Research map / Adversarial robustness and security / Certified and empirical robustness
Adversarial Parameter Attack on Deep Neural Networks
Lijia Yu, Yihan Wang, Xiao-Shan Gao
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 |
|---|---|---|---|
| Random L_inf parameter perturbations (no adversarial optimization) (ablation) | VGG19 · CIFAR-10 | adversarial rate: AR 0.13 at ratio 0.02 to 0.24 at 0.10 with accuracy kept | No clear effect |
| Adversarial parameter optimization (L_inf attack) instead of random perturbation (ablation) | VGG19 · CIFAR-10 | adversarial accuracy: AA drops from 45% to 8% at ratio 0.10 with AC 77% | Improved |
| Increase L_inf perturbation ratio from 0.02 to 0.10 (ablation) | VGG19 adversarial parameter attack · CIFAR-10 | adversarial rate: AR rises from 0.15 to 0.79 | Improved |
| Increase network depth (more layers at same width) (ablation) | VGG19 adversarial parameter attack · CIFAR-10 | adversarial rate at gamma=0.04: AR 0.32 for VGG19 to 0.53 for 32 extra layers | Improved |
| Increase network width (1.25x to 2.5x) (ablation) | VGG19 adversarial parameter attack · CIFAR-10 | adversarial rate: AR roughly 0.26-0.31 at gamma=0.04 vs 0.32 baseline | No clear effect |
| Adversarial parameter attack on ResNet56 vs VGG19 at same ratio | ResNet56 · CIFAR-10 | adversarial rate: ResNet56 gets AR 0.98 at 0.10 but random perturbations are far less effective on ResNet56 than VGG19 | Mixed |
| L_0 sparse weight swapping attack instead of L_inf (ablation) | VGG19 · CIFAR-10 | adversarial rate: AR 0.36-0.47 with accuracy drop to 72-76% | No clear effect |
Closest papers by meaning
- On Detecting Adversarial Perturbations (ICLR 2017 · similarity 0.88)
- Adaptive Perturbation for Adversarial Attack (TPAMI 2024 · similarity 0.87)
- Adversarial Training and Robustness for Multiple Perturbations (NeurIPS 2019 · similarity 0.86)
- Defense Against Universal Adversarial Perturbations (CVPR 2018 · similarity 0.86)
- SuperDeepFool: a new fast and accurate minimal adversarial attack (NeurIPS 2024 · similarity 0.86)
- Combating Adversaries with Anti-adversaries (AAAI 2022 · 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.
