摘要
由于金融信息的隐私性、保密性、安全性等特点,金融产业存在数据信息难以共享、对用户数据信息保护不足等问题,大数据及机器学习在金融领域的实际应用受到一定限制。联邦学习作为一种保护数据安全的机器学习方法,无需各参与方共享数据即可实现联合建模,为大数据及机器学习在金融领域的应用提供了可以实现数据安全要求的潜在方案。文本对联邦学习的概念、主要应用领域、其在金融领域的应用愿景及可能存在的问题等进行介绍和分析,以期为联邦学习在未来金融产业的应用提供一定帮助或借鉴。
Due to the confidentiality of finance-related information,sharing of financial data is highly restricted in order to protect user’s privacy.Therefore,the application of Big Data and machine learning in finance is limited.As a decentralized machine learning method,federated learning does not require data collection,thus providing a data-secured potential solution for applications of machine learning in financial industry.This paper introduces federated learning and its current main fields of application,including its applications in finance,to provide basis for further studies.
作者
孙昊颖
SUN Hao-ying(College of Economy and Trade,Zhongkai University of Agriculture and Engineering,Guangzhou Guangdong 510225)
出处
《新疆师范大学学报（哲学社会科学版）》
CSSCI
北大核心
2022年第3期140-148,共9页
Journal of Xinjiang Normal University(Philosophy and Social Sciences)
关键词
联邦学习
金融
反洗钱
保险定价
普惠金融
Federated Learning
Finance
Anti-money Laundering
Insurance Pricing
Inclusive Finance
