Data security sharing method, storage medium and computing equipment

A technology of data security and local data, applied in patient-specific data, medical data mining, computer-aided medical procedures, etc., can solve problems such as not considering evil, leaking data privacy, not considering collaborative learning, etc.

Pending Publication Date: 2020-06-05
JINAN UNIVERSITY
View PDF7 Cites 7 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

But they do not take into account that the parameters shared in the collaborative learning process may also leak local data privacy, such as Hitaj et al. [Hitaj, B., Ateniese, G., & Perez-Cruz, F. (2017, October). Deep models under the GAN:information leakage from collaborative deep learning.In Proceedings of the2017 ACM SIGSAC Conference on Computer and Communications Security(pp.603-618).ACM.]Proposed use of generative confrontation learning method for data implementation of other participating institutions in the collaborative learning process inference attack
Also, they don't take into account the cases of participating agencies doing evil

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Data security sharing method, storage medium and computing equipment
  • Data security sharing method, storage medium and computing equipment
  • Data security sharing method, storage medium and computing equipment

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0061] This embodiment discloses a data security sharing method, such as figure 1 with figure 2 shown, including the following steps:

[0062] S1. Multiple user sets Ω are determined, and each user set includes multiple participating organizations that communicate with the three servers.

[0063] Identify learning objectives for participating institutions and training conditions for collaborative deep learning training.

[0064] Among them, the participating institutions have local data sets and server environments for local collaborative deep learning training. The local data sets have local data with classification labels. Invalid model parameters can be uploaded during the process. In this embodiment, the participating institutions are medical institutions, the local data is medical image data, and each medical image carries a corresponding disease classification label. The learning objective of participating institutions is to train a model to classify diseases.

[0...

Embodiment 2

[0106] This embodiment discloses a storage medium that stores a program, and when the program is executed by a processor, implements the data security sharing method described in Embodiment 1, specifically as follows:

[0107] S1. Determine multiple user sets Ω, each user set includes multiple participating organizations that communicate with 3 servers;

[0108] Identify learning objectives for participating institutions and training conditions for collaborative deep learning training;

[0109] S2. For each server, send the parameter secret share owned by itself to each participating organization in the user set;

[0110] S3. In each participating organization, reconstruct the received current parameter secret shares and carry out collaborative deep learning training to obtain updated parameters, and then divide the updated parameters into 3 update parameter secret shares and send them to 3 respectively server;

[0111] S4. For each server, after receiving the update paramet...

Embodiment 3

[0116] This embodiment discloses a computing device, including a processor and a memory for storing a program executable by the processor. When the processor executes the program stored in the memory, it implements the data security sharing method described in Embodiment 1, specifically as follows :

[0117] S1. Determine multiple user sets Ω, each user set includes multiple participating organizations that communicate with 3 servers;

[0118] Identify learning objectives for participating institutions and training conditions for collaborative deep learning training;

[0119] S2. For each server, send the parameter secret share owned by itself to each participating organization in the user set;

[0120] S3. In each participating organization, reconstruct the received current parameter secret shares and carry out collaborative deep learning training to obtain updated parameters, and then divide the updated parameters into 3 update parameter secret shares and send them to 3 res...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a data security sharing method, a storage medium and computing equipment. The method comprises the following steps: firstly, determining a plurality of user sets and learning targets and training conditions of participating mechanisms in the user sets; dividing parameters of each server into current parameter secret shares and sending the current parameter secret shares tothe participating mechanisms; reconstructing, by the participating mechanism, the current parameter secret share and performs cooperative deep learning training, dividing the obtained updated parameters into updated parameter secret shares and sending the updated parameter secret shares to the server; verifying, by the server, whether the parameters are legal or not according to the updated parameter secret shares, and if yes, adding the updated parameter secret shares into the parameter secret shares currently owned by the server; and when all participation mechanisms reach the learning target, dividing the latest parameters of each server into latest parameter secret shares, and then sending the latest parameter secret shares to the participating mechanisms, thereby completing data security sharing. According to the invention, the parameter security can be ensured while the privacy is protected, and the data security sharing among different participating mechanisms is realized.

Description

technical field [0001] The invention relates to the technical field of privacy protection data sharing, in particular to a data security sharing method, a storage medium and a computing device. Background technique [0002] With the continuous maturity of artificial intelligence technology, more and more fields have begun to apply artificial intelligence technology. For example, in the 1970s, attempts to use artificial intelligence applications in the medical field began to appear abroad. Our country is also constantly making various attempts in the field of artificial intelligence + medical care. For example, specific applications such as Baidu Medical Brain and Ali Health Medical AI System continue to provide solutions for the development of intelligent medical care. The application scenarios of artificial intelligence in the medical field are very extensive, and the direction of artificial intelligence medical imaging is one of the main applications of artificial intelli...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G16H10/60G16H50/70
CPCG16H10/60G16H50/70
Inventor 翁健董彩芹刘志全刘家男杨雅希成玉丹赵红霞
Owner JINAN UNIVERSITY
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap