Research map / Adversarial robustness and security / Attacks on text and graphs
How Does Data Augmentation Affect Privacy in Machine Learning?
Da Yu, Huishuai Zhang, Wei Chen, Jian Yin, Tie-Yan Liu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use mean loss across augmented instances instead of a single-instance loss | M_loss membership inference · CIFAR10, WRN16-8, k=10 | MI success rate (%): 61.9 -> 68.9 | Improved |
| Use permutation-invariant raw-moment features instead of mean loss | M_mean membership inference · CIFAR10, WRN16-8, k=10 | MI success rate (%): 68.9 -> 70.1 | Improved |
| Use augmented-instance losses for membership inference | M_loss · CIFAR10, WRN16-8, augmentation-trained model versus unaugmented model | MI success rate (%): 62.9 -> 61.9 | Improved |
| Use concatenated raw losses as neural-network features rather than permutation-invariant moments (ablation) | M_moments · CIFAR10, WRN16-8, k=10 | MI success rate (%): 70.1 -> 63.1 | Hurt |
| Use mean loss instead of concatenated raw losses as neural-network features (ablation) | M_NN_loss · CIFAR10, WRN16-8, k=10 | MI success rate (%): 63.1 -> 68.9 | Improved |
| Increase the number of augmented attack instances k (ablation) | M_loss, M_mean, and M_moments · CIFAR10 and CIFAR100 | MI success rate (%) | Hurt |
Closest papers by meaning
- Confidence Is All You Need for MI Attacks (Student Abstract) (AAAI 2024 · similarity 0.88)
- Causal Evaluation of Membership Inference Attacks (NeurIPS 2026 · similarity 0.86)
- Low-Cost High-Power Membership Inference Attacks (ICML 2024 · similarity 0.86)
- Chameleon: Increasing Label-Only Membership Leakage with Adaptive Poisoning (ICLR 2024 · similarity 0.86)
- Assessing Per-Sample Membership Inference Vulnerability without Retraining (NeurIPS 2026 · similarity 0.85)
- Impact of Dataset Properties on Membership Inference Vulnerability of Deep Transfer Learning (NeurIPS 2025 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
