by Hui Zhang
JD iCity, an intelligent brand of JD Digits, as one of the earliest Chinese enterprises to study and productization of federal learning(FL), had participated setting standard about the framework and application of federated machine learning. Titled IEEE P3652.1, the publication was recently approved by the Institute of Electrical and Electronics Engineers (IEEE) for release at the end of this year.
Federated learning is a type of machine learning that utilizes lower latency, less power consumption and higher privacy, by eliminating the need to store training data in the cloud. The IEEE P3652.1 guide will be the first in the world dedicated to industry standard-setting for the technology.
At present, FL has been widely implemented in the areas of finance, healthcare and smart cities. The introduction of the first international standard will enable more enterprises and institutions that want to apply and are applying FL to further expand cooperation and build a more complete and powerful federal ecosystem together.
As one of the contributors to the IEEE P3652.1 guide, JD iCity’s self-developed federal digital gateway product is committed to providing solutions for both government and enterprise consumers to solve the problem of data sharing difficulty.
As a leader in smart city construction, JD iCity’s federal digital gateway product has been applied in various areas including credit, smart address selection, precise marketing, and municipal governance modernization. The product has been deployed in over nine projects involving government entities and enterprises.
In the scenario of intelligent site selection, JD iCity analyzed data from multiple sides to not only help enterprises with site selection, but also to provide monitoring of people flow to assist in store operations, and to attract consumers to the stores, thus providing comprehensive support.
With regards to credit, the Federal Digital Gateway product uses multi-party data to build a credit scorecard model, which can be used to quantitatively analyze the credit information of enterprises or individuals, thus reducing the risk of credit application.