A blockchain-based federated learning system and method

Abstract

The present invention provides a blockchain-based federated learning system and method. The system includes: a model training module for updating machine learning models in the federated learning process and aggregation of change values ​​of the machine learning models; The contract module is used to provide decentralized control functions and key management functions in the process of the federated learning; the storage module based on the IPFS protocol is used to provide decentralized Information storage mechanism: The model training module, the smart contract module based on blockchain technology, and the storage module based on the IPFS protocol are simultaneously run on each node participating in federated learning. To achieve complete decentralization of the entire system, the failure and exit of any node will not affect other nodes to continue federated learning, which is more robust.

Description

A blockchain-based federated learning system and method technical field The present invention relates to the technical field of federated learning, in particular to a blockchain-based federated learning system and method. Law. Background technique [0002] Federated learning enables participating institutions to collaborate on training machines without directly exchanging raw data. machine learning model. For those companies or institutions with insufficient data, it can allow them to unite and obtain better The model does not expose the original data at the same time, and achieves mutual benefit and win-win. In the existing engineering technology, the collaborative training of various institutions depends on the Rely on centralized third-party collaborative nodes for control, aggregation and key management. The existing centralized methods have the following shortcomings point: [0003] 1) The cooperative node will continuously obtain the information uploaded by ...

AI Technical Summary

Problems solved by technology

[0004] 2) When the cooperative node fails, the entire system will crash and cannot continue to run

Due to the single point of failure of the centralized collaboration node, the federated learning of each institution will be forcibly terminated and the collaborative training cannot continue

[0005] The federated learning method in the prior art has the problems of privacy risk and single point of failure

Images