基于联邦学习的地方征信平台建设路径探讨  被引量:7

Discussion on the Path for Construction of Federated Learning-based Local Credit Information Platform

在线阅读下载全文

作  者:杨阳[1] Yang Yang(Maanshan Central Sub-branch of the People’s Bank of China,Maanshan 243000,Anhui,China)

机构地区:[1]中国人民银行马鞍山市中心支行,安徽马鞍山243000

出  处:《征信》2022年第9期49-54,共6页Credit Reference

摘  要:地方征信平台建设是解决信息不对称问题,推进中小企业等主体将“信用”向“信贷”转换,促进普惠金融发展的重要抓手。近年来,地方政府、人民银行对地方征信平台建设进行积极探索和有益尝试并取得积极成效,但仍面临信息共享和应用不足、信息安全保障难等问题,建设成效有待提升。从地方征信平台建设工作实践出发,探讨利用联邦学习技术对信息“分别持有,联合使用”的优势,突破地方征信平台建设面临的瓶颈,提出由地方政府统筹推进,依托联邦学习技术框架建设地方征信平台,发挥平台信用赋能作用,拓展数据信息多元化应用的路径和相关建议。The construction of local credit information platform is an important starting point to solve the problem of information asymmetry, promote the transformation of “credit” to “credit loan” by small and medium-sized enterprises and other entities, and promote the development of inclusive finance. In recent years,local governments and the People’s Bank of China have made positive explorations and beneficial attempts on the construction of local credit information platform, and achieved positive results. However, they still face such problems as insufficient information sharing and application, and difficulties in information security guarantee, and the construction effects necessary to be improved. From local credit information platform construction practice, this paper discusses the use of federated learning technology to“hold information separately and use it jointly”to break through the bottlenecks faced by the construction of local credit information platform, and proposes a path and relevant suggestions for the overall promotion of local governments, to rely on the framework of federated learning technology to build a local credit information platform, to play the role of platform credit empowerment, and to expand the diversified application of data and information..

关 键 词:地方征信平台 联邦学习 信息安全 普惠金融 

分 类 号:F832.4[经济管理—金融学]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象