A Federated Learning Method Based on Trusted Execution Environment

Abstract

The invention belongs to the field of data security, and is a federated learning method based on a trusted execution environment. The data and the initialized model parameters returned by the cloud are loaded into the safe area, the gradient of the model parameters after training is obtained, and a digital signature is generated. The local user identity authentication is uploaded to the cloud; the cloud verifies the local user identity authentication, and after successful verification, the uploaded model parameter gradient and model integration algorithm are obtained, placed in the cloud security area, the model is integrated, and the model parameter gradient is updated. The invention utilizes the trusted execution environment to generate the security zone, and the user cannot bypass the training process and directly give the training result, thereby realizing the training integrity and user privacy protection.

Description

technical field [0001] The invention belongs to the field of data security, in particular to a federated learning method based on a trusted execution environment. Background technique [0002] Machine learning based on private data has achieved good results in practical applications. Many companies, such as Google, Facebook, Apple, etc., collect massive training data from users and apply powerful GPU computing power to deploy deep learning algorithms. In order to get a deeper model, many companies are willing to collect complementary data and collaborative training. However, this method of training machine learning models by directly assembling multi-user raw data sets hides huge challenges: privacy data security, data poisoning, data volume misreporting and other issues. [0003] The invention patent with the publication number CN108717514A published on October 30, 2018 discloses a data privacy protection method and system in machine learning, which first selects the encry...

AI Technical Summary

Problems solved by technology

The Trusted Execution Environment executes the machine learning code using at least one data oblivious program, wherein the data oblivious program is any of memory access patterns, disk access patterns, and network access The pattern makes it impossible to predict the process of classified data from the pattern

Similarly, its system protects private data very well, but it also does not solve the problem of the authenticity of training results and the amount of false data

Images