Federated learning method based on adaptive differential privacy and client selection optimization

doi:10.11772/j.issn.1001-9081.2024020162

Journal of Computer Applications ›› 2025, Vol. 45 ›› Issue (2): 482-489.DOI: 10.11772/j.issn.1001-9081.2024020162

• Cyber security • Previous Articles

Federated learning method based on adaptive differential privacy and client selection optimization

Chao XU¹^,², Shufen ZHANG¹^,²^,³(), Haitian CHEN¹^,⁴, Lulu PENG¹^,⁴, Shuaihua ZHANG¹^,⁴

^1.College of Science，North China University of Science and Technology，Tangshan Hebei 063210，China
^2.Hebei Key Laboratory of Data Science and Application （North China University of Science and Technology），Tangshan Hebei 063210，China
^3.Tangshan Key Laboratory of Big Data Security and Intelligent Computing，Beijing Jiaotong University，Tangshan Hebei 063210，China
^4.Tangshan Key Laboratory of Data Science （North China University of Science and Technology），Tangshan Hebei 063210，China

Received:2024-02-21 Revised:2024-03-23 Accepted:2024-04-01 Online:2024-06-04 Published:2025-02-10
Contact: Shufen ZHANG
About author:XU Chao， born in 1998， M. S. candidate. His research interests include data security， differential privacy， federated learning.
CHEN Haitian， born in 1998， M. S. candidate. His research interests include data security， privacy protection.
PENG Lulu， born in 1994， M. S. candidate. His research interests include data security， privacy protection.
ZHANG Shuaihua， born in 1999， M. S. candidate. His research interests include data security， network security， privacy protection.
Supported by:
National Natural Science Foundation of China(U20A20179)

基于自适应差分隐私与客户选择优化的联邦学习方法

徐超¹^,², 张淑芬¹^,²^,³(), 陈海田¹^,⁴, 彭璐璐¹^,⁴, 张帅华¹^,⁴

^1.华北理工大学理学院，河北唐山 063210
^2.河北省数据科学与应用重点实验室（华北理工大学），河北唐山 063210
^3.北京交通大学唐山市大数据安全与智能计算重点实验室，河北唐山 063210
^4.唐山市数据科学重点实验室（华北理工大学），河北唐山 063210

通讯作者: 张淑芬
作者简介:徐超（1998—），男，河南驻马店人，硕士研究生，CCF会员，主要研究方向：数据安全、差分隐私、联邦学习
陈海田（1998—），男，湖南娄底人，硕士研究生，CCF会员，主要研究方向：数据安全、隐私保护
彭璐璐（1994—），女，河南驻马店人，硕士研究生，CCF会员，主要研究方向：数据安全、隐私保护
张帅华（1999—），男，河北石家庄人，硕士研究生，CCF会员，主要研究方向：数据安全、网络安全、隐私保护。
基金资助:
国家自然科学基金资助项目(U20A20179)

Abstract

Abstract:

The method of applying differential privacy to federated learning has been one of the key techniques for protecting the privacy of training data. Addressing the issue that most previous works do not consider the heterogeneity of parameters， resulting in pruning training parameters uniformly， leading to uniform noise addition in each round， thus affecting model convergence and the quality of training parameters， an adaptive noise addition scheme based on gradient clipping was proposed. Considering the heterogeneity of gradients， adaptive gradient clipping was executed for different clients in different rounds， thereby allowing for the adaptive adjustment of noise magnitude. At the same time， to further improve model performance， different from traditional client random sampling methods， a client sampling method that combines roulette and elite preservation was proposed. Combining the aforementioned two methods， a Client Selection and Adaptive Gradient Clipping Differential Privacy_Federated Learning （CS&AGC DP_FL） was proposed. Experimental results demonstrate that， when the privacy budget is 0.5， compared to the Federated Learning method based on Adaptive Differential Privacy （Adapt DP_FL）， the proposed method improves the final model’s classification accuracy by 4.9 percentage points under the same level of privacy constraints. Additionally， in terms of convergence speed， the proposed method requires 4 to 10 fewer rounds to reach convergence compared to the methods to be compared.

