GRILL: Restoring Gradient Signal in Ill-Conditioned Layers for More Effective Adversarial Attacks on Autoencoders
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_ | 1866917287118766080 |
|---|---|
| author | Ramanaik, Chethan Krishnamurthy Roy, Arjun Callies, Tobias Ntoutsi, Eirini |
| author_facet | Ramanaik, Chethan Krishnamurthy Roy, Arjun Callies, Tobias Ntoutsi, Eirini |
| contents | Adversarial robustness of deep autoencoders (AEs) has received less attention than that of discriminative models, although their compressed latent representations induce ill-conditioned mappings that can amplify small input perturbations and destabilize reconstructions. Existing white-box attacks for AEs, which optimize norm-bounded adversarial perturbations to maximize output damage, often stop at suboptimal attacks. We observe that this limitation stems from vanishing adversarial loss gradients during backpropagation through ill-conditioned layers, caused by near-zero singular values in their Jacobians. To address this issue, we introduce GRILL, a technique that locally restores gradient signals in ill-conditioned layers, enabling more effective norm-bounded attacks. Through extensive experiments across multiple AE architectures, considering both sample-specific and universal attacks under both standard and adaptive attack settings, we show that GRILL significantly increases attack effectiveness, leading to a more rigorous evaluation of AE robustness. Beyond AEs, we provide empirical evidence that modern multimodal architectures with encoder-decoder structures exhibit similar vulnerabilities under GRILL. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_03646 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | GRILL: Restoring Gradient Signal in Ill-Conditioned Layers for More Effective Adversarial Attacks on Autoencoders Ramanaik, Chethan Krishnamurthy Roy, Arjun Callies, Tobias Ntoutsi, Eirini Machine Learning Artificial Intelligence Computer Vision and Pattern Recognition Adversarial robustness of deep autoencoders (AEs) has received less attention than that of discriminative models, although their compressed latent representations induce ill-conditioned mappings that can amplify small input perturbations and destabilize reconstructions. Existing white-box attacks for AEs, which optimize norm-bounded adversarial perturbations to maximize output damage, often stop at suboptimal attacks. We observe that this limitation stems from vanishing adversarial loss gradients during backpropagation through ill-conditioned layers, caused by near-zero singular values in their Jacobians. To address this issue, we introduce GRILL, a technique that locally restores gradient signals in ill-conditioned layers, enabling more effective norm-bounded attacks. Through extensive experiments across multiple AE architectures, considering both sample-specific and universal attacks under both standard and adaptive attack settings, we show that GRILL significantly increases attack effectiveness, leading to a more rigorous evaluation of AE robustness. Beyond AEs, we provide empirical evidence that modern multimodal architectures with encoder-decoder structures exhibit similar vulnerabilities under GRILL. |
| title | GRILL: Restoring Gradient Signal in Ill-Conditioned Layers for More Effective Adversarial Attacks on Autoencoders |
| topic | Machine Learning Artificial Intelligence Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2505.03646 |