Federated learning incentive method, device and equipment based on block chain and storage medium

Abstract

The embodiment of the invention provides a federated learning incentive method, device and equipment based on a block chain and a storage medium, and the method comprises the steps: obtaining the local data and local model of each participant; training local data of all participants based on a federation learning framework to obtain a federation model; respectively calculating the performance index of each local model, and calculating the performance index of the federated model; for each participant, calculating the data contribution degree of the local data of the participant to the federated learning framework according to the performance index of the local model of the participant and the performance index of the federated model; and according to the data contribution degree of each participant, allocating an excitation value to each participant, and storing related data on the block chain. According to the scheme, participants are promoted to more actively provide data for federated learning, optimization of a federated model is facilitated, popularization of a federated learning framework is facilitated, and the method has accuracy and reliability.

Description

Technical field [0001] The present invention relates to the field of artificial intelligence technology, particularly to study the Federal excitation method, apparatus, device, and a storage medium based on the block chain. Background technique [0002] Currently, the federal framework is a distributed learning artificial intelligence model training framework that allows multiple participants to continue learning machine under the premise of legal compliance requirements to protect data privacy, meet to address data silos, silos for different companies data, provides a good value of data mining platform, the federal model can be obtained by federal learning than any single company alone model the effect of their training data better. [0003] However, there is no active participants in the creation of power usage data, is not conducive to optimizing the federal model, is not conducive to the sustainable development of the federal framework for learning, learning is not conducive ...

AI Technical Summary

Problems solved by technology

[0004] The embodiment of the present invention provides a block chain-based federated learning incentive method to solve the technical problems in the prior art that the optimization of the federated model is limited and the promotion of the federated learning framework is limited

Images