Enhancing All-to-X Backdoor Attacks with Optimized Target Class Mapping
This work addresses a security threat in machine learning systems by enhancing backdoor attacks, which is incremental as it builds on existing A2X attack research.
The paper tackles the problem of All-to-X backdoor attacks with multiple target classes, which are often assumed to have low success rates, and demonstrates that these attacks are robust against state-of-the-art defenses while improving the attack success rate by up to 28% on datasets like CIFAR10, CIFAR100, and Tiny-ImageNet.
Backdoor attacks pose severe threats to machine learning systems, prompting extensive research in this area. However, most existing work focuses on single-target All-to-One (A2O) attacks, overlooking the more complex All-to-X (A2X) attacks with multiple target classes, which are often assumed to have low attack success rates. In this paper, we first demonstrate that A2X attacks are robust against state-of-the-art defenses. We then propose a novel attack strategy that enhances the success rate of A2X attacks while maintaining robustness by optimizing grouping and target class assignment mechanisms. Our method improves the attack success rate by up to 28%, with average improvements of 6.7%, 16.4%, 14.1% on CIFAR10, CIFAR100, and Tiny-ImageNet, respectively. We anticipate that this study will raise awareness of A2X attacks and stimulate further research in this under-explored area. Our code is available at https://github.com/kazefjj/A2X-backdoor .