MSRA-Fed: A Communication-Efficient Federated Learning Method Based on Model Split and Representation Aggregate

doi:10.12142/ZTECOM.202203005

ZTE Communications ›› 2022, Vol. 20 ›› Issue (3): 35-42.DOI: 10.12142/ZTECOM.202203005

收稿日期:2022-06-20 出版日期:2022-09-13 发布日期:2022-09-14

MSRA-Fed: A Communication-Efficient Federated Learning Method Based on Model Split and Representation Aggregate

LIU Qinbo^1,², JIN Zhihao¹, WANG Jiabo¹, LIU Yang^1,³(), LUO Wenjian^1,³

^1.School of Computer Science and Technology, Harbin Institute of Technology, Shenzhen 518055, China
^2.Guangdong Provincial Key Laboratory of Novel Security Intelligence Technologies, Shenzhen 518055, China
^3.Peng Cheng Laboratory, Shenzhen 518055, China

Received:2022-06-20 Online:2022-09-13 Published:2022-09-14
Contact: LIU Yang
About author:LIU Qinbo received his BS degree in mathematics and physics basic science from the School of Mathematical Sciences, University of Electronic Science and Technology of China in 2021. He is currently pursuing his ME degree with the School of Computer Science and Technology, Harbin Institute of Technology, Shenzhen, China. His research interests include federated learning and GNNs.|JIN Zhihao received his BE degree in computer science and technology from the School of Computer Science and Technology, Harbin Institute of Technology, Shenzhen, China in 2022. His research interests include federated learning.|WANG Jiabo received his BE degree in software engineering from the School of Information Science and Technology, Dalian Maritime University, China in 2021. He is currently pursuing his ME degree with the School of Computer Science and Technology, Harbin Institute of Technology, Shenzhen, China. His research interests include federated learning.|LUO Wenjian received his BS and PhD degrees from the Department of Computer Science and Technology, University of Science and Technology of China, in 1998 and 2003, respectively. He is currently a professor with the School of Computer Science and Technology, Harbin Institute of Technology, Shenzhen, China. His current research interests include computational intelligence and applications, network security and data privacy, machine learning, and data mining. Dr. LUO is also a senior member of the Association for Computing Machinery (ACM) and the China Computer Federation (CCF). He has been a member of the organizational team of more than ten academic conferences, in various functions, such as the program chair, the symposium chair and the publicity chair. He also serves as the chair of the IEEE CIS ECTC Task Force on Artificial Immune Systems. He also serves as an associate editor or an editorial board member for several journals, including Information Sciences, Swarm and Evolutionary Computation, Journal of Information Security and Applications, Applied Soft Computing, and Complex & Intelligent Systems.
Supported by:
his work is supported by Shenzhen Basic Research (General Project)(JCYJ20190806142601687);Shenzhen Stable Supporting Program (General Project)(GXWD20201230155427003-20200821160539001);Guangdong Provincial Key Laboratory of Novel Security Intelligence Technologies(2022B1212010005);Shenzhen Basic Research (Key Project)(JCYJ20200109113405927)

摘要/Abstract

Abstract:

Recent years have witnessed a spurt of progress in federated learning, which can coordinate multi-participation model training while protecting the data privacy of participants. However, low communication efficiency is a bottleneck when deploying federated learning to edge computing and IoT devices due to the need to transmit a huge number of parameters during co-training. In this paper, we verify that the outputs of the last hidden layer can record the characteristics of training data. Accordingly, we propose a communication-efficient strategy based on model split and representation aggregate. Specifically, we make the client upload the outputs of the last hidden layer instead of all model parameters when participating in the aggregation, and the server distributes gradients according to the global information to revise local models. Empirical evidence from experiments verifies that our method can complete training by uploading less than one-tenth of model parameters, while preserving the usability of the model.

Key words: federated learning, communication load, efficient communication, privacy protection

. [J]. ZTE Communications, 2022, 20(3): 35-42.

