Research map / Adversarial robustness and security / Black-box and transfer attacks
A Frank-Wolfe Framework for Efficient and Effective Adversarial Attacks
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 Frank-Wolfe updates instead of PGD | White-box targeted attack · MNIST | Distortion: 0.277 -> 0.256 | Helped |
| Use Frank-Wolfe updates instead of MI-FGSM | White-box targeted attack · MNIST | # Iterations: 4.0 -> 3.3 | Helped |
| Use Frank-Wolfe updates instead of PGD | White-box targeted attack · ImageNet | Distortion: 0.049 -> 0.019 | Helped |
| Use Frank-Wolfe updates instead of MI-FGSM | White-box targeted attack · ImageNet | # Iterations: 5.0 -> 4.8 | Helped |
| Use Frank-Wolfe black-box attack with sphere sensing instead of NES-PGD | Black-box targeted attack · MNIST | ASR(%): 96.8 -> 99.9 | Helped |
| Use Frank-Wolfe black-box attack with sphere sensing instead of NES-PGD | Black-box targeted attack · MNIST | Queries: 5349.0 -> 1132.6 | Helped |
| Use Frank-Wolfe black-box attack with Gaussian sensing instead of NES-PGD | Black-box targeted attack · ImageNet | ASR(%): 88.0 -> 98.4 | Helped |
| Use Frank-Wolfe black-box attack with Gaussian sensing instead of NES-PGD | Black-box targeted attack · ImageNet | Queries: 26302.8 -> 15099.4 | Helped |
| Use Gaussian rather than sphere sensing (ablation) | Frank-Wolfe black-box attack · ImageNet | ASR(%): 97.2 -> 98.4 | Helped |
| Use Gaussian rather than sphere sensing (ablation) | Frank-Wolfe black-box attack · MNIST | Queries: 1132.6 -> 1144.4 | Hurt |
| Use Gaussian rather than sphere sensing (ablation) | Frank-Wolfe black-box attack · ImageNet | Queries: 15424.0 -> 15099.4 | Helped |
Closest papers by meaning
- Simple Black-box Adversarial Attacks (ICML 2019 · similarity 0.90)
- Adversarial Eigen Attack on Black-Box Models (CVPR 2022 · similarity 0.90)
- Efficient Black-box Adversarial Attacks via Bayesian Optimization Guided by a Function Prior (ICML 2024 · similarity 0.88)
- BRUSLEATTACK: A QUERY-EFFICIENT SCORE- BASED BLACK-BOX SPARSE ADVERSARIAL ATTACK (ICLR 2024 · similarity 0.87)
- Towards Efficient Training and Evaluation of Robust Models against $l_0$ Bounded Adversarial Perturbations (ICML 2024 · similarity 0.86)
- $\sigma$-zero: Gradient-based Optimization of $\ell_0$-norm Adversarial Examples (ICLR 2025 · similarity 0.86)
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.