A Versatile Framework for Designing Group-Sparse Adversarial Attacks
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866911219881869312 |
|---|---|
| author | Heshmati, Alireza Roudi, Saman Soleimani Amini, Sajjad Ghaemmaghami, Shahrokh Marvasti, Farokh |
| author_facet | Heshmati, Alireza Roudi, Saman Soleimani Amini, Sajjad Ghaemmaghami, Shahrokh Marvasti, Farokh |
| contents | Existing adversarial attacks often neglect perturbation sparsity, limiting their ability to model structural changes and to explain how deep neural networks (DNNs) process meaningful input patterns. We propose ATOS (Attack Through Overlapping Sparsity), a differentiable optimization framework that generates structured, sparse adversarial perturbations in element-wise, pixel-wise, and group-wise forms. For white-box attacks on image classifiers, we introduce the Overlapping Smoothed L0 (OSL0) function, which promotes convergence to a stationary point while encouraging sparse, structured perturbations. By grouping channels and adjacent pixels, ATOS improves interpretability and helps identify robust versus non-robust features. We approximate the L-infinity gradient using the logarithm of the sum of exponential absolute values to tightly control perturbation magnitude. On CIFAR-10 and ImageNet, ATOS achieves a 100% attack success rate while producing significantly sparser and more structurally coherent perturbations than prior methods. The structured group-wise attack highlights critical regions from the network's perspective, providing counterfactual explanations by replacing class-defining regions with robust features from the target class. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_16637 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | A Versatile Framework for Designing Group-Sparse Adversarial Attacks Heshmati, Alireza Roudi, Saman Soleimani Amini, Sajjad Ghaemmaghami, Shahrokh Marvasti, Farokh Cryptography and Security Machine Learning Image and Video Processing Existing adversarial attacks often neglect perturbation sparsity, limiting their ability to model structural changes and to explain how deep neural networks (DNNs) process meaningful input patterns. We propose ATOS (Attack Through Overlapping Sparsity), a differentiable optimization framework that generates structured, sparse adversarial perturbations in element-wise, pixel-wise, and group-wise forms. For white-box attacks on image classifiers, we introduce the Overlapping Smoothed L0 (OSL0) function, which promotes convergence to a stationary point while encouraging sparse, structured perturbations. By grouping channels and adjacent pixels, ATOS improves interpretability and helps identify robust versus non-robust features. We approximate the L-infinity gradient using the logarithm of the sum of exponential absolute values to tightly control perturbation magnitude. On CIFAR-10 and ImageNet, ATOS achieves a 100% attack success rate while producing significantly sparser and more structurally coherent perturbations than prior methods. The structured group-wise attack highlights critical regions from the network's perspective, providing counterfactual explanations by replacing class-defining regions with robust features from the target class. |
| title | A Versatile Framework for Designing Group-Sparse Adversarial Attacks |
| topic | Cryptography and Security Machine Learning Image and Video Processing |
| url | https://arxiv.org/abs/2510.16637 |