LIU Qinbo, JIN Zhihao, WANG Jiabo, LIU Yang, LUO Wenjian. MSRA-Fed: A Communication-Efficient Federated Learning Method Based on Model Split and Representation Aggregate[J]. ZTE Communications, 2022, 20(3): 35-42.

图/表 4

参考文献 25

1	TALUKDER A, HAAS R. AIoT: AI meets IoT and web in smart healthcare [C]//13th ACM Web Science Conference 2021. ACM, 2021: 92–98. DOI: 10.1145/3462741.3466650 DOI
2	ALKHATIB S, WAYCOTT J, BUCHANAN G, et al. Privacy and the Internet of Things (IoT) monitoring solutions for older adults: a review [J]. Studies in health technology and informatics, 2018, 252: 8–14
3	LI T, SAHU A K, TALWALKAR A, et al. Federated learning: challenges, methods, and future directions [J]. IEEE signal processing magazine, 2020, 37(3): 50–60. DOI: 10.1007/978-3-030-85559-8_13 DOI
4	NGUYEN D C, DING M, PATHIRANA P N, et al. Federated learning for internet of things: a comprehensive survey [J]. IEEE communications surveys & tutorials, 2021, 23(3): 1622-1658. DOI: 10.1109/COMST.2021.3075439 DOI
5	MCMAHAN B, MOORE E, RAMAGE D, et al. Communication-efficient learning of deep networks from decentralized data [C]//20th International Conference on Artificial Intelligence and Statistics (AISTATS). PMLR, 2017: 1273–1282. DOI: 10.48550/arXiv.1602.05629 DOI
6	SHAHID O, POURIYEH S, PARIZI R M, et al. Communication efficiency in federated learning: Achievements and challenges [EB/OL]. (2021-07-23)[2022-05-01].
7	CHAI Z, ALI A, ZAWAD S, et al. TiFL: a tier-based federated learning system [C]//29th International Symposium on High-Performance Parallel and Distributed Computing. ACM, 2020: 125–136. DOI: 10.1145/3369583.3392686 DOI
8	TAN Y, LONG G, LIU L, et al. Fedproto: federated prototype learning across heterogeneous clients [C]//AAAI Conference on Artificial Intelligence. AAAI, 2022: 8432–8440. DOI: 10.1609/aaai.v36i8.20819 DOI
9	THAPA C, CHAMIKARA M A P, CAMTEPE S, et al. Splitfed: when federated learning meets split learning [EB/OL]. (2022-02-16)[2022-05-01].
10	KONEČNÝ J, MCMAHAN H B, RAMAGE D, et al. Federated optimization: distributed machine learning for on-device intelligence [EB/OL]. (2016-10-08)[2022-05-01].
11	KHAN L U, SAAD W, HAN Z, et al. Federated learning for internet of things: recent advances, taxonomy, and open challenges [J]. IEEE communications surveys & tutorials, 2021, 23(3): 1759–1799. DOI: 10.1109/COMST.2021.3090430 DOI
12	WANG L, WANG W, LI B. CMFL: mitigating communication overhead for federated learning [C]//IEEE 39th International Conference on Distributed Computing Systems (ICDCS). IEEE, 2019: 954–964. DOI: 10.1109/ICDCS.2019.00099 DOI
13	SATTLER F, WIEDEMANN S, MÜLLER K R, et al. Robust and communication-efficient federated learning from non-IID data [J]. IEEE transactions on neural networks and learning systems, 2019, 31(9): 3400–3413. DOI: 10.1109/TNNLS.2019.2944481 DOI
14	KONEČNÝ J, MCMAHAN H B, YU F X, et al. Federated learning: strategies for improving communication efficiency [EB/OL]. (2017-10-30)[2022-05-01].
15	SURESH A T, FELIX X Y, KUMAR S, et al. Distributed mean estimation with limited communication [C]//International conference on machine learning. PMLR, 2017: 3329–3337. DOI: 10.48550/arXiv.1611.00429 DOI
16	CALDAS S, KONEČNY J, MCMAHAN H B, et al. Expanding the reach of federated learning by reducing client resource requirements [EB/OL]. (2019-01-08)[2022-05-01].
17	DEAN J, CORRADO G, MONGA R, et al. Large scale distributed deep networks [C]//25th International Conference on Neural Information Processing Systems, NIPS. 2012: 1223–1231
18	HUANG Y, CHENG Y, BAPNA A, et al. GPipe: efficient training of giant neural networks using pipeline parallelism [C]//33rd International Conference on Neural Information Processing Systems. NIPS, 2019: 103–112
19	JIANG L, WANG Y, ZHENG W, et al. LSTMSPLIT: Effective SPLIT Learning based LSTM on Sequential Time-Series Data [EB/OL]. (2022-03-08)[2022-05-01].
20	CHEN Y, SUN X, JIN Y. Communication-efficient federated deep learning with layerwise asynchronous model update and temporally weighted aggregation [J]. IEEE transactions on neural networks and learning systems, 2019, 31(10): 4229–4238. DOI: 10.1109/TNNLS.2019.2953131 DOI
21	THAPA C, CHAMIKARA M A P, CAMTEPE S A. Advancements of federated learning towards privacy preservation: from federated learning to split learning [M]//Federated Learning Systems. Springer, Cham, 2021: 79–109
22	WANG W, FENG F, HE X, et al. Denoising implicit feedback for recommendation [C]//14th ACM International Conference on Web Search and Data Mining. ACM, 2021: 373–381. DOI: 10.1145/3437963.3441800 DOI
23	SINGH A, VEPAKOMMA P, GUPTA O, et al. Detailed comparison of communication efficiency of split learning and federated learning [EB/OL]. (2019-01-08)[2022-05-01].
24	WAINAKH A, VENTOLA F, MÜßIG T, et al. User-level label leakage from gradients in federated learning [J]. Proceedings on privacy enhancing technologies, 2022(2): 227–244. DOI: 10.2478/popets-2022-0043 DOI
25	KHOSRAVY M, NAKAMURA K, HIROSE Y, et al. Model inversion attack: Analysis under gray-box scenario on deep learning based face recognition system [J]. KSII transactions on internet and information systems (TIIS), 2021, 15(3): 1100–1118. DOI: 10.3837/tiis.2021.03.015 DOI

