Encipherment protection system and method for artificial neural network model
An artificial neural network and protection system technology, applied in biological neural network models, neural learning methods, digital data protection, etc.
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0050] A kind of encryption protection system for artificial neural network model comprises data input module 1, encryption module 2, encrypted data input module 3, artificial neural network model module 4 and data output module 5, and described data input module 1 and encryption module 2 signal The encryption module 2 is connected to the encrypted data input module 3 by signal, the encrypted data input module 3 is connected to the artificial neural network model module 4 by signal, and the artificial neural network model module 4 is connected to the data output module 5 by signal.
[0051] The data input module 1 is used to provide the original input data D to the encryption module 2 .
[0052] The encryption module 2 is used to encrypt the original input data D provided by the data input module 1, and output the encrypted input data D ， Described encryption module 2 is by one group of password group S that has N digits as control amount, all original input data D that data i...
Embodiment 2
[0061] A kind of encryption protection method for artificial neural network model, it is characterized in that: comprise the following steps:
[0062] A. Provide raw input data D;
[0063] S. Encrypt the original input data D to generate encrypted input data D ， ;
[0064] B. will encrypt the input data D. ， Input to the artificial neural network model, and the artificial neural network model encrypts the input data D ， perform calculations and obtain results;
[0065] C. Output the calculated results.
[0066] Described step S specifically comprises:
[0067] S1. Provide a group of cipher group S with N digits as the control amount, the cipher group S specifically defines the encryption process for the original input data D; wherein, the length of the cipher group S is determined by the encryption method and encryption requirements in the encryption process Defined; each of the N digits of the password group S is selected from one of the Arabic numerals 0-9;
[0068] S...
Embodiment 3
[0079] For a system structure diagram of an encryption protection system 100 for an artificial neural network model, please refer to the appendix figure 1 , the system 100 includes an encryption module 2, the encryption module 2 is set before the artificial neural network model module 4, the original input data D provided by the data input module 1 first enters the encryption module 2 for encryption processing, and generates encrypted input data D ， , and output to the encrypted data input module 3, the artificial neural network model module 4 is based on the encrypted input data D provided by the encrypted data input module 3 ， Perform calculations. The encryption module 2 is controlled by a group of cipher groups S with N digits, and adjusts the original input data D in a unified manner, wherein the length of the cipher group S is determined by the encryption method and the encryption method in the encryption process implemented by the encryption module 2 Restricted by the ...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 
PatSnap Eureka turns technology decisions into work you can execute. Powered by our Innovation Knowledge Graph, it runs expert workflows across engineering, life sciences, materials and intellectual property. Get your review-ready output in minutes.
