Model training method, device, equipment and storage medium for federated learning

Abstract

The embodiment of the present application discloses a federated learning model training method, device, equipment and storage medium, belonging to the technical field of machine learning. The method includes: generating the i-th scalar operator based on the i-th model parameter of the i-th sub-model and the i-th first-order gradient; sending the i-th fusion operator to the next node device based on the i-th scalar operator; Step gradient scalar, i-th model parameters, and i-th first-order gradient, determine the i-th second-order gradient descent direction of the i-th sub-model; update the i-th sub-model based on the i-th second-order gradient descent direction. In the embodiment of this application, the node devices pass the fusion operator to jointly calculate the second-order gradient descent direction of the sub-models to complete the iterative model training, and can use the second-order gradient descent method to train the machine learning model without relying on third-party nodes. , can avoid the problem of centralized security risks at a single point, enhance the security of federated learning, and facilitate the practical application.

Description

technical field [0001] The embodiments of the present application relate to the technical field of machine learning, and in particular to a federated learning model training method, device, device, and storage medium. Background technique [0002] Federated machine learning is a machine learning framework that can combine data sources from multiple parties to train machine learning models while ensuring that the data does not leave the domain, so that on the basis of meeting privacy protection and data security, multi-party data sources can be used to improve Model performance. [0003] In related technologies, the model training phase of federated learning requires a trusted third party as the central coordinating node to send the initial model to each participant, and collect the models trained by each participant using local data, so as to coordinate the aggregation of the models of all parties , and then send the aggregated model to each participant for iterative traini...

AI Technical Summary

Problems solved by technology

[0004] However, relying on a third party for model training enables the third party to obtain the model parameters of all other participants, but there is still the problem of leaking private data, the security of model training is low, and it is difficult to find a trusted third party. High, making it difficult to apply the solution

Images