Federal learning method and device based on client classification and information entropy

Abstract

The invention discloses a federal learning method and device based on client classification and information entropy, and relates to the technical field of machine learning, and the method comprises the steps: classifying a client into a first server or a second server based on the bias degree of the client in a non-independent identically distributed data scene; training the client in the corresponding server to obtain a trained client model, determining a local model parameter of the client model, and correspondingly updating a first model parameter of the first server and a second model parameter of the second server based on the local model parameter; and determining that the first server and the second server meet the interaction condition, and updating the central model parameter of the central server based on the weights corresponding to the first model parameter and the second model parameter respectively. According to the invention, the model accuracy of federal learning can be improved, so that the federal learning is suitable for Non-IID scenes with different mixing degrees.

Description

technical field [0001] The invention relates to the technical field of machine learning, in particular to a federated learning method and device based on client classification and information entropy. Background technique [0002] Machine Learning (ML) is a field of computer science that gives computers the ability to learn without being explicitly programmed. A machine learning model can be trained to implement a complex function that generates one or more predicted outputs based on a set of inputs. [0003] Federated Learning is a distributed machine learning framework that can train machine learning models on the premise of ensuring data privacy and security, and can effectively help multiple institutions meet the requirements of user privacy protection, data security and government regulations Next, perform data usage and machine learning modeling. [0004] At present, federated learning mostly focuses on Non-Independent Identically Distribution (Non-IID) scenarios wit...

AI Technical Summary

Problems solved by technology

[0005] In view of this, the embodiment of the present invention provides a federated learning method and device based on client classification and information entropy to solve the problem that the performance of the current federated learning model cannot meet the requirements in the scenario of non-IID data with different mixing degrees The problem

Images