Key words: federated learning, differential privacy, adaptive noise, roulette, elite reservation

摘要：

将差分隐私应用于联邦学习的方法是保护训练数据隐私的关键技术之一。针对之前多数工作未考虑参数的异质性，对训练参数均匀裁剪使每轮加入的噪声都是均匀的，从而影响模型收敛和训练参数质量的问题，提出一种基于梯度裁剪的自适应噪声添加方案。考虑梯度的异质性，在不同轮次为不同客户端执行自适应的梯度裁剪，从而使噪声大小自适应调整；同时，为进一步提升模型性能，对比传统的客户端随机采样方式，提出一种结合轮盘赌与精英保留的客户端采样方法。结合上述2种方法，提出一种结合客户端选择的自适应差分隐私联邦学习（CS&AGC DP_FL）方法。实验结果表明，在隐私预算为0.5时，相较于自适应差分隐私的联邦学习方法（Adapt DP_FL），所提方法能在相同级别的隐私约束下使最终的模型分类准确率提升4.9个百分点，并且在收敛速度方面，所提方法相较于对比方法进入收敛状态所需的轮次减少了4~10轮。

关键词: 联邦学习, 差分隐私, 自适应噪声, 轮盘赌, 精英保留

CLC Number:

TP309

Chao XU, Shufen ZHANG, Haitian CHEN, Lulu PENG, Shuaihua ZHANG. Federated learning method based on adaptive differential privacy and client selection optimization[J]. Journal of Computer Applications, 2025, 45(2): 482-489.

徐超, 张淑芬, 陈海田, 彭璐璐, 张帅华. 基于自适应差分隐私与客户选择优化的联邦学习方法[J]. 《计算机应用》唯一官方网站, 2025, 45(2): 482-489.

Figures/Tables 10

Fig. 1 Federated learning training process

Fig. 2 Training framework

Fig. 3 Accuracy on MNIST and FashionMNIST datasets

Fig. 4 Performance comparison of different methods under different ε values on MNIST dataset

Tab. 1 Accuracies of different privacy budgets on MNIST dataset

ε	准确率/%
ε	DP_FL	Adapt DP_FL	CS&AGC DP_FL
0.5	73.93	78.13	83.03
0.6	78.50	82.17	85.67
0.7	79.37	86.27	88.53
0.8	79.83	88.03	88.73

Fig. 5 Performance comparison of different methods under different ε values on FashionMNIST dataset

Tab. 2 Accuracies of different privacy budgets on FashionMNIST dataset

ε	准确率/%
ε	DP_FL	Adapt DP_FL	CS&AGC DP_FL
0.5	70.57	73.73	78.53
0.6	72.07	76.63	78.67
0.7	72.17	78.47	78.97
0.8	74.13	79.13	80.10

Tab. 3 Partial rounds’ accuracies when ε=0.5

数据集	方法	第5轮	第10轮	第15轮	第20轮
MNIST	DP_FL	51.60	57.30	62.10	68.73
	Adapt DP_FL	42.50	54.50	62.57	68.57
	CS&AGC DP_FL	70.60	76.23	79.20	80.57
FashionMNIST	DP_FL	58.57	62.13	67.73	65.93
	Adapt DP_FL	59.57	63.43	68.03	69.60
	CS&AGC DP_FL	70.80	74.73	75.83	76.50

Fig. 6 Performance comparison of different methods under different k values on MNIST dataset

Fig. 7 Performance comparison of different methods under different k values on FashionMNIST dataset

