Skip to main content
Springer Nature Link
Log in

Differentially private federated learning with non-IID data

  • Regular Paper
  • Published:
Computing Aims and scope Submit manuscript

Abstract

In Differentially Private Federated Learning (DPFL), gradient clipping and random noise addition disproportionately affect statistically heterogeneous data. As a consequence, DPFL has a disparate impact: the accuracy of models trained with DPFL tends to decrease more on these data. If the accuracy of the original model decreases on heterogeneous data, DPFL may degrade the accuracy performance more. In this work, we study the utility loss inequality due to differential privacy and compare the convergence of the private and non-private models. Specifically, we analyze the gradient differences caused by statistically heterogeneous data and explain how statistical heterogeneity relates to the effect of privacy on model convergence. In addition, we propose an improved DPFL algorithm, called R-DPFL, to achieve differential privacy at the same cost but with good utility. R-DPFL adjusts the gradient clipping value and the number of selected users at beginning according to the degree of statistical heterogeneity of the data, and weakens the direct proportional relationship between the differential privacy and the gradient difference, thereby reducing the impact of differential privacy on the model trained on heterogeneous data. Our experimental evaluation shows the effectiveness of our elimination algorithm in achieving the same cost of differential privacy with satisfactory utility. Our code is publicly available at https://github.com/chengshuyan/R-DPFL.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Subscribe and save

Springer+ Basic
$34.99 /Month
  • Get 10 units per month
  • Download Article/Chapter or eBook
  • 1 Unit = 1 Article or 1 Chapter
  • Cancel anytime
Subscribe now

Buy Now

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9

Similar content being viewed by others

Explore related subjects

Discover the latest articles, news and stories from top researchers in related subjects.

References

  1. Li G, Yu Y (2018) Contrast-oriented deep neural networks for salient object detection. IEEE Trans Neural Netw Learn Syst 29:6038–6051

    Article  Google Scholar 

  2. Zhao Z, Zheng P, Xu S, Wu X (2019) Object detection with deep learning: a review. IEEE Trans Neural Netw Learn Syst 30:3212–3232

    Article  Google Scholar 

  3. Kaissis G, Ziller A, Passerat-Palmbach J et al (2021) End-to-end privacy preserving deep learning on multi-institutional medical imaging. Nat Mach Intell 3:473–484

    Article  Google Scholar 

  4. Ziller A, Usynin D, Braren R, et al. (2021) Medical imaging deep learning with differential privacy. Scientific Reports 11

  5. McMahan B, Moore E, Ramage D, Hampson S, Arcas B (2017) Communication-efficient learning of deep networks from decentralized data. Proc Mach Learn Res 54:1273–1282

    Google Scholar 

  6. Kairouz P, McMahan H, Avent B, et al (2021) Advances and open problems in federated learning. Found Trends® Mach Learn 14(1–2):1–210

  7. Li T, Sahu A, Talwalkar A, Smith V (2020) Federated learning: challenges, methods, and future directions. IEEE Signal Process Mag 37(3):50–60

    Article  Google Scholar 

  8. Jamali-Rad H, Abdizadeh M, Singh A (2022) Federated learning with taskonomy for non-IID data. arXiv:2103.15947

  9. Heaven D (2019) Why deep-learning is are so easy to fool. Nature 574:163–166

    Article  Google Scholar 

  10. Shokri R, Stronati M, Song C, Shmatikov V (2017) membership inference attacks against machine learning models. In: 2017 IEEE symposium on security and privacy (SP), pp 3–18

  11. Su J, Vargas D, Sakurai K (2019) One pixel attack for fooling deep neural networks. IEEE Trans Evolut Comput 23:828–841

    Article  Google Scholar 

  12. Yuan X, Ma X, Zhang L, Fang Y, Wu D (2022) Beyond class-level privacy leakage: breaking record-level privacy in federated learning. IEEE Internet Things J 9:2555–2565

    Article  Google Scholar 

  13. Zhu L, Liu Z, Han S (2019) Deep leakage from gradients. arXiv:2301.02621

  14. Zhao B, Mopuri K, Bilen H (2020) iDLG: improved deep leakage from gradients. arXiv:2001.02610

  15. Zheng W, Popa R, Gonzalez J, Stoica I (2019) Helen: maliciously secure coopetitive learning for linear models. In: 2019 IEEE symposium on security and privacy (SP), pp 724–738

  16. Lu Y, Huang X, Dai Y, Maharjan S, Zhang Y (2020) Differentially private asynchronous federated learning for mobile edge computing in urban informatics. IEEE Trans Ind Inf 16(3):2134–2143

    Article  Google Scholar 

  17. Huang X, Ding Y, Jiang Z et al (2020) DP-FL: a novel differentially private federated learning framework for the unbalanced data. World Wide Web 23(4):2529–2545

    Article  Google Scholar 

  18. Zheng Q, Chen S, Long Q, Su W (2021) Federated f-differential privacy. Proc Mach Learn Res 130:2251–2259

    Google Scholar 

  19. Bonawitz K, Ivanov V, Kreuter B, et al (2017) Practical secure aggregation for privacy-preserving machine learning. In: Proceedings of the 2017 ACM SIGSAC conference on computer and communications security, pp 1175–1191

  20. Chase M, Ran G, Laine K, Lauter K, Rindal P (2017) Private collaborative neural network learning. IACR Cryptol. ePrint Arch

  21. Fereidooni H, Marchal S, Miettinen M (2021) SAFELearn: secure aggregation for private FEderated learning. In: 2021 IEEE security and privacy workshops (SPW), pp 56–62

  22. Geyer C, Klein T, Nabi M (2017) Differentially private federated learning: a client level perspective. arXiv1712.07557

  23. Li T, Sahu A, Zaheer M et al (2020) Federated optimization in heterogeneous networks. arXiv:1812.06127

  24. Huang L, Yin Y, Fu Z, Zhang S, Deng H, Liu D (2020) Loadaboost: loss-based adaboost federated machine learning with reduced computational complexity on IID and non-IID intensive care data. PLoS One 15(4):e0230706

    Article  Google Scholar 

  25. Lin T, Kong L, Stich S, Jaggi M (2020) Ensemble distillation for robust model fusion in federated learning. arXiv2006.07242

  26. Agarwal N, Suresh A, Yu F, Kumar S, Mcmahan H (2018) cpSGD: communication-efficient and differentially-private distributed SGD. arXiv:1805.10559

  27. Asoodeh S, Chen W, Calmon P, Ozgur A (2021) Differentially private federated learning: an information-theoretic perspective. In: 2021 IEEE international symposium on information theory (ISIT), pp 344–349

  28. Liang Z, Wang B, Gu Q, Osher S, Yao Y (2021) Exploring private federated learning with laplacian smoothing. arXiv2005.00218

  29. McMahan H, Ramage D, Talwar K, Zhang L (2018) Learning differentially private recurrent language models. arXiv:1710.06963

  30. Malekzadeh M, Hasircioglu B, Mital N, et al (2021) Dopamine: differentially private federated learning on medical data. arXiv:2101.11693

  31. Zhu Y, Yu X, Tsai Y, et al (2020) Voting-based approaches for differentially private federated learning. arXiv:2010.04851

  32. Wei K, Li J, Ding M et al (2020) Federated learning with differential privacy: algorithms and performance analysis. IEEE Trans Inf Forensics Secur 15:3454–3469

    Article  Google Scholar 

  33. Zhang Z, Guan C, Chen H, Yang X, Gong W, Yang A (2022) Adaptive privacy-preserving federated learning for fault diagnosis in internet of ships. IEEE Internet Things J 9:6844–6854

    Article  Google Scholar 

  34. Xie C, Long Y, Chen P, Kenthapadi K, Li B (2021) Certified robustness for free in differentially private federated learning. In: Annual conference on neural information processing systems

  35. Xie C, Chen M, Chen P, Li B (2021) CRFL: certifiably robust federated learning against backdoor attacks. In: 38th international conference on machine learning

  36. Truex S, Baracaldo N, Anwar A et al (2019) A hybrid approach to privacy-preserving federated learning. Inf Spektrum 42:356–357

    Article  Google Scholar 

  37. Zhou J, Wu N, Wang Y et al (2023) A differentially private federated learning model against poisoning attacks in edge computing. IEEE Trans Dependable Secure Comput 20:1941–1958

    Google Scholar 

  38. Cui L, Ma J, Zhou Y, Yu S (2023) Boosting accuracy of differentially private federated learning in industrial IoT with sparse responses. IEEE Trans Ind Inf 19:910–920

    Article  Google Scholar 

  39. Gu X, Li M, Xiong L (2023) DP-BREM: differentially-private and byzantine-robust federated learning with client momentum. arXiv:2306.12608

  40. Li Y, Wang S, Chi C, Quek T (2023) Differentially private federated learning in edge networks: the perspective of noise reduction. IEEE Netw 36(5):167–172

    Article  Google Scholar 

  41. Wang S, Tuor T, Salonidis T, Leung K, Makaya C, He T, Chan K (2019) Adaptive federated learning in resource constrained edge computing systems. IEEE J Sel Areas Commun 37:1205–1221

    Article  Google Scholar 

  42. Li X, Huan K, Yang W, Wang S, Zhang Z (2020) On the convergence of FedAvg on non-IID data. In: Eighth international conference on learning representations

  43. Vanhaesebrouck P, Bellet A, Tommasi M (2017) Decentralized collaborative learning of personalized models over networks. In: 20th international conference on artificial intelligence and statistics, vol 54, pp 509–517

  44. Amiri S, Belloum A, Nalisnick E, Klous S, Gommans L (2022) On the impact of non-IID data on the performance and fairness of differentially private federated learning. In: 52nd annual IEEE/IFIP international conference on dependable systems and networks workshops (DSN-W), pp 52–58

  45. Chiaro D, Prezioso E, Ianni M, Giampaolo F (2023) FL-enhance: a federated learning framework for balancing non-IID data with augmented and shared compressed samples. Inf Fusion 98:101836

    Article  Google Scholar 

  46. Arafeh M, Ould-Slimane H, Otrok H, Mourad A, Talhi C, Damiani E (2023) Data independent warmup scheme for non-IID federated learning. Inf Sci 623:342–360

    Article  Google Scholar 

Download references

Acknowledgements

The subject is sponsored by the National Natural Science Foundation of P. R. China (No. 61872196, No. 61872194, No. 61902196, No. 62102194 and No. 62102196), Six Talent Peaks Project of Jiangsu Province (No. RJFW-111), Postgraduate Research and Practice Innovation Program of Jiangsu Province (No. KYCX22_1019).

Author information

Authors and Affiliations

  1. School of Computer Science, Nanjing University of Posts and Telecommunications, Wenyuan Road, Nanjing, 210023, China

    Shuyan Cheng, Peng Li, Ruchuan Wang & He Xu

Authors
  1. Shuyan Cheng
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Peng Li
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Ruchuan Wang
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. He Xu
    View author publications

    You can also search for this author inPubMed Google Scholar

Corresponding author

Correspondence to Peng Li.

Ethics declarations

Conflict of interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Cheng, S., Li, P., Wang, R. et al. Differentially private federated learning with non-IID data. Computing 106, 2459–2488 (2024). https://doi.org/10.1007/s00607-024-01257-2

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00607-024-01257-2

Keywords

Mathematics Subject Classification