Please wait a minute...
Frontiers of Computer Science

ISSN 2095-2228

ISSN 2095-2236(Online)

CN 10-1014/TP

Postal Subscription Code 80-970

2018 Impact Factor: 1.129

Front. Comput. Sci.    2024, Vol. 18 Issue (5) : 185348    https://doi.org/10.1007/s11704-024-3767-z
Artificial Intelligence
FedTop: a constraint-loosed federated learning aggregation method against poisoning attack
Che WANG1, Zhenhao WU1, Jianbo GAO1,2(), Jiashuo ZHANG1, Junjie XIA3, Feng GAO3, Zhi GUAN4, Zhong CHEN1,2
1. School of Computer Science, Peking University, Beijing 100871, China
2. Peking University Chongqing Research Institute of Big Data, Chongqing 401329, China
3. China Unicom, Beijing 100033, China
4. National Engineering Research Center for Software Engineering, Peking University, Beijing 100871, China
 Download: PDF(1870 KB)   HTML
 Export: BibTeX | EndNote | Reference Manager | ProCite | RefWorks
Corresponding Author(s): Jianbo GAO   
Just Accepted Date: 28 April 2024   Issue Date: 14 June 2024
 Cite this article:   
Che WANG,Zhenhao WU,Jianbo GAO, et al. FedTop: a constraint-loosed federated learning aggregation method against poisoning attack[J]. Front. Comput. Sci., 2024, 18(5): 185348.
 URL:  
https://academic.hep.com.cn/fcs/EN/10.1007/s11704-024-3767-z
https://academic.hep.com.cn/fcs/EN/Y2024/V18/I5/185348
  
Aggregation methodsMNISTCIFAR-10YELP
IIDNon-IIDIIDNon-IIDIIDNon-IID
FedAvg [6]1.5501.5601.2831.3421.0811.156
Median [3]1.5611.5861.2821.3071.0991.120
Multi-Krum [7]1.5841.6591.3041.3861.1201.775
Zeno [5]1.5521.5671.3241.3941.1021.182
FedTop1.5481.5640.9431.0691.0891.118
Tab.1  Loss of different methods in normal environments
Aggregation methods Scale up Mix Scale down
3 8 12 3 8 12 3 8 12
FedAvg [6] 2.021 2.056 2.045 1.968 2.027 2.033 1.779 2.078 2.235
Median [3] 1.695 2.046 2.089 1.657 1.998 2.085 1.656 2.245 2.250
Multi-Krum [7] 1.702 2.136 2.129 1.713 2.010 2.116 1.725 2.260 2.268
Zeno [5] 1.610 2.062 2.055 1.620 1.817 2.024 1.851 2.238 2.279
FedTop 1.597 1.576 1.624 1.598 1.585 1.579 1.581 1.589 1.602
Tab.2  Loss of different methods with YELP in malicious environments
1 Q, Yang Y, Liu T, Chen Y Tong . Federated machine learning: concept and applications. ACM Transactions on Intelligent Systems and Technology, 2019, 10(2): 12
2 P, Gupta K, Yadav B B, Gupta M, Alazab T R Gadekallu . A novel data poisoning attack in federated learning based on inverted loss function. Computers & Security, 2023, 130: 103270
3 K, Pillutla S M, Kakade Z Harchaoui . Robust aggregation for federated learning. IEEE Transactions on Signal Processing, 2022, 70: 1142−1154
4 C, Xie M, Chen P Y, Chen B Li . CRFL: certifiably robust federated learning against backdoor attacks. In: Proceedings of the 38th International Conference on Machine Learning. 2021, 11372–11382
5 C, Xie S, Koyejo I Gupta . Zeno++: robust fully asynchronous SGD. In: Proceedings of the 37th International Conference on Machine Learning. 2020, 972
6 X, Li K, Huang W, Yang S, Wang Z Zhang . On the convergence of fedAvg on non-IID data. In: Proceedings of the 8th International Conference on Learning Representations. 2020
7 P, Blanchard Mhamdi E M, El R, Guerraoui J Stainer . Machine learning with adversaries: Byzantine tolerant gradient descent. In: Proceedings of the 31st International Conference on Neural Information Processing Systems. 2017, 118–128
8 L Deng . The MNIST database of handwritten digit images for machine learning research [Best of the Web]. IEEE Signal Processing Magazine, 2012, 29(6): 141−142
9 A Krizhevsky . Learning multiple layers of features from tiny images. Toronto: University of Toronto, 2009
10 E S, Alamoudi N S Alghamdi . Sentiment classification and aspect-based sentiment analysis on yelp reviews using deep learning and word embeddings. Journal of Decision Systems, 2021, 30(2−3): 259–281
[1] FCS-23767-OF-CW_suppl_1 Download
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed