Deeply differential privacy protection method based on generative adversarial network
A privacy protection and differential privacy technology, applied in the field of deep learning and privacy protection, can solve the problem of leaking sensitive user information, and achieve the effect of speeding up training, reducing the selection range, and optimizing accuracy
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Example Embodiment
[0028] The present invention will be described below in conjunction with the drawings and specific embodiments. With figure 1 Describes the processing process of the deep differential privacy protection method based on the generative confrontation network.
[0029] Such as figure 1 As shown, the specific implementation steps of the present invention:
[0030] (1) Calculate the upper bound of the privacy budget based on the size of the input data set, query sensitivity, and the probability obtained by the attacker. The calculation method of the upper bound of the privacy budget:
[0031]
[0032] Where ε is the privacy budget, n is the potential data set of the input data set (the potential data set refers to, assuming that the input data set is D, the possible value method of the neighboring data set D'of the data set is n, where D and D 'There is only one piece of data), △q is the sensitivity of the query function q for data sets D and D', △v is the maximum difference between the...
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
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