Federation modeling method and device based on backward method and readable storage medium

Abstract

The invention discloses a federation modeling method and device based on a backward method and a readable storage medium. The federation modeling method based on the backward method comprises the steps of conducting federation modeling, obtaining complete model to be trained, receiving a tagging party characteristic data set fed back by third equipment coordinated and associated with the first equipment; determining each to-be-rejected label feature corresponding to the label party feature data set, based on the features of each to-be-rejected label, rejecting the to-be-rejected labels; performing longitudinal federation modeling through second equipment federated with the first equipment, and executing a preset cyclic training process on each to-be-trained complete model to select each cyclic target feature meeting a preset contribution degree standard from each to-be-eliminated label feature, and eliminating each cyclic target feature in the to-be-trained complete model to obtain a target modeling model. The technical problems that longitudinal federation modeling is low in efficiency and poor in effect are solved.

Description

technical field [0001] The present application relates to the technical field of machine learning of financial technology (Fintech), and in particular to a federated modeling method, device and readable storage medium based on the backward method. Background technique [0002] With the continuous development of financial technology, especially Internet technology and finance, more and more technologies (such as distributed, blockchain, artificial intelligence, etc.) Requirements, such as the distribution of corresponding to-do items in the financial industry also have higher requirements. [0003] With the continuous development of Internet technology and finance, the application of federated learning is becoming more and more extensive. In the vertical federated learning scenario, both the label provider and the feature provider participating in the federated learning will provide training features, and based on these training features through the Federated learning is use...

AI Technical Summary

Problems solved by technology

[0004] The main purpose of this application is to provide a federated modeling method, device and readable storage medium based on the backward method, aiming to solve the technical problems of low efficiency and poor effect of vertical federated modeling in the prior art

Images