联邦学习的公平性研究综述

doi:10.11959/j.issn.2096-0271.2022088

大数据 ›› 2024, Vol. 10 ›› Issue (1): 62-85.doi: 10.11959/j.issn.2096-0271.2022088

• 研究 • 上一篇

联邦学习的公平性研究综述

朱智韬¹^,², 司世景¹, 王健宗¹, 程宁¹, 孔令炜¹, 黄章成¹, 肖京¹

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:2024-01-01 发布日期:2024-01-01
作者简介:朱智韬（1996- ），男，中国科学技术大学硕士生，中国计算机学会会员，现任平安科技（深圳）有限公司算法工程师，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，英国帝国理工学院博士，深圳市海外高层次人才，美国杜克大学人工智能博士后，中国计算机学会会员，现任平安科技（深圳）有限公司资深算法研究员，中国科学技术大学研究生企业导师。至今累计发表机器学习、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。美国佛罗里达大学人工智能博士后，中国计算机学会（CCF）理事、杰出会员，CCF大数据专家委员会委员，曾任美国莱斯大学电子与计算机工程系研究员，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。
程宁（1981- ），男，中国科学院自动化研究所博士，中国科学院计算技术研究所博士后，现任平安科技（深圳）有限公司数据挖掘专家，清华大学、中国科学技术大学研究生企业导师。主持广东省自然科学基金项目一项，至今累计发表语音识别和语音合成领域国际核心论文50余篇。
孔令炜（1995- ），男，平安科技（深圳）有限公司联邦学习团队算法工程师，中国计算机学会会员，主要研究方向为联邦学习系统和安全通信等。
黄章成（1990- ），男，平安科技（深圳）有限公司联邦学习团队资深算法工程师，人工智能专家，中国计算机学会会员，主要研究方向为联邦学习、分布式计算及系统和加密通信等。
肖京（1972- ），男，博士，美国卡耐基梅隆大学博士，国家特聘专家。国家新一代普惠金融人工智能开放创新平台技术负责人、深圳市政协委员、深圳市决策咨询委员会委员，兼中国计算机学会深圳分部副主席、广东省人工智能与机器人学会副理事长、深圳市人工智能行业协会会长、深圳市人工智能学会副理事长，清华大学、上海交通大学、同济大学等客座教授。肖京长期从事人工智能与大数据分析挖掘相关领域研究，先后在爱普生美国研究院及美国微软公司担任高级研发管理职务，现任平安集团首席科学家，技术研究院院长，负责人工智能技术研发及在金融、医疗、智慧城市等领域的应用，带领团队树立了多项传统行业智能化经营的标杆。肖京己发表学术论文249篇，美国授权专利101项，中国发明专利155项，参与及承担国家级项目8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)

A survey on the fairness of federated learning

Zhitao ZHU¹^,², Shijing SI¹, Jianzong WANG¹, Ning CHENG¹, Lingwei KONG¹, Zhangcheng HUANG¹, Jing XIAO¹

¹ Ping An Technology (Shenzhen) Co., Ltd., Shenzhen 518063, China
² University of Science and Technology of China, Hefei 230026, China

Online:2024-01-01 Published:2024-01-01
Supported by:
The Key Research and Development Program of Guangdong Province(2021B0101400003)

摘要/Abstract

摘要：

联邦学习使用来自多个参与者提供的数据协同训练全局模型，近年来在促进企业间数据合作方面发挥着越来越重要的作用。另外，联邦学习训练范式常常面临数据不足的困境，因此为联邦学习参与者提供公平性保证以激励更多参与者贡献他们宝贵的资源是非常重要的。针对联邦学习的公平性问题，首先依据公平目标不同，从模型表现均衡、贡献评估公平、消除群体歧视出发进行了联邦学习公平性的3种分类；然后对现有的公平性促进方法进行了深入介绍与比较，旨在帮助研究者开发新的公平性促进方法；最后通过对联邦学习落地过程中的需求进行剖析，提出了未来联邦学习公平性研究的5个方向。

关键词: 联邦学习, 公平性, 表现均衡, 贡献衡量

Abstract:

Federated learning uses data from multiple participants to collaboratively train global models and has played an increasingly important role in recent years in facilitating inter-firm data collaboration.On the other hand, the federal learning training paradigm often faces the dilemma of insufficient data, so it is important to provide assurance of fairness to motivate more participants to contribute their valuable resources.This paper illustrates the issue of fairness in federated learning.Firstly, three classifications of fairness based on different equity goals, from model performance balance, contribution assessment equity, and elimination of group discrimination are proposed, and then we provide indepth introduction and comparison of existing fairness promotion methods, aiming to help researchers develop new fairness promotion methods.Finally, by dissecting the needs in the process of federal learning implementation, five directions for future federated learning fairness research are proposed.

Key words: federated learning, fairness, balance in performance, measure of contributions

中图分类号:

朱智韬, 司世景, 王健宗, 程宁, 孔令炜, 黄章成, 肖京. 联邦学习的公平性研究综述[J]. 大数据, 2024, 10(1): 62-85.

Zhitao ZHU, Shijing SI, Jianzong WANG, Ning CHENG, Lingwei KONG, Zhangcheng HUANG, Jing XIAO. A survey on the fairness of federated learning[J]. Big Data Research, 2024, 10(1): 62-85.

图/表 8

表1

图1

表2

客户端表现均衡方法的比较"

参考文献	技术方案	公平指标	优势	局限性	帕累托最优
Yang et al.^[9]	组合多臂老虎机	测试准确度	为弱势客户端提供参与机会	仅考虑参与频率
RBCS-F^[10]	上下文组合多臂老虎机	选择分布参数	实现了长期公平约束	牺牲训练效率	√
FMAB^[11]	多臂老虎机	-	考虑客户端抽样过程的不确定性	在探测与通信时产生损耗
FedCS^[8]	估计等待时间指导客户选择	聚合更新客户端数	考虑了设备异质性问题	依赖诚实且有评估能力的客户端
TRA^[12]	容忍丢包损失	测试准确度方差	保证了一定丢包比例下的个性化和公平性	依赖诚实且有评估能力的客户端
PRLC^[14]	本地补偿	拉动操作数	有更好的扩展性	容易遭遇模型发散问题	√
Fed-ZDA^[15]	本地补偿	测试准确度方差	零样本数据增强	耗费更多的算力与时间
AFL^[16]	最小化最大期望损失	验证准确度标准差	防止模型对任何特定客户的过度拟合	仅适用于小规模客户，缺乏灵活性
FedMGDA+^[17]	多目标优化	训练损失平均方差	不牺牲任何参与者性能	难以识别恶意客户端	√
FCFL^[18]	多目标优化	人口均等、机会均等	考虑所有客户端优化目标	只关注客户端级别的群体公平性	√
q-FFL^[19]	放大高损失客户端的损失	测试准确度方差	更均匀的精度分布	需要多次试验确定最佳权重分配参数且对异构数据效果差
α-FedAvg^[20]	提升较低准确度客户端的权重	Jain’s 指数	可通过算法在训练之前确定参数α的取值	无法抵御膨胀损失攻击
DRFL^[21]	由客户端损失动态调整权重	准确度均匀程度	更方便地调整公平参数	没有解决公平参数对不同数据集的适应问题
PG-FFL^[22]	强化学习	基尼系数	自适应学习聚合权重	训练成本较高
TrustFed^[24]	维持信誉分数剔除异常设备	-	引入区块链抵御对抗性攻击	增加传输数据量
FedProx^[26]	允许客户端训练个性化模型	收敛稳定性	有助于减少系统异质性的负面影响	-
Ditto^[78]	多任务学习	测试准确度标准差	插件式设计方便改造现有模型	增加计算量
FedFV^[30]	使用梯度投影缓和梯度冲突	测试准确度方差	避免牺牲部分客户端模型准确性	过时的梯度估计可能导致模型发散	√
Cali3F^[31]	梯度校正技术与更新共享	NDCG标准差	避免本地模型发散，提高推荐性能均匀性	在推荐性能上非最优
FD^[27]	随机丢弃子模型激活节点	-	降低了通信和本地计算成本	需要多次试验选取最佳暂退率	√
AFD^[28]	维护激活分数图生成子模型	-	自动选取激活比例	-	√
FedAMP^[29]	自适应消息传递框架	测试准确度	增量优化	效果依赖于深度神经网络提高了计算量

表2

图2

图3

图4

表3

图5

参考文献 91

[1]	王健宗, 孔令炜, 黄章成 ,等. 联邦学习算法综述[J]. 大数据, 2020,6(6): 64-82.
	WANG J Z , KONG L W , HUANG Z C ,et al. Research review of federated learning algorithms[J]. Big Data Research, 2020,6(6): 64-82.
[2]	ZENG R F , ZENG C , WANG X W ,et al. A comprehensive survey of incentive mechanism for federated learning[J]. ArXiv e-Prints, 2021,arXiv:2106.15406.
[3]	ZHAN Y F , ZHANG J , HONG Z C ,et al. A survey of incentive mechanism design for federated learning[J]. IEEE Transactions on Emerging Topics in Computing, 2022,10(2): 1035-1044.
[4]	SHI Y , YU H , LEUNG C . A survey of fairness-aware federated learning[J]. CoRR, 2021:abs/2111.01872.
[5]	CATON S , HAAS C . Fairness in machine learning:a survey[J]. ACM Computing Surveys, 2023,arXiv:2010.04053.
[6]	MEHRABI N , MORSTATTER F , SAXENA N ,et al. A survey on bias and fairness in machine learning[J]. ACM Computing Surveys, 2022,54(6): 1-35.
[7]	KONG L W , TAO H T , WANG J Z ,et al. Network coding for federated Learning Systems[M]// Neural Information Processing. Cham: Springer, 2020: 546-557.
[8]	NISHIO T , YONETANI R . Client selection for federated learning with heterogeneous resources in mobile edge[C]// Proceedings of the ICC 2019 - 2019 IEEE International Conference on Communications. Piscataway:IEEE Press, 2019: 1-7.
[9]	YANG M , WANG X M , ZHU H B ,et al. Federated learning with class imbalance reduction[C]// Proceedings of the 2021 29th European Signal Processing Conference. Piscataway:IEEE Press, 2021: 2174-2178.
[10]	HUANG T S , LIN W W , WU W T ,et al. An efficiency-boosting client selection scheme for federated learning with fairness guarantee[J]. IEEE Transactions on Parallel and Distributed Systems, 2021,32(7): 1552-1564.
[11]	SHI C , SHEN C . Federated multi-armed bandits[C]// Proceedings of the 35th AAAI Conference on Artificial Intelligence,33rd Conference on Innovative Applications of Artificial Intelligence,The 11th Symposium on Educational Advances in Artificial Intelligence. New York:AAAI Press, 2021: 9603-9611.
[12]	ZHOU P , FANG P , HUI P . Loss tolerant federated learning[J]. CoRR, 2021:abs/2105.03591.
[13]	XIA J C , ZENG G X , ZHANG J X ,et al. Rethinking transport layer design for distributed machine learning[C]// Proceedings of the 3rd Asia-Pacific Workshop on Networking. New York:ACM, 2019: 22-28.
[14]	WANG H Z , QU Z H , GUO S ,et al. Intermittent pulling with local compensation for communication-efficient distributed learning[J]. IEEE Transactions on Emerging Topics in Computing, 2022,10(2): 779-791.
[15]	HAO W T , EL-KHAMY M , LEE J ,et al. Towards fair federated learning with zero-shot data augmentation[C]// Proceedings of the 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops. Piscataway:IEEE Press, 2021: 3305-3314.
[16]	MOHRI M , SIVEK G , SURESH A T . Agnostic federated learning[C]// Proceedings of the 36th International Conference on Machine Learning.[S.l.:s.n.], 2019: 4615-4625.
[17]	HU Z , SHALOUDEGI K , ZHANG G ,et al. FedMGDA+:federated learning meets multi-objective optimization[J]. CoRR, 2020:abs/2006.11489.
[18]	CUI S , PAN W , LIANG J ,et al. Addressing algorithmic disparity and performance inconsistency in federated learning[C]// Proceedings of Advances in Neural Information Processing Systems 34:Annual Conference on Neural Information Processing Systems 2021.[S.l.:s.n.], 2021: 26091-26102.
[19]	LI T , SANJABI M , BEIRAMI A ,et al. Fair resource allocation in federated learning[EB]. arXiv preprint, 2019,arXiv:1905.10497.
[20]	田家会, 吕锡香, 邹仁朋 ,等. 一种联邦学习中的公平资源分配方案[J]. 计算机研究与发展, 2022,59(6): 1240-1254.
	TIAN J H , Lü X X , ZOU R P ,et al. A fair resource allocation scheme in federated learning[J]. Journal of Computer Research and Development, 2022,59(6): 1240-1254.
[21]	ZHAO Z Y , JOSHI G . A dynamic reweighting strategy for fair federated learning[C]// Proceedings of the ICASSP 2022 - 2022 IEEE International Conference on Acoustics,Speech and Signal Processing. Piscataway:IEEE Press, 2022: 8772-8776.
[22]	SUN Y Q , SI S J , WANG J Z ,et al. A fair federated learning framework with reinforcement learning[C]// Proceedings of the 2022 International Joint Conference on Neural Networks. Piscataway:IEEE Press, 2022: 1-8.
[23]	LI T , BEIRAMI A , SANJABI M ,et al. Tilted Empirical Risk Minimization[C]// Proceedings of the 9th International Conference on Learning Representations.[S.l.:s.n.], 2021.
[24]	REHMAN M H U , DIRIR A M , SALAH K ,et al. TrustFed:a framework for fair and trustworthy cross-device federated learning in IIoT[J]. IEEE Transactions on Industrial Informatics, 2021,17(12): 8485-8494.
[25]	陈乃月, 金一, 李浥东 ,等. 基于区块链的公平性联邦学习模型[J]. 计算机工程, 2022,48(6): 33-41.
	CHEN N Y , JIN Y , LI Y D ,et al. Federated learning model with fairness based on blockchain[J]. Computer Engineering, 2022,48(6): 33-41.
[26]	LI T , SAHU A K , ZAHEER M ,et al. Federated optimization in heterogeneous networks[EB]. arXiv preprint, 2018,arXiv:1812.06127.
[27]	CALDAS S , KONE\VCNY J , MCMAHAN H B ,et al. Expanding the reach of federated learning by reducing client resource requirements[J]. ArXiv e-Prints, 2018:arXiv:1812.07210.
[28]	BOUACIDA N , HOU J H , ZANG H ,et al. Adaptive federated dropout:improving communication efficiency and generalization for federated learning[C]// Proceedings of the IEEE INFOCOM 2021 - IEEE Conference on Computer Communications Workshops. Piscataway:IEEE Press, 2021: 1-6.
[29]	HUANG Y , CHU L , ZHOU Z ,et al. Personalized cross-silo federated learning on non-IID data[C]// Proceedings of 35th AAAI Conference on Artificial Intelligence,AAAI 2021,33rd Conference on Innovative Applications of Artificial Intelligence,IAAI 2021,The 11th Symposium on Educational Advances in Artificial Intelligence. New York:AAAI Press, 2021: 7865-7873.
[30]	WANG Z , FAN X L , QI J Z ,et al. Federated learning with fair averaging[C]// Proceedings of the Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence. California:International Joint Conferences on Artificial Intelligence Organization, 2021: 1615-1623.
[31]	ZHU Z T , SI S J , WANG J Z ,et al. Cali3F:calibrated fast fair federated recommendation system[C]// Proceedings of the 2022 International Joint Conference on Neural Networks. Piscataway:IEEE Press, 2022: 1-8.
[32]	杨秀清, 彭长根, 刘海 ,等. 基于数据质量评估的公平联邦学习方案[J]. 计算机与数字工程, 2022,50(6): 1278-1285.
	YANG X Q , PENG C G , LIU H ,et al. Fair federated learning based on data quality evaluation[J]. Computer ＆ Digital Engineering, 2022,50(6): 1278-1285.
[33]	ZHANG J F , LI C , ROBLES-KELLY A , et al ,et al. Hierarchically fair federated learning[EB]. arXiv preprint, 2020,arXiv:2004.10386.
[34]	LYU L J , YU J S , NANDAKUMAR K ,et al. Towards fair and privacy-preserving federated deep models[J]. IEEE Transactions on Parallel and Distributed Systems, 2020,31(11): 2524-2541.
[35]	KANG J W , XIONG Z H , NIYATO D ,et al. Incentive design for efficient federated learning in mobile networks:a contract theory approach[C]// Proceedings of the 2019 IEEE VTS Asia Pacific Wireless Communications Symposium. Piscataway:IEEE Press, 2019: 1-5.
[36]	SARIKAYA Y , ERCETIN O . Motivating workers in federated learning:a stackelberg game perspective[J]. IEEE Networking Letters, 2020,2(1): 23-27.
[37]	THI LE T H , TRAN N H , TUN Y K ,et al. An incentive mechanism for federated learning in wireless cellular networks:an auction approach[J]. IEEE Transactions on Wireless Communications, 2021,20(8): 4874-4887.
[38]	DENG Y H , LYU F , REN J ,et al. FAIR:quality-aware federated learning with precise user incentive and model aggregation[C]// Proceedings of the IEEE INFOCOM 2021 - IEEE Conference on Computer Communications. Piscataway:IEEE Press, 2021: 1-10.
[39]	ZENG R F , ZHANG S X , WANG J Q ,et al. FMore:an incentive scheme of multi-dimensional auction for federated learning in MEC[C]// Proceedings of the 2020 IEEE 40th International Conference on Distributed Computing Systems. Piscataway:IEEE Press, 2020: 278-288.
[40]	CONG M S , YU H , WENG X ,et al. A VCG-based fair incentive mechanism for federated learning[EB]. arXiv preprint, 2020,arXiv:2008.06680.
[41]	WANG G , DANG C X , ZHOU Z Y . Measure contribution of participants in federated learning[C]// Proceedings of the 2019 IEEE International Conference on Big Data. Piscataway:IEEE Press, 2019: 2597-2604.
[42]	NISHIO T , SHINKUMA R , MANDAYAM N B . Estimation of individual device contributions for incentivizing federated learning[C]// Proceedings of the 2020 IEEE Globecom Workshops. Piscataway:IEEE Press, 2020: 1-6.
[43]	SONG T S , TONG Y X , WEI S Y . Profit allocation for federated learning[C]// Proceedings of the 2019 IEEE International Conference on Big Data. Piscataway:IEEE Press, 2019: 2577-2586.
[44]	WEI S Y , TONG Y X , ZHOU Z M ,et al. Efficient and fair data valuation for horizontal federated learning[M]// YANG Q,FAN L,YU H. Federated Learning. Cham: Springer, 2020: 139-152.
[45]	WANG T H , RAUSCH J , ZHANG C ,et al. A principled approach to data valuation for federated learning[M]// YANG Q,FAN L,YU H. Federated Learning. Cham: Springer, 2020: 153-167.
[46]	JIA R , DAO D , WANG B ,et al. Towards efficient data valuation based on the Shapley value[C]// Proceedings of the 22nd International Conference on Artificial Intelligence and Statistics,AISTATS.[S.l.:s.n.], 2019: 1167-1176.
[47]	RADANOVIC G , FALTINGS B , JURCA R . Incentives for effort in crowdsourcing using the peer truth serum[J]. ACM Transactions on Intelligent Systems and Technology, 2016,7(4): 1-28.
[48]	SHYN S K , KIM D , KIM K ,et al. FedCCEA:a practical approach of client contribution evaluation for federated learning[EB]. arXiv preprint, 2021.
[49]	YU H , LIU Z L , LIU Y ,et al. A fairness-aware incentive scheme for federated learning[C]// Proceedings of the Proceedings of the AAAI/ACM Conference on AI,Ethics,and Society. New York:ACM, 2020: 393-399.
[50]	LYU L J , XU X Y , WANG Q ,et al. Collaborative fairness in federated learning[M]// Lecture notes in computer science. Cham: Springer, 2020: 189-204.
[51]	KANG J W , XIONG Z H , NIYATO D ,et al. Reliable federated learning for mobile networks[J]. IEEE Wireless Communications, 2020,27(2): 72-80.
[52]	ZHANG J W , WU Y Z , PAN R . Incentive mechanism for horizontal federated learning based on reputation and reverse auction[C]// Proceedings of the Proceedings of the Web Conference 2021. New York:ACM, 2021: 947-956.
[53]	ZHU Z W , HU X , CAVERLEE J . Fairnessaware tensor-based recommendation[C]// Proceedings of the Proceedings of the 27th ACM International Conference on Information and Knowledge Management. New York:ACM, 2018: 1153-1162.
[54]	DíAZ M , JOHNSON I , LAZAR A ,et al. Addressing age-related bias in sentiment analysis[C]// Proceedings of the Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence. California:International Joint Conferences on Artificial Intelligence Organization, 2019: 6146-6150.
[55]	KLARE B F , BURGE M J , KLONTZ J C ,et al. Face recognition performance:role of demographic information[J]. IEEE Transactions on Information Forensics and Security, 2012,7(6): 1789-1801.
[56]	GU J T , HASSAN H , DEVLIN J ,et al. Universal neural machine translation for extremely low resource languages[C]// Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics:Human Language Technologies,Volume 1 (Long Papers). Stroudsburg:Association for Computational Linguistics, 2018: 344-354.
[57]	KAIROUZ P , MCMAHAN H B , AVENT B ,et al. Advances and open problems in federated learning[J]. Foundations and Trends? in Machine Learning, 2021,14(1/2): 1-210.
[58]	ECKHOUSE L , LUM K , CONTI-COOK C , ,et al. Layers of bias:a unified approach for understanding problems with risk assessment[J]. Criminal Justice and Behavior, 2019,46(2): 185-209.
[59]	RICHARDSON R , SCHULTZ J , CRAWFORD K . Dirty data,bad predictions:how civil rights violations impact police data,predictive policing systems,and justice[R]. New York University Law Review Online, 2019.
[60]	LIAO J C , KAIROUZ P , HUANG C ,et al. Learning generative adversarial representations under fairness and censoring constraints[EB]. arXiv preprint, 2019
[61]	BRISIMI T S , CHEN R D , MELA T ,et al. Federated learning of predictive models from federated Electronic Health Records[J]. International Journal of Medical Informatics, 2018,112: 59-67.
[62]	CHANG K , BALACHANDAR N , LAM C ,et al. Distributed deep learning networks among institutions for medical imaging[J]. Journal of the American Medical Informatics Association, 2018,25(8): 945-954.
[63]	MARTIN A R , KANAI M , KAMATANI Y ,et al. Current clinical use of polygenic scores will risk exacerbating health disparities[J]. bioRxiv, 2019,doi:10.1101/441261.
[64]	BUOLAMWINI J , GEBRU T . Gender Shades:intersectional accuracy disparities in commercial gender classification[C]// Proceedings of Conference on Fairness,Accountability and Transparency.[S.l.:s.n.], 2018: 77-91.
[65]	DU W , XU D P , WU X T ,et al. Fairnessaware agnostic federated learning[M]// DEMENICONI C,DAVIDSON I,eds.Proceedings of the 2021 SIAM international conference on data mining. Philadelphia: Society for Industrial and Applied Mathematics, 2021: 181-189.
[66]	ZENG Y C , CHEN H X , LEE K ,et al. Improving fairness via federated learning[EB]. arXiv preprint, 2021,arXiv:2110.15545.
[67]	RODRíGUEZ-GáLVEZ B , GRANQVIST F , VAN DALEN R ,et al. Enforcing fairness in private federated learning via the modified method of differential multipliers[EB]. arXiv preprint, 2021,arXiv:2109.08604.
[68]	CHU L Y , WANG L J , DONG Y J ,et al. FedFair:training fair models in cross-silo federated learning[EB]. arXiv preprint, 2021,arXiv:2109.05662.
[69]	EZZELDIN Y H , YAN S , HE C Y ,et al. FairFed:enabling group fairness in federated learning[EB]. arXiv preprint, 2021,arXiv:2110.00857.
[70]	CUMMINGS R , GUPTA V , KIMPARA D ,et al. On the compatibility of privacy and fairness[C]// Proceedings of the UMAP’19 Adjunct:Adjunct Publication of the 27th Conference on User Modeling,Adaptation and Personalization. New York:ACM, 2019: 309-315.
[71]	CUMMINGS R , KREHBIEL S , LAI K A ,et al. Differential privacy for growing databases[C]// Proceedings of the Proceedings of the 32nd International Conference on Neural Information Processing Systems. New York:ACM, 2018: 8878-8887.
[72]	JAGIELSKI M , KEARNS M J , MAO J ,et al. Differentially private fair learning[C]// Proceedings of the 36th International Conference on Machine Learning.[S.l.:s.n.], 2019: 3000-3008.
[73]	AVENT B , KOROLOVA A , ZEBER D ,et al. BLENDER:enabling local search with a hybrid differential privacy model[EB]. arXiv preprint, 2017,arXiv:1705.00831.
[74]	WANG J Z , HUANG Z C , KONG L W ,et al. Modeling without sharing privacy:federated neural machine translation[M]// Web Information Systems Engineering -WISE 2021. Cham: Springer, 2021: 216-223.
[75]	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.
[76]	XUE A S , ZHU X H , WANG J Z ,et al. Diversified point cloud classification using personalized federated learning[C]// Proceedings of the 2021 International Joint Conference on Neural Networks. Piscataway:IEEE Press, 2021: 1-8.
[77]	ZENG Q S , DU Y Q , HUANG K B ,et al. Energy-efficient radio resource allocation for federated edge learning[C]// Proceedings of the 2020 IEEE International Conference on Communications Workshops. Piscataway:IEEE Press, 2020: 1-6.
[78]	LI T , HU S Y , BEIRAMI A ,et al. Ditto:fair and robust federated learning through personalization[EB]. arXiv preprint, 2020,arXiv:2012.04221.
[79]	朱智韬, 司世景, 王健宗 ,等. 联邦推荐系统综述[J]. 大数据, 2022,8(4): 105-132.
	ZHU Z T , SI S J , WANG J Z ,et al. Survey on federated recommendation systems[J]. Big Data Research, 2022,8(4): 105-132.
[80]	GOLLAPUDI S , KOLLIAS K , PANIGRAHI D ,et al. Profit sharing and efficiency in utility games[C]// Proceedings of 25th Annual European Symposium on Algorithms.[S.l.:s.n.], 2017: 1-14.
[81]	BOLTON P , DEWATRIPONT M . Contract theory[M]. Cambridge: MIT Press, 2004.
[82]	CORMODE G , JHA S , KULKARNI T ,et al. Privacy at scale:local differential privacy in practice[C]// Proceedings of the Proceedings of the 2018 International Conference on Management of Data. New York:ACM, 2018: 1655-1658.
[83]	ZHAO J , ZHU X H , WANG J Z ,et al. Efficient client contribution evaluation for horizontal federated learning[C]// Proceedings of the ICASSP 2021-2021 IEEE International Conference on Acoustics,Speech and Signal Processing. Piscataway:IEEE Press, 2021: 3060-3064.
[84]	YANG Q , LIU Y , CHEN T J ,et al. Federated machine learning:concept and applications[J]. ACM Transactions on Intelligent Systems and Technology, 2019,10(2): 12.
[85]	DWORK C , HARDT M , PITASSI T ,et al. Fairness through awareness[C]// Proceedings of the Proceedings of the 3rd Innovations in Theoretical Computer Science Conference. New York:ACM, 2012: 214-226.
[86]	ZAFAR M B , VALERA I , GOMEZRODRIGUEZ M ,et al. Fairness constraints:mechanisms for fair classification[C]// Proceedings of the 20th International Conference on Artificial Intelligence and Statistics.[S.l.:s.n.], 2017: 962-970.
[87]	MITCHELL S , POTASH E , BAROCAS S ,et al. Prediction-based decisions and fairness:a catalogue of choices,assumptions,and definitions[EB]. arXiv preprint, 2018,arXiv:1811.07867.
[88]	ROH Y , LEE K , WHANG S E ,et al. FairBatch:batch selection for model fairness[EB]. arXiv preprint, 2020,arXiv:2012.01696.
[89]	ZHOU Z R , CHU L Y , LIU C X ,et al. Towards fair federated learning[C]// Proceedings of the Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery ＆ Data Mining. New York:ACM, 2021: 4100-4101.
[90]	BENGIO Y , LOURADOUR J , COLLOBERT R ,et al. Curriculum learning[C]// Proceedings of the Proceedings of the 26th Annual International Conference on Machine Learning. New York:ACM, 2009: 41-48.
[91]	LIU G Y , MA X Q , YANG Y ,et al. FedEraser:enabling efficient client-level data removal from federated learning models[C]// Proceedings of the 2021 IEEE/ACM 29th International Symposium on Quality of Service. Piscataway:IEEE Press, 2021: 1-10.

公平性分类	分类标准	相关工作
表现均衡公平性	客户端选择	[8-15]
	权重分配	[16-25]
	个性化模型	[26-31]
贡献评估公平性	资源条件	[32-40]
	效用影响	[41-47]
	验证精度	[48-52]
模型公平性	不公平来源	[53-60]
	联邦学习优势	[61-64]
	公平与隐私	[18,65-73]

参考文献	关键技术	衡量标准	分配结果	要求诚实客户端	抗膨胀攻击
[33]	分级别训练	本地数据条件	分级模型	√
[34]	信誉机制	可信度、承诺水平	分级模型	√
[35]	契约理论	本地数据	参与机会	√
[36]	Stacklberg博弈	CPU功率	参与机会	√
[37]	拍卖理论	多种资源	参与机会	√
[38]	拍卖理论	学习质量	参与机会	√
[39]	拍卖理论	相互评估	参与机会	√	√
[40]	VCG机制	计算成本、数据质量	金钱激励	√
[41]	边际损失	特征重要性	-		√
[42,80]	边际损失	边际损失	-		√
[43,45-47]	夏普利值	梯度信息	-		√
[48]	采样测试方法	采样数据大小	-	√	√
[49]	动态收益共享	期望损失、等待时间	金钱激励	√
[50]	信誉机制	验证数据集上准确度	-		√
[51]	信誉机制	本地计算时间和数据集大小	-		√
[52]	信誉机制	验证数据集上准确度	-		√

联邦学习的公平性研究综述

A survey on the fairness of federated learning

在线阅读

PDF下载

可视化

摘要/Abstract

引用本文

使用本文

图/表 8

参考文献 91

相关文章 8

Metrics

推荐阅读 0

[1]	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.
[2]	张传尧, 司世景, 王健宗, 肖京. 联邦元学习综述[J]. 大数据, 2023, 9(2): 122-146.
[3]	张燕, 杨一帆, 伊人, 罗圣美, 唐剑飞, 夏正勋. 隐私计算场景下数据质量治理探索与实践[J]. 大数据, 2022, 8(5): 55-73.
[4]	尹虹舒, 周旭华, 周文君. 纵向联邦线性模型在线推理过程中成员推断攻击的隐私保护研究[J]. 大数据, 2022, 8(5): 45-54.
[5]	吴建汉, 司世景, 王健宗, 肖京. 联邦学习攻击与防御综述[J]. 大数据, 2022, 8(5): 12-32.
[6]	朱智韬, 司世景, 王健宗, 肖京. 联邦推荐系统综述[J]. 大数据, 2022, 8(4): 105-132.
[7]	王健宗, 孔令炜, 黄章成, 陈霖捷, 刘懿, 卢春曦, 肖京. 联邦学习隐私保护研究进展[J]. 大数据, 2021, 7(3): 130-149.
[8]	王健宗, 孔令炜, 黄章成, 陈霖捷, 刘懿, 何安珣, 肖京. 联邦学习算法综述[J]. 大数据, 2020, 6(6): 64-82.