论文标题
分布式黑盒攻击：不要高估黑盒攻击
Distributed Black-box Attack: Do Not Overestimate Black-box Attacks
论文作者
论文摘要
随着云计算的普遍性，深度学习模型将部署在云服务器上，然后作为API提供给最终用户。但是，黑盒对手攻击可以欺骗图像分类模型，而无需访问模型结构和权重。最近的研究报告说，攻击成功率超过95％，查询不到1,000个。然后出现问题：黑框攻击是否已成为对云API的真正威胁？为了阐明这一点，我们的研究表明，由于几个常见的错误高估了黑盒攻击的效率，因此黑盒攻击对云API的有效程度不如研究论文中提出的有效。为了避免类似的错误，我们直接在云API而不是本地模型上进行黑框攻击。
As cloud computing becomes pervasive, deep learning models are deployed on cloud servers and then provided as APIs to end users. However, black-box adversarial attacks can fool image classification models without access to model structure and weights. Recent studies have reported attack success rates of over 95% with fewer than 1,000 queries. Then the question arises: whether black-box attacks have become a real threat against cloud APIs? To shed some light on this, our research indicates that black-box attacks are not as effective against cloud APIs as proposed in research papers due to several common mistakes that overestimate the efficiency of black-box attacks. To avoid similar mistakes, we conduct black-box attacks directly on cloud APIs rather than local models.
