federated graph learning.

Key words: Federated learning, Graph neural network, Federated graph learning, Privacy computing

中图分类号:

TP181

王鑫, 熊书博, 孙凌云. 图联邦学习:问题、方法与挑战[J]. 计算机科学, 2025, 52(1): 362-373. https://doi.org/10.11896/jsjkx.240500118

WANG Xin, XIONG Shubo, SUN Lingyun. Federated Graph Learning:Problems,Methods and Challenges[J]. Computer Science, 2025, 52(1): 362-373. https://doi.org/10.11896/jsjkx.240500118

参考文献

[1]MITTONE G,SVOBODA F,ALDINUCCI M,et al.A Federated Learning Benchmark for Drug-Target Interaction[C]//Companion Proceedings of the ACM Web Conference 2023.New York,NY,USA:Association for Computing Machinery,2023:1177-1181.
[2]QI T,CHEN L,LI G,et al.FedAGCN:A traffic flow prediction framework based on federated learning and Asynchronous Graph Convolutional Network[J].Applied Soft Computing,2023,138:110175.
[3]MEI G,GUO Z,LIU S,et al.SGNN:A Graph Neural Network Based Federated Learning Approach by Hiding Structure[C]//2019 IEEE International Conference on Big Data(Big Data).2019:2560-2568.
[4]GUAN Z L,DU J P,XUE Z,et al.Personalized Public SafetyEmergency Detection Method Based on Enhanced Federated Graph Neural Network [J].Journal of Software,2024,35(4):1774-1789.
[5]ZHU W,LUO J,WHITE A D.Federated learning of molecular properties with graph neural networks in a heterogeneous setting[J].Patterns,2022,3(6):100521.
[6]HE C,BALASUBRAMANIAN K,CEYANI E,et al.Fed-GraphNN:A Federated Learning System and Benchmark for Graph Neural Networks[J].arXiv:2104.07145,2021.
[7]LIU R,XING P,DENG Z,et al.Federated Graph Neural Networks:Overview,Techniques and Challenges[J].arXiv:2202.07256,2024.
[8]WU Z,PAN S,CHEN F,et al.A Comprehensive Survey onGraph Neural Networks[J].IEEE Transactions on Neural Networks and Learning Systems,2021,32(1):4-24.
[9]KAIROUZ P,MCMAHAN H B,AVENT B,et al.Advancesand Open Problems in Federated Learning[M]//Now Foundations and Trends.2021:1-210.
[10]ZHANG H,SHEN T,WU F,et al.Federated Graph Learning -- A Position Paper[J].arXiv:2105.11099 2021.
[11]ZHAO G,HUANG Y,TSAI C H.FedGSL:Federated Graph Structure Learning for Local Subgraph Augmentation[C]//2022 IEEE International Conference on Big Data(Big Data).2022:818-824.
[12]CHEN S,XUE D,CHUAI G,et al.FL-QSAR:a federatedlearning-based QSAR prototype for collaborative drug discovery[J].Bioinformatics(Oxford,England),2021,36(22-23):5492-5498.
[13]CHEN M,ZHANG W,YUAN Z,et al.FedE:EmbeddingKnowledge Graphs in Federated Setting[C]//Proceedings of the 10th International Joint Conference on Knowledge Graphs.New York,NY,USA:Association for Computing Machinery,2022:80-88.
[14]PENG H,LI H,SONG Y,et al.Differentially Private Federated Knowledge Graphs Embedding[C]//Proceedings of the 30th ACM International Conference on Information & Knowledge Management.New York,NY,USA:Association for Computing Machinery,2021:1416-1425.
[15]WU C,WU F,LYU L,et al.FedGNN:A federated graph neural network framework for privacy-preserving personalization[J].Nature Communications,2022,13(1):3091.
[16]BAEK J,JEONG W,JIN J,et al.Personalized Subgraph Federated Learning[J].arXiv:2206.10206,2023.
[17]ZHANG K,YANG C,LI X,et al.Subgraph Federated Learning with Missing Neighbor Generation[J].arXiv:2106.13430,2021.
[18]PENG L,WANG N,DVORNEK N,et al.FedNI:FederatedGraph Learning With Network Inpainting for Population-Based Disease Prediction[J].IEEE Transactions on Medical Imaging,2023,42(7):2032-2043.
[19]XIE H,XIONG L,YANG C.Federated Node Classification over Graphs with Latent Link-type Heterogeneity[C]//Proceedings of the ACM Web Conference 2023.Austin TX USA:ACM,2023:556-566.
[20]BONAWITZ K,IVANOV V,KREUTER B,et al.Practical Secure Aggregation for Privacy-Preserving Machine Learning[C]//Proceedings of the 2017 ACM SIGSAC Conference on Computer and Communications Security.New York,NY,USA:Association for Computing Machinery,2017:1175-1191.
