IBD-PSC: Input-level Backdoor Detection via Parameter-oriented Scaling Consistency
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866909214475026432 |
|---|---|
| author | Hou, Linshan Feng, Ruili Hua, Zhongyun Luo, Wei Zhang, Leo Yu Li, Yiming |
| author_facet | Hou, Linshan Feng, Ruili Hua, Zhongyun Luo, Wei Zhang, Leo Yu Li, Yiming |
| contents | Deep neural networks (DNNs) are vulnerable to backdoor attacks, where adversaries can maliciously trigger model misclassifications by implanting a hidden backdoor during model training. This paper proposes a simple yet effective input-level backdoor detection (dubbed IBD-PSC) as a `firewall' to filter out malicious testing images. Our method is motivated by an intriguing phenomenon, i.e., parameter-oriented scaling consistency (PSC), where the prediction confidences of poisoned samples are significantly more consistent than those of benign ones when amplifying model parameters. In particular, we provide theoretical analysis to safeguard the foundations of the PSC phenomenon. We also design an adaptive method to select BN layers to scale up for effective detection. Extensive experiments are conducted on benchmark datasets, verifying the effectiveness and efficiency of our IBD-PSC method and its resistance to adaptive attacks. Codes are available at \href{https://github.com/THUYimingLi/BackdoorBox}{BackdoorBox}. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2405_09786 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | IBD-PSC: Input-level Backdoor Detection via Parameter-oriented Scaling Consistency Hou, Linshan Feng, Ruili Hua, Zhongyun Luo, Wei Zhang, Leo Yu Li, Yiming Machine Learning Cryptography and Security Deep neural networks (DNNs) are vulnerable to backdoor attacks, where adversaries can maliciously trigger model misclassifications by implanting a hidden backdoor during model training. This paper proposes a simple yet effective input-level backdoor detection (dubbed IBD-PSC) as a `firewall' to filter out malicious testing images. Our method is motivated by an intriguing phenomenon, i.e., parameter-oriented scaling consistency (PSC), where the prediction confidences of poisoned samples are significantly more consistent than those of benign ones when amplifying model parameters. In particular, we provide theoretical analysis to safeguard the foundations of the PSC phenomenon. We also design an adaptive method to select BN layers to scale up for effective detection. Extensive experiments are conducted on benchmark datasets, verifying the effectiveness and efficiency of our IBD-PSC method and its resistance to adaptive attacks. Codes are available at \href{https://github.com/THUYimingLi/BackdoorBox}{BackdoorBox}. |
| title | IBD-PSC: Input-level Backdoor Detection via Parameter-oriented Scaling Consistency |
| topic | Machine Learning Cryptography and Security |
| url | https://arxiv.org/abs/2405.09786 |