《计算机应用研究》|Application Research of Computers

基于差分进化的缺陷样本生成算法

Defect sample generation algorithm based on differential evolution

免费全文下载 (已被下载 次)  
获取PDF全文
作者 范纯龙,宿彤,滕一平,王翼新,丁国辉
机构 沈阳航空航天大学 计算机学院 辽宁省大规模分布式系统实验室,沈阳 110136
统计 摘要被查看 次,已被下载
文章编号 1001-3695(2021)01-045-0227-05
DOI 10.19734/j.issn.1001-3695.2019.10.0623
摘要 黑盒攻击主要是基于决策的攻击,但普遍存在查询次数多、敏感点难选择的问题,因此提出了基于差分进化的缺陷样本生成算法。算法将黑盒攻击定义为一个无约束优化问题,利用差分进化搜索图像敏感点,并优化基于深度学习模型决策定义的损失函数以及梯度计算方法,实现有效的黑盒攻击。在攻击成功率相同的条件下,在MNIST和CIFAR10数据集上的平均查询次数分别减少了28.3%和14.8%。
关键词 缺陷样本; 深度学习; 优化算法; 查询次数
基金项目 国家自然科学基金资助项目(61902260)
本文URL http://www.arocmag.com/article/01-2021-01-045.html
英文标题 Defect sample generation algorithm based on differential evolution
作者英文名 Fan Chunlong, Su Tong, Teng Yiping, Wang Yixin, Ding Guohui
机构英文名 Large-scale Distributed System Laboratory in Liaoning Province,School of Computer,Shenyang Aerospace University,Shenyang 110136,China
英文摘要 Black box attack is mainly focuses on the decision information, but there are many query times and the selection of sensitive pixels is difficult. Therefore, this paper proposed a defect sample generation algorithm based on differential evolution. By defining the black box attack as an unconstrained optimization problem, the algorithm used differential evolution to search image sensitive points, and optimized the loss function of decision definition and gradient calculation based on the deep learning model to achieve an effective black box attack. On MNIST and CIFAR10 datasets, the average query times of the proposed algorithm decreased by 28.3% and 14.8% respectively under the same attack success rate.
英文关键词 defect sample; deep learning; optimization algorithm; query times
参考文献 查看稿件参考文献
 
收稿日期 2019/10/7
修回日期 2019/11/21
页码 227-231
中图分类号 TP301.6
文献标志码 A