[ "叶剑(1976- ),男,中国联合网络通信有限公司软件研究院高级工程师,主要研究方向为大数据技术、数据挖掘、数据安全、人工智能、数据治理等。" ]
[ "李文(1991- ),女,中国联合网络通信有限公司西安软件研究院助理工程师,主要研究方向为可视化大数据分析、大数据治理、数据安全等。" ]
网络首发:2023-11,
纸质出版:2023-11-15
移动端阅览
叶剑, 李文. 支持互联互通的隐私计算网关设计与实现[J]. 大数据, 2023,9(6):28-38.
Jian YE, Wen LI. Design and implementation of trusted gateway for privacy-preserving interconnection[J]. Big data research, 2023, 9(6): 28-38.
叶剑, 李文. 支持互联互通的隐私计算网关设计与实现[J]. 大数据, 2023,9(6):28-38. DOI: 10.11959/j.issn.2096-0271.2023072.
Jian YE, Wen LI. Design and implementation of trusted gateway for privacy-preserving interconnection[J]. Big data research, 2023, 9(6): 28-38. DOI: 10.11959/j.issn.2096-0271.2023072.
基于国内外隐私计算发展现状,总结了隐私计算互联互通研究进程。运用系统架构视角,阐述互联互通技术的“应用层、协议层、通信层”三层次实现路径。针对目前互联互通平台计算原理复杂、架构多样化等特点,创新性地提出Adaptation机制互联互通框架。通过关键技术的设计和实现,在保证原有功能实现的基础上,解决了不同架构的兼容问题。通过传统机器学习、横向联邦、纵向联邦具体实验场景,针对数据量、特征分布等维度,验证了Adaptation框架下互联互通可信网关的有效性和合理性。
Based on the current development status of privacy computing both domestically and internationally
this paper summarizes the research progress in the field of privacy computing interconnectivity.Utilizing a systemic architecture perspective
it elaborates on the implementation pathways of interconnectivity technology across three levels: the"application layer
protocol layer
and communication layer".Considering the complexities arising from the current characteristics of interconnectivity platforms
including intricate computational principles and diverse architectures
an innovative Adaptation mechanism is proposed as an interconnectivity framework.Through the design and implementation of key technologies
this framework not only ensures the realization of existing functionalities but also addresses the compatibility issues posed by different architectures.By means of experiments conducted in various scenarios encompassing traditional machine learning
horizontal federated learning
and vertical federated learning
and considering dimensions such as data volume and feature distribution
the effectiveness and viability of the trustworthy interconnectivity gateway under the Adaptation framework have been demonstrated.
UN Global Working Group on Big Data . UN handbook on privacy- preserving computation techniques [R ] . 2019 .
隐私计算联盟 , 中国信息通信研究院 . 隐私计算应用研究报告(2022年) [R ] . 2022 .
Privacy Computing Alliance , China Academy of Information and Communications Technology . Research report on privacy computing applications (2022) [R ] . 2022 .
王思源 , 闫树 . 隐私计算面临的挑战与发展趋势浅析 [J ] . 通信世界 , 2022 ( 2 ): 19 - 21 .
WANG S Y , YAN S . Challenges and development trends of privacy computing [J ] . Communications World , 2022 ( 2 ): 19 - 21 .
隐私计算联盟 , 中国信息通信研究院 . 隐私计算白皮书 [R ] . 2021 .
Privacy Computing Alliance , China Academy of Information and Communications Technology . Privacy computing white paper [R ] . 2021 .
吕艾临 , 闫树 . 隐私计算跨平台互联互通的若干思考 [J ] . 信息通信技术与政策 , 2022 ( 5 ): 2 - 6 .
LYU A L , YAN S . Some thoughts on cross-platform interconnection of privacy preserving computing [J ] . Information and Communications Technology and Policy , 2022 ( 5 ): 2 - 6 .
姚明 , 何浩 , 李博 , 等 . 隐私计算跨平台互联互通研究与实践 [J ] . 中国科技信息 , 2022 ( 16 ): 140 - 143 .
YAO M , HE H , LI B , et al . Research and practice on cross-platform interconnection of privacy computing [J ] . China Science and Technology Information , 2022 ( 16 ): 140 - 143 .
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 .[S.l.:s.n. ] , 2017 : 1273 - 1282 .
LIU Y , FAN T , CHEN T J , et al . FATE:an industrial grade platform for collaborative learning with data protection [J ] . Journal of Machine Learning Research , 2021 , 22 ( 226 ): 1 - 6 .
KAIROUZ P , AVENT B , MCMAHAN H B , et al . Advances and open problems in federated learning [J ] . Foundations and Trends® in Machine Learning , 2021 , 14 ( 1/2 ): 1 - 210 .
FENG S , YU H . Multi-participant multiclass vertical federated learning [EB ] . arXiv preprint,2020 , 2020 ,arXiv:2001.11154.
0
浏览量
288
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构
京公网安备11010802024621