[21]ACAR A,AKSU H,ULUAGAC A S,et al.A Survey on Homomorphic Encryption Schemes:Theory and Implementation[J].ACM Computing Surveys,2018,51(4):79:1-79:35.
[22]GEISLER S,ZÜGNER D,GÜNNEMANN S.Reliable GraphNeural Networks via Robust Aggregation[C]//Advances in Neural Information Processing Systems.2020:13272-13284.
[23]CHEN C,ZHOU J,ZHENG L,et al.Vertically FederatedGraph Neural Network for Privacy-Preserving Node Classification[J].arXiv:2005.11903,2022.
[24]JIANG M,JUNG T,KARL R,et al.Federated Dynamic GNN with Secure Aggregation[J].arXiv:2009.07351,2022.
[25]GÜRLER Z,REKIK I.Federated Brain Graph Evolution Prediction Using Decentralized Connectivity Datasets With Temporally-Varying Acquisitions[J].IEEE Transactions on Medical Imaging,2023,42(7):2022-2031.
[26]CHEN C,HU W,XU Z,et al.FedGL:Federated Graph Lear-ning Framework with Global Self-Supervision[J].arXiv:2105.03170,2021.
[27]MAI P,PANG Y.Vertical Federated Graph Neural Network for Recommender System[J].arXiv:2303.05786,2021.
[28]NI X,XU X,LYU L,et al.A Vertical Federated LearningFramework for Graph Convolutional Network[J].arXiv:2106.11593,2021.
[29]ZHANG X,YIN W,HONG M,et al.Hybrid federated learning:Algorithms and implementation[J].arXiv:2012.12420,2020.
[30]GAO H,GE S,CHANG T H.FedHD:Communication-efficient federated learning from hybrid data[J].Journal of the Franklin Institute,2023,360(12):8416-8454.
[31]JANG J,KLABJAN D,MENDIRATTA V,et al.HybridFedGraph:An efficient hybrid federated learning algorithm using graph convolutional neural network[J].arXiv:2404.09443,2024.
[32]RIZK E,SAYED A H.A Graph Federated Architecture with Privacy Preserving Learning[C]//2021 IEEE 22nd International Workshop on Signal Processing Advances in Wireless Communications(SPAWC).2021:131-135.
[33]MCMAHAN B,MOORE E,RAMAGE D,et al.Communica-tion-efficient learning of deep networks from decentralized data[C]//Artificial Intelligence and Statistics.PMLR,2017:1273-1282.
[34]HU K,WU J,LI Y,et al.FedGCN:Federated Learning-Based Graph Convolutional Networks for Non-Euclidean Spatial Data[J].Mathematics,2022,10(6):1000.
[35]DINH C T,VU T T,TRAN N H,et al.A New Look and Convergence Rate of Federated Multi-Task Learning with Laplacian Regularization[J].arXiv:2102.07148,2022.
[36]PEI Y,MAO R,LIU Y,et al.Decentralized federated graph neural networks[C]//International Workshop on Federated and Transfer Learning for Data Sparsity and Confidentiality in Conjunction with IJCAI.2021.
[37]LI X,WU Z,ZHANG W,et al.FedGTA:Topology-aware Ave-raging for Federated Graph Learning[J].arXiv:2401.11755,2024.
[38]DENG Z,HUANG X,LI D,et al.FedGraph:an AggregationMethod from Graph Perspective[J].arXiv:2210.02733,2022.
[39]LALITHA A,KILINC O C,JAVIDI T,et al.Peer-to-peer Fe-derated Learning on Graphs[J].arXiv:2210.02733,2022.
[40]HE C,CEYANI E,BALASUBRAMANIAN K,et al.Spread-GNN:Decentralized Multi-Task Federated Learning for Graph Neural Networks on Molecular Data[J].Proceedings of the AAAI Conference on Artificial Intelligence,2022,36(6):6865-6873.
[41]CHOI B,YONG S J,HAN D J,et al.Communication-Computation Efficient Secure Aggregation for Federated Learning[J].arXiv:2012.05433,2021.
[42]GOGINENI V C,WERNER S,HUANG Y F,et al.Decentra-lized Graph Federated Multitask Learning for Streaming Data[C]//2022 56th Annual Conference on Information Sciences and Systems(CISS).2022:101-106.
[43]MALINOVSKIY G,KOVALEV D,GASANOV E,et al.From local SGD to local fixed-point methods for federated learning[C]//International Conference on Machine Learning.PMLR,2020:6692-6701.
[44]CHEN F,LI P,MIYAZAKI T,et al.FedGraph:FederatedGraph Learning with Intelligent Sampling[J].arXiv:2111.01370,2021.
[45]YAO Y,JIN W,RAVI S,et al.FedGCN:Convergence-Communication Tradeoffs in Federated Training of Graph Convolutional Networks[J].arXiv:22041.12433,2022.
[46]ZHANG C,ZHANG S,YU J J Q,et al.FASTGNN:A Topolo-gical Information Protected Federated Learning Approach for Traffic Speed Forecasting[J].IEEE Transactions on Industrial Informatics,2021,17(12):8464-8474.
[47]MENG C,RAMBHATLA S,LIU Y.Cross-Node FederatedGraph Neural Network for Spatio-Temporal Data Modeling[J].arXiv:2160.05223,2021.
[48]CHEN F,LONG G,WU Z,et al.Personalized Federated Lear-ning With Graph[J].arXiv:2203.00829,2022.
[49]XING P,LU S,WU L,et al.BiG-Fed:Bilevel Optimization Enhanced Graph-Aided Federated Learning[EB/OL].https://fli-cml.github.io/2021/papers/FL-ICML21_paper_74.pdf.
[50]LEE H,BERTOZZI A L,KOVACˇEVIĆ J,et al.Privacy-Preserving Federated Multi-Task Linear Regression:A One-Shot Linear Mixing Approach Inspired By Graph Regularization[C]//ICASSP 2022－2022 IEEE International Conference on Acoustics,Speech and Signal Processing(ICASSP).2022:5947-5951.
[51]ZHU H,XU J,LIU S,et al.Federated Learning on Non-IID Data:A Survey[J].arXiv:2106.06843,2022.
[52]XIE H,MA J,XIONG L,et al.Federated Graph Classification over Non-IID Graphs[C]//Advances in Neural Information Processing System.2021:18839-18852.
[53]ZHENG L,ZHOU J,CHEN C,et al.ASFGNN:AutomatedSeparated-Federated Graph Neural Network[J].arXiv:2011.03248,2020.
[54]HE H,BAI Y,GARCIA E A,et al.ADASYN:Adaptive syn-thetic sampling approach for imbalanced learning[C]//2008 IEEE International Joint Conference on Neural Networks(IEEE World Congress on Computational Intelligence).2008:1322-1328.
[55]LIN Y,CHEN C,CHEN C,et al.Improving Federated Rela-tional Data Modeling via Basis Alignment and Weight Penalty[J].arXiv:2011.11369,2020.
[56]WANG B,LI A,LI H,et al.GraphFL:A Federated LearningFramework for Semi-Supervised Node Classification on Graphs[J].arXiv:2012.04187,2020.
[57]TAN Y,LIU Y,LONG G,et al.Federated Learning on Non-IID Graphs via Structural Knowledge Sharing[J].Proceedings of the AAAI Conference on Artificial Intelligence,2023,37(8):9953-9961.
[58]HANZELY F,RICHTÁRIK P.Federated Learning of a Mixture of Global and Local Models[J].arXiv:2002.05516,2021.
[59]DINH C,TRAN N,NGUYEN J.Personalized Federated Lear-ning with Moreau Envelopes[C]//Advances in Neural Information Processing Systems.2020:21394-21405.
[60]ARIVAZHAGAN M G,AGGARWAL V,SINGH A K,et al.Federated Learning with Personalization Layers[J].arXiv:1912.00818,2019.
[61]SATTLER F,MÜLLER K R,SAMEK W.Clustered Federated Learning:Model-Agnostic Distributed Multi-Task Optimization under Privacy Constraints[J].arXiv:1910.01991,2019.
[62]ZHANG K,XIE H,GU Z,et al.Subgraph federated learningover heterogeneous graphs[J].arXiv:2106.13430,2022.
[63]WEBER M,CHEN J,SUZUMURA T,et al.Scalable GraphLearning for Anti-Money Laundering:A First Look[J].arXiv:1812.00076,2018.
[64]DU H,SHEN M,SUN R,et al.Malicious Transaction Identification in Digital Currency via Federated Graph Deep Learning[C]//IEEE Conference on Computer Communications Workshops.2022:1-6.
[65]CHEN Z,LI W,XING X,et al.Medical federated learning with joint graph purification for noisy label learning[J].Medical Image Analysis,2023,90:102976.
[66]AHMED U,LIN J C W,SRIVASTAVA G.Hyper-Graph Attention Based Federated Learning Methods for Use in Mental Health Detection[J].IEEE Journal of Biomedical and Health Informatics,2023,27(2):768-777.
[67]LIU L,TIAN Y,CHAKRABORTY C,et al.Multilevel Federated Learning-Based Intelligent Traffic Flow Forecasting for Transportation Network Management[J].IEEE Transactions on Network and Service Management,2023,20(2):1446-1458.
[68]CHEN J,HUANG G,ZHENG H,et al.Graph-Fraudster:Adversarial Attacks on Graph Neural Network-Based Vertical Fe-derated Learning[J].IEEE Transactions on Computational Social Systems,2023,10(2):492-506.
[69]LI R C,ZHENG H B,ZHAO W H,et al.Data Reconstruction Attack for Vertical Graph Federated Learning[J].Computer Science,2023,50(7):332-338.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed