Universal Adversarial Perturbations are image-agnostic and model-independent noise that when added with any image can mislead the trained Deep Convolutional Neural Networks into the wrong prediction. Since these Universal Adversarial Perturbations can seriously jeopardize the security and integrity of practical Deep Learning applications, existing techniques use additional neural networks to detect the existence of these noises at the input image source. In this paper, we demonstrate an attack strategy that when activated by rogue means (e.g., malware, trojan) can bypass these existing countermeasures by augmenting the adversarial noise at the AI hardware accelerator stage. We demonstrate the accelerator-level universal adversarial noise attack on several deep Learning models using co-simulation of the software kernel of Conv2D function and the Verilog RTL model of the hardware under the FuseSoC environment.
翻译:通用反向扰动是图像认知和模型独立的噪音,如果加上任何图像,就会将受过训练的深演神经网络误导到错误的预测中。由于这些通用反演扰动可能严重危害实际深学习应用的安全性和完整性,现有技术使用额外的神经网络来检测输入图像源是否存在这些噪音。在本文中,我们展示了一种攻击战略,即如果以无赖手段(如恶意软件、trojan)启动,就可以绕过这些现有的反措施,在AI硬件加速器阶段增加对抗性噪音。我们展示了对几个深层学习模型的加速器级普遍对立声攻击,同时使用了Conv2D功能软件核心和FuseSoC环境中的Verilog RTL硬件模型。