Research map / Adversarial robustness and security / Black-box and transfer attacks
Adaptive Perturbation for Adversarial Attack
Zheng Yuan, Jie Zhang, Zhaoyan Jiang, Liangliang Li, Shiguang Shan
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 |
|---|---|---|---|
| replace sign function with fixed scaling factor (APAA_f) | TIM · ImageNet, Inc-v3 | attack success rate: 86.5 -> 88.6 | Improved |
| replace sign function with fixed scaling factor (APAA_f) | VT · ImageNet, Inc-v3 | attack success rate: 89.1 -> 92.3 | Improved |
| replace sign function with fixed scaling factor (APAA_f) | MIFGSM · CIFAR10, Res-18 | attack success rate: 88.8 -> 92.7 | Improved |
| replace sign function with fixed scaling factor (APAA_f) | SIM · CIFAR10, Res-18 | attack success rate: 93.2 -> 96.5 | Improved |
| use adaptive scaling factor generator instead of fixed factor | SIM · CIFAR10, RegNet target | attack success rate: 91.6 -> 93.8 | Improved |
| use adaptive scaling factor generator instead of fixed factor | MIFGSM · CIFAR10, RegNet target | attack success rate: 85.6 -> 86.5 | Improved |
| replace sign with arctanh (ablation) | MIFGSM · CIFAR10, Res-18 | attack success rate: 88.8 -> 55.8 | Hurt |
| replace sign with Adam (ablation) | MIFGSM · CIFAR10, Res-18 | attack success rate: 88.8 -> 92.2 | Mixed |
| increase gamma from 0.2 to 1.0 on normal models (ablation) | TIM w/ APAA_f · ImageNet, Inc-v3 | attack success rate: 84.6 -> 89.7 | Improved |
| increase gamma from 0.2 to 1.0 on adversarially trained models (ablation) | TIM w/ APAA_f · ImageNet, Inc-v3_adv | attack success rate: 69.5 -> 57.1 | Hurt |
| add APAA_f and use 10 steps rather than baseline 100 steps | MIFGSM · CIFAR10, RegNet target | attack success rate: 94.4 -> 96.3 | Improved |
| train scaling generator with a single classifier (ablation) | APAA_a · transferability | attack success rate | Hurt |
Closest papers by meaning
- Making Adversarial Examples More Transferable and Indistinguishable (AAAI 2022 · similarity 0.89)
- On the Convergence of an Adaptive Momentum Method for Adversarial Attacks (AAAI 2024 · similarity 0.89)
- Nesterov Accelerated Gradient and Scale Invariance for Adversarial Attacks (ICLR 2020 · similarity 0.89)
- Towards Transferable Adversarial Attacks with Centralized Perturbation (AAAI 2024 · similarity 0.88)
- Adversarial Initialization with Universal Adversarial Perturbation: A New Approach to Fast Adversarial Training (AAAI 2024 · similarity 0.88)
- Boosting Adversarial Transferability through Augmentation in Hypothesis Space (CVPR 2025 · similarity 0.87)
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.
