Data security access method based on block chain in federated learning environment

Abstract

The invention provides a data security access method based on a block chain in a federated learning environment, and mainly solves the problem that a user does not have a complete control right to owndata in existing federated learning. The method comprises the implementation steps of 1) performing initialization operation on a blockchain, a smart contract and a KGA; 2) the user performs local training on the data, stores the trained data by using IPFS, encrypts the data through CP-ABE, and uploads the encrypted data to a deployed blockchain; 3) the server interacts with the KGA to obtain theuser uploaded data of the required version for centralized training, and updates the final data after training to the blockchain; and 4) the user acquires final data uploaded by the server and updates the local training model according to the data information.The learning process of the method is open and transparent, the control right of the user on the own data is ensured, data backtracking andmodel rollback can be conveniently performed, and version management is facilitated.

Description

technical field [0001] The invention belongs to the field of communication technology, relates to information security and machine learning technology, and further relates to a block chain-based data security access method in a federated learning environment, which can perform data access control in various scenarios where federated learning is used. Background technique [0002] With the advent of the era of big data, the amount of users' personal data has grown exponentially, resulting in a gradual increase in the cost of local machine learning data storage and computing. When the local computing power is not enough to support the user's business requirements, federated learning occurs. Federated learning is a collaborative artificial intelligence technology proposed by Google. Its initial purpose was to solve the problem of mobile phone users updating models locally. The main idea is to use a distributed architecture to update local datasets and models as a sub-update to b...

AI Technical Summary

Problems solved by technology

As a decentralized, non-tamperable, open and transparent distributed platform, the blockchain has been widely concerned, especially the Turing-complete smart contract based on the Ethereum structure, which makes up for the inability of early platforms such as Bitcoin to implement complex functions. Insufficient functions, but the storage limitation of the blockchain makes it problematic to support large-scale data

[0004] In the application number CN201910774155.3 titled "Federated Learning-Based Disease Data Processing Method, Device, Equipment and Medium" and the application number CN201910872625.X titled "A Method and System for Enhancing the Privacy of Federated Learning Training Data" In the two patent application documents, the application of federated learning and the security of interactive data were discussed, but the access control to data in a multi-party environment was not mentioned.

Images