Label	1st Neuron	2nd Neuron	3rd Neuron	4th Neuron	5th Neuron	6th Neuron	7th Neuron	8th Neuron	9th Neuron	10th Neuron	11th Neuron	12th Neuron
Label-3	1.00	2.94	1.98	11.82	1.80	2.16	0.01	9.60	0.28	4.29	3.84	11.03
Label-6	4.82	0.00	3.98	1.88	0.00	3.65	1.17	0.03	36.54	8.05	3.65	0.72
Label-7	0.00	0.01	1.37	13.00	4.92	5.20	3.33	0.10	17.91	12.54	3.82	0.08
Blended	52.11	34.38	18.51	24.65	16.18	11.99	3.99	35.46	17.24	18.57	27.22	23.68

Label	1st Neuron	2nd Neuron	3rd Neuron	4th Neuron	5th Neuron	6th Neuron	7th Neuron	8th Neuron	9th Neuron	10th Neuron	11th Neuron	12th Neuron
Label-3	1.00	2.94	1.98	11.82	1.80	2.16	0.01	9.60	0.28	4.29	3.84	11.03
Label-6	4.82	0.00	3.98	1.88	0.00	3.65	1.17	0.03	36.54	8.05	3.65	0.72
Label-7	0.00	0.01	1.37	13.00	4.92	5.20	3.33	0.10	17.91	12.54	3.82	0.08
Blended	52.11	34.38	18.51	24.65	16.18	11.99	3.99	35.46	17.24	18.57	27.22	23.68

MSRA-Fed: A Communication-Efficient Federated Learning Method Based on Model Split and Representation Aggregate

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

图/表 4

参考文献 25

相关文章 0

编辑推荐

Metrics