Edge computing privacy protection system and method based on joint learning

Abstract

The invention discloses an edge computing privacy protection system and method based on joint learning. The system comprises a client and a server, the client is used for local training, adding disturbance to updated parameters, encrypting the updated parameters and sending the encrypted parameters to the server, and the server receives encrypted data sent by the client, decrypts the encrypted data and updates local parameters to update a deep learning model; the protection method comprises the following steps: step 1, adding disturbance to parameters at a client; step 2, encrypting the data at a client; and step 3, decrypting the data at the server. According to the method, each participant can safely submit data on the premise of not needing any trusted aggregator; noise disturbance is added to local updating in a distributed mode, updating of the disturbance is encrypted through a Paillier homomorphic cryptosystem, safety and performance analysis shows that the PPFL protocol can guarantee privacy of client data and learning accuracy at the same time, and the conflict problem of privacy protection and learning accuracy is solved.

Description

technical field [0001] The present invention relates to the technical fields of distributed computing, differential privacy, data encryption and deep learning, in particular to an edge computing privacy protection system based on device joint learning, and also relates to an edge computing privacy protection method based on device joint learning. Background technique [0002] With the rapid development of Internet of Things (IoT) and mobile network applications, deep learning technology has achieved great success in many emerging applications such as computer vision and speech recognition. Machine learning, especially deep learning, is envisioned as an indispensable service that is generally available. However, this intelligent computing scheme requires users to outsource their sensitive data to unknown third parties for machine learning services, which puts high demands on privacy protection in this case without compromising service quality. As a collaborative form of deep...

AI Technical Summary

Problems solved by technology

However, this operation will lead to severe cumulative noise and greatly damage the accuracy of the global model

In addition, Bonawitz et al. proposed a high-dimensional data aggregation protocol using a secret sharing method, however, generating perturbations for each user's data in this aggregation protocol requires high communication costs and is also vulnerable to malicious servers

Images