大数据 ›› 2022, Vol. 8 ›› Issue (4): 105-132.doi: 10.11959/j.issn.2096-0271.2022032

• 研究 • 上一篇    下一篇

联邦推荐系统综述

朱智韬1,2, 司世景1, 王健宗1, 肖京1   

  1. 1 平安科技(深圳)有限公司,广东 深圳 518063
    2 中国科学技术大学,安徽 合肥 230026
  • 出版日期:2022-07-15 发布日期:2022-07-01
  • 作者简介:朱智韬(1996- ),男,中国科学技术大学硕士生,平安科技(深圳)有限公司算法工程师,中国计算机学会(CCF)会员,主要研究方向为人工智能、联邦学习和推荐系统等
    司世景(1988- ),男,博士,平安科技(深圳)有限公司资深算法研究员,中国科学技术大学硕士生企业导师, CCF会员。发表机器学习、大数据和人工智能领域国际核心论文20余篇
    王健宗(1983- ),男,博士,平安科技(深圳)有限公司副总工程师、资深人工智能总监。CCF理事、杰出会员, CCF大数据专家委员会委员,主要研究方向为联邦学习、深度学习、云计算、物联网和元宇宙
    肖京(1972- ),男,博士,平安科技(深圳)有限公司首席科学家,深圳市政协委员,中国计算机学会深圳会员活动中心副主席,清华大学、上海交通大学、同济大学、香港中文大学、深圳大学、上海纽约大学客座教授,长期从事人工智能与大数据分析挖掘相关领域研究工作,发表计算机图形学、自动驾驶、3D显示、医疗诊断、联邦学习等领域国际核心论文230余篇,授权专利220余项
  • 基金资助:
    广东省重点领域研发计划“新一代人工智能”重大专项(2021B0101400003)

Survey on federated recommendation systems

Zhitao ZHU1,2, Shijing SI1, Jianzong WANG1, Jing XIAO1   

  1. 1 Ping An Technology (Shenzhen) Co., Ltd., Shenzhen 518063, China
    2 University of Science and Technology of China, Hefei 230026, China
  • Online:2022-07-15 Published:2022-07-01
  • Supported by:
    The Key Research and Development Program of Guangdong Province(2021B0101400003)

摘要:

在联邦学习范式中,原始数据被本地存储在独立的用户客户端中,而脱敏数据被发送到中心服务器中加以聚合,这给众多领域提供了一种新颖的设计思路。考虑到传统推荐系统的研究方向集中于提高推荐效果,在资源节约、跨领域推荐、隐私保护等方面还具有很大改进空间,如何将联邦学习与推荐系统结合以解决这些问题成为当前的一个研究热点。对近年来基于联邦学习的推荐系统进行了全面的总结、比较与分析,首先介绍了推荐系统的传统实现方式及面临的瓶颈;然后引入了联邦学习范式,描述了联邦学习在隐私保护、利用多领域用户数据两方面给推荐系统带来的增益,以及二者结合的技术挑战,进而详细说明了现有的联邦推荐系统部署方式;最后,对联邦推荐系统未来的研究进行了展望与总结。

关键词: 联邦学习, 推荐系统, 隐私保护, 协同过滤, 深度学习

Abstract:

In the federated learning (FL) paradigm, the original data are stored in independent clients while masked data are sent to a central server to be aggregated, which proposes a novel design approach to numerous domains.Given the wide application of recommendation systems (RS) in diverse domains, combining RS with FL techniques has been gaining momentum to reduce the computational cost, do cross-domain recommendation and protect users’ privacy while maintaining recommendations performance as traditional RS.The federated learning-based recommendation systems in recent years were comprehensively summarized.The difference between traditional and federated recommendation systems was analyzed, and the main research direction and progress of federated recommendation systems were demonstrated with comparison and analysis.Firstly, the traditional recommendation systems and their bottleneck were summarized.Then the federated learning paradigm was introduced.Furthermore, the advantages of combining federated learning with recommendation systems were depicted in two aspects: privacy protection and usage of multi-domain user information, along with the technical challenges during the combination.At the same time, the existing deployment of federated recommendation systems was illustrated in detail.Finally, future research on federated recommendation systems was prospected and summarized.

Key words: federated learning, recommendation system, privacy-preserving, collaborative filtering, deep learning

中图分类号: 

No Suggested Reading articles found!