- Library Home /
- Search Collections /
- Open Collections /
- Browse Collections /
- UBC Theses and Dissertations /
- A kernel-based approach to differentially private image...
Open Collections
UBC Theses and Dissertations
UBC Theses and Dissertations
A kernel-based approach to differentially private image generation Jalali Asadabadi, Milad
Abstract
The gold standard privacy notion, differential privacy (DP), has gained widespread adoption in academic research, industry products, and government databases due to its mathematically provable privacy guarantee. However, the composability property of DP leads to privacy degradation with multiple accesses to the same data. Differentially private data generation has emerged as a solution, creating synthetic datasets resembling private data while allowing repeated access without additional privacy loss. Existing methods often assume specific use cases for synthetic data, limiting flexibility. This thesis addresses the challenge of producing flexible synthetic data by leveraging deep generative modeling and addressing privacy loss in other methods such as generative adversarial networks (GAN). we propose utilizing public data to learn perceptual features (PFs) for comparing real and synthetic data distributions, employing a non-adversarial generator training scheme based on Maximum Mean Discrepancy (MMD) to mitigate privacy loss. Experimental results reveal the efficacy of our method. it successfully generates samples for CIFAR-10, CelebA, MNIST, and FashionMNIST. Theoretical analysis of our privacy-preserving loss function clarifies the privacy-accuracy trade-offs.
Item Metadata
Title |
A kernel-based approach to differentially private image generation
|
Creator | |
Supervisor | |
Publisher |
University of British Columbia
|
Date Issued |
2023
|
Description |
The gold standard privacy notion, differential privacy (DP), has gained widespread adoption in academic research, industry products, and government databases due to its mathematically provable privacy guarantee. However, the composability property of DP leads to privacy degradation with multiple accesses to the same data. Differentially private data generation has emerged as a solution, creating synthetic datasets resembling private data while allowing repeated access without additional privacy loss. Existing methods often assume specific use cases for synthetic data, limiting flexibility.
This thesis addresses the challenge of producing flexible synthetic data by leveraging deep generative modeling and addressing privacy loss in other methods such as generative adversarial networks (GAN). we propose utilizing public data to learn perceptual features (PFs) for comparing real and synthetic data distributions, employing a non-adversarial generator training scheme based on Maximum Mean Discrepancy (MMD) to mitigate privacy loss.
Experimental results reveal the efficacy of our method. it successfully generates samples for CIFAR-10, CelebA, MNIST, and FashionMNIST. Theoretical analysis of our privacy-preserving loss function clarifies the privacy-accuracy trade-offs.
|
Genre | |
Type | |
Language |
eng
|
Date Available |
2023-10-18
|
Provider |
Vancouver : University of British Columbia Library
|
Rights |
Attribution-ShareAlike 4.0 International
|
DOI |
10.14288/1.0437191
|
URI | |
Degree | |
Program | |
Affiliation | |
Degree Grantor |
University of British Columbia
|
Graduation Date |
2023-11
|
Campus | |
Scholarly Level |
Graduate
|
Rights URI | |
Aggregated Source Repository |
DSpace
|
Item Media
Item Citations and Data
Rights
Attribution-ShareAlike 4.0 International