对抗攻击归因：发现对抗性ML攻击中的可归因信号
Adversarial Attack Attribution: Discovering Attributable Signals in Adversarial ML Attacks
摘要：
众所周知，机器学习（ML）模型容易受到对抗性输入的影响，研究人员已经证明，即使是自动驾驶汽车和ML即服务产品等生产系统也很容易受到影响。这些系统是不良行为者的目标。它们的中断可能造成真正的物质和经济损害。当对生产ML系统的攻击发生时，将攻击归因于负责的威胁组的能力是制定响应并追究攻击者责任的关键步骤。我们提出以下问题：可以adversarially扰动输入归因于特定的方法来产生的攻击？换句话说，有没有办法在这些攻击中找到暴露攻击算法、模型架构或攻击中使用的超参数的信号？我们引入了对抗攻击归因的概念，并创建了一个简单的监督学习实验框架，以检查发现归因信号在对抗攻击的可行性。我们发现，在CIFAR-10和MNIST数据集上区分不同攻击算法、模型和超参数生成的攻击是可能的。
英文摘要：
Machine Learning (ML) models are known to be vulnerable to adversarial inputs and researchers have demonstrated that even production systems, such as self-driving cars and ML-as-a-service offerings, are susceptible. These systems represent a target for bad actors. Their disruption can cause real physical and economic harm. When attacks on production ML systems occur, the ability to attribute the attack to the responsible threat group is a critical step in formulating a response and holding the attackers accountable. We pose the following question: can adversarially perturbed inputs be attributed to the particular methods used to generate the attack? In other words, is there a way to find a signal in these attacks that exposes the attack algorithm, model architecture, or hyperparameters used in the attack? We introduce the concept of adversarial attack attribution and create a simple supervised learning experimental framework to examine the feasibility of discovering attributable signals in adversarial attacks. We find that it is possible to differentiate attacks generated with different attack algorithms, models, and hyperparameters on both the CIFAR-10 and MNIST datasets.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
