From Head to Tail: Efficient Black-box Model Inversion Attack via Long-tailed Learning
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910886509150208 |
|---|---|
| author | Li, Ziang Zhang, Hongguang Wang, Juan Chen, Meihui Hu, Hongxin Yi, Wenzhe Xu, Xiaoyang Yang, Mengda Ma, Chenjun |
| author_facet | Li, Ziang Zhang, Hongguang Wang, Juan Chen, Meihui Hu, Hongxin Yi, Wenzhe Xu, Xiaoyang Yang, Mengda Ma, Chenjun |
| contents | Model Inversion Attacks (MIAs) aim to reconstruct private training data from models, leading to privacy leakage, particularly in facial recognition systems. Although many studies have enhanced the effectiveness of white-box MIAs, less attention has been paid to improving efficiency and utility under limited attacker capabilities. Existing black-box MIAs necessitate an impractical number of queries, incurring significant overhead. Therefore, we analyze the limitations of existing MIAs and introduce Surrogate Model-based Inversion with Long-tailed Enhancement (SMILE), a high-resolution oriented and query-efficient MIA for the black-box setting. We begin by analyzing the initialization of MIAs from a data distribution perspective and propose a long-tailed surrogate training method to obtain high-quality initial points. We then enhance the attack's effectiveness by employing the gradient-free black-box optimization algorithm selected by NGOpt. Our experiments show that SMILE outperforms existing state-of-the-art black-box MIAs while requiring only about 5% of the query overhead. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_16266 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | From Head to Tail: Efficient Black-box Model Inversion Attack via Long-tailed Learning Li, Ziang Zhang, Hongguang Wang, Juan Chen, Meihui Hu, Hongxin Yi, Wenzhe Xu, Xiaoyang Yang, Mengda Ma, Chenjun Cryptography and Security Model Inversion Attacks (MIAs) aim to reconstruct private training data from models, leading to privacy leakage, particularly in facial recognition systems. Although many studies have enhanced the effectiveness of white-box MIAs, less attention has been paid to improving efficiency and utility under limited attacker capabilities. Existing black-box MIAs necessitate an impractical number of queries, incurring significant overhead. Therefore, we analyze the limitations of existing MIAs and introduce Surrogate Model-based Inversion with Long-tailed Enhancement (SMILE), a high-resolution oriented and query-efficient MIA for the black-box setting. We begin by analyzing the initialization of MIAs from a data distribution perspective and propose a long-tailed surrogate training method to obtain high-quality initial points. We then enhance the attack's effectiveness by employing the gradient-free black-box optimization algorithm selected by NGOpt. Our experiments show that SMILE outperforms existing state-of-the-art black-box MIAs while requiring only about 5% of the query overhead. |
| title | From Head to Tail: Efficient Black-box Model Inversion Attack via Long-tailed Learning |
| topic | Cryptography and Security |
| url | https://arxiv.org/abs/2503.16266 |