Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

Neural network model encryption protection system and method involving domain transformation data encryption

A neural network model and data encryption technology, applied in the field of artificial neural network protection mechanism

Active Publication Date: 2018-11-13
CHENGDU PANOAI INTELLIGENT TECH CO LTD
View PDF4 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] Aiming at the above-mentioned defects of the existing artificial neural network protection mechanism, the present invention provides a neural network model encryption protection system involving domain transformation data encryption

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
  • Neural network model encryption protection system and method involving domain transformation data encryption
  • Neural network model encryption protection system and method involving domain transformation data encryption
  • Neural network model encryption protection system and method involving domain transformation data encryption

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0069] For a system structure diagram of a neural network model encryption protection system 100 involving domain transformation data encryption, please refer to the appendix figure 1 For the encryption principle of the encryption module 2 of the encryption protection system 100, please refer to the appended figure 2 , the encryption protection system 100 includes a data input module 1 , an encryption module 2 , an encrypted data input module 3 , an artificial neural network model module 4 and a data output module 5 . The data input module 1 is connected to the signal of the encryption module 2, the signal of the encryption module 2 is connected to the signal of the encrypted data input module 3, the signal of the encrypted data input module 3 is connected to the signal of the artificial neural network model module 4, and the signal of the artificial neural network model module 4 is connected to the signal of the data output module 5 connected.

[0070] Specifically, the dat...

Embodiment 2

[0088] For the method flow chart of a neural network model encryption protection method involving domain transformation data encryption, please refer to the appendix of the specification. Figure 4 , the method includes the following steps:

[0089] Step S1. Provide original input data As;

[0090] Step S2. For the original input data A s Perform encryption processing to generate encrypted input data A s ’ ;

[0091] Step S3. Receive encrypted input data A s ’ , and input it into the artificial neural network model;

[0092] Step S4. The artificial neural network model encrypts the input data A s ’ Carry out calculations and obtain calculation results;

[0093] Step S5. Output the calculation result.

[0094] Further, the step S2 includes the following steps:

[0095] Step S21. The original input data A s Convert to a two-dimensional structure, and the original input data A after structure conversion s Carry out domain transformation, transform the original input ...

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 belongs to the field of artificial neural network protection mechanisms, and relates to a neural network model encryption protection system and method involving domain transformation data encryption. The system comprises an encryption module; a frequency domain transformation module in the encryption module is used for transforming original input data into a two-dimensional structure, conducting domain transformation on the original input data obtained after structure transformation and transforming the original input data in the space domain into original frequency domain data in the frequency domain; a frequency domain data processing module in the encryption module is used for processing the original frequency domain data in a specific processing mode to generate encryptedfrequency domain data; and a frequency domain inverse transformation module in the encryption module is used for conducting inverse domain transformation corresponding to above-mentioned domain transformation on the encrypted frequency domain data and transforming the encrypted frequency domain data in the frequency domain into the encrypted input data in the space domain. According to the methodof achieving input data encryption through domain transformation, the concealment and the anti-attacking performance of password information are improved, and the security of a trained artificial neural network model can be improved.

Description

technical field [0001] The invention belongs to the field of artificial neural network protection mechanism, and in particular relates to a neural network model encryption protection system and method related to domain transformation data encryption. Background technique [0002] Deep learning is the main technical solution for current artificial intelligence applications. The artificial neural network model trained with deep learning technology is the result of the labor of the original developers. However, in the process of publishing and applying the artificial neural network model, its network structure and node weights will be completely exposed to the outside world. After the artificial neural network model is released and / or applied by a third party, it is easy to be copied, re-developed or modified, resulting in damage to the rights and interests of the original developer. The protection schemes applicable to artificial neural network models in the prior art mainly...

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
IPC IPC(8): G06F21/60G06N3/02
CPCG06F21/602G06N3/02
Inventor 尹愚
Owner CHENGDU PANOAI INTELLIGENT TECH CO LTD
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
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