Concept-Based Masking: A Patch-Agnostic Defense Against Adversarial Patch Attacks
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866914076192407552 |
|---|---|
| author | Mehrotra, Ayushi Peng, Derek Bhusal, Dipkamal Rastogi, Nidhi |
| author_facet | Mehrotra, Ayushi Peng, Derek Bhusal, Dipkamal Rastogi, Nidhi |
| contents | Adversarial patch attacks pose a practical threat to deep learning models by forcing targeted misclassifications through localized perturbations, often realized in the physical world. Existing defenses typically assume prior knowledge of patch size or location, limiting their applicability. In this work, we propose a patch-agnostic defense that leverages concept-based explanations to identify and suppress the most influential concept activation vectors, thereby neutralizing patch effects without explicit detection. Evaluated on Imagenette with a ResNet-50, our method achieves higher robust and clean accuracy than the state-of-the-art PatchCleanser, while maintaining strong performance across varying patch sizes and locations. Our results highlight the promise of combining interpretability with robustness and suggest concept-driven defenses as a scalable strategy for securing machine learning models against adversarial patch attacks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_04245 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Concept-Based Masking: A Patch-Agnostic Defense Against Adversarial Patch Attacks Mehrotra, Ayushi Peng, Derek Bhusal, Dipkamal Rastogi, Nidhi Computer Vision and Pattern Recognition Artificial Intelligence Adversarial patch attacks pose a practical threat to deep learning models by forcing targeted misclassifications through localized perturbations, often realized in the physical world. Existing defenses typically assume prior knowledge of patch size or location, limiting their applicability. In this work, we propose a patch-agnostic defense that leverages concept-based explanations to identify and suppress the most influential concept activation vectors, thereby neutralizing patch effects without explicit detection. Evaluated on Imagenette with a ResNet-50, our method achieves higher robust and clean accuracy than the state-of-the-art PatchCleanser, while maintaining strong performance across varying patch sizes and locations. Our results highlight the promise of combining interpretability with robustness and suggest concept-driven defenses as a scalable strategy for securing machine learning models against adversarial patch attacks. |
| title | Concept-Based Masking: A Patch-Agnostic Defense Against Adversarial Patch Attacks |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2510.04245 |