References 29

1	POUYANFAR S， SADIQ S， YAN Y， et al. A survey on deep learning： algorithms， techniques， and applications［J］. ACM Computing Surveys， 2018， 51（5）： No.92.
2	McMAHAN H B， MOORE E， RAMAGE D， et al. Communication-efficient learning of deep networks from decentralized data［C］// Proceedings of the 20th International Conference on Artificial Intelligence and Statistics. New York： JMLR.org， 2017： 1273-1282.
3	ZHU L， LIU Z， HAN S. Deep leakage from gradients［C］// Proceedings of the 33rd International Conference on Neural Information Processing Systems. Red Hook： Curran Associates Inc.， 2019： 14774-14784.
4	SONG C， RISTENPART T， SHMATIKOV V. Machine learning models that remember too much［C］// Proceedings of the 2017 ACM SIGSAC Conference on Computer and Communications Security. New York： ACM， 2017： 587-601.
5	DWORK C， McSHERRY F， NISSIM K， et al. Calibrating noise to sensitivity in private data analysis［C］// Proceedings of the 2006 Theory of Cryptography Conference， LNCS 3876. Berlin： Springer， 2006： 265-284.
6	RIVEST R L， ADLEMAN L， DERTOUZOS M L. On data banks and privacy homomorphisms［M］// DeMILLO R A， DOBKIN D P， JONES A K， et al. Foundations of secure computation. Cambridge： Academia Press， 1978： 169-179.
7	LI Y， ZHOU Y， JOLFAEI A， et al. Privacy-preserving federated learning framework based on chained secure multiparty computing［J］. IEEE Internet of Things Journal， 2021， 8（8）： 6178-6186.
8	AHMAD A， LUO W， ROBLES-KELLY A. Robust federated learning under statistical heterogeneity via Hessian spectral decomposition［J］. Pattern Recognition， 2023， 141： No.109635.
9	WANG Z， HU Q， ZOU X. Can we trust the similarity measurement in federated learning？［EB/OL］. ［2023-12-23］..
10	ABADI M， CHU A， GOODFELLOW I， et al. Deep learning with differential privacy［C］// Proceedings of the 2016 ACM SIGSAC Conference on Computer and Communications Security. New York： ACM， 2016： 308-318.
11	GEYER R C， KLEIN T， NABI M. Differentially private federated learning： a client level perspective［EB/OL］. ［2023-11-16］..
12	ZHENG Q， CHEN S， LONG Q， et al. Federated f-differential privacy［C］// Proceedings of the 24th International Conference on Artificial Intelligence and Statistics. New York： JMLR.org， 2021： 2251-2259.
13	SINGH I， ZHOU H， YANG K， et al. Differentially-private federated neural architecture search［EB/OL］. ［2023-10-18］..
14	TRUEX S， LIU L， CHOW K H， et al. LDP-Fed： federated learning with local differential privacy［C］// Proceedings of the 3rd ACM International Workshop on Edge Systems， Analytics and Networking. New York： ACM， 2020： 61-66.
15	WEI K， LI J， DING M， et al. User-level privacy-preserving federated learning： analysis and performance optimization［J］. IEEE Transactions on Mobile Computing， 2022， 21（9）： 3388-3401.
16	康海燕，冀源蕊. 基于本地化差分隐私的联邦学习方法研究［J］. 通信学报， 2022， 43（10）： 94-105.
	KANG H Y， JI Y R. Research on federated learning approach based on local differential privacy［J］. Journal on Communications， 2022， 43（10）： 94-105.
17	VAN DER VEEN K L， SEGGERS R， BLOEM P， et al. Three tools for practical differential privacy［EB/OL］. ［2023-10-09］..
18	FU J， CHEN Z， HAN X. Adapt DP-FL： differentially private federated learning with adaptive noise［C］// Proceedings of the 2022 IEEE International Conference on Trust， Security and Privacy in Computing and Communications. Piscataway： IEEE， 2022： 656-663.
19	XU Z， SHI S， LIU A X， et al. An adaptive and fast convergent approach to differentially private deep learning［C］// Proceedings of the 2020 IEEE Conference on Computer Communications. Piscataway： IEEE， 2020： 1867-1876.
20	YU L， LIU L， PU C， et al. Differentially private model publishing for deep learning［C］// Proceedings of the 2019 IEEE Symposium on Security and Privacy. Piscataway： IEEE， 2019： 332-349.
21	张少波，张激勇，朱更明，等. 基于Bregman散度和差分隐私的个性化联邦学习方法［J］. 软件学报， 2024， 35（11）： 5249-5262.
	ZHANG S B， ZHANG J Y， ZHU G M， et al. Personalized federated learning method based on Bregman divergence and differential privacy［J］. Journal of Software， 2024， 35（11）： 5249-5262.
22	尹春勇，屈锐. 基于个性化差分隐私的联邦学习算法［J］. 计算机应用， 2023， 43（4）：1160-1168.
	YIN C Y， QU R. Federated learning algorithm based on personalized differential privacy［J］. Journal of Computer Applications， 2023， 43（4）：1160-1168.
23	GOETZ J， MALIK K， BUI D， et al. Active federated learning［EB/OL］. ［2023-11-01］..
24	XIE Y， ZHANG L. Federated learning with personalized differential privacy combining client selection［C］// Proceedings of the 8th International Conference on Big Data Computing and Communications. Piscataway： IEEE， 2022： 79-87.
25	LI T， SANJABI M， BEIRAMI A， et al. Fair resource allocation in federated learning［EB/OL］. ［2024-01-01］..
26	YANG C， ZHAO X. Study on the selection method of federated learning clients for smart manufacturing［J］. Electronics， 2023， 12（11）： No.2532.
27	YANG Q， LIU Y， CHENG Y， et al. Federated learning， SLAIML［M］. Cham： Springer， 2020： 1-189.
28	DWORK C. Differential privacy： a survey of results［C］// Proceedings of the 2008 International Conference on Theory and Applications of Models of Computation， LNCS 4978. Berlin： Springer， 2008： 1-19.
29	DWORK C， ROTH A. The algorithmic foundations of differential privacy［J］. Foundations and Trends® in Theoretical Computer Science， 2014， 9（3/4）： 211-407.

