Data processing method and system based on neural network and federated learning

Abstract

The embodiment of the invention provides a data processing method and system based on a neural network and federated learning. The method comprises the following steps: determining an encryption and decryption neural network according to a joint modeling task, wherein the encryption and decryption neural network comprises M encryption neural networks and a decryption neural network; training the encryption and decryption neural network according to randomly generated training data until the encryption and decryption neural network converges; sending the M trained encrypted neural networks to Mclients, so that each client determines encrypted data by taking owned local data and model parameters of the joint modeling task as input data of each encrypted neural network; and receiving encrypted data output by the encrypted neural network of each client, inputting the encrypted data into the decrypted neural network, and determining gradient update data of the model parameters. According to the method provided by the embodiment of the invention, the joint modeling is realized while the data privacy is ensured, the operation is simple and convenient, and the data processing efficiency is improved.

Description

technical field [0001] The invention relates to the technical field of data processing, in particular to a data processing method and system based on neural network and federated learning. Background technique [0002] The data of each data company exists in the form of isolated islands. The most direct way to solve data isolated islands is to integrate data for centralized processing. For example, a typical data processing model is to collect data from one party and then transfer it to another party for processing. Clean and model, and finally sell the model to a third party. In fields such as finance and the Internet, user data often involves user privacy, and centralized processing of data will face the risk of privacy leakage. Federated learning is an effective means to break data islands and protect user data privacy. [0003] Federated learning includes horizontal federated learning, vertical federated learning and migration federated learning. In horizontal federat...

AI Technical Summary

Problems solved by technology

[0005] Embodiments of the present invention provide a data processing method and system based on neural network and federated learning to solve problems in vertical federated learning and migration federated learning

Images