Federated learning method and device for vertical data

Abstract

The invention relates to a federated learning method and device for vertical data. The method is applied to a coordination server, and comprises the following steps: receiving Bloom filters of sampledata identifiers of at least two data providers; matching the Bloom filters of the sample data identifiers of the at least two data providers, and obtaining sample alignment information and common data indication information corresponding to the at least two data providers; encrypting the common data indication information by using a public key to obtain encrypted indication information; sending the sample alignment information, the public key, the encryption indication information and the model parameters to a corresponding data provider; receiving encryption model gradient information of atleast two data providers; and updating model parameters corresponding to the at least two data providers according to the encryption model gradient information of the at least two data providers. According to the method, fuzzy matching of a plurality of identifiers can be realized, the applicability of an application scene is improved, and model parameters can be updated under the condition of ensuring data privacy.

Description

technical field [0001] The present disclosure relates to the technical field of big data security processing, in particular to a vertical data federated learning method and device. Background technique [0002] Federated learning is a machine learning framework proposed in recent years, first proposed by Google in 2016. It was born to solve the problem of updating models locally for end users of Android phones. In traditional machine learning, all data is collected on one architecture terminal or distributed across multiple terminals, and then machine learning is performed on the data. The existing federated learning design concept can ensure that the data will not leave the Android mobile phone user's local area, so that machine learning can be performed locally using the mobile phone chip, and finally the updated model parameters will be returned to the server. This makes the application scenarios of federated learning focus more on the relationship between enterprises a...

AI Technical Summary

Problems solved by technology

Moreover, most of the existing federated learning is in the theoretical stage, and the data alignment part is also used as the default condition in practical research applications.

Images