A model training method and a device based on shared data

Abstract

A model training method and apparatus based on shared data are disclosed. On the one hand, joint training can be carried out according to the data provided by multiple data providers, so as to obtaina more accurate and comprehensive data model; On the other hand, the operations (such as data decryption operation, local model parameter update operation, etc.) involved in the privacy data during model training are encapsulated in a trusted execution environment of the data provider. Data plaintext cannot be acquired outside the trusted execution environment.

Description

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine

Login to View More

AI Technical Summary

Benefits of technology

This technology allows users to train their collaborative team with better accuracy through various types of interactions between different parties involved during collaboration sessions or when they want access protected resources like databases. By embedding sensitive data into an operating system's secure shelling layer called Trusted Execution Environment (TEE), these technologies help protect both inside and external sources from being compromised.

Problems solved by technology

The technical issue addressed in this patented text relates to improving the efficiency or safety of accessing shared datasets while also maintaining user privacy rights during collaborative work between different parties involved.

Images