Generalization and Membership Inference Attack a Practical Perspective
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866910156678234112 |
|---|---|
| author | Rahmani, Fateme Siavoshani, Mahdi Jafari Rohban, Mohammad Hossein |
| author_facet | Rahmani, Fateme Siavoshani, Mahdi Jafari Rohban, Mohammad Hossein |
| contents | With the emergence of new evaluation metrics and attack methodologies for Membership Inference Attacks (MIA), it becomes essential to reevaluate previously accepted assumptions. In this paper, we revisit the longstanding debate regarding the correlation between MIA success rates and model generalization using an empirical approach. We focused on employing augmentation techniques and early stopping to enhance model generalization and examined their impact on MIA success rates. We found that utilizing advanced generalization techniques can significantly decrease attack performance, potentially by up to 100 times. Moreover, combining these methods not only improves model generalization but also reduces attack effectiveness by introducing randomness during training. Additionally, our study confirmed the direct impact of generalization on MIA performance through an analysis of over 1K models in a controlled environment. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_19936 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Generalization and Membership Inference Attack a Practical Perspective Rahmani, Fateme Siavoshani, Mahdi Jafari Rohban, Mohammad Hossein Machine Learning Artificial Intelligence With the emergence of new evaluation metrics and attack methodologies for Membership Inference Attacks (MIA), it becomes essential to reevaluate previously accepted assumptions. In this paper, we revisit the longstanding debate regarding the correlation between MIA success rates and model generalization using an empirical approach. We focused on employing augmentation techniques and early stopping to enhance model generalization and examined their impact on MIA success rates. We found that utilizing advanced generalization techniques can significantly decrease attack performance, potentially by up to 100 times. Moreover, combining these methods not only improves model generalization but also reduces attack effectiveness by introducing randomness during training. Additionally, our study confirmed the direct impact of generalization on MIA performance through an analysis of over 1K models in a controlled environment. |
| title | Generalization and Membership Inference Attack a Practical Perspective |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2604.19936 |