[1]	Liang ZHU, Jingzhe MU, Hongqiang ZUO, Jingzhong GU, Fubao ZHU. Location privacy-preserving recommendation scheme based on federated graph neural network [J]. Journal of Computer Applications, 2025, 45(1): 136-143.
[2]	Yan YAN, Xingying QIAN, Pengbin YAN, Jie YANG. Federated learning-based statistical prediction and differential privacy protection method for location big data [J]. Journal of Computer Applications, 2025, 45(1): 127-135.
[3]	Shufen ZHANG, Hongyang ZHANG, Zhiqiang REN, Xuebin CHEN. Survey of fairness in federated learning [J]. Journal of Computer Applications, 2025, 45(1): 1-14.
[4]	Tingwei CHEN, Jiacheng ZHANG, Junlu WANG. Random validation blockchain construction for federated learning [J]. Journal of Computer Applications, 2024, 44(9): 2770-2776.
[5]	Zhizheng ZHANG, Xiaojian ZHANG, Junqing WANG, Guanghui FENG. Federated spatial data publication method with differential privacy and secure aggregation [J]. Journal of Computer Applications, 2024, 44(9): 2777-2784.
[6]	Zheyuan SHEN, Keke YANG, Jing LI. Personalized federated learning method based on dual stream neural network [J]. Journal of Computer Applications, 2024, 44(8): 2319-2325.
[7]	Wei LUO, Jinquan LIU, Zheng ZHANG. Dual vertical federated learning framework incorporating secret sharing technology [J]. Journal of Computer Applications, 2024, 44(6): 1872-1879.
[8]	Xuebin CHEN, Zhiqiang REN, Hongyang ZHANG. Review on security threats and defense measures in federated learning [J]. Journal of Computer Applications, 2024, 44(6): 1663-1672.
[9]	Sunjie YU, Hui ZENG, Shiyu XIONG, Hongzhou SHI. Incentive mechanism for federated learning based on generative adversarial network [J]. Journal of Computer Applications, 2024, 44(2): 344-352.
[10]	Peng PENG, Zhiwei NI, Xuhui ZHU, Qian CHEN. Interference trajectory publication based on improved glowworm swarm algorithm and differential privacy [J]. Journal of Computer Applications, 2024, 44(2): 496-503.
[11]	Rui GAO, Xuebin CHEN, Zucuan ZHANG. Dynamic social network privacy publishing method for partial graph updating [J]. Journal of Computer Applications, 2024, 44(12): 3831-3838.
[12]	Zucuan ZHANG, Xuebin CHEN, Rui GAO, Yuanhuai ZOU. Federated learning client selection method based on label classification [J]. Journal of Computer Applications, 2024, 44(12): 3759-3765.
[13]	Jie WU, Xuezhong QIAN, Wei SONG. Personalized federated learning based on similarity clustering and regularization [J]. Journal of Computer Applications, 2024, 44(11): 3345-3353.
[14]	Xuebin CHEN, Changsheng QU. Overview of backdoor attacks and defense in federated learning [J]. Journal of Computer Applications, 2024, 44(11): 3459-3469.
[15]	Shuaihua ZHANG, Shufen ZHANG, Mingchuan ZHOU, Chao XU, Xuebin CHEN. Malicious traffic detection model based on semi-supervised federated learning [J]. Journal of Computer Applications, 2024, 44(11): 3487-3494.

Federated learning method based on adaptive differential privacy and client selection optimization

基于自适应差分隐私与客户选择优化的联邦学习方法

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 10

References 29

Related Articles 15

Recommended Articles

Metrics