Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Method, device and storage medium for encrypting neural network model

A neural network model and neural network technology, applied in the field of neural network model encryption, can solve problems such as easy leakage of model information, achieve the effect of meeting offline deployment requirements, improving transmission efficiency, and improving efficiency

Active Publication Date: 2022-03-11
BEIJING XIAOMI MOBILE SOFTWARE CO LTD
View PDF10 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The application of the neural network model often involves security issues. When the neural network model is deployed on the terminal device side, the model information is easy to leak

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
  • Method, device and storage medium for encrypting neural network model
  • Method, device and storage medium for encrypting neural network model
  • Method, device and storage medium for encrypting neural network model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0048] There are generally two ways to deploy the neural network model: one is to deploy on the server side and provide online services through the network; the other is to deploy on the terminal device side, including smart terminals such as mobile phones, and use the accelerator on the terminal device side to offline Execute the model. The model is deployed on the terminal device side, and all the information of the model is stored in the model file, and the model information can be obtained directly through the model file, which has low security.

[0049] The method for encrypting the neural network model may at least include a symmetric encryption method and an asymmetric encryption method. There are two types of symmetric encryption methods: stream cipher encryption and block cipher encryption. Stream cipher encryption refers to encrypting one letter or number at a time; block cipher encryption refers to encrypting a data block including multiple bits into a unit, and it...

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 present disclosure relates to a method, device, and storage medium for encrypting a neural network model or obtaining output information of a neural network model. The method includes: encrypting at least a part of the model information in the original model file describing the target neural network model , to obtain an encrypted model file; according to the encrypted model file, generate a model program code describing the target neural network model; based on the request of the terminal device, send an application installation package containing the model program code to the terminal device . Through the technical solution of the present disclosure, after encrypting at least part of the model information in the neural network model, the neural network model is converted into program code, thereby increasing the difficulty of cracking, improving the security performance of the neural network model, and sending the application to the terminal device There is no need to transfer the original model file when the program installs the package, thus improving the transfer efficiency.

Description

technical field [0001] The present disclosure relates to information processing technology, and in particular to a method and device for encrypting a neural network model, and a storage medium. Background technique [0002] The neural network model is a mathematical model that simulates the actual human neural network, and has a wide range of applications in many fields such as system identification, pattern recognition, and artificial intelligence. With the continuous maturity of neural network technology, neural network models have been applied in many terminal equipment application products. The application of the neural network model often involves security issues. When the neural network model is deployed on the terminal device side, the model information is easy to leak. Therefore, finding an encryption method suitable for the neural network model is one of the important topics in the application of the neural network model. Contents of the invention [0003] The d...

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 Patents(China)
IPC IPC(8): G06F21/60G06F21/62
CPCG06F21/602G06F21/6209G06N3/02G06F21/60H04L63/0428G06F8/41G06F8/61G06F8/76G06N3/04H04L9/0643
Inventor 刘琦叶剑武何亮亮
Owner BEIJING XIAOMI MOBILE SOFTWARE CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com