摘要
通过对各种文献的查询和研讨,分析了联邦学习技术在金融领域的应用现状及其所面临的挑战。联邦学习技术基于分布式学习理念,已被应用于反欺诈、风险管理、股票推荐等金融领域,并取得了一定成效。然而,由于数据异构性、隐私保护、模型融合等问题,联邦学习在金融领域仍然面临诸多挑战。未来的研究方向包括改进模型融合算法、提升安全性与隐私保护技术等。
This paper analyzes the current status and challenges of federated learning technology in the financial domain through literature review and case analysis.Based on the concept of distributed learning,federated learning techniques have been applied to financial domains such as anti-fraud,risk management,stock recommendation,etc.,and have achieved certain results.However,federated learning still faces many challenges in the financial domain due to data heterogeneity,privacy protection,model fusion,and other issues.Future research directions include improving model fusion algorithms,enhancing security and privacy protection techniques,and so on.
作者
聂璇
王殊
刘渊
NIE Xuan;WANG Shu;LIU Yuan(Bank of Sanxiang,Changsha 410017,China)
出处
《电脑与信息技术》
2024年第3期45-50,85,共7页
Computer and Information Technology
关键词
金融
联邦学习
隐私保护
finance
federated learning
privacy preserving