Please wait a minute...
 首页  期刊介绍 期刊订阅 联系我们 横山亮次奖 百年刊庆
最新录用  |  预出版  |  当期目录  |  过刊浏览  |  阅读排行  |  下载排行  |  引用排行  |  横山亮次奖  |  百年刊庆
清华大学学报(自然科学版)  2022, Vol. 62 Issue (5): 832-841    DOI: 10.16511/j.cnki.qhdxxb.2022.26.002
  专题:漏洞分析与风险评估 本期目录 | 过刊浏览 | 高级检索 |
宋宇波1,2, 朱靖恺1,2, 赵灵奇1,2, 胡爱群2,3
1. 东南大学 网络空间安全学院, 江苏省计算机网络技术重点实验室, 南京 211189;
2. 紫金山实验室, 南京 211189;
3. 东南大学 信息科学与工程学院, 移动通信国家重点实验室, 南京 211189
Centralized federated learning model based on model accuracy
SONG Yubo1,2, ZHU Jingkai1,2, ZHAO Lingqi1,2, HU Aiqun2,3
1. Jiangsu Key Laboratory of Computer Networking Technology, School of Cyber Science and Engineering, Southeast University, Nanjing 211189, China;
2. Purple Mountain Laboratories, Nanjing 211189, China;
3. State Key Laboratory of Mobile Communications, School of Information Science and Engineering, Southeast University, Nanjing 211189, China
全文: PDF(6739 KB)   HTML
输出: BibTeX | EndNote (RIS)      
摘要 现有的联邦学习存在恶意中央服务器和恶意参与者发布虚假数据毒害模型等问题。针对此情况,该文提出了一种去中心化的联邦学习模型,该模型将聚合工作由中央服务器移至参与者本地,各个参与者依据聚合算法将训练之后的模型参数写入交易,生成区块发布到区块链网络中。采用一种基于模型准确率的Byzantine容错共识算法构建共识小组,通过建立节点信息表实现节点动态加入。对所提的链上去中心化联邦学习模型的吞吐量、时延等性能进行了相关测试,结果表明:在相同条件下,基于模型准确率的高性能Byzantine容错共识算法相较于传统的Byzantine容错共识算法,吞吐量提升60%,系统平均时延从6 s减少到1 s。
E-mail Alert
关键词 联邦学习区块链共识机制模型准确率去中心化学习    
Abstract:Existing federated learning models have problems due to malicious central servers and malicious participants publishing false data that poisons the model. A decentralized federated learning model was developed to address these problems by moving the aggregation work from the central server to the participants' computers. Each participant uses the aggregation algorithm to write the trained model parameters into the transaction and generates blocks that are then published to the blockchain network. A Byzantine fault-tolerant consensus algorithm based on model accuracy is used to build a consensus group and the nodes are dynamically joined by establishing a node information table. The results show that under the same conditions, compared with the traditional Byzantine fault-tolerant consensus algorithm, the throughput of the high-performance Byzantine fault-tolerant consensus algorithm based on model accuracy is increased by 60%, and the average system delay is reduced from 6 s to 1 s.
Key wordsfederal learning    blockchain    consensus mechanism    model accuracy    decentralized learning
收稿日期: 2021-08-27      出版日期: 2022-04-26
作者简介: 宋宇波(1977—),男,副教授。
宋宇波, 朱靖恺, 赵灵奇, 胡爱群. 基于模型准确率的链上去中心化联邦学习模型[J]. 清华大学学报(自然科学版), 2022, 62(5): 832-841.
SONG Yubo, ZHU Jingkai, ZHAO Lingqi, HU Aiqun. Centralized federated learning model based on model accuracy. Journal of Tsinghua University(Science and Technology), 2022, 62(5): 832-841.
链接本文:  或
[1] 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 (AISTATS). Fort Lauderdale, USA:JMLR, 2017:1273-1282.
[2] ZHANG J L, CHEN J J, WU D, et al. Poisoning attack in federated learning using generative adversarial nets[C]//Proceedings of the 18th IEEE International Conference on Trust, Security and Privacy in Computing and Communications/13th IEEE International Conference on Big Data Science and Engineering (TrustCom/BigDataSE). Rotorua, New Zealand:IEEE, 2019:374-380.
[3] LIU Y, YU J J Q, KANG J W, et al. Privacy-preserving traffic flow prediction:A federated learning approach[J]. IEEE Internet of Things Journal, 2020, 7(8):7751-7763.
[4] 赵灵奇, 宋宇波, 张克落, 等. 基于区块链和分层加密的物流隐私保护机制[J]. 应用科学学报, 2019, 37(2):224-234. ZHAO L Q, SONG Y B, ZHANG K L, et al. Logistics information privacy protection based on blockchain and hierarchical encryption[J]. Journal of Applied Sciences, 2019, 37(2):224-234. (in Chinese)
[5] FINN C, ABBEEL P, LEVINE S. Model-agnostic meta-learning for fast adaptation of deep networks[C]//Proceedings of the 34th International Conference on Machine Learning. Sydney, Australia:JMLR, 2017:1126-1135.
[6] HUBERT E, TOMER K, BRENDAN M, et al. Semi-cyclic stochastic gradient descent[C]//Proceedings of the 36th International Conference on Machine Learning. Long Beach, USA:PMLR 97, 2019:1764-1773.
[7] WANG H Y, YUROCHKIN M, SUN Y K, et al. Federated learning with matched averaging[C]//Proceedings of the 8th International Conference on Learning Representations (ICLR). Addis Ababa, Ethiopia:ICLR, 2020:1-16.
[8] MENG W Z, TISCHHAUSER E W, WANG Q J, et al. When intrusion detection meets blockchain technology:A review[J]. IEEE Access, 2018(6):10179-10188.
[9] LU Y L, HUANG X H, DAI Y Y, et al. Blockchain and federated learning for privacy-preserved data sharing in industrial IoT[J]. IEEE Transactions on Industrial Informatics, 2020, 16(6):4177-4186.
[10] NASR M, SHOKRI R, HOUMANSADR A. Comprehensive privacy analysis of deep learning:Passive and active white-box inference attacks against centralized and federated learning[C]//Proceedings of the 2019 IEEE Symposium on Security and Privacy (SP). San Francisco, USA:IEEE, 2019:739-753.
[11] NAKAMOTO S. Bitcoin:A peer-to-peer electronic cash system[Z/OL].[2021-07-10].
[12] 宋宇波, 张仕奇, 宋睿. 一种基于投票权竞争的区块链共识机制[J]. 山东大学学报(理学版), 2020, 55(3):43-50. SONG Y B, ZHANG S Q, SONG R. A blockchain consensus mechanism based on voting rights competition[J]. Journal of Shandong University (Natural Science), 2020, 55(3):43-50. (in Chinese)
[13] AMOUSSOU-GUENOU Y, DEL POZZO A, POTOP-BUTUCARU M, et al. Dissecting tendermint[C]//Proceedings of the 7th International Conference on Networked Systems. Marrakech, Morocco:Springer, 2019:166-182.
No related articles found!
Full text



版权所有 © 《清华大学学报(自然科学版)》编辑部
本系统由北京玛格泰克科技发展有限公司设计开发 技术支持