| Author: | Song, Xiaolin |
| Title: | Property inference attacks and defenses in deep neural networks : a comparative study |
| Advisors: | Hu, Haibo (EEE) |
| Degree: | M.Sc. |
| Year: | 2024 |
| Department: | Department of Electrical and Electronic Engineering |
| Pages: | 42 pages : color illustrations |
| Language: | English |
| Abstract: | Machine learning is widely used in various fields and promotes the development of various industries. At the same time, the security of machine learning is of great concern. More and more research is focused on the security of machine learning because machine learning models require massive amounts of data for training, and the data during the training and prediction process may contain users' private information, which makes its security a challenge, so more and more research is focused on the security of machine learning. Attribute inference attack is a method of attacking the training set of a model in machine learning species. This method is widely used in scenarios such as fully-connected neural networks and federated learning, which pose a great threat to the privacy and security of users. In order to protect the user's privacy and security, this dissertation introduces different attacks and defense methods in machine learning, and provides a detailed description of the attribute inference attack in fully connected row neural networks, and proposes a defense method against attribute inference attack and compares it to verify the effect. |
| Rights: | All rights reserved |
| Access: | restricted access |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| 8288.pdf | For All Users (off-campus access for PolyU Staff & Students only) | 1.45 MB | Adobe PDF | View/Open |
Copyright Undertaking
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
- I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.
Please use this identifier to cite or link to this item:
https://theses.lib.polyu.edu.hk/handle/200/13881