Machine Learning (ML) has proven to be effective in many application domains. However, ML methods can be vulnerable to adversarial attacks, in which an attacker tries to fool the classification/prediction mechanism by crafting the input data. In the case of ML-based Network Intrusion Detection Systems (NIDSs), the attacker might use their knowledge of the intrusion detection logic to generate malicious traffic that remains undetected. One way to solve this issue is to adopt adversarial training, in which the training set is augmented with adversarial traffic samples. This paper presents an adversarial training approach called GADoT, which leverages a Generative Adversarial Network (GAN) to generate adversarial DDoS samples for training. We show that a state-of-the-art NIDS with high accuracy on popular datasets can experience more than 60% undetected malicious flows under adversarial attacks. We then demonstrate how this score drops to 1.8% or less after adversarial training using GADoT.

翻译：机器学习(ML) 在许多应用领域证明是有效的。 但是, ML 方法可能很容易受到对抗性攻击, 攻击者试图通过编造输入数据来愚弄分类/ 防范机制。 在基于 ML 的网络入侵探测系统(NIDS ) 中, 攻击者可能利用他们对入侵探测逻辑的知识来生成仍未被发现的恶意交通。 解决这个问题的一个办法是采用对抗性训练, 使训练内容增加对抗性交通样本。 本文展示了一种称为GADoT 的对抗性训练方法, 利用一个GADoT(GAN) 来生成对抗性 DDoS 样本来进行培训。 我们显示,在流行数据集中具有高度精准性的最新NIDS 可以在对抗性攻击中经历60%以上未被发现的恶意流动。 我们然后演示在使用 GADoT 进行对抗性训练后如何得分跌至1.8%或更少。