PFGuard: A Generative Framework with Privacy and Fairness Safeguards
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910849078132736 |
|---|---|
| author | Kim, Soyeon Roh, Yuji Heo, Geon Whang, Steven Euijong |
| author_facet | Kim, Soyeon Roh, Yuji Heo, Geon Whang, Steven Euijong |
| contents | Generative models must ensure both privacy and fairness for Trustworthy AI. While these goals have been pursued separately, recent studies propose to combine existing privacy and fairness techniques to achieve both goals. However, naively combining these techniques can be insufficient due to privacy-fairness conflicts, where a sample in a minority group may be represented in ways that support fairness, only to be suppressed for privacy. We demonstrate how these conflicts lead to adverse effects, such as privacy violations and unexpected fairness-utility tradeoffs. To mitigate these risks, we propose PFGuard, a generative framework with privacy and fairness safeguards, which simultaneously addresses privacy, fairness, and utility. By using an ensemble of multiple teacher models, PFGuard balances privacy-fairness conflicts between fair and private training stages and achieves high utility based on ensemble learning. Extensive experiments show that PFGuard successfully generates synthetic data on high-dimensional data while providing both DP guarantees and convergence in fair generative modeling. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_02246 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | PFGuard: A Generative Framework with Privacy and Fairness Safeguards Kim, Soyeon Roh, Yuji Heo, Geon Whang, Steven Euijong Machine Learning Artificial Intelligence Generative models must ensure both privacy and fairness for Trustworthy AI. While these goals have been pursued separately, recent studies propose to combine existing privacy and fairness techniques to achieve both goals. However, naively combining these techniques can be insufficient due to privacy-fairness conflicts, where a sample in a minority group may be represented in ways that support fairness, only to be suppressed for privacy. We demonstrate how these conflicts lead to adverse effects, such as privacy violations and unexpected fairness-utility tradeoffs. To mitigate these risks, we propose PFGuard, a generative framework with privacy and fairness safeguards, which simultaneously addresses privacy, fairness, and utility. By using an ensemble of multiple teacher models, PFGuard balances privacy-fairness conflicts between fair and private training stages and achieves high utility based on ensemble learning. Extensive experiments show that PFGuard successfully generates synthetic data on high-dimensional data while providing both DP guarantees and convergence in fair generative modeling. |
| title | PFGuard: A Generative Framework with Privacy and Fairness Safeguards |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2410.02246 |