论文标题
对光学神经网络的对抗攻击
Adversarial attacks on an optical neural network
论文作者
论文摘要
对抗性攻击已经对机器学习系统进行了广泛的研究，包括数字域中的深度学习。但是，以前很少考虑对光学神经网络（ONN）的对抗性攻击。在这项工作中，我们首先使用互连的Mach-Zhhnder干涉仪（MZI）的网格构建具有ONN的精确图像分类器。然后，首次提出了相应的对抗攻击方案。受攻击的图像在视觉上与原始图像非常相似，但是ONN系统在大多数时候都会出现故障并生成错误的分类结果。结果表明，对抗攻击也是光学机器学习系统的重要问题。
Adversarial attacks have been extensively investigated for machine learning systems including deep learning in the digital domain. However, the adversarial attacks on optical neural networks (ONN) have been seldom considered previously. In this work, we first construct an accurate image classifier with an ONN using a mesh of interconnected Mach-Zehnder interferometers (MZI). Then a corresponding adversarial attack scheme is proposed for the first time. The attacked images are visually very similar to the original ones but the ONN system becomes malfunctioned and generates wrong classification results in most time. The results indicate that adversarial attack is also a significant issue for optical machine learning systems.
