Decentralized federated learning method, device and system

Abstract

The invention discloses a decentralized federated learning method, device and system, and belongs to the field of federated learning, and the method comprises the steps: building a global communication network among a plurality of clients participating in federated learning, and enabling a communication path to exist between any two clients; each client respectively receiving the model parameters of other clients directly communicating with the client at the previous moment, respectively calculating the sum of the products of the obtained model parameters at the previous moment and the corresponding weight coefficients, and calculating the first product of the loss function gradient of the local model at the previous moment and a preset adaptive learning rate; updating a model parameter at the current moment to a difference value between the model parameter and the first product; and repeatedly executing the iteration updating operation until the loss function of the local model of each client is not higher than the corresponding threshold value, or until the number of times of repeated execution reaches the maximum number of iterations. While privacy and data security of each client are protected, each local model is globally trained.

Description

technical field [0001] The invention belongs to the field of federated learning, and more specifically relates to a decentralized federated learning method, device and system. Background technique [0002] Federated learning aims to establish a federated learning model based on distributed data sets to deal with the problem of data islands. With the application of artificial intelligence in various industries, people's attention to privacy and data security continues to increase. How to solve the problems of data fragmentation and data isolation under stricter and new privacy protection regulations is the primary challenge in current artificial intelligence research and practice. People's loss of control over data and opacity about the distribution of benefits has exacerbated the severity of so-called data fragmentation and island distribution. In order to ensure user privacy and data security, the process of exchanging model information between clients will be carefully d...

AI Technical Summary

Problems solved by technology

[0004] The existing decentralized federated learning method lacks unified analysis in three aspects: non-independent and identical distribution of data between clients, global shared model training, and mitigation of communication pressure, and does not take into account the attacks between users Behavior, this mutual distrust will lead to attackers attacking other people's devices through the network, which will eventually lead to data leakage of each participant

Images