Self-adaptive asynchronous federated learning method with local privacy protection

Abstract

The invention discloses a self-adaptive asynchronous federal learning method with local privacy protection, which comprises the following steps that: a central server initializes a global model and broadcasts global model parameters, a gradient cutting standard, a noise mechanism and a noise variance to all participating users; each user firstly uses samples extracted from local data to train a global model and cuts and disturbs the gradients one by one, then the disturbed gradients are sent to a central server, the central server selects the first K disturbed gradients from a buffer queue to perform average aggregation, the averaged gradient is substituted into a stochastic gradient descent formula to update global model parameters, and a gradient cutting standard, a noise variance and a learning rate are adaptively adjusted according to the number of iterations in a preset stage; and then the central server broadcasts the updated global model parameters, the gradient cutting standard and the noise variance to K users participating in updating in the last round, and the local users and the central server repeat the operation until the number of global iterations reaches a given standard.

Description

technical field [0001] The invention belongs to the field of safe federated learning, and in particular relates to an adaptive asynchronous federated learning method with local privacy protection. Background technique [0002] Today's society has entered the era of big data. In-depth analysis and mining of data through artificial intelligence, machine learning, and big data technology can maximize the release of the value of data, thereby promoting the rapid development of social economy. However, since data is naturally distributed or stored in different user devices (including personal devices or enterprise devices) and these data are highly sensitive, direct aggregation, analysis, and mining of user data will result in leakage of user privacy information. Federated learning ensures that user data is kept locally, and the training of the global model is completed only through multiple rounds of intermediate information exchange and aggregation between the user and the cent...

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to overcome the shortcomings of the above existing methods and provide an adaptive asynchronous federated learning method with local privacy protection, which solves the security of user privacy information and the utility of the final model in asynchronous federated learning

Images