智能系统学报2025，Vol.20Issue(3)：594-604,11.DOI:10.11992/tis.202405037
基于L1-mask约束的对抗攻击优化方法
Adversarial attack optimization method based on L1-mask constraint
摘要
Abstract
The existing adversarial attack methods generally utilize infinite or L2 norms to measure distance.However,these methods can be improved in terms of imperceptibility.Moreover,the L,norm,as a conventionally employed met-ric method in sparse learning,has not been extensively studied in terms of improving the imperceptibility of adversarial samples.To address this research gap,an adversarial attack method based on the L,norm constraint is proposed,and it focuses limited perturbations on more crucial features by performing feature differentiation processing.Additionally,an L1-mask constraint method based on saliency analysis is proposed to improve attack targeting by masking low-saliency features.The results reveal that these improvements enhance the imperceptibility of adversarial samples and reduce the risk of overfitting alternative models with adversarial samples,thereby enhancing the transferability of adversarial at-tacks.Experiments using the ImageNet compatible dataset reveal that the imperceptibility FID index of the L1-con-strained adversarial attack methods is approximately 5.7％lower than that of the infinite norm while maintaining the same success rate for black box attacks.Conversely,the FID index of L1-mask-constrained adversarial attack methods is approximately 9.5％lower.关键词
对抗攻击/L1范数/遮盖/显著性/不可察觉性/迁移性/稀疏/约束Key words
adversarial attack/L1 norm/mask/saliency/imperceptibility/transferability/sparse/constraint分类
信息技术与安全科学引用本文复制引用
周强,陈军,陶卿..基于L1-mask约束的对抗攻击优化方法[J].智能系统学报,2025,20(3):594-604,11.基金项目
国家自然科学基金项目(62076252). （62076252）
