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

Intellectual property protection method and system for deep neural network and terminal

A deep neural network and neural network technology, applied in neural learning methods, biological neural network models, neural architectures, etc., can solve problems such as poor timeliness, high cost, inability to prevent network copying and illegal use, etc.

Active Publication Date: 2019-06-21
FANDWORK TECH NANJING CO LTD
View PDF6 Cites 9 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This watermark network can prove its ownership, but cannot prevent the copying and illegal use of the network itself
The rights protection of the watermarking network must go through legal channels through the production of evidence, which is time-sensitive and expensive

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
  • Intellectual property protection method and system for deep neural network and terminal
  • Intellectual property protection method and system for deep neural network and terminal
  • Intellectual property protection method and system for deep neural network and terminal

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0082] refer to figure 1 , shows a flow chart of steps of a method for protecting intellectual property rights of an arbitrary deep neural network by using a self-locking mechanism according to an embodiment of the present invention.

[0083] In this embodiment, the method for protecting the intellectual property rights of any deep neural network by using the self-locking mechanism includes the following steps.

[0084] Step S100, according to the target network structure information to be locked and self-locking performance requirements, prepare and obtain a specific digital key that meets the requirements.

[0085] Step S102, according to the specific digital key, self-locking performance requirements, training data samples, and basic functions of the target network, prepare and obtain a deep neural network with self-locking function that meets the basic functional requirements.

[0086] Step S104, according to the provided correct digital key, open the basic functions of t...

Embodiment 2

[0091] refer to figure 2 , shows a flow chart of steps of a method for preparing and obtaining a deep neural network with a self-locking mechanism according to a specific digital key according to Embodiment 2 of the present invention.

[0092] In this embodiment, the method for preparing and obtaining a deep neural network with a self-locking mechanism according to a specific digital key includes the following steps.

[0093] Step S1020, according to the data samples with or without label information, train and obtain the first part of parameters of the deep neural network.

[0094] Step S1022, according to the undisclosed digital key and the value of the first part of the network parameter, calculate and obtain the value of the second part of the network parameter during the training process.

[0095] Step S1024, according to the network performance, iteratively trains and updates the parameters of the first part and the second part of the deep neural network until the netw...

Embodiment 3

[0097] refer to image 3 , shows a flow chart of steps of using a deep neural network method with a self-locking mechanism according to the correct digital key according to Embodiment 3 of the present invention.

[0098] In this embodiment, the method of using a deep neural network with a self-locking mechanism according to a specific digital key includes the following steps.

[0099] Step S1040, according to the provided correct digital key and the value of the first part of the network parameter, calculate and obtain the correct value of the second part of the network parameter.

[0100] In step S1042, according to the first part of the parameters of the deep neural network and the correct value of the second part of the parameters, the basic functions of the open network are used to process the input data, and a normal data processing result is obtained.

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 intelligence. Through a skillfully designed self-locking mechanism, the intellectual property of any deep neural network is protected, and an illegal infringement behavior is stopped in time. The self-locking mechanism comprises two core processes: firstly, when a neural network is trained, a specific digital key is used for locking processing; Andsecondly, when the neural network is used, the processed neural network can play a normal role only by key matching. In terms of the timeliness and the right protection cost of right protection, the self-locking mechanism provided by the invention has obvious advantages that a pirate network without key matching cannot play a normal function and cannot compete with a legal authorized network. Dueto the fact that a stealer cannot benefit from the motivation machine, infringement does not exist. On the contrary, a traditional digital watermark protection mechanism can only prove the ownership of the neural network, rights protection must pass legal ways, the timeliness is poor, and the cost is high.

Description

[0001] 1. Technical field [0002] Embodiments of the present invention relate to the field of artificial intelligence, and in particular to a method, system and terminal for intellectual property protection of deep neural networks. [0003] 2. Background technology [0004] Deep neural network: also referred to as neural network or network, refers to the artificial intelligence technology that made a major breakthrough in the early 21st century. Its basic feature is to learn the intelligent processing ability of the input signal through multi-level feature extraction from local to global. The input signal can be a one-dimensional voice signal, or a high-dimensional image, video and other multimedia digital signals. For a given digital signal sample, the deep learning algorithm masters the required data processing capabilities by adjusting and optimizing the massive parameters of the deep neural network. Different digital samples and network structures determine the different...

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): G06N3/04G06N3/08G06Q50/18
CPCG06Q50/18G06N3/04G06N3/08
Inventor 范力欣
Owner FANDWORK TECH NANJING 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