Abstract
Mining graph data has gained wide attention in modern applications. With the explosive growth of graph data, it is common to see many of them collected and stored in different distinction systems. These local graphs can not be directly shared due to privacy and bandwidth concerns. Thus, Federated Learning approach needs to be considered to collaboratively train a powerful generalizable model. However, these local subgraphs are usually heterogeneously distributed. Such heterogeneity brings challenges for subgraph federated learning. In this work, we analyze subgraph federated learning and find that sub-optimal objectives under the FedAVG training setting influence the performance of GNN. To this end, we propose InfoFedSage, a federated subgraph learning framework guided by Information bottleneck to alleviate the non-iid issue. Experiments on public datasets demonstrate the effectiveness of InfoFedSage against heterogeneous subgraph federated learning.
J. Guo and S. Li—Equal contribution.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsReferences
Achille, A., Soatto, S.: Information dropout: learning optimal representations through noisy computation. IEEE Trans. Pattern Anal. Mach. Intell. 40, 2897–2905 (2018)
Alemi, A.A., Fischer, I., Dillon, J.V., Murphy, K.: Deep variational information bottleneck. arXiv preprint arXiv:1612.00410 (2016)
Blondel, V.D., Guillaume, J.L., Lambiotte, R., Lefebvre, E.: Fast unfolding of communities in large networks. J. Stat. Mech: Theory Exp. 2008, 10008 (2008)
Blum, A., Haghtalab, N., Procaccia, A.D.: Variational dropout and the local reparameterization trick. In: NIPS (2015)
Gao, C., et al.: Graph neural networks for recommender systems: challenges, methods, and directions. ArXiv abs/2109.12843 (2021)
Guo, J.N., Li, S., Zhao, Y., Zhang, Y.: Learning robust representation through graph adversarial contrastive learning. In: Bhattacharya, A., et al. (eds.) DASFAA 2022. LNCS, vol. 13245, pp. 682–697. Springer, Cham (2022). https://doi.org/10.1007/978-3-031-00123-9_54
Guo, J., et al.: Learning multi-granularity user intent unit for session-based recommendation. In: Proceedings of the 15’th ACM International Conference on Web Search and Data Mining, WSDM 2022 (2022)
Guo, J., Zhang, P., Li, C., Xie, X., Zhang, Y., Kim, S.: Evolutionary preference learning via graph nested GRU ode for session-based recommendation. In: Proceedings of the 31st ACM International Conference on Information & Knowledge Management (2022)
Hamilton, W.L., Ying, Z., Leskovec, J.: Inductive representation learning on large graphs. In: NIPS (2017)
Hartigan, J.A., Wong, M.A.: A k-means clustering algorithm (1979)
He, C., et al.: FedGraphNN: a federated learning system and benchmark for graph neural networks. ArXiv abs/2104.07145 (2021)
Kingma, D.P., Welling, M.: Auto-encoding variational bayes. CoRR abs/1312.6114 (2014)
Li, D., Wang, J.: FedMD: heterogenous federated learning via model distillation. ArXiv abs/1910.03581 (2019)
Li, T., Sahu, A.K., Talwalkar, A.S., Smith, V.: Federated learning: challenges, methods, and future directions. IEEE Signal Process. Mag. 37, 50–60 (2020)
Li, T., Sahu, A.K., Zaheer, M., Sanjabi, M., Talwalkar, A., Smith, V.: Federated optimization in heterogeneous networks. Proc. Mach. Learn. Syst. 2, 429–450 (2020)
Li, X., Huang, K., Yang, W., Wang, S., Zhang, Z.: On the convergence of FedAvg on non-IID data. ArXiv abs/1907.02189 (2020)
Lin, T., Kong, L., Stich, S.U., Jaggi, M.: Ensemble distillation for robust model fusion in federated learning. ArXiv abs/2006.07242 (2020)
McMahan, H.B., Moore, E., Ramage, D., Hampson, S., y Arcas, B.A.: Communication-efficient learning of deep networks from decentralized data. In: AISTATS (2017)
Namata, G., London, B., Getoor, L., Huang, B.: Query-driven active surveying for collective classification (2012)
Qiu, Y., Huang, C., Wang, J., Huang, Z., Xiao, J.: A privacy-preserving subgraph-level federated graph neural network via differential privacy. In: Memmi, G., Yang, B., Kong, L., Zhang, T., Qiu, M. (eds.) KSEM 2022. LNCS, vol. 13370, pp. 165–177. Springer, Cham (2022). https://doi.org/10.1007/978-3-031-10989-8_14
Sahu, A.K., Li, T., Sanjabi, M., Zaheer, M., Talwalkar, A.S., Smith, V.: Federated optimization in heterogeneous networks. arXiv Learning (2020)
Sen, P., Namata, G., Bilgic, M., Getoor, L., Gallagher, B., Eliassi-Rad, T.: Collective classification in network data (2008)
Shchur, O., Mumme, M., Bojchevski, A., Günnemann, S.: Pitfalls of graph neural network evaluation. ArXiv abs/1811.05868 (2018)
Tishby, N., Zaslavsky, N.: Deep learning and the information bottleneck principle. In: 2015 IEEE Information Theory Workshop (ITW), pp. 1–5 (2015)
Wasserman, S., Faust, K.: Social network analysis - methods and applications. In: Structural Analysis in the Social Sciences (2007)
Wu, L., et al.: Graph neural networks for natural language processing: a survey. ArXiv abs/2106.06090 (2021)
Wu, T., Ren, H., Li, P., Leskovec, J.: Graph information bottleneck. ArXiv abs/2010.12811 (2020)
Xie, H., Ma, J., Xiong, L., Yang, C.: Federated graph classification over non-IID graphs. Adv. Neural. Inf. Process. Syst. 34, 18839–18852 (2021)
Zhang, K., Yang, C., Li, X., Sun, L., Yiu, S.M.: Subgraph federated learning with missing neighbor generation. ArXiv abs/2106.13430 (2021)
Zhang, P., et al.: Efficiently leveraging multi-level user intent for session-based recommendation via atten-mixer network. arXiv preprint arXiv:2206.12781 (2022)
Zhu, Z., Hong, J., Zhou, J.: Data-free knowledge distillation for heterogeneous federated learning. Proc. Mach. Learn. Res. 139, 12878–12889 (2021)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
A Appendix
A Appendix
1.1 A.1 Proof for Proposition 4.1
We first state a lemma.
Lemma A.1. Given X have n states (\(x_1,x_2, \cdots x_n\)) and \(x_1\) can be divided into k sub states (\(x_{11},x_{12}, \cdots x_{1n}\)), Y has m states (\(y_1,y_2, \cdots y_n\)), we have
Proof. The mutual information between (\(x_{11},x_{12}, \cdots x_{1n}\)) and Y:
Then, we have
Then, we can get Corollary A.1:
We restate Proposition 4.1: For \(Z_X' = Z_{X1} \cup \cdots \cup Z_{Xm}\) and \(Y' = Y_{1} \cup \cdots \cup Y_{m}\), we have
Proof. We first consider the first inequality, since the definition of mutual information has the form
Notice that the entropy of labels \(H(Y_i)\) is independent of our optimization procedure and can be ignored. Therefore, we have
Then, we consider the second inequality. Directly, based on our Corollary A.1, we can get
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Guo, J., Li, S., Zhang, Y. (2023). An Information Theoretic Perspective for Heterogeneous Subgraph Federated Learning. In: Wang, X., et al. Database Systems for Advanced Applications. DASFAA 2023. Lecture Notes in Computer Science, vol 13943. Springer, Cham. https://doi.org/10.1007/978-3-031-30637-2_50
Download citation
DOI: https://doi.org/10.1007/978-3-031-30637-2_50
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-30636-5
Online ISBN: 978-3-031-30637-2
eBook Packages: Computer ScienceComputer